Multi-Objective Planning Techniques in Distribution Networks - MDPI

energies Review

Multi-Objective Planning Techniques in Distribution Networks: A Composite Review Syed Ali Abbas Kazmi *, Muhammad Khuram Shahzad and Dong Ryeol Shin * Department of Electrical and Computer Engineering, College of Information and Communication Engineering (CICE), Sungkyunkwan University (SKKU), Suwon 440-746, Korea; [email protected] * Correspondence: [email protected] (S.A.A.K.); [email protected] (D.R.S.); Tel.: +82-10-5947-6376 (S.A.A.K.); +82-31-290-7125 (D.R.S.) Academic Editor: Francesco Calise Received: 13 July 2016; Accepted: 31 January 2017; Published: 12 February 2017

Abstract: Distribution networks (DNWs) are facing numerous challenges, notably growing load demands, environmental concerns, operational constraints and expansion limitations with the current infrastructure. These challenges serve as a motivation factor for various distribution network planning (DP) strategies, such as timely addressing load growth aiming at prominent objectives such as reliability, power quality, economic viability, system stability and deferring costly reinforcements. The continuous transformation of passive to active distribution networks (ADN) needs to consider choices, primarily distributed generation (DG), network topology change, installation of new protection devices and key enablers as planning options in addition to traditional grid reinforcements. Since modern DP (MDP) in deregulated market environments includes multiple stakeholders, primarily owners, regulators, operators and consumers, one solution fit for all planning scenarios may not satisfy all these stakeholders. Hence, this paper presents a review of several planning techniques (PTs) based on mult-objective optimizations (MOOs) in DNWs, aiming at better trade-off solutions among conflicting objectives and satisfying multiple stakeholders. The PTs in the paper spread across four distinct planning classifications including DG units as an alternative to costly reinforcements, capacitors and power electronic devices for ensuring power quality aspects, grid reinforcements, expansions, and upgrades as a separate category and network topology alteration and reconfiguration as a viable planning option. Several research works associated with multi-objective planning techniques (MOPT) have been reviewed with relevant models, methods and achieved objectives, abiding with system constraints. The paper also provides a composite review of current research accounts and interdependence of associated components in the respective classifications. The potential future planning areas, aiming at the multi-objective-based frameworks, are also presented in this paper. Keywords: active distribution network (ADN); distributed generation (DG); distributed energy resources (DERs); distribution network planning (DP); multi-objective optimization (MOO); multi-criteria decision analysis (MCDA); distributed generation placement (DGP); volt-ampere reactive power (VAR) compensation and power quality (VPQ); component reinforcement and up gradation (CRU); network (distribution) topology change and reconfiguration (NTR); planning techniques (PT); multiple objective planning (MOP); multi-objective planning techniques (MOPTs); future distribution networks (FDNs)

1. Introduction Electrical power grids (as hierarchical networks) are traditionally responsible for the unidirectional flow of power from centralized generation sources via transmission networks (TNWs) to distribution networks (DNWs) for ultimate electricity consumption. Increasing load demands, fewer expansion Energies 2017, 10, 208; doi:10.3390/en10020208

www.mdpi.com/journal/energies

Energies 2017, 10, 208

2 of 44

options, increasingly competitive electricity market scenarios and environmental concerns result in stressed operating conditions for both TNWs and DNWs. In comparison, DNWs gets a higher share of stressed conditions due to their design, planning and operation limitations [1]. Traditionally DNWs were purposely planned to operate in a radial configuration to maintain one-way power flow. Such a setup was preferred largely due to simple protection equipment, reduced short circuit currents (SCCs) of the networks, easy control requirements, a safe and economical operation for the end consumer. However, the rapid growth of DNWs and associated loads over large geographical areas results in technical issues like voltage instability, increased system losses, low network reliability, compromised power quality and capacity enhancement concerns, respectively [2,3]. The factors above are motivating system planners to find alternative methods rather than traditional ones, to meet the increased demand and concerned issues promptly. The possible solution strategy calls for planning modifications on short/medium/long-term basis in the potential areas of generation capacity enhancement, improving power quality, load management, emissions control and ensuring overall system reliability. 1.1. Traditional Versus Modern Distribution Planning Traditional distribution planning (TDP) methods had the narrow aim of finding economically feasible solutions based on single objective optimization techniques. The respective methods usually focus on grid reinforcements with optimal location and capacity of future substations (SSs), feeders (Fr), and conductors (branches) to address future load demands (across the planning horizon). These evaluations usually favor one decision maker (distribution companies) regarding decision support [4]. Major planning restrictions are also faced within TDP when high distributed generation (DG) penetration exists. The DNW design complications may result in further deterioration of the aforesaid technical issues rather than solving them. The conventional “fit-and-forget” approach has resulted in unfeasible costly reinforcements, to address various technical issues and retain DNW within operation limits. Modern distribution planning (MDP) methods are somewhat both better and complex than their traditional counterparts in various aspects. The most important feature aims at planning with “active network management” (ANM) for increased DG penetration abiding with system operational limits rather than the “fit-and-forget” approach. The concept of ANM introduces new planning concepts in modern planning paradigms, predominately renewable energy sources (RES) integration, distributed storage technologies (STs) and electrical vehicles (EVs); supported by communication, intelligent metering, active demand side management (DSM) and advance distributed automation (ADA) [5]. Another significant feature of MDP methods is that they can exploit various multi-objective planning techniques (MOPTs) to sort out viable trade-off (compromised) solutions among conflicting objectives that satisfy multiple (diverse) stakeholders [6,7]. 1.2. Potential Planning Techniques in Modern Distribution Planning The distribution network planning (DP) problems are becoming more complicated with the active participation of several stakeholders in the competitive energy market. The achievement of acceptable solutions ensuring economic viability, acceptable power quality, utmost reliability and improved operational aspects (better voltage stability and reduced power losses) among major market participants is one of the key motives of modern planning studies. Since the MDP problem essentially needs multi-objective optimization (MOO) methods to find a feasible solution in the setup above, therefore it can be established that MDP is a multi-objective planning (MOP) problem [7–10]. Numerous distribution planning techniques proposed in the literature since the last decade address the complex nature of DP problems from MDP perspectives. Major research accounts are nowadays more focused on DG planning options, conventional solution techniques and modified grid reinforcement strategies [6–17]. Wang et al. [6] provided a review of multi-criteria decision making (MCDM) for decision aids in MOP problems till 2009. A critical review of “state-of-the-art survey” was presented by

Energies 2017, 10, 208

3 of 44

Rodriguez et al. [7] regarding MO-based DG/DER planning methods till 2009. Georgilakis and Hatziargyriou [8,10] presented two notable planning surveys for optimal DG placement and grid reinforcements, respectively. The multi-objective planning (MOP) problem had partially addressed in both works. Kalambe in [9] offered a bibliographic survey regarding single objective techniques focusing on loss minimization for DG, capacitor placement, and network reconfiguration (NTR) respectively. DG as an attractive planning alternative for utilities regarding loss minimization, voltage stability and deferring costly grid reinforcements has advocated in [10–12]. The integration of RES-based DGs (REG) to achieved clean energy and emission reduction are among the notable motives for researchers. However planning with only high penetration of REG is not very productive due to the limitations in traditional operational (inadequate reactive power support at unity power factor, power quality) and protection capabilities of current DNWs [13]. Hence, volt-amphere reactive (VAR) power compensation and power quality issues have been addressed with planned allocation of capacitors [14] and advanced power electronic devices [15]. The protection issue has addressed with planned allocation of protection and automation devices; to ensure reliability and system stability under both normal and emergency scenarios [16]. Also, a few new MDP strategies are employed to address load management issues by employing coordinated planning of multiple components (DG units, capacitors, protection devices) with NR (and/or topology alteration) [17–19]. A review of the literature shows that most of the planning techniques address single objectives optimization on a large scale. However, little attention is paid to various planning techniques (PTs) in multi-objective framework [20], which can be interdependent, integrated and coordinated with traditional planning options from a MDP viewpoint. Hence the potential planning techniques in this paper are distinctly classified into four categories. The first category covers planning components, primarily DG units (of various types and concepts), STs and EVs. The second category comprises capacitors (types) and power electronic devices, to ensure VAR compensation and power quality. The third category contains protection, automation devices, and traditional grid reinforcements to guarantee reliability and system stability under new or expansion planning scenarios [8,18,19]. Finally, network topology optimization with alteration and/or NTR (to retain the radial nature of distribution systems) has been considered as a planning option. 1.3. Paper Contribution The paper presents a composite review of prominent planning techniques applied to distribution systems. Most notably, the review indicates the interdependence and coordination of planning components in MDP. The core focus of this paper will remain on four MOPTs, associated methods, achieved objectives and associated taxonomy (consisting of 80 papers published in the last decade as a whole, since 2005 in particular and more specifically after 2010) on a relatively large number of works. The PT categorizations have designated in this paper by DG placement (DGP) [21–52]. Followed by VAR compensation and power quality (VPQ) [53–70]. Moreover, component (protection and automation devices) placement, modern grid reinforcement and upgrades (CRU) [71–89]. Finally a change of topology and/or NTR has discussed as a planning option in [90–100]. The details in each category are discussed in later sections. Besides distinct classifications, the interdependence of individual planning components in each PT will also be presented. This work compliments the existing works by: (1) (2) (3)

Offering a composite review for researchers, planning engineers and distribution companies, regarding multiple planning techniques aiming at MDP under MO framework. Presenting planning components’ interdependence, interaction, coordination (in each PT) and addressing the contributions of each PT from the perspective of MOO in DNWs. Discussing objective attainment, methods, test systems, challenges/requirements and future research directions.

Energies 2017, 10, 208

4 of 44

It is a well-established fact that real world planning problems are multi-objective (MO) in nature. Also, more investigation is needed to carry out realistic (MO) planning by upgrading conventional to smart DNWs or redesigning them from the beginning. Hence, the core aim of this paper is to provide a background for future distribution network (FDN) planning on the basis of the limitations perceived in available works and from the perspectives of MO achievement. The paper is organized as follows: Section 2 covers classification details of planning techniques, objectives, MO formulations and major considerations for MOP models. A composite review of studied works has presented regarding decision variables, constraints, objectives, test systems, methods and considered models within PT taxonomies in Section 3. Section 4 outlines MO method classification. The contribution of the reviewed work has presented in Section 5. Future research directions have been suggested in Section 6. The paper concludes in Section 7. 2. Classifications of Planning Techniques and Key Enablers The primary objective from the perspective of planning, operation, and upgrading of any DNW is to meet the demands of end consumers in a better, safe, economical, reliable and timely manner (planning horizon). The general DP problem is considered a mixed integer nonlinear programming problem (MINLP) [8,10]. The TDP problem focuses on minimization of economic cost objectives for traditional reinforcements subject to a set of constraints. The MDP problem focuses on planning techniques (PT) from various aspects, multiple objectives, decision variables, load models while abiding with system constraints. Based on our literature survey, four PTs are presented in Sections 2.1–2.4. 2.1. Distributed Generation Placement (DGP) The DG is defined as “a small-scale electrical power generation source connected directly to the DNW or on the consumer side of the meter”. The DG type by connection and voltage level includes DGs on the DNW side with a medium voltage (MV) or low voltage (LV), whereas usually LV is on the consumer side. The DG units connected directly to DNW (close to load centers) are considered as an attractive planning alternative for utilities worldwide. However, high DG penetration has changed the nature of DNW from passive to an active one with bidirectional power flows. The prominent benefits attributed to DGP-based planning include voltage support, loss minimization, an alternative to costly reinforcements, and reduction of GHG (with REGs). Broader DG concepts by types are covered in [11–13] as follows:

• • • • •

Conventional (synchronous) DG units, for example, microturbines and diesel generators. Non-conventional, for example, fuel cells (FCs), EVs, plug-in EV (PEV). Wind (asynchronous) and photovoltaic (PV/electronic converter)-based REG units. Distributed energy resources (DERs) concepts (including DG, ST, and RL). Various types of storage technologies (ST) and concepts like DSS.

2.2. Volt-Ampere Reactive Power Compensation and Power Quality (VPQ) Capacitor (Cap) placement is one of the oldest techniques, still in use today (as shunt capacitor banks). They primarily provide VAR compensation within DNWs (located at SS or distributed in the field) to reduce reactive power losses, power factor correction and voltage stability. Moreover, improvement in power electronics, particularly FACTS devices and power filters, make room for modifications in planning approach, which is meeting load demands ensuring both VAR compensation and power quality during the planning process. The broad VPQ category, comprising of planning components, are arranged as per following types [14,15]:

• •

Capacitor types: Fixed, switched and combined fixed/switched (VRCs). Power electronic devices types: FACTS (SVC and STATCOM) and power filters (passive, active or hybrid).

Energies 2017, 10, 208

5 of 44

2.3. Component Reinforcements/Allocations and Upgradations (CRU) The main motive of this technique is to find a most economical solution for future substations, feeders, and upgraded devices to meet future requirements ensuring the quality of service promptly. It can be divided into two sub-classes as follows [10,16]. 2.3.1. Grid Reinforcement Component (GRC) This sub-class mainly deals with the traditional planning aspects for network expansions and component replacement for a static or dynamic period [4,10]. Main GRCs include: (1) branches or cables or conductors (Br); (2) feeders (Fr); (3) transformers (TF); (4) substations (SS). 2.3.2. Grid Upgrades with Devices/Components (GUC) This sub-class deals with the grid upgrading with devices or components [14,15], vital for necessary upgrades, such as: (1) automatic-reclosers (RA) for automation/protection; (2) normally open (NO)/tie-switches (TSW) and normally closed (NC)/sectionalizing switches (SSW) placement in DNW for load management; (3) devices placement like voltage regulators (VR) or automatic voltage regulators (AVRs) and online tap changers (OLTCs). 2.4. Distribution Network Topology Alteration and Reconfiguration (NTR) This technique refers to the operational planning with variation in the DNW topology [17–19]. Topology changes are realized by changing the open/close status of (SSW/NC and TSW/NO) switches to ensure primarily radial configuration, load management and system protection (with unidirectional power flow) in current DNWs. The topology can be changed to loops by simply closing TSW between two radial feeders with an advanced or upgraded protection scheme. However, the key motive is to find the best switching combination that ensures system loss reduction, cost minimization, ensuring the quality of service and reliability. Other benefits attributed to NTR include load balancing and planning maintenance outages. 2.5. Classification of Objectives Any planning process aims towards achieving maximization (↑↑) and minimization (↓↓) of certain objectives respectively. The objectives associated with the planning have broadly classified into four major types, which have presented in the Sections 2.5.1–2.5.4, respectively. 2.5.1. Technical Objectives (1)

(2)

(3)

(4) (5)

Network power losses (NPLs): The minimization (↓↓) of system losses in the literature have been addressed as energy losses in distribution lines (EL), real/resistive (P-loss), reactive/inductive (Q-loss), P-loss index (ILP), Q-loss index (ILQ) and reactive power deviation (QPD). Voltage stability (VS): The voltage objective from the maximization (↑↑) aspect in the literature has been evaluated regarding magnitude of profile (VMP), profile index (VPI), stability level (VSL), stability index (VSI), stability margin or load-ability limit (VSM) and high-level limit (MaVL). Also, voltage criteria regarding minimization (↓↓) have been addressed as sag level (VSgL), deviation/drop (VD) and total variation (TVV). Furthermore, an error at power buses (VEPB), unbalance profile (VUBP), deviation index (IVD), minimum limit (MiVL) and level at DG (VLDG) have been minimized. Power quality (PQ): Objectives have achieved with minimization (↓↓) of total harmonic distortions (THD); voltage THD (VTHD) (↓↓), P-loss under THD (PLHD) and reactive current component (QIC). Capacity enhancement: Maximization (↑↑) with DG is designated with penetration (CEDGP) and penetration level (DGPL) respectively. Load balancing (LB): Must be achieve with a minimum number of switching actions for NTR (↑↑).

Energies 2017, 10, 208

(6)

(7)

(8)

6 of 44

Reinforcement components: The performance is normally evaluated as the possibility of overload (OL) at SS, Fr, loads nodes (LN) (OLSSFrLN) (↓↓), feeder current flow (FCF), reserve capacity of conductor (RCC) and RCC index (RCCI). Other performance indicators are capacity security margin of TFs and feeders (CSM) (↑↑), network capacity release (NCR), current carrying capacity of branch (CCC) (↑↑), CCC index (CCI) (↑↑), line loading Index (LLI) (↑↑) and line flow limit index (IC). Protection: Primarily evaluated with performance based indices. Notably, short circuit level (SCL) (↓↓), short circuit index (SCI) (↓↓), three-phase short circuit (3-ph-SC), single-phase-ground (1-ph-G), fault current level (FCLL) due to DG (FCLLDG) (↓↓). Overall system stability: Besides voltage, frequency and phase angle; is evaluated with critical clearing time (CCT) for transient stability (↓↓) and average network security index (ANSI) (↑↑).

2.5.2. Economic Objectives (1)

(2)

(3) (4) (5)

(6) (7)

(8)

Project planning costs: They have been addressed in the literature; aiming at minimization of (↓↓) payback year (PBY), net present value (NPV) of components/systems, project installation (ItC), network upgrading (CNU) and annual (AC) costs. The objectives are also targeted at maximization of (↑↑) time deferral in new installations and economic index (EI), respectively. System running (operation) costs: These objectives comprise cost minimizations (↓↓) in investment (InvC), operation (OC), maintenance (MC), O&M (OMC), investment and operation (IOC) and capacity adequacy cost (CAC). Stakeholder economics: They concern objective maximization (↑↑) of profits, net savings (NS), the benefit to cost ratio and DG owner income (DGOI). Technical costs: These involve cost minimization (↓↓) of power (CPL) or energy (CEL) losses and overall power losses during operations (OCPL) (↓↓). Cost of reliability: This has been evaluated regarding overall cost minimization (↓↓) with the reliability indices, such as energy not supplied (CENS), system average interruption frequency index (CSAIDI), interruption (IntC) non-distributed energy (NDE) and customer service interruptions (CSI). Also, cost objective minimization (↓↓) concerning customer interruption (CIC) and DG unavailability (DGUI) have also considered in the literature. Market economics: Market-based objectives concern the minimization (↓↓) of economic risks in electricity market price (EMP) and cost of purchased energy (CPE) (↓↓). Planning component costs: The monetary functions, focusing on a wide range of planning components, have been evaluated as cost (or investment) minimization (↓↓) of: (1) equipment (EC); (2) switching (SWC); (3) NTR operations (NRC); (4) capacitor (CC); (5) DG installation (CDG); (6) DG investment (DGI); (7) voltage regulator (VR); (8) VRC (CVR/C); (9) DG and PPF (ICDGPPF); (10) switch purchase and maintenance (SPMC); (11) reinforcements (IRC); (12) energy loss reduction (ELC); (13) DG and capacitors (ICAll ); (14) monetary risk (MnC) and (15) system upgrades (SUCs). Other planning-related costs: The addressed objectives in this subcategory deal with cost minimization (↓↓) of investment, EL and O&M cost (IELOM); system planning (SCP), overall fixed and variable cost (OFVC) and overall new and old devices (COD). Also, includes operation and investment (EOI); total operation cost (TOC); expected global cost (EGC) and overall complete system (OCS) (including installations, O&M, power losses, reliability excluding profits).

2.5.3. Techno-Economic Objectives (1) (2) (3)

System efficiency (↑↑). Optimizing spinning reserve (SR) (↓↓). Social benefits have addressed in the literature improving electricity service quality (↑↑), reducing the unit rate for consumers (↓↓) and consumer interruption level (CILL) (↓↓).

Energies 2017, 10, 208

(4) (5)

(6) (7)

7 of 44

Electrical service security and control (↑↑). System reliability maximization (↑↑) can be achieved with objective minimization (↓↓) of: (1) system average interruption frequency index (SAIFI); (2) system average interruption unavailability index (SAIUI); (3) system average interruption duration index (SAIDI); (4) energy not supplied (ENS); (5) ENS for average case (AENS); (6) expected ENS (EENS); and (7) overall contingency load loss index (CLLI). Techno-economic hazards minimization (↓↓) has considered as objectives addressing system failure, malfunctions and conditional value at risk (CVaR). Reliability of DG (DGR) (↑↑).

2.5.4. Environmental Objectives (1)

(2) (3) (4) (5)

(6)

Greenhouse gases (GHGs): The minimization (↓↓) of GHG emissions (GHGE) are among the key objectives of modern planning. In literature, emissions have addressed with minimization (↓↓) of average (annual) GHG (ACE), grid (GHGG) and DG (GHGDG) respectively. Penetration of REGs (↑↑). Energy diversity with REG (↑↑). Cost and quantity of fossil fuel saving with less costly alternatives (↓↓). REG-based objectives: They principally deal with minimization (↓↓) of external cost of energy (ECE); power buying (PB) from SS (PBSS) and DG owner (PBDG); energy imported from the grid (EIG) and distributed storage system (DSS) energy losses (DEL). Also, energy export from DG (REG) to grid (EEDG2G) needs to be maximized (↑↑). Health care costs due to GHGs and particle emissions needs to be minimized (↓↓).

2.6. Classification of MOP Formulations The classic TDP techniques were single objective, single stakeholder, and single dimensional approaches. In contrast, MDP considers diverse stakeholder participation in the presence of various planning techniques, active network management and ownership (distribution system operator, DG operator, consumer, etc.) has led to planning objectives that are conflicting in nature. For example; switching radial DNW topology to loop will increase DG penetration, improve voltage and increase reliability, however, system losses increases. Therefore MOO can be employed to bring a compromise solution among conflicting objectives, abiding system constraints to satisfy all stakeholders. Finding a single solution for MOP involves two steps, comprising of optimization and decision making (DM). Depending on the order, in which these steps have performed, MO formulations can be classified as two core approaches (classes), shown as follows [6,7,20]. 2.6.1. MO + W or Priori Class In this type of formulation, DM precedes optimization. The multiple objectives transform into the single objective function, and the individual weights are assigned to each objective by user-defined (decision maker) preferences before execution of optimization algorithms. This class is also known for a priori articulation of preferences. The optimization of the single objective function is more qualitative in nature (with preferred weights) and results in a single optimized solution. Also, a great deal of background knowledge is required to evaluate the required weights. Key methods like goal programming and MO performance index (IMO) can be included in this classification. The DM approaches utilized in these formulations are generally from the family of multi-criteria decision analysis (MCDA). Major MCDA-based DM techniques employed in MO formulations include weighted sum method (WSM), weighted product method (WPM), analytical hierarchal process (AHP), max-min and fuzzy DM (FDM) approaches. For simplicity, this type of problem is shown with term “MO + W” in the rest of the paper.

Energies 2017, 10, 208

8 of 44

2.6.2. MO-P or Posteriori Class In this class, optimization is preferred over DM to achieve realistic solutions. An optimization algorithm determines a set of potential solutions (Pareto frontier), also called Pareto optimal set solutions, and are usually non-dominated (or non-inferior) in nature. The decision maker then chooses a solution from the resulting Pareto set on the basis of respective preferences, also called posteriori articulation of preferences. It is important to note that MO solutions obtained with no articulation of preferences are related arbitrary to the class of Pareto optimal set. Such approach is more quantitative in nature, and evolutionary (meta-heuristic) algorithms have normally utilized as optimization processes. This type of problem formulation has shown with the term “MO-P”. 2.7. Classification of Models in MOP The associated classifications of key enablers aiming at MOP problem, have presented in Sections 2.7.1–2.7.6, respectively. 2.7.1. Key Decision Variables (DV) for PT

• • •

• •

DGP: (1) Location (L); (2) capacity (C); (3) L + C; (4) L + C + Type (T) (DG/DER or storage (ST) or others like EV); (5) L + C + Number (N) (single (S) or multiple (M)); (6) L + C + T + N. VPQ: (1) L; (2) C; (3) L + C; (4) L + C + T (capacitor, FACTS devices and power filters); (5) L + C + N; (6) L + C + T + N. CRU/GRC: (1) Feeder (Fr.) location (L); (2) Fr. location (L) and capacity(C) (L + C); (3) substation (SS) location (L) + SS capacity (C) + Fr. (L); (4) SS (C) + Fr. (L + C); (5) SS (C) + Fr. (L + C); (6) SS (L + C) + Fr. (L + C); (7) SS (L) + Fr. (C) + Load (Ld) allocation; (8) SS (L + C) + Fr. (C) + Ld allocation; (9) SS (C) + Fr. (C) + capacity of reinforcement planning components (SS, TF, Fr. and Br.) and up-gradation devices (DG, capacitor, device, switches) or simply component capacity (Cc); (10) SS (C) + Fr. (C) + Cc + Ld allocation; (11) SS (L + C) + Fr. (L + C) + Cc + Ld allocation; CRU/GUC: (1) L; (2) C; (3) L + C; (4) L + C + T (VR, SWs, RAs); (5) L + C + N; (6) L + C + T + N. NTR: (1) L; (2) N; (3) N + L; (4) Switching status (NO/NC); (5) Radial (retained) topology (RT); (6) N + L + (NO/NC); (7) N + L + (NO/NC) + RT; (8) Loop Topology (LT); (9) N + L + (NO/NC) + LT.

2.7.2. Multi-objective Planning Types by Planning Period/Horizon The general classification of planning can be divided into: (1) short; (2) medium and (3) long term basis. Planning process on the basis of the application comprises of: (1) new; and (2) expansion type considering load growth over the planning period/horizon (PP). A broad classification of DP problems on PP basis includes: (1) static type (one-step/single stage); (2) dynamic type (multistage). 2.7.3. Multi-objective Planning Types by Planning Components Coordination The optimization problem type can-be related to the planning components or combination of them (interdependence) as follows: (1) optimal DG placement (DGP); (2) optimal capacitor, FACTS devices and PF placement (VPQ); (3) optimal grid component reinforcement, allocation and upgradations (CRU); (4) network topology and/or reconfiguration (NTR); (5) A + B; (6) A + C; (7) A + D; (8) B + C; (9) B + D; (10) C + D; (11) A + B + C; (12) A + B + D; (13) A + C + D; (14) B + C + D and (15) A + B + C + D. 2.7.4. Major Constraints The most important constraints considered on broader scale includes: (a) Bus voltage drop limit; (b) Power flow equality constraint; (c) Branch/Feeder capacity; (d) Transformer overloading limit; (e) Harmonics limit; (f) Radial operation constraint; (g) Short circuit current limit; (h) Power-factor limit for utilities; (i) Reliability limits; (j) Protection limits; (k) Types of load; (l) Load flow constraint; (m) DG capacity limit; (n) Unique parameter selection limit; (o) DISCO limitation; (p) Cost constraints; (q) Budget constraint; (r) Standard size of components; (s) Planning periods; (t) Weight factor;

Energies 2017, 10, 208

9 of 44

(u) Number of components (DG, Cap, devices, switches); (v) Thermal limits; (w) Stability limit; (x) Vector constraint; (y) Real (P) and reactive (Q) power compensation limit; and (z) Future constraints (incorporating new, ANM and other related constraints). 2.7.5. Classification of Load Variables (Models and Profiles)

•

Load Models (LdM): The load (Ld) variables are usually modeled as: (1)

Balanced three-phase loads represented by a single phase load: (a) (b)

Distributed type loads (across the branches). Concentrated type loads (on the nodes/buses). i. ii. iii. iv.

(2) (3) (4) (5) (6)

•

Constant loads (CLd). Variable load (VLd). (Time-varying or voltage dependent). Fuzzy load (FLd). Probabilistic load (PLd).

Balanced three-phase loads (BL). Unbalanced three phase loads (UL). Combined three and single phase loads (UBL). Controllable (responsive or flexible) loads (CL). Non-controllable/non-linear loads (NL).

Load Profiles (LdP): The general load profiles have modeled as: (1) (2) (3) (4) (5) (6)

Single load level (SLL). Multiple load levels (MLL). Fuzzy load level (FLL). Probabilistic load level (PLL). Time-varying load level (TVLL). Critical load level (CLL).

2.7.6. Test Distribution System Types The test distribution system by voltage level includes: (1) primary at MV level (6.6 KV–34.5 KV); (2) secondary at LV level (110 V–600 V); (3) combination of both (primary and secondary) DNWs. The test system on application basis may include: (1) real; (2) test DNW, as considered by researchers. 3. Composite Review of MOP Techniques with Taxonomy In this section, a composite review of related research aiming at PT (individual, integrated and interdependent) is presented from the perspective of the MOP problem. Furthermore, from the readers’ comfort viewpoint, the information about planning components (in each PT), decision variables, major constraints, considered objectives, the test DNWs, MO classification, algorithms or methods, planning periods, online year, load models, profiles and concerned information have arranged in tabular format throughout the Tables 1–4, respectively. Planning components associated to each PT in reviewed work are designated with symbols A, B, C, and D, respectively, also shown in an overarching diagram as in Figure 1. The color coding allocated to each PT in the arrangement order includes green (A) for DGP [21–52]; orange (B) for VPQ [53–70]; blue (C) for CRU [71–89] and purple (D) for NTR [90–100]. It is important to note that date of publication in each table represents the online date of the article. In each table, refer to Section 2.7.1 for decision variables (DVs). Major constraints have provided in Section 2.7.4. The objective classifications with abbreviations are provided in Section 2.5. The MO classifications are provided in Section 2.6; planning algorithms are discussed in Section 4 and finally load models in Section 2.7.5.

Energies 2017, 10, 208

10 of 44

Table 1. MO Planning with DGP. Ref.

PT

(DV) (L,C, N, T)

Major Constraint

Objective Function (OF)/Considered Objectives

Test DNW

MO Class/Algorithm/Planning Period (PP)/ Others

Year (Online)/LdM + LdP/Others/

[21]

A.

M/(DG) (L + C)

a), b), c), f),

Cost [Upgradation; purchased energy; losses; energy not supplied] (↓↓)

Real Italian DNW

(MO-P)/GA + ε constrained/Dynamic (20 years)/

May 2005/CLd + PLL/

[22]

A.

M/(DG) (L + C)

a), b), c), e), g), f),

[Cost of energy losses; Voltage deviation; VTHD] (↓↓)

18 Bus Test DNW

(MO-P)/GA + ε constrained/Dynamic (10 years)

Jul. 2005/CLd + PLL/

[23]

A.

M/(DG) (L + C + T)

a), b), c), f), l), q),

[Power losses (↓↓); Voltage Profile (↑↑)]

IEEE 34 Bus DNW

(MO-P)/MCS + GA/Static (1 year)

Mar. 2007/VLd + TVLL/ Mar. 2008/FLd + FLL/

[24]

A.

M/(DG) (L + C)

a), b), c), f), h), m)

[IELOM, OLSSFrLN, EMP] (↓↓)

9 Bus DNW

(MO-P)/NSGA-II + max-min/Dynamic (30 years)

[25]

A.

S/(DG) (L)

a), b), c), f), v)

Single OF (SOF) with weights: [P-loss; Q-loss; VD; CRC; 3-ph-SC; 1-ph-G]

IEEE 34 Bus DNW

(MO+W)/ES/Static (1 yr.)/Wt. (1–6): [0.33; 0.10; 0.15; 0.07; 0.07; 0.15]

Apr. 2008/CLd + TVLL/

[26]

A.

M/(DG) (L + C)

a), b), c), f), m), u), s), k)

[P-loss (↓↓); Voltage Deviation (↓↓)]

IEEE 34 Bus DNW

(MO + W)/Fuzzy goal programming (FGP) + GA/Dynamic (5 years)

Apr. 2008/VLd + MLL

[27]

A.

M/(DG) (L)

a), b), c), f), v)

[EEDG2G (↑↑); P-loss (↓↓); SCL (↓↓)]

IEEE 34 Bus DNW

(MO-P)/NSGA/Static (1 year)

Sep. 2008/CLd + TVLL Feb. 2009/VLd + MLL/

[28]

A.

S/(DG) (L + C)

a), b), c), f),

Multi-objective performance index (IMO): [P-loss; Q-loss; CCI; VP]

16 Bus.; 37 Bus Test DNW.

(MO + W)/ES [GA vs. ES]/Static/Wt.(1–4) [0.40; 0.20; 0.25; 0.15]

[29]

A. D.

M/(DG) (L + C + T)

a), b), c), f), g),

Single OF (SOF) with weights: FMOI = [VDI; PLI; QLI; LLI; SCI]

IEEE 33; IEEE 69; 25 Bus Test DNW;

(MO + W)/Fuzzy GA + WSM/Static/Wt.(1–5) [0.25; 0.40; 0.15; 0.10; 0.10]/

Jul. 2010/CLd + SLL/

[30]

A.

M/(DG) (L + C + T)

a), b), c), f), l), m), p), q), r),

SOF for total cost of DG: [IC; MC; OC; CAC, NLC]

IEEE 37 Bus DNW

(MO + W)/GA + MCS, AHP/Chance constrained programming (CCP) model/Dynamic (3 years)/Wt.(1–5) [0.34; 0.34; 0.11; 0.11; 0.10]/PEV

Oct. 2011/PLd + PLL/

[31]

A.

M/(DG) (L + C)

a), b), c), f), l), m), n), v)

[P-loss (↓↓); VP(↑↑); VSI(↑↑)]

IEEE 33; IEEE 69 Bus DNW;

(MO + W)/Hybrid GA + PSO [GA: DG (L); PSO: DG (C)]/Static

Jan. 2012/CLd + SLL

[32]

A.

M/(DG) (L + C + T)

a), b), c), f), g),

GMOI(↓↓): [VDI (↓↓); PLI (↓↓); QLI (↓↓); LLI (↓↓); SCI (↓↓)]

IEEE 33Bus DNW; 25Bus Test DNW;

(MO + W)/GA + Goal programing (Goal P)/Static/

Feb. 2012/CLd + SLL/

[33]

A.

M/(DG) (L + C + T)

a), b), c), f), m),

Minimize OF: [EL ; VD] (↓↓)

24 Bus rural DNW

(MO + W)/Two-stage (heuristic iterative method + ES)/Dynamic (3 years)

Sep. 2012/VLd + TVLL

[34]

A

M/(DG) (L + C + T)

a), b), c), k), l), m),v), y),

[OCS (↓↓); DGR (↑↑)] MS * = Maintenance schedule

IEEE 37 bus DNW

(MO-P)/GA + MCOM (Multi-criteria optimization model)/Dynamic (20 years)

Jan. 2013/PLd + PLL/

[35]

A.

M/(DG) (L + C)

a), b), c), f), h), i),

[P-loss (↓↓); VP (↑↑)]


(MO-P)/IMOHS framework/Static

Mar. 2013/VLd + TVLL/

[36]

A.

M/(DG) (L + C + T)

a), b), c), m), u)

[P-loss (br. & TF); VD; CCT] (↓↓)

IEEE 33Bus DNW

(MO-P)/Hybrid PSO-SFL, FDM/Static/DIgSILENT® /Wt. 0.4–0.9

Jun. 2013/[CLd , VLd (voltage)] + MLL/

[37]

A

M/(DG) (L + C)

a), b), c), f), h), v),w),y)

[P-loss (↓↓); VSI (↑↑); TVV (↓↓)]


(MO-P)/PFDE/Static

Nov. 2013/CLd + SLL/

Energies 2017, 10, 208

11 of 44

Table 1. Cont. Ref.

PT

(DV) (L,C, N, T)

Major Constraint

Objective Function (OF)/Considered Objectives

Test DNW



[38]

A

M/(DG) (L + C + T *)

a),b),c),f),h),l), m),p),s),u),z *)

[SCP; GHGE] (↓↓); z * = Plug-in EV (PEV) and REGs

Test 38 bus DNW

(MO-P)/MCS + NSGA II (MINLP)/Dynamic (20 years (Yr.))


[39]

A

M/(DG) (L + C + T)

a), b), c), d), f), m), v),

[MnC; GHGE] (↓↓);

9 bus Test DNW

(MO-P)/(Augmented ε-constrained + FCM/MCS) + MINLP framework/Dynamic (10 years)


[40]

A

M/(DG + DR + SR) (L + T)

a), b), c), l), m), r), z1), z2),

[TOC; GHGE] (↓↓); z1 = DRP * constraint z2 = spinning reserve (SR) constraint.

69 Bus DNW

(MO-P)/(AεCM)/Dynamic (24 h (Hr.)) for Short-term planning (STP)/ DRP *: Demand response provider

Mar. 2014/PLd + PLL/

[41]

A.

M/(DG *) (L + C + T *)

a), b), c), f), p), q), u)

[ENS (with CVaR); EGC] (↓↓), * DG types: PV, wind, ST, EV

IEEE 13 Bus DNW

(MO-P)/Fast NSGA II, MCS-OPF framework/Static/

Jun. 2014/PLd + TVLL/

[42]

A.

M/(DG) (L + C)

a), b), c), f), m), v)

MOF = min(OF1 + PC1 OF2 + PC2 OF3) [P-loss (↓↓); VD(↓↓); VSI (↑↑)]

IEEE 33; IEEE 69; 118 Bus DNW;

(MO-P)/QOTLBO/Static/PC1 = 0.35; PC2 = 0.65

Jul. 2014/CLd + SLL/

OF1 = DG owner cost: [DGOI(↑↑); DGI(↓↓); MC(↓↓); OC(↓↓)] OF2 = DISCO owner cost: [PBSS (↓↓); PBDG (↓↓); CIC (ENS) (↓↓)]

IEEE 33Bus DNW;

(MO-P)/MOPSO + FDM /Dynamic (20 years)/Goal1: Cost of DISCO (↓↓); Goal2: DG owner benefit (↑↑);

Aug. 2014/VLd + TVLL

[43]

A.

M/(DG) (L + C)

a), b), c), d), o), p), q)

[44]

A

M/(DG) (L + C + T)

a), b), c), k), l), m)

[MnC; GHGE] (↓↓);

IEEE 33Bus DNW;

(MO-P)/IMOPSO-PS/Static (1 year)

Aug. 2014/CL + CLL

IEEE 34 bus DNW

(MO + W)/MISOCP + AHP/Dynamic (5 years)/Wt. (1–5): [0.0562; 0.0396; 0.2535; 0.6421; 0.0086]/

Sep. 2014/VLd + TVLL/

[45]

A

M/(DG) (L + C + T)

a), b), c), m), p)

Single OF with weights: [VD; FCF; NPL; ECE; DEL] DSS *: Distributed storage system.

[46]

A.

S/M (DG) (L + C)

a), b), c), d), f), l), m), v)

IMO: [P-loss; Q-loss; VP; RCCI;]

IEEE 69; IEEE 123 Bus DNW

(MO + W)/S-BB-BC algorithm/Static/Wt. (1–4): [0.40; 0.20; 0.15; 0.25]

Nov. 2014/CLd + SLL for 69 bus; UB + TVLL for 123;

[47]

A

M/(DG) (L + C)

a), b), c), f), v)

Single OF with weights: [ILP; ILQ; VSI; IC; IVD]

Test 38 bus; IEEE 69 bus DNW

(MO + W)/CABC/Static/Wt. (1–5): [0.35; 0.15; 0.10; 0.25; 0.15]

Nov. 2014 /VLd + MLL/

[48]

A.

M/(DG) (L + C)

a), b), c), m), r), z *)

Single objective function: [PBY (↓↓); P-loss (↓↓); VSL (↑↑)] z * = RES penetration constraint

28 Bus rural DNW

(MO + W)/MOPSO + FDM/Static (1 year)/Focus on DSM.

Jan. 2015/VLd + TVLL/

[49]

A.

S/M (DG) (L + C)

a), b), c), f), m)

[P-loss; CDG] (↓↓)

15 Bus DNW

(MO-P)/SQP + WSM, FDM/Dynamic (20 years)/

Apr. 2015/CLd + SLL

[50]

A

M/(DG) (L + C)

a), b), c), f), m), v)

[P-loss (↓↓); VD (↓↓); VSM (↑↑)]

IEEE 33; Real 292; Real 588 bus DNW;

(MO-P)/Improved-NSGA-II (INSGA-II) + FDM/Static (1 year)

Apr. 2015/VLd + MLL

[51]

A.

S/M (DG) (L + C)

a), b), c), m), u)

[NPL; VD; CDG] (↓↓)

IEEE 33Bus Radial DNW

(MO-P)/(MOShBAT)/Static/

Jan. 2016/VLd (Voltage) + MLL

[52]

A. B.

M/(DG + Cap) (L + C + T)

a), b), c), d), f), m), u), y)

[EI (↑↑); VSI (↑↑); ANSI (↑↑); ACE (↓↓)]

28 bus rural DNW

(MO-P)/MOPSO + FDM/Dynamic (5 years)

Apr. 2016/VLd + MLL/

Notes: PT: in Sections 2.1–2.4; DV: Decision Variables (Location, capacity, numbers and types); Constraints in Section 2.7.4; Objectives in Section 2.5; MO Class in Section 2.6; LdM, LdP in Section 2.7.5. Also, “*” indicates any additional information about a particular decision variable, constraint and objective, respectively.

Energies 2017, 10, 208

12 of 44

Table 2. MO Planning with VPQ. Ref.

PT

(DV) (L, C, N, T)

Major Constraint

Objective Function/Considered Objectives

Test DNW

MO Class/Algorithm/Planning Period (PP)/Others


[53]

B.

M/(Cap) (L + C)

a), b), c), f),

[P-loss; CC] (↓↓)

Portuguese 94 bus DNW

(MO-P)/Tabu Search (TS)/Static

May 2005/VLd + MLL/ Aug. 2007/CLd + MLL

[54]

B.

M/(Cap) (L + C + T)

a), b, f),

[NS (↑↑) (with EL (↓↓)); VD (↓↓)]

IEEE 69 Bus DNW

MO + W/Fuzzy-GA (MO) approach/Static/wt(1–2): [0.5;0.5]

[55]

B.

M/(Cap) (L + C + T)

a), b), c), d), f), p),

[CCI (↓↓); P-loss (↓↓); VD (↓↓); CSM (↑↑)]

IEEE 69 Bus DNW/Two stage approach

(MO-P)/1). IA; 2). Compromise programming (TSIAECP)/Static

Mar. 2008/CLd + MLL

[56]

B.

M/(Cap) (L + C + T)

a), b), c), f), m),

[EL; VD] (↓↓)

12 Bus DNW Radial

(MO-P)/Hybrid SA-PSO, (IBVT) approach/Static (1 year)

Jan. 2009/CLd + MLL

[57]

B. C.

[M/(Cap) + S/(AVR)] (L + T)

a), b), c), f), p), u), x),

[EL; VD; OCS] (↓↓)

IEEE 69 Bus DNW Radial

(MO-P)/(SPEA2) improved with fuzzy logic/Static (1 year)

Aug. 2010/CLd + MLL

[58]

B.

M/(Cap) (L+C)

a), b), c), f), y),

[P-loss; CC] (↓↓)

Portuguese 94 bus DNW

(MO-P)/Elitist NSGA-II (ENSGA-II)/Static/

Jun. 2012/CLd + MLL/

[59]

B. C.

[M/(Cap) + M/(VR)] (L+ C + T)

a), b), c), f), m),

[(CVR/C + OCPL); VD] (↓↓)

IEEE 69; Test 136 bus DNW

(MO-P)/MILP model / Static (1 year)

Jan. 2013/CLd + MLL

(MO-P)/SAMHBMO/Static (1 year)

May 2013/PLd + PLL/

[60]

B.

M/(Cap) (L + C + T)

a), b), f), m),

[AC (↓↓); P-loss (↓↓); VD (↓↓)]

IEEE 9; IEEE 34 Bus DNW

[61]

B.

M/(Cap) (L + C + T)

a), b), c), f), h), l), m), y),

[P-loss (↓↓); VSI (↑↑); NS (↑↑);]

IEEE 34; Test 118 bus DNW

(MO + W)/APSO/Static (1 year)

May 2013/CLd + SLL: 34 bus CLd + MLL: 118 bus

[62]

B.

M/(Cap) (L + C + T)

a), b), c), f), h), l), m), y),

[P-loss (↓↓); VSI (↑↑); NS (↑↑);]

IEEE 34; Portuguese 94 bus DNW

(MO + W)/ABC/Static (1 year)/

Jul. 2013/CLd + SLL

[63]

B.

M/(Cap) (L + C + T)

a), b), c), f), m),

[AC (↓↓); P-loss (↓↓); VD (↓↓)]

IEEE 9; 34; 69 bus DNW;

(MO-P)/AMHBMO/Static (1 year)/

Aug. 2013/CLd + SLL

[64]

A. B.

M/[DG +Cap] (L + C + T)

a), b), c), f), k), l), p), q),

[CENS; CSAIDI; ELC; ICAll ] (↓↓)

115 bus practical Iranian DNW

(MO-P)/IDEA/Static (1 year)/

Dec.–Jan. 2013–14/VLd [Time and Voltage] + TVLL

[65]

A. B.

M/[DG + Cap] (L + C)

a), b), c), f), h), m),

[P-loss (↓↓); VSI (↑↑); SCI (↓↓)]

IEEE 33; Portuguese 94 bus DNW

(MO-P)/Fuzzy MOPSO + FDM/Dynamic (10 years)/

Dec. 2014/FLd + FLL/Uncertainty

[66]

B.

M/(Cap) (L + C + T)

a), b), c), f), h),

[QIC (↓↓); P-loss (↓↓); CCC(↑↑); MiVL (↓↓); MaVL (↑↑)]

51 Bus Test; IEEE 69 Bus DNW

(MO + W)/Fuzzy-GA (MO) approach/Dynamic (20 years)/

Dec. 2015/VLd + MLL

[67]

B.

M/(Cap) (L + C + T)

a), b), c), f), h), l), m), y),

Strategy 1: Minimize cost: [CCP; CCI; OMC; EL] (↓↓) Strategy 2: Technical: [P-loss (↓↓); VMP (↑↑); VUBP (↓↓)]

60 bus real unbalanced Australian MV DNW

(MO + W)/MIP problem with BSFS load flow based PSO method + WSM/Dynamic (30 years)/

Jan. 2016/UBL + TVLL

[68]

A. B.

M/[DG+PPF] (L + C)

a), b), c), f), l), m), w),

[THD; PLHD; ICDGPPF] (↓↓)

IEEE 34 Bus DNW

(MO + W)/ABFO/Static (1 year)/PPF: Passive power filter

Jan. 2016/NL + MLL

[69]

A. B.

M/[DG + RA + D-FACTS](L + C)

a), b), c), f), r),

[OFVC; CLLI] (↓↓)

54 bus test DNW

(MO-P)/MOSOA + max-min approach/Static (1 year)

Apr. 2016/CLd + SLL/

[70]

B. C. D.

M/Fr.(L + C)+ [RA + D-FACTS] (L)

a), b), c), f), p), s),

[OFVC; CLLI] (↓↓)

54 Bus Test DNW

MO-P/MOSOA + max-min approach/Static (1 year)/ADA: Advance distributed automation

May 2016/CLd + SLL/(ADA)

Notes: PT: in Sections 2.1–2.4; DV: Decision Variables (Location, capacity, numbers and types); Constraints in 2.7.4; Objectives in Section 2.5; MO Class in Section 2.6; LdM/LdP in 2.7.5.

Energies 2017, 10, 208

13 of 44

Table 3. MO Planning with CRU. Ref.

PT

(DV) (L, C, N, T)

Major Constraint


Test DNW



[71]

C. D.

M/(SS, Fr.) (L + C + T)

a), b), c), m), p), r),

[OFVC; ENS] (↓↓)

41 nodes,2 SSs, 73 routes; Radial and Mesh (10 × RF *);

MO-P/NSGA + SPEA, FCM)/Static/ RF * = Reserve feeders for loop


MO + W/ MINLP + GA/Dynamic (5 years)/Main problem into three subproblems

Dec. 2008/CLd + SLL

[72]

B. C.

M/[(SS, Fr, SSW)] + [M/(Cap)] (L + C)

a), b), c), f), i), m), n),

[IOC; IOBDRA; CIC] (↓↓)

Urban electric DNW expansion planning Radial;

[73]

C.

M/(SSW) (L)

a), b), c), f), m), p),

[EENS; COD(SSW)] (↓↓)

Rural Billinton; Iranian DNW;

MO + W/ACO + FDM/Static (1 year)

Jan. 2009/CLd + SLL Mar. 2009/CLd + TVLL

[74]

A. C.

M/(DG + Fr) (L + C + T)

a), b), c), f), p), q),

[IOC (DG/DER); DSOROC] (↓↓)

UKGDS 355 bus Radial DNW;

MO-P/Hybrid SPEA2 + OPF/Static (New DNW) (20 years)

[75]

C. D.

M/[SW + PD] (L + T)

a), b), c), f), h), i), j), l),

[COD; SAIFI and SAIDI] (↓↓)

18 Bus; Practical 51 Section DNW;

MO-P/MACO/Static (1 year) PD *: Protection devices

Mar. 2009/Dist. Load (DL) + PLL

[76]

A. C. D.

M/[(DG + (SS, Fr) + (SSW)] (L + C + T)

a), b), c), r),

Stage I: [(OCS + TOC); CLLI] (↓↓); Stage II: [{(OCS + TOC); CLLI}; P-loss (↓↓); DGPL (↑↑)]

21 bus; 100 bus DNW;

MO-P/MOPSO/Static (1 year)

Aug. 2010/CLd + SLL

[77]

C.

M/(SS + TF + Fr + Br + SSWs)(L + C)

a), b), c), f), p), q), s), u),

[EOI; CENS] (↓↓)

DNW with 180 load buses: 49 exists, 131 added;

MO-P/MINLP + MORTS + GA/Static (1 year)/RTS: Reactive Tabu Search

Jun. 2011/VLd + MLL

IEEE 123 bus DNW;

(MO + W)/MSFLA + Fuzzy approach/Static (1 year)/

Oct. 2011/VLd + TVLL

IEEE 33; Actual 177 bus system;

(MO-P)/MOGA/Dynamic (5 years)

Dec. 2011–12/VLd + MLL Dec. 2011–12/CLd + SLL

[78]

C. D.

M/(A&M SWs *) (L)

a), b), c), f), i), p),

[CIC; SPMC] (↓↓) A&M SWs *: Auto and manual SWs

[79]

A. C. D.

M/(DG + SWs + Br.) (L + C + T)

a), b), c), d), f), n), r),

[CEL; NDE; IRC; EIG] (↓↓)

[80]

C. D.

M/(Fr + SWs + TL *) (L + C)

a), b), c), f), j), i), p), q),

[OCS; CLLI] (↓↓)TL *: Tie-Line

21 Bus; 54 Bus; 100 Bus DNW;

MO-P/S1: SPEA2-MOPSO; S2: SPEA2-BMOPSO/Static (1 year)

[81]

A. C.

[M/(SS + Fr + DG] (L + C)

a), b), c), f), i), m), n), p), q),

[IOC; CIC] (↓↓)

Urban DNW 30000 Consumers;

MO + W/IGDSEP model as SCMINLP AGA & IAGA / Dynamic (5 years)

Mar. 2012/VLd + TVLL/Wt.: 19

[82]

A. C.

M/(DG + Br) (L + C)

a), b), c), d), f), m),

[IOC; ENS] (↓↓)

Actual DNW 2-SS, 16 LP, 24 Br;

(MO-P)/Hybrid PSO-SFL/Dynamic (4 years)/

Aug. 2012/CLd + MLL/

[83]

C.

M/(Fr + Br) (L + C)

a), b), c), f), r), t),

[IOC; IntC] (↓↓)

21 Bus; 54 Bus; 100 Bus DNW;

MO + W/Dynamic Programming (DynP)/Static (1 year)

Nov. 2012/CLd (BL) + MLL

[84]

C.

M/[SS+SWs + Fr + Br] (L + C)

a), b), c), d), f), m), n), r),

[COD; NSEC(as AIC)] (↓↓)

54 Bus DNW;

MO-P/MO Tabu search (MOTS)/Dynamic (15 years)/

Jun. 2013/VLd + MLL

[85]

C. D.

M/Fr(L + C)] +M/(RA)(L)

a), b), c), f), r),

[OCS; CLLI; P-loss; VD] (↓↓)

54 Bus DNW; 100 Bus DNW;

MO + W/MOSOA/Static (1 year)/ADA

Jun. 2014/CLd + SLL/ Apr. 2015/CLd + SLL/

[86]

C. D.

M/[RA + SW] (L + C)

a), b), c), p), s),

[OCS; CLLI] (↓↓)

54 Bus DNW;

MO-P/MOSOA + Fuzzy cardinal priority ranking/Static (10 year)/ADA

[87]

C.

M/(SWD) (L + C)

a), b), c), f), t),

[1.EC; 2.SAIDI and SAIFI; 3.DGUI] (↓↓) A1: Min [1 + 2 + 3]; A2: Min [1 + (2 + 3)]

94 Bus Portuguese DNW;

MO-P/NSGA-II/Static (1 year)/Approach 2 (A2) > (A1)/

Sep. 2015/CLd + SLL

[88]

A. C.

M/(DG + FCL) (L + C + T)

a), b), c), d), f), g), j), r),

[FCLLDG; VSgL; EC(FCL *)] (↓↓) FCL *: Fault current limiters

Canadian benchmark test; IEEE 69 bus DNW;

(MO-P)/PSO + NLP/Static


[89]

A. C.

S/DG [L + C] + S/OLTC(L)

a), b), c), f), i), z),

[VEPB; QPD; VLDG] (↓↓) z = Using dynamic weights.


MO + W/MOGA/Static

Dec. 2015–16/UBL + TVLL

Note: “*” indicates any additional information about a particular decision variable, constraint and objective, respectively.

Energies 2017, 10, 208

14 of 44

Table 4. MO Planning with NTR. Ref.

PT

(DV) (L, C, N, T)

Major Constraint


Test DNW

MO Class/Algorithm/Planning Period (PP)/Others


[90]

D.

M/(SW St.) (L + T)

a), b), c), f), j), m), v)

[P-loss; Reliability indices: (SAIFI, SAIUI, SAIDI, ENS)] (↓↓)

Civanlar 17 bus; Baran 33 bus; Real 172 bus DNW

MO-P/Micro-GA (mGA)/Static/

Apr. 2009/CLd + SLL

[91]

D.

M/(SW St.) (L + T)

a), b), c), f), i),

[P-loss; CILL] (↓↓)

Actual Medium DNW, 96 Br, 28 SWs.

(MO-P)/Fuzzy MCDM in “Proof of concept” tool/ Static (1 year)

May 2009/CLd + MLL/

[92]

C. D.

M/(Fr + SW St. *) (L + T)

a), b), c), f), y)

[CIC; NLC; SWC] (↓↓) SW St.*: Switch Status

3 Fr, 16 bus; 8 Fr DNW.

MO + W/ Binary PSO (BPSO)/Static (1 year)/

May 2009/VLd + TVLL/

[93]

D.

M/(SW St.) (L + T)

a), b), c), d), f), v)

[NLC; ENS] (↓↓)


MO-P/BPSO/Static (1 year)/

Apr. 2012/PLd + PLL

[94]

A. C. D.

M/[DG (L+C)] + M/Fr (L)] + TL

a), b), c), h), w)

[VD; P-loss; CEDGP] (↓↓) TL *: Tie-Line

Sample DNW (3 Fr)

MO + W/GA + MCDA weight/Static/DER control

Nov. 2012/CLd + SLL

[95]

D.

M/(SW St.) (L + T)

a), b), c), d), f), i), m), v)

[P-loss; SAIFI] (↓↓)

33 bus; 67 bus TPC;

MO-P/NSGA-II/Static (1 year)

Mar. 2013/CL + BL

[96]

A. C. D.

M/[DG + Br + SW)] (L + C + T)

a), b), c), f), j), l), m), n), r), s), u)

NPV (↓↓) OF1: [SUC (lines); CPL; NRC; O&M; CDG]; NPV (↓↓) OF2: [CDGG; GHGDG]

Test 38; Test 119 Bus DNW;

(MO-P)/(NSGA)/Dynamic (20 year)/

Aug. 2013/PLd + TVLL

[97]

C. D.

M/(Tie-SW + RA) (L + T)

a), b), c), f), j), v)

[P-loss; ENS; IC (Tie-SWs & new devices)] (↓↓)

IEEE 69 Bus Radial DNW

MO-P/NSGA-II+ non dominated MCS/Static (1 year)/

Oct. 2013/VLd + MLL

[98]

A. D.

M/[DG + SW St. *] (L + C + T)

a), b), c), f), v)

[P-loss; OCS; IC; SAIFI; AENS] (↓↓) SW St. *: Switch Status

Baran & Wu 32 bus test DNW

MO-P/SAMBA/Static (1 year)

Jan. 2014/PLd + PLL

[99]

A. B. C. D.

S/[DG + FACTS] + M/[(TL + SW St. *)]; (L + C + T) *

a), b), c), f), v)

[VP(↑↑); P-loss(↓↓); LB(↑↑)] SW St. *: Switch Status

IEEE 33 Bus; TPC Test DNW

MO + W/Fuzzy-ACO/Static/Wt.(1–3): (0.33 each)

Jan. 2015/CLD + MLL/

[100]

A. D.

M/(DG + SW St.] (L + C + T)

a), b), c), f), h), m), v), w)

[P-loss(Br and TFs); OC (SS + DG); CCT] (↓↓)

33 Bus DNW Radial

MO-P/EGSA + FDM/Static/

Mar. 2016/ VLd + TVLL/

Notes: PT: in Sections 2.1–2.4; DV: Decision Variables (Location, capacity, numbers and types); Constraints in Section 2.7.4; Objectives in Section 2.5; MO Class in Section 2.6; LdM, LDP in Section 2.7.5. Also, “*” indicates any additional information about a particular decision variable, constraint and objective, respectively.

Energies 2017, 10, 208

15 of 44

Energies 2017, 10, 208

15 of 44

Figure 1. An overarching diagram of the paper. Figure 1. An overarching diagram of the paper.

4. Planning Techniques based on Evaluation Methods in Multi-objective Planning 4. Planning Techniques based on Evaluation Methods in Multi-objective Planning The methods employed in MOP problems spread across four PT (as indicated in taxonomies of The employed in MOP problems spread meta-heuristics, across four PT (as indicated in taxonomies of Tables methods 1–4) can be broadly categorized into numerical, hybrid and decision-making Tables 1–4) can be broadly categorized into numerical, meta-heuristics, hybrid and decision-making methods respectively. In addition, main contributions of reviewed work are shown chronologically methods respectively. In addition, main contributions of reviewed work are shown chronologically in in Tables 5–8. Tables 5–8.

Energies 2017, 10, 208

16 of 44

Energies 2017, 10, 208

16 of 44

Table 5.Energies The contribution of MODGP-based planning methods and related works. (PTC: Components in planning techniques; MODGP: based DG placement). 2017, 10, 208 16 of MO 44 Table 5. The contribution of MODGP-based planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement).

Energies 2017, 10, 208

16 of 44

Table 5. TheMO contribution Components in planning techniques; MODGP: MO based DG placement). Ref. Year PTC Aof MODGP-based B BC D C planning Contribution of MODGP-based Planning Ref. Energies Year D methods and related works. (PTC: Contribution of MODGP-Based Planning 2017, MO 10, 208 PTC 16 of 44 MODGP formulation based on GA and ε constrained method is proposed to solve the best compromise (tradeoff) solutionplacement). for DM Table 5. The contribution andformulation related works. (PTC: Components in planning techniques; MODGP: MOthe based Year Aof MODGP-based B- -C D Contribution of MODGP-based Planning ✓ - - planning [21]2005 2005 4 1 PTC 1 [21] Ref. 4 MO - methods MODGP based on GA and ε constrained method is proposed to solve best DG compromise (tradeoff) solution for DM (DISCO). Energies 2017, 10, 208

(DISCO).

16 of 44

(DISCO). along with max-min approach solves MODGP problem considering future load uncertainties and risk management NSGA-II

16 of 44

MODGP formulation based on GA and ε constrained method is proposed to solve the best compromise (tradeoff) solution for DM Table 5. The contribution planning methods andwith related (PTC: Components in planning techniques; MODGP: MO to based DG placement). Ref. Year MO Aof MODGP-based B- -C D of MODGP-based Planning ✓ - - Similarly, [21] 2005 4 1 [22] Energies 3 208 - with Similarly, the works. same formulation (GAmethod), and ε constrained method), a double tradeoff method isalternative. presented [22]20052017, 3 1 PTC the same formulation (GA and ε Contribution constrained a double tradeoff method is presented find the best 10, 16 ofto 44find the best alternative. (DISCO). MODGP formulation based on GA and ε constrained method is proposed to solve the best compromise (tradeoff) solution for DM ✓ [23] 2007 2 1 MCS embedded in GA planning methodology is proposed to improve the accuracy for stochastic DG integration with tradeoff solution. Table 5. The contribution of MODGP-based planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). A B- -C D of MODGP-based Planning ✓ - - Similarly, [21] 2005 2 MO 4 1 [23] Ref. - with MCS embedded in GA methodology is proposed to improve the forfind stochastic integration with tradeoff solution. [22]2007 Year 3 1 PTC the same formulation (GAplanning and ε Contribution constrained method), a double tradeoff method is accuracy presented to the best DG alternative. Energies [24] 2017, 200810, 2083

1

✓

-

-

-

MODGP formulation on GA and ε constrained method is proposed solve the compromise (tradeoff) solution for DM ✓ --[23] 2007 2 1 embedded in GAbased planning methodology isContribution proposed to improve the to accuracy forbest stochastic DG integration with tradeoff solution. Table 5. The contribution planning methods and related works. (PTC: Components inMODGP planning techniques; MODGP: MO to based DG placement). Aof MODGP-based B-- -C D of MODGP-based Planning ✓ [21] 2005 4 1 [24] Ref. 3 MO - MCS - with NSGA-II along with max-min approach solves problem considering future load uncertainties and [22]2008 Year 3 1 PTC Similarly, the same formulation (GA and constrained method), a double tradeoff method is presented find the best alternative. EA is employed to solve IMO considering bothεtime-varying generation and demand behavior aiming at various technical impacts of risk management (DISCO). ✓ [24] 200810, 2083 1 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management [25] 2017, 6 Energies 16 of 44 MODGP formulation based on GA and ε constrained method is proposed to solve the best compromise (tradeoff) solution for DM ✓ [23] 2007 2 1 MCS embedded in GA planning methodology is proposed to improve the accuracy for stochastic DG integration with tradeoff solution. Table 5. The contribution of MODGP-based planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). A B- -C D of MODGP-based Planning ✓ - - DGP [21] 2005 6 MO 4 1 [25] Ref. - with EA is employed to solve IMO considering both time-varying generation demand aiming at various technical impacts of DGP [22]2008 Year 3 1 PTC Similarly, the same formulation (GA and ε Contribution constrained method), a double tradeoff method and is presented to behavior find the best alternative.

EA is employed to solve IMO considering both time-varying generation and demand behavior aiming at various technical impacts of (DISCO). ✓ ---2008 3 1 NSGA-II alongprogramming with max-min approach problem considering future load uncertainties risk management 6 ✓ GA with goal methodology findsMODGP a solution with uncertainties among MO andand constraints. 2008 2 1 MODGP formulation based on GA and εsolves constrained method is associated proposed to solve the best compromise (tradeoff) solution for DM ✓ 16 of 44 (DISCO). ✓ [24] 2008 3 1 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management [25] 6 ✓ GA with goal programming methodology findsisaproposed solution with associated uncertainties among MO and constraints. [26] 2008 2 1 MODGP formulation based on GA and ε constrained method is proposed to solve the best compromise (tradeoff) solution for DM ✓ [23] 2007 2 1 MCS embedded in GA planning methodology to improve the accuracy for stochastic DG integration with tradeoff solution. 2017, 2084 16 of 44 contrasting objectives. ✓ - -- - DGP [21] 200510, 1 An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. [27] Energies 2008 3 1 The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among [22] 3 Similarly, with the same formulation (GA and ε constrained method), a double tradeoff method is presented to find the best alternative. EA is employed to solve IMO considering both time-varying generation and demand behavior aiming at various technical impacts of Ref. Year PTC Aof MODGP-based B-- 5. The C D Contribution ofworks. MODGP-based Planning ✓ The proposed NSGA algorithm finds arrangements for wind-based regarding compromise solution among contrasting objectives. [27] Table 2008 3 1 (DISCO). [28] 2009 4 ✓ --- contribution --[24] 2008 3 1 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management 5. TheMO contribution Table planning of MODGP-based methods and related planning works. methods (PTC: and Components related inDGs planning (PTC: Components techniques; MODGP: inMO planning MO techniques; based DG placement). MODGP: MO based DG placement). [25] 6 ✓ GA with goal programming methodology finds solution with associated uncertainties among constraints. [26] 2008 2 1 GA isembedded also used for comparison ofmethodology results and advocated for DNW systems even with suboptimal solutions. ✓ -[23] 2007 1 MCS in GA planning isaproposed tolarge improve the accuracy forbest stochastic DGand integration with tradeoff solution. DGP Energies 2017, 10, 2082 16 of 44 GA is also used for MODGP formulation based onconsidering GA and ε constrained isMODGP proposed to solve the compromise (tradeoff) solution for DM An exhaustive search analysis (ES) has presented withmethod IMOpresented for problem under variable load conditions. ✓ An exhaustive search analysis (ES) has IMO for MODGP problem under variable load conditions. [22] 2005 3 Similarly, withNSGA the same formulation (GA and εtime-varying constrained method), awith double tradeoff method is presented to find the best alternative. EA isproposed employed to solve IMO both generation and demand behavior aiming atset various technical impacts of The algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27]2009 2008 3 1 1 ✓ [21] 2005 [28] 2009 4 1 [28] Ref. 4 ✓ [29] 2010 5 2 A fuzzy embedded GA is employed to solve fuzzy weighted single objective function employing fuzzy theory for MODGP. ✓ [24] 2008 3 1 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management Table 5. The contribution of MODGP-based planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). [25] 6 ✓ GA methodology finds aproposed solutionfor with uncertainties among MO constraints. [26] 2008 2 1 Ref. ✓ Year MO PTC A Year B-C D PTCwith Agoal Bprogramming D ofassociated MODGP-based Contribution Planning ofsuboptimal MODGP-based Planning (DISCO). GA isembedded also used for comparison ofmethodology results and advocated for DNW systems even with solutions. comparison of results and advocated large DNW systems even with suboptimal solutions. -- MO -[23] 2007 2 1 MCS inCGA planning isContribution tolarge improve the accuracy for stochastic DGand integration with tradeoff solution. DGP An exhaustive search analysis (ES)tohas presented with IMO for MODGP and problem under variable load conditions. ✓ MCS-embedded GA isalgorithm presented solve chance constrained programming framework considering future uncertainties of impacts loads/DGs. [30] 2011 5 1 EA isproposed employed to solve IMO considering both time-varying generation demand behavior aiming atset various technical Energies 2017, 208contribution 16 of 44 solution for DM The NSGA finds arrangements formethod wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 MODGP formulation based on MODGP GA and formulation εsolves constrained based onwith GA isand proposed εplanning constrained to solve method the best iscompromise proposed to (tradeoff) solve the solution best compromise for DM of(tradeoff) [28] 2009 4 ✓ --[22] 2005 3 1 Similarly, with the same (GA and εMODGP method), a double tradeoff method is presented to find the best alternative. ✓ [29] 2010 5 2 A fuzzy embedded GA isformulation employed to solve fuzzy weighted single function employing fuzzy theory for MODGP. [24] 2008 NSGA-II along with max-min approach problem considering future load uncertainties and risk management Table 5.10, The planning methods and related works. (PTC: Components inobjective techniques; MODGP: MO based DG placement). [25] 6 1 GA with methodology aconstrained solution uncertainties among MO and constraints. [26]2010 2008 2 2 PTC Year Aof MODGP-based B---- -C Contribution ofassociated MODGP-based Planning ✓ ✓goal -programming -iscomparison - embedded [21] 2005 4 1 [21] ✓ 2005 1hybrid GA is also used for ofsolve results and advocated for large DNW systems even with suboptimal solutions. [29] Ref. 5 MO A fuzzy GA isfinds employed to solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. ✓ -- 4D -- - DGP A GA-PSO proposed to MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 1 An exhaustive search analysis (ES) has presented with IMOtofor MODGP problem under variable load conditions. (DISCO). (DISCO). ✓ [23] 2007 2 1 MCS embedded in GA planning methodology is proposed improve the accuracy for stochastic DG integration with tradeoff solution. MCS-embedded GA is presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 EA is employed to solve IMO considering both time-varying generation and demand behavior aiming at various technical impacts of ✓ The proposed NSGA algorithm finds arrangements formethod wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 3 MODGP formulation based on GA and ε constrained is proposed to solve the best compromise (tradeoff) solution for DM [28] 2009 4 1 ✓ ✓ [29] 2010 5 2 A fuzzy embedded GA is employed to solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. ✓ [25] 2008 6 1 A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a single objective GA with methodology finds aconstrained solution with associated uncertainties among MO and constraints. [26]2011 2 1 PTC Ref. Year MO Aof MODGP-based B-- -C Contribution of MODGP-based Planning -- 3D -- - DGP [21] 2005 4 1 GA is also used for-is comparison results and advocated for large DNW systems even with suboptimal solutions. ✓ [30] Energies 5 208 - goal MCS-embedded GA isMODGP presented to solve chance programming framework considering future of alternative. loads/DGs. -programming - formulation [22] 3 2005 Similarly, 1hybrid with the same Similarly, (GA with and the εMODGP same formulation method), (GA aconstrained and double ε constrained tradeoff method method), is apresented double tradeoff to find the method alternative. is presented ✓ [24] 2008 1 [22] ✓ NSGA-II along with max-min approach solves problem considering future load uncertainties and risk management Table 5.10, The contribution planning methods and related works. (PTC: Components inDG planning techniques; MODGP: MO based DGbest placement). A GA-PSO proposed toofsolve problem regarding location (with GA) and capacity (with PSO) respectively. [31] 2012 3 [32] 5 2017, 16 uncertainties ofto 44find the best An exhaustive search analysis (ES) presented with IMO for MODGP problem under variable load conditions. (DISCO). ✓ ---MCS-embedded GA isalgorithm presented tohas solve chance constrained programming framework considering future uncertainties of for loads/DGs. [30] 2011 5 1 function, which is solved by GA ensure optimal DG (wind) planning. ✓ 2007 The proposed NSGA finds formethod wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 3 1 MODGP formulation based on GA and εtoconstrained is proposed to solve the best compromise (tradeoff) solution DM [28] 2009 4 [29] 2010 5 2 A embedded GA employed toarrangements solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. ✓ --- 2✓ --programming -solve -isIMO [23] 2 1fuzzy MCS infurther GA planning MCS embedded inadvocated istime-varying GA proposed planning toprogramming methodology improve the accuracy isdemand proposed forbehavior stochastic to improve DG the integration accuracy for with stochastic tradeoff DG solution. integration with tradeoff solution. EA is employed to considering both generation and aiming at various technical impacts of(with PSO) A goal attainment method (GoA) has presented, where goal transforms multiple objective functions into abest single objective ✓ GA with goal methodology finds aconstrained solution with associated uncertainties among MO and constraints. [26]2012 2007 2008 3 2 1 1 1 [23] ✓ [21] 4 GA isembedded also used for comparison ofmethodology results and for large DNW systems even with suboptimal solutions. ✓ [31] Ref. A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity respectively. [22] 2005 3 1 Similarly, with the same formulation (GA and ε method), a double tradeoff method is presented to find the alternative. A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 ✓ [25] 2008 6 1 [32] 5 MODGP is solved by twotostage heuristic iterative method including clustering (outer) and EA (inner) respectively, Year MO PTC A 2008 BC D Contribution ofMODGP MODGP-based Planning An exhaustive search analysis (ES) presented with IMO for problem under variable load conditions. (DISCO). MCS-embedded GA isalgorithm presented solve constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 ✓problem -- 3✓ --NSGA - GA -is employed [24] 2008 3 1 1fuzzy NSGA-II along with max-min NSGA-II approach along solves with MODGP max-min problem approach considering solves MODGP future load problem uncertainties considering and future riskoptimizations management load uncertainties and risk management DGP function, which is solved byhas GA tochance ensure optimal DG (wind) planning. ✓ [33] Table 2012 2 The proposed finds for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 3 1 [24] ✓ [28] 2009 4 [29] 2010 5 2 Aplanning embedded toarrangements solve fuzzy weighted single function fuzzy set theory for MODGP. ✓ -[23] 2007 2 1 MCS embedded infurther GA planning methodology is where proposed to improve the accuracy foremploying stochastic DG integration with tradeoff solution. 5.10, The of MODGP-based methods and related works. (PTC: Components inobjective planning techniques; MODGP: MO based DG placement). A goal attainment method (GoA) has presented, goal programming transforms multiple objective functions into abest single objective to find time varying voltage magnitude loss sensitivity factor at each node. Energies 2017, 208contribution of 44 technical MODGP formulation based on GA and εand constrained method is associated proposed to solve the best compromise (tradeoff) solution for DM GA is also used for comparison of results and advocated for large DNW systems even with suboptimal solutions. ✓ A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a single -[22] 3 Similarly, with the same formulation (GA and εsolve method), aboth double tradeoff method is presented to find the alternative. A hybrid GA-PSO is proposed to MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. 2012 3 1 EA is employed to solve IMO EA considering issolve employed both to time-varying IMO considering generation and time-varying demand behavior generation aiming and at(inner) demand various technical behavior impacts aiming at of16 various impactsobjective of [32] 5 GA with goal programming methodology finds aconstrained solution with uncertainties among MO and constraints. [26]2012 2005 2008 5 2 1 1 MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA optimizations respectively, An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. - - --- 6--- - function, [21] 2005 4 1 [25] ✓ [32] [31] MCS-embedded GA is presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 [24] 3 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management ✓ ✓ [25] 2008 6 1 2008 1 which is further solved by GA to ensure optimal DG (wind) planning. [33] 2012 2 The multi-criteria model aims to achieve contrasting objectives (cost and reliability) forobjective DGP. GA further finds set ofsolution. non[28] 2009 4 (DISCO). ✓ [29] 2010 5 2 A fuzzy embedded GA is which employed toarrangements solve fuzzy weighted single objective function employing fuzzy set theorycontrasting for MODGP. function, is further solved by wind-based GA to optimal DG (wind) planning. ✓ --[23] 2007 2 1 MCS embedded inplanning GA planning methodology isContribution proposed to improve the accuracy for stochastic DG integration with tradeoff A goal attainment method (GoA) has presented, where goal programming transforms multiple functions into aasingle objective DGP DGP to find time varying voltage magnitude and loss sensitivity factor at each node. ✓ The proposed NSGA algorithm finds for DGs regarding compromise solution among objectives. [27] 2008 3 1 Ref. Year MO PTC A B--C D ofensure MODGP-based Planning [34] 2013 2 GA is also used for comparison of results and advocated for large DNW systems even with suboptimal solutions. ✓ A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 1 EA is employed to solve IMO considering both time-varying generation and demand behavior aiming at various technical impacts of [32] 5 MODGP problem issame solved by approach twotostage heuristic iterative method clustering (outer) and EA (inner) optimizations respectively, Energies 2017, 208contribution 16 of 44 dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. ✓ --- 2-[22] 2005 3 1 Similarly, with the (GA and εMODGP method), afinds double tradeoff method is presented to find the best alternative. MCS-embedded isbased presented solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 [24] 2008 NSGA-II along with max-min solves problem considering future load uncertainties and risk management 5.10, The of MODGP-based planning methods and related works. (PTC: Components inincluding planning techniques; MODGP: MO based DG placement). [25] 6 1 which isGA further solved byaims GA toconstrained ensure optimal DG (wind) planning. ✓ -programming - GA -isformulation GA with goal GA methodology with programming finds aconstrained solution methodology with associated uncertainties asolve solution with among associated MO and uncertainties constraints. among MO and constraints. [26] 2008 2 2008 1fuzzy [33] Table 2012 The multi-criteria planning model achieve contrasting objectives (cost and reliability) forobjective DGP. GA further finds set ofDM nonAn exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. MODGP formulation on GA and εto method is proposed to the best compromise (tradeoff) solution for ✓ ✓ function, [29] 2010 5 2 [26] ✓ A embedded employed togoal solve fuzzy weighted single function employing fuzzy set theory for MODGP. MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, A goal attainment method (GoA) has presented, where goalto programming transforms multiple functions into aasingle objective DGP to find time varying voltage magnitude and loss sensitivity factor atobjective each node. [34] 2013 2 2 1 1 [28] 2009 ✓ -- - -2005 4 [35] An MO planning has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. [33] [21] - problem [23] 2 MCS embedded inframework GA planning methodology istime-varying proposed improve the accuracy forbehavior stochastic DG integration with tradeoff solution. A GA-PSO proposed toof solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 1 EA is employed to solve considering both generation and demand aiming at(inner) various technical impacts of among contrasting objectives. [32] 5 MODGP is solved by two stage heuristic iterative method including clustering (outer) and EA optimizations respectively, ✓ 2008 ✓ - 3--- - function, -NSGA -is - IMO The proposed algorithm The finds proposed arrangements NSGA algorithm for wind-based finds arrangements DGs regarding for compromise wind-based DGs solution regarding among compromise contrasting objectives. solution [27]2012 2007 2008 3 1 [27] ✓ 1hybrid dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. GA is also used for comparison results and advocated for large DNW systems even with suboptimal solutions. (DISCO). to find time varying voltage magnitude and loss sensitivity factor at each node. ✓ MCS-embedded GA is presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 1 [25] 6 which is further solved by GA to ensure optimal DG (wind) planning. ✓ GA with goal programming methodology finds a solution with associated uncertainties among MO and constraints. [26] 2008 2 1 [33] 2012 EnRef. o Year MO PTC Aof MODGP-based BC D Contribution of MODGP-based Planning The multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) for DGP. GA further finds asingle set of objective nonAn MO optimization problem for multiple micro-turbines (DG) placement and sizes is solved by hybrid PSO and SFL algorithms based ✓ [24] 2008 3 1 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management Table 5. The contribution planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a DGP to find time varying voltage magnitude and loss sensitivity factor atpresented each node. [34] exhaustive search analysis An (ES) exhaustive has presented search with analysis IMO (ES) for has MODGP problem with under IMO for variable MODGP load problem conditions. variable load conditions. ✓ [35] 2013 2 An MO planning has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. [29] 2010 5 2 fuzzy embedded GA isformulation employed to solve fuzzy weighted single objective function employing fuzzy set theory forthe MODGP. ✓ ---[22] 2005 3 1 Similarly, with theframework same (GA and ε constrained method), a double tradeoff method is presented tounder find best objectives. alternative. [36] A GA-PSO proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 [32] 5 1 MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting [27] 2008 3 ✓ 2009 ✓goal dominant solutions for wind-based DG sizing, siting, and maintenance schedules. -programming -is - fuzzy [28] 2009 4 1 [28] ✓ 1hybrid MODGP formulation based on GA and εunits method is proposed to solve the best compromise (tradeoff) solution forrespectively, DM The multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) for DGP. GAoffurther finds a set of non-dominant framework, followed by decision-making to select the most preferred Pareto optimal solution satisfying competing objectives. EA is employed to solve IMO considering both time-varying generation and demand behavior aiming at various technical impacts is further solved by GA toconstrained ensure optimal DG (wind) planning. ✓ ---- 4--- - function, GA with methodology finds atool solution with associated uncertainties among MO constraints. [26] 2 [33] 2012 The multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) for DGP. GA further finds aof set of nonGA is also used for comparison GA of ismethodology results also used and for advocated comparison for of large results DNW and systems advocated even with large suboptimal DNW systems solutions. even with solutions. [21]2013 2008 2005 2 4 1 1 [34] EnRef. - which An MO optimization for multiple micro-turbines (DG) placement and sizes isfor solved by hybrid PSO andwith SFL algorithms based MCS-embedded GA isproblem presented to solve chance constrained programming framework considering future uncertainties loads/DGs. [30] 2011 5 [23] 2007 2 1 MCS embedded in GA planning isContribution proposed to improve the accuracy for stochastic DGand integration tradeoff solution. A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into asuboptimal single objective [25] 2008 6 to find time varying voltage magnitude and loss sensitivity factor at each node. o ✓ - ba - edDGP Year MO PTC Ao MODGP B--- -C D of MODGP-based Planning [34] 2013 2 1 on✓ An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. [35] MO planning framework has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. (DISCO). solutions for wind-based DG units sizing, siting, and maintenance schedules. [36] 2013 3 [37] The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). ✓ [32] 2012 5 1 Tab e 5 The on bu p ann ng me hod and e a ed wo k PTC Componen n p ann ng e hn que MODGP MO ba ed DG p a emen MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, The proposed algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 3 ✓ 2010 dominant solutions for wind-based DG units sizing, and maintenance schedules. [28] 2009 4 1 ✓ ✓ -NSGA -isGA [29] 2010 5 2 [29] ✓ 2fuzzy embedded isfuzzy employed A approach fuzzy to embedded solve fuzzy GA weighted issiting, single to solve objective fuzzy function weighted employing single objective fuzzyGA set function theory for employing MODGP. fuzzy set theory for MODGP. framework, followed by decision-making tool toemployed select the most preferred Pareto optimal solution satisfying competing objectives. A hybrid GA-PSO proposed to solve MODGP problem regarding DG location (with GA) and capacity (with respectively. [31] 2012 --- 5✓ -[24] 2008 3 1 NSGA-II along with max-min solves MODGP problem considering future load uncertainties and riskPSO) management function, which is planning further solved GA toconstrained ensure optimal DG (wind) planning. [33] 2012 2 The multi-criteria model aims achieve contrasting objectives (cost and reliability) for DGP. further finds a set of nonGA is also used for comparison ofby results and advocated for large DNW systems even with suboptimal solutions. MODGP formulation based on GA and εto method is proposed to solve the best compromise (tradeoff) solution for DM ✓ An MO optimization problem for multiple micro-turbines (DG) placement and sizes is solved by hybrid PSO and SFL algorithms based [22] 2005 3 1 Similarly, with the same formulation (GA and ε constrained method), a double tradeoff method is presented to find the best alternative. The DGP problem for location, size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), in which ✓ GA with goal programming methodology finds a solution with associated uncertainties among MO and constraints. [26] 2008 2 1 to find time varying voltage magnitude and loss sensitivity factor at each node. [34] 2013 An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. ✓ 2011 - 5-[35] 2 MO planning framework has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. [21] 2005 4 1 [30] ✓ [35] EnR 2013 2 1 An MO p ann ng amewo k ha deve oped name y mp oved mu ob e ve ha mony ea h MOHS wh h an evaofua e he DGP [36] 2013 3 ✓ MCS-embedded GA is presented MCS-embedded to solve chance GA is constrained presented programming to solve chance framework constrained considering programming future framework uncertainties considering of loads/DGs. future uncertainties loads/DGs. [30] 2011 5 1 [37] The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a single objective [38] 2014 2 1 EA is employed to solve IMO considering both time-varying generation and demand behavior aiming at various technical impacts of MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, o ✓ solutions for wind-based DG units sizing, siting, and schedules. Ao MODGP B--C D on o maintenance MODGP b ng dPareto P hn nn ngcompany [28] TabY 2009 4on PTC 1 on✓ ✓ (DISCO). [29] 2010 5 2 A embedded GA isfuzzy employed toarrangements solve fuzzy weighted single function employing fuzzy set theory for MODGP. framework, followed by decision-making tool tobu select the most optimal solution satisfying competing objectives. --- ba -- eddominant [23] 2007 2 1 MCS embedded inplanning GA planning methodology isCon proposed to improve the accuracy for stochastic DG(LDC). integration with tradeoff solution. e 5 TheMO bu pfuzzy ann ng me hod and ecompromise afor ed wo kand PTC Componen nobjective ppreferred ann eeven que MODGP MO ba ed DG aobjectives. emen [32] 2012 5 an NSGA-II is NSGA used to obtain (Pareto frontier) solution for a systems local distribution [25] 2008 6 [33] 2 1 ✓ The proposed algorithm finds for wind-based DGs regarding compromise solution among contrasting [27] 3 multi-criteria model aims to achieve contrasting objectives (cost and reliability) for DGP. GA further finds apset of nonGA is also used of results advocated for large DNW with suboptimal solutions. An MO optimization problem multiple micro-turbines (DG) placement and sizes isregarding solved bycapacity hybrid PSO and SFL algorithms ✓ --- 3--- for -on - solved A GA-PSO iscomparison proposed A hybrid to solve GA-PSO MODGP isproblem proposed regarding to solve MODGP DG location problem and DG location (with (with GA) respectively. and capacity (with PSO) respectively. [31] 2012 3 1 1hybrid The DGP problem for size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), inbased which function, which isGA further by GA to ensure DG (wind) planning. DGP to find time varying voltage magnitude and loss sensitivity factor at each node. ✓ 2012 [34] 2013 2 1 [31] ✓ [35] An MO planning framework has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. MODGP owith mu blocation, d Pareto on GA nd on noptimal dframework m hod p(PFDE) opo d one o v(with hload bGA) omp om dDG oPSO) o u on oand DM ✓ An MO op m za on p ob em o mu p e m o u b DG p a emen and ze o ved by hyb d PSO and SFL a go hm ba ed [36] 2013 ---[22] 2005 3 1 Similarly, the same formulation (GA and ε constrained method), a double tradeoff method is presented to find the best alternative. MCS-embedded is presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 [37] The methodology based on frontier differential evolution algorithm is proposed for optimal MODGP (sizing location). [24] 2008 NSGA-II along with max-min approach solves MODGP problem considering future uncertainties and risk management ✓ [38] 2014 2 1 The MO probabilistic (mathematical programming) has proposed for various DER planning (six types, size, and location) dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. ✓ ✓ [29]2013 2008 2010 3 5 1 1 2 A fuzzy embedded GA isfuzzy employed topresented, solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. 36 R framework, followed by decision-making tobu select the most preferred Pareto optimal solution satisfying competing objectives. A goal attainment method Acompromise goal has attainment method where (GoA) goal programming has presented, transforms where goal multiple programming objective transforms functions multiple into aasingle objective objective functions a single an NSGA-II isprogramming used to obtain (Pareto frontier) solution foroo a (cost local distribution company (LDC). MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and (inner) optimizations respectively, ---GA with goal finds atool solution with associated uncertainties among MO and constraints. [26] 2 The multi-criteria planning model aims to achieve contrasting objectives for GA further finds of nonAo MODGP B-C Con on o MODGP b d Preliability) nn ng [28] TabY 2009 4on PTC D SCO An MO optimization problem multiple micro-turbines (DG) placement and sizes isbehavior solved by hybrid PSO and SFL algorithms based amewo k(GoA) oconsidering owed by uzzy de on ng o eand ee(with he mo paiming eDGP. eEA ed Pa eba o can op ma u on a y nginto ompe ng objective ob e ve ✓ [23] 2007 2 1 embedded in GA planning methodology istime-varying proposed tomak improve accuracy for stochastic DG integration with tradeoff solution. A GA-PSO is proposed to solve problem regarding location GA) and capacity (with PSO) respectively. [31] 2012 3 The DGP problem location, size, and (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), in which ✓ - for -for - IMO [32] 5 [32] 2012 5D e 5 TheMO bu on✓ p1hybrid ann ng me hod and emethodology afor ed wo ktype PTC Componen nDG p the ann ng hn que MODGP MO ed DG pset aoand emen EA is employed to solve both generation and demand at various technical impacts of ✓ --- ba -- edMCS [33] 2012 2 1 [39] 2014 that aims atused DISCOs contribution in theMODGP competitive electricity market. Moreover modified augmented ε-constrained method along with [34] 2013 GA is also comparison of results and advocated for large DNW systems even with suboptimal solutions. [35] 2 An MO planning framework has developed, namely improved multi-objective harmony search (IMOHS), which evaluate the DGP. [36] 2013 3 ✓ MCS-embedded GA is presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 1 [37] The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing location). [38] 2014 2 function, which is further solved function, by GA which to ensure is further optimal solved DG by (wind) GA to planning. ensure optimal DG (wind) planning. ✓ [25] 2008 6 1 The MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) to find time varying voltage magnitude and loss sensitivity factor at each node. proposed algorithm finds arrangements for wind-based DGs solution among contrasting [27] 3 dominant solutions for wind-based units siting, and maintenance En[24] MODGP ow mu on bobtain d(GoA) on GA nd on nfrontier) d hod p opo dregarding o schedules. odistribution vsolutions. hload ompobjective om d oo management ou oobjectives. DM ✓ framework, followed by fuzzy decision-making tool tomthe select the most preferred Pareto optimal solution competing objectives. S m y along h for m ocompromise mu onDG GA ndsizing, on nproblem dset m hod doub dcompromise ob muncertainties hod p (LDC). nsatisfying drisk ndinto h aon bsingle n v o 3 NSGA-II with max-min approach solves MODGP considering future and A goal attainment method has presented, where goal programming transforms multiple functions objective an NSGA-II ishNSGA used to compromise (Pareto solution for aoptimal local company DGP FDM has utilized best solution among ofddefined Pareto ✓ [29]2013 2008 2010 3 5 1 1 2 fuzzy embedded GA is employed to solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. An MO optimization problem for multiple micro-turbines (DG) placement and sizes is solved by hybrid PSO and SFL algorithms based 37 R The me hodo ogy ba ed on Pa e o on e e en a evo u on PFDE a go hm p opo ed o op ma MODGP z ng and o a on ✓ A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 1 The DGP problem for location, size, and type (REG and PEV) is as MO mixed integer nonlinear programming (MINLP), in which [32] 5 ✓ MODGP problem is solved by MODGP two stage problem heuristic is solved iterative by method two stage including heuristic clustering iterative method (outer) and including EA (inner) clustering optimizations (outer) and respectively, EA (inner) optimizations respectively, [39] 2014 2 1 that aims at DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with The multi-criteria planning model aims to achieve contrasting objectives (cost and DGP. GA further finds adset of nonY MO PTC A BC D function, Con bu on omp MODGP b demand dP ng An exhaustive search analysis (ES) has presented IMO MODGP problem under variable load conditions. [35] 2013 MO planning has developed, namely improved multi-objective harmony search (IMOHS), can the DGP. D1 SCO [36] 3 1 [33] ✓ [37] 2013 The methodology based Pareto frontier differential evolution (PFDE) algorithm isnn for optimal MODGP (sizing MCS mb dd-programming d nframework GA(mathematical p- on nn ng m hodo ogy pwith opo dplanning ofor ov hand ufor yreliability) oproposed oDER h for DG nat gwhich ontypes, w hevaluate oand olocation). uofon [38] 2014 which is further solved by GA tochance ensure optimal DG (wind) planning. EA is employed IMO considering both time-varying generation behavior aiming various technical impacts The MO probabilistic programming) framework has proposed various planning (six DG size, and location) ✓ -solve [33] 2012 ✓ ---- 2--GA with goal methodology finds atool solution with associated uncertainties among MO and constraints. [26] 2008 2 1 [34] 2013 [40] 2014 MO augmented εto constrained method proposed for DGP on short term basis for future DNWs with ANM functionalities. [28] 2009 4 framework, followed by fuzzy decision-making to the select the most preferred Pareto optimal solution satisfying competing objectives. ✓ 2012 MCS-embedded GA is presented to solve constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 1 A goal attainment method has presented, where programming transforms multiple functions into aon single objective an NSGA-II ishused to obtain compromise (Pareto solution for aoptimal local company (LDC). ✓ [25] 2008 6 1 to find time varying voltage to magnitude find time and varying loss sensitivity voltage magnitude factor at and each loss node. sensitivity factor atobjective each node. FDM has utilized for best compromise solution among set of Pareto dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. GA is also used for comparison of results and advocated for large DNW systems even with suboptimal solutions. MODGP ow mu on blocation, d(GoA) on GA nd on nfrontier) d mgoal hod p opo d ong odistribution vsolutions. hois b omp om d ba o o DG u o respectively, DM An MO optimization problem for multiple micro-turbines (DG) placement and sizes solved by hybrid PSO and SFL algorithms based ✓ The DGP p ob em o o a on ze and ype REG and PEV defined a MO m xed n ege non nea p og amm ng M NLP n wh h an S m y h m o mu on GA nd on n d m hod doub d m hod p n d o nd h b n v The DGP problem for size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), in which NSGA ong w h m x m n pp o h o v MODGP p ob m on d ng u u o d un n nd k m n g m n [32] 2012 5 1 Tab e 5 The on bu on o MODGP ba ed p ann ng me hod and e a ed wo k PTC Componen n p ann e hn que MODGP MO ed p a emen DGP ✓ MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations [39] 2014 2 that aims at DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 3 o MOO planning on non-dominated sorting genetic algorithm II proposed (NSGA-II) along with MCS (forrespectively. uncertain operation ✓ [36] 2013 38 En[33] 2014 2012 1 1111 [37] The methodology based on Pareto frontier differential evolution (PFDE) algorithm is for optimal MODGP (sizing location). A hybrid GA-PSO isframework proposed tobased solve MODGP problem regarding DG location (with GA) andplanning capacity (with PSO) [31] 2012 3 ✓ [38] 2014 function, which is further solved by GA to ensure optimal DG (wind) planning. The MO probabilistic (mathematical programming) framework proposed various (six DG types, size, and ------2 multi-criteria planning model The multi-criteria aims to achieve planning contrasting model objectives aims toon achieve (cost and contrasting reliability) objectives (cost and further reliability) finds adfor setoand DGP. of nonGA [40] 2014 2 MO augmented εonframework constrained method for DGP planning on short term for future DNWs with ANM functionalities. [35] 2013 2 An MO planning has developed, namely multi-objective harmony search (IMOHS), can the DGP. [41] Dfuzzy SCO [29] 2010 5 2 [34] ✓ A embedded GA employed to solve fuzzy weighted single objective function fuzzy set theory MODGP. framework, followed by decision-making to select the most preferred Pareto optimal solution satisfying competing objectives. NSGA u ed ohas obproposed a (Pareto n omp om estorage ehas oat euncertainties o ubasis on ovariable o MO aDGP. dng bu on ompany LDC ✓ MCS mb dd GA p-isfuzzy nn ng hodo ogy pwith opo oPa mp ov ufor ynd obemploying oDER haofor DG nGA on wfor hevaluate olocation) uo on further finds a set of nonan NSGA-II is-programming used to obtain compromise frontier) solution for ahoptimal local distribution company (LDC). EA mp oy d oaddresses v MO onm dproblem ng bo h m vimproved yd ng g n on nd d msolutions. h v m vgwhich ou hn mp to find time varying voltage magnitude and loss sensitivity factor each node. FDM has utilized for best compromise solution among the set of Pareto ✓ 2013 - 2✓ GA with goal methodology finds atool solution with associated among and constraints. [26] 2008 2 1 ✓ -for [34] 2013 1goal An exhaustive search analysis (ES) presented IMO for MODGP problem under load conditions. scenarios) and OPF, DGP of REG and integration within DNW. The DGP problem location, size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), in which A attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a single objective ✓ MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, [39] 2014 2 1 that aims at DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with dominant solutions for wind-based dominant units sizing, for wind-based siting, and maintenance DG units sizing, schedules. and maintenance schedules. MOO planning framework based on genetic algorithm IInn (NSGA-II) along with MCS (for operation R Y MO PTC A B-C D Con bu on om MODGP band P ng --[28] 2009 4 1 An problem for multiple micro-turbines (DG) placement sizes solved by hybrid PSO and SFL ✓ S mMO yoptimization w hw hεGA m o mu on GA on nsorting d mprogramming hod dreliability) oisproposed m hod poptimal n(six d ok m nd huncertain baofset nbased v o MCS-embedded is presented to solve chance constrained framework considering future uncertainties 5 [37] 2013 3 The methodology based Pareto frontier differential evolution (PFDE) algorithm is for MODGP (sizing and location). [38] 2014 NSGA ong hcomparison m xon nlearning-based pp oDG hsolutions onon-dominated vnd MODGP ob on ddoub ng ud usiting, d un ned nd n galgorithms m nloads/DGs. [32] 2012 1 MO probabilistic (mathematical programming) framework has proposed for various DER planning DG types, size, and location) DGP [33] 2 The multi-criteria planning model aims to achieve contrasting objectives (cost and for DGP. GA further finds ofng non---[40] 2014 1 MO augmented constrained proposed for DGP planning short term basis for future DNWs with ANM functionalities. proposed algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 En[30] [41] Tab2011 GA is also used for of results advocated large DNW systems even with suboptimal solutions. [36] 2013 1 on✓ Quasi-oppositional teaching optimization (QOTLBO) a ovariant of TLBO as MOO The MO pm obab ma hema a the ppfor og amm ng amewo k ha p opo oproposed va ou DER pregarding ann x DG ype ze and o a on ha an NSGA-II is-NSGA used to obtain compromise (Pareto frontier) solution for aoptimal local distribution company (LDC). which isframework further solved by GA toand ensure optimal DG (wind) planning. e 5 The 2323on bu o MODGP p1hybrid ann ng me hod and eAn amethod ed wo k PTC Componen non pmethodology, ann e(with hn MODGP MO ba ed DG emen to find time varying voltage magnitude and loss sensitivity factor at each node. FDM has utilized for best compromise solution among set of --- ba [34] 2013 1 ✓ - fuzzy [35] 2013 2-- edfunction, An MO planning has MO developed, planning namely framework improved has developed, multi-objective namely improved search multi-objective (IMOHS), which harmony can evaluate search the scenarios) and addresses DGP problem of REG integration within MODGP osolutions mu b d GA nd on delectricity mselect hod pPareto opo dd ong vsolutions. hDNW. bque omp om d ou oon ohn u onp o(IMOHS), DM framework, followed by decision-making tool to the most preferred Pareto optimal solution satisfying objectives. [42] 2014 ✓ MCS mb dd GA p nnon m hodo ogy pnwith opo dstorage ofor ov has uo harmony ynd o(NSGA-II) omong DG nwith wcompeting h da oinDGP. uowhich onwhich can evaluate the DGP. A GA-PSO is proposed to solve MODGP problem regarding DG location GA) and capacity (with PSO) respectively. [31] 2012 1 [35] ✓ The DGP problem for location, size, and type (REG and PEV) is defined MO mixed integer nonlinear programming (MINLP), EA mp oy on-on o v MO on dfor ng bo h m v y ng g n on nd d m bemploying hvariable vh oeove m ng vgon mp 2 that aims at DISCOs contribution in the competitive market. Moreover modified augmented ε-constrained method along with ✓ GA w hlocation go pdOPF, og mm mng hodo ogy nd oweighted uand on w hmp oobjective un n(OPF) MO nd n(for dominant for wind-based DG units sizing, siting, and maintenance schedules. The MOO planning framework based on non-dominated sorting genetic algorithm IIke along MCS uncertain operation An exhaustive search analysis (ES) has presented IMO MODGP problem under load conditions. ✓ [29]2014 2014 2010 2 5 1 1 2 A fuzzy embedded GA is ng employed to solve fuzzy single function fuzzy set theory for MODGP. optimal and sizing of DGP solving multi-objective optimal power flow problem for radial DNW. 39 [39] a m a D SCO on bu on n he ompe ve e e y ma Mo mod fied augmen ed ε on a ned me hod a ong w h FDM ha ✓ [38] 2014 2 1 MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, ✓ The multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) for DGP. GA further finds a set of non[40] 2014 2 MO augmented ε constrained method proposed for DGP planning on short term basis for future DNWs with ANM functionalities. [41] [28] 2009 4 An MO optimization problem An for MO multiple optimization micro-turbines problem (DG) for multiple placement micro-turbines anddistribution sizes is (DG) solved placement by hybrid and PSO sizes and isSFL solved algorithms by hybrid based PSO and SFL algorithms based Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, a variant of TLBO proposed as MOO regarding D SCO [37] 2013 3 1 The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). an NSGA-II is used to obtain compromise (Pareto frontier) solution for a local company (LDC). ✓ NSGA ong w h m x m n pp o h o v MODGP p ob m on d ng u u o d un n nd k m n g m n A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a single objective DGP [33] 2012 2 1 FDM has utilized for best compromise solution among the set of Pareto optimal solutions. [34] 2013 ✓ Th p opo d-NSGA go nd ng m noconstrained oimproved w nd bprogramming d DGMoreover dharmony omp om ou mong on ngloads/DGs. ob vwith [35] 2 1 [36] An MO planning framework has developed, namely multi-objective search which can evaluate the DGP. scenarios) and OPF, addresses DGP problem of REG and storage integration within DNW. R Ao MODGP B--C Con bu on omarket. MODGP dng P nn ng GA is also used comparison of results and for large DNW even suboptimal solutions. ✓ -has - o hm [36] 2013 3on PTC 2013 3D [42] TabY 2014 ✓ --- ba --- eddominant MCS-embedded GA isand presented to solve chance framework considering future uncertainties of 2011 5 1 MOPSO algorithm been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DGDG owner profit) in o ufor zed be omp om eadvocated uelectricity on among he esystems obg ng Pa ohn opwith ma o(IMOHS), u onon [32] 2012 ✓ [39] 2014 2 1 that aims at DISCOs contribution in the competitive modified augmented ε-constrained along e 5 TheMO bu p1m ann ng me hod eframework, aDGP ed wo kfollowed PTC Componen n ppreferred ann eto que MODGP MO ba ed emen to find time varying voltage magnitude and loss sensitivity factor at each node. solutions for wind-based DG units sizing, siting, and maintenance schedules. The MOO planning framework based on non-dominated genetic algorithm II along with (for uncertain operation En[30] framework, followed by fuzzy decision-making tool by to fuzzy select decision-making the most tool Pareto select optimal the most solution preferred satisfying Pareto competing optimal solution objectives. satisfying competing objectives. optimal location sizing of solving multi-objective optimal power flow problem for radial DNW. S w hpd h m o on GA nd on nsorting d m hod doub d o(NSGA-II) m n dMCS o types, hmethod bp a n owhich v [43] 2014 4,3 1 on✓ The DGP problem for size, and type (REG and PEV) is defined as MO mixed integer programming (MINLP), in function, which is further solved by GA to ensure optimal DG planning. MO probabilistic (mathematical programming) has proposed for various DER planning (six size, location) EA mp oconstrained o vlocation, MO on dfor ng bo h m v y ng n on d m nd bemploying hvmong vhod oof m von ou hn mp ✓ GA w hyoptimization go og m hodo ogy nd oweighted u on w h(wind) oon un n(OPF) MO nnd -----[40] 2014 2 1 MO augmented εand method proposed for DGP planning short term basis for future DNWs with ANM functionalities. [41] An xh u vmu hmm nis ng ydmu ES h p nfuzzy dproblem hmthe MO og MODGP pnd ob m und bnonlinear opng dnd ond on MO problem for multiple micro-turbines placement and sizes solved by hybrid PSO and SFL algorithms based ✓ ✓ Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, TLBO proposed as MOO regarding MODGP ooy on b on GA nd on nw dframework hod pPareto opo dd o o vsolutions. haisvariant bGA) omp om dDG oPSO) o uMODGP. on oand DM [29] 2010 5 2 fuzzy embedded GA employed to solve single objective function fuzzy set theory for addition to finding generated electricity prices in a(DG) competitive electrical market. A hybrid GA-PSO isoptimal proposed to solve MODGP regarding DG location (with and capacity (with respectively. [31] 3 [38] 2014 2 FDM has utilized for best compromise solution among set of optimal The multi-criteria planning model aims to a achieve contrasting objectives (cost and reliability) for DGP. GA further finds ad set of non--- 3--[35] 2013 2 1 An MO planning framework has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. scenarios) and OPF, addresses DGP problem ofbased REG and storage integration within DNW. [36] 3 1 1 ✓ [42]2014 2012 2014 2 ✓ [37] 2013 1 [37] ✓ 2013 The 1 methodology based on Pareto The methodology frontier differential on evolution Pareto frontier (PFDE) differential algorithm evolution is proposed (PFDE) for optimal algorithm MODGP is proposed (sizing for and optimal location). and location). MOPSO algorithm has been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DG owner profit) 40 R MO augmen ed ε on ned me hod p opo ed o DGP p ann ng on ho e m ba o u u e DNW w h ANM (sizing un ona e MCS mb dd d n GA p nn ng m hodo ogy p opo d o mp ov h u y o o h DG n g on w h o o u on MODGP an NSGA-II is used to obtain compromise (Pareto frontier) solution for a local distribution company (LDC). [39] 2014 that aims at DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with DGP MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, [34] 2013 2 1 ✓ Th p opo d MOPSO NSGA gofuzzy hm ng nwhere o bu w bprogramming d DGpreferred g d om o for ufuture on mong on ngloads/DGs. obobjectives. vin The MOO framework based on non-dominated sorting genetic algorithm IIomp along with MCS (for uncertain operation Y MO PTC A B-C D Con onprogramming oproposed dng P nn ngh GA o planning u followed d o omp onmethod o nd upresented, ndmmulti-objective dvo d ond g MODGP DNW yb m v(OPF) n(NSGA-II) w ubop m osatisfying uuncertainties on framework, by decision-making tool to select the most Pareto optimal solution competing optimal location and sizing of DGP for solving optimal power flow problem radial DNW. D SCO ✓ MCS-embedded GA is presented to solve chance constrained framework considering of [30] 2011 5 1 [43] 2014 4,3 An improved with preference strategy (IMPSO-PS) is for MODGP with the aim to achieve optimal capacity and A goal attainment method (GoA) has goal transforms multiple objective functions into a single objective ✓ [33] 2012 2 1 [40] 2014 MO augmented ε constrained proposed for DGP planning on short term basis for future DNWs with ANM functionalities. dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. [41] An MO optimization problem multiple micro-turbines placement and sizes is solved by hybrid PSO and SFL algorithms Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, aPEV) variant of TLBO proposed as MOO regarding addition to finding optimal generated electricity prices in a(DG) competitive electrical market. The DGP problem for location, size, DGP and problem (REG for location, and PEV) size, is defined and type as (REG MO and integer is defined as MO programming mixed integer (MINLP), nonlinear inbased which programming (MINLP), in which [44] 2014 --- ba -- edNSGA ong w hhas m xon m n pp ohwo hsolution ovon vnd MODGP pstorage ob m on dann ng umixed uvarious o dgene un n nd k NSGA m n gMODGP mp naaoand 1 MO probabilistic (mathematical programming) framework has proposed for DER planning (six DG types, size, and location) FDM has utilized for best compromise among the set Pareto optimal solutions. e 5 The 52323on bu on✓ o MODGP p1m ann ng me hod and eThe afor ed ktype n p ng eo hn que MODGP MO ba ed DG emen to find time varying voltage magnitude and loss sensitivity factor at each GA w h✓ go pOPF, ogd mm ng m hodo ogy oComponen u on w hof oob d un n MO nd on nnd scenarios) and DGP problem of REG and integration within DNW. An xh ualgorithm vmu haddresses n ydp ES p nPTC d w h MO onon MODGP pnode. ob m und vmong bnonlinear oupgo d ond on En[32] oMCS o un [36] 1 [38] [42] Tab2012 ✓ MODGP o on b on GA nd n d m hod p opo d o o v h b omp om d o u on DM u y mb dd GA mp oy d o o u y w gh d ng v un on mp oy ng y h y o [37] 2013 The methodology based Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing location). ✓ MOPSO been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DG owner profit) in The MOO ann ng amewo k ba ed on dom na ed ng a hm ong w h e a n ope a on S y w h h m o mu on GA nd on n d m hod doub d o m hod n d o h b n v A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 1 [38] 2014 2014 2 locations. function, which is further solved by GA to ensure optimal DG (wind) planning. The MOO framework based on sorting genetic algorithm II (NSGA-II) along with MCS (for uncertain operation ---2013 2 4,3 2 1 An MO planning framework has developed, namely multi-objective harmony search (IMOHS), which can evaluate the DGP. framework, followed by fuzzy decision-making tool to select most preferred Pareto optimal solution satisfying competing objectives. optimal location sizing of DGP solving multi-objective optimal power flow (OPF) problem for radial DNW. ✓ [43] 41 [35] an NSGA-II is used to obtain an compromise NSGA-II isnon-dominated (Pareto used to frontier) obtain compromise solution for (Pareto a (cost local distribution frontier) solution company for (LDC). local distribution company (LDC). An improved with preference strategy (IMPSO-PS) isgthe proposed for MODGP with the aim to achieve optimal capacity and [39] 2 that aims atplanning DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along EA mp oy d o oomp vlocation, on dfor bo m vimproved y ng on nd dm m bk hn v od m vmong ou hn mp owith [40] MO augmented εand method proposed for DGP planning on short term basis for future DNWs with ANM The multi-criteria planning model to achieve contrasting objectives reliability) for DGP. GA finds set of non✓ Th p opo d MOPSO NSGA hm nd ng m nwhere o w nd b DG g d ng om oon uang on ng [41]2014 2014 2014 2 1 1 1 GA o u d o om nd d o gnog y vnd n w m oh on Quasi-oppositional teaching learning-based (QOTLBO) methodology, aomp of TLBO proposed as MOO regarding Dgoal SCO GA non d oaims oupresented, vng h nh on n d p mm ng m wo on ng u u unDNW nfunctionalities. dobjective DG addition to finding optimal generated electricity prices in aob competitive market. ena opgoMO and OPF add eoptimization edvo p em odDNW REG and oP age eg aand won n The DGP problem for size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), ✓ MCS mb dd d nconstrained GA nn ng hodo ogy pDGP opo d oprogramming mp ov helectrical uand y oisvariant oh hubop DG n gufurther on w h algorithms d oob oin uvwhich on [44] 2014 2 1 A attainment method (GoA) has goal transforms multiple objective functions into aaosingle MODGP problem is solved by two stage heuristic iterative method including clustering (outer) EA (inner) optimizations respectively, ✓ ------[34] 2013 2 1 scenarios) and OPF, addresses DGP problem of REG and storage integration within DNW. R Y MO PTC A B---C D Con bu on oof MODGP b d nn ng An MO optimization problem for multiple micro-turbines (DG) placement and sizes solved by hybrid PSO and SFL based [42] ✓ [37] 2013 3 1 The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). MOPSO algorithm has been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DG owner profit) in [38] 2014 2 locations. ✓ [32] 2012 5 1 The MO probabilistic (mathematical The MO programming) probabilistic (mathematical framework has programming) proposed for framework various DER has planning proposed (six for DG various types, DER size, planning and location) (six DG types, size, and location) FDM has utilized for best compromise solution among the set Pareto optimal solutions. DGP [33] 2 dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. The MOO planning framework based on non-dominated sorting genetic algorithm II (NSGA-II) along with MCS (for uncertain operation An xhytime u vvarying hvoltage n ymp ES p nPTC d MO o (wind) p ob me und vun bng op En[36] - edfunction, 2013 3on bu 1 on✓ optimal location sizing of for solving multi-objective optimal flow (OPF) problem ✓ ✓ uann mb dd dfurther GA oy dohwo ok v(Pareto uMODGP yw wh gh dfactor ng vng un mp oy upfor yradial ynd [43] Tab2014 2014 S yng wong h mu on on np(QOTLBO) d m hod don hod nnd okoPSO hgMODGP b regarding an NSGA-II ishPSO used to obtain frontier) solution for adoub local distribution company (LDC). A hyb d GA pm opo dcompromise oed o voGA MODGP p ob m gis dMODGP ng DG o on w GA nd yond wdhDNW. hon pm yn v o An improved MOPSO with preference (IMPSO-PS) proposed for MODGP with the aim tod achieve optimal capacity and ✓ which isand solved by GA tooh ensure planning. NSGA w hoptimal m x om pp hstrategy vnd ob m on dpower umarket. dm n m n naov emen e 5 The 4,3 o MODGP pm me hod and enmagnitude aDGP Componen nob p ann ng hn MODGP MO edocompeting DG to find and loss sensitivity at each --- ba [41] 2 1 [39] Quasi-oppositional teaching learning-based optimization methodology, aohovariant of TLBO proposed as MOO framework, followed by decision-making tool to select the most preferred optimal solution satisfying objectives. MODGP oand mu blocation, d generated on GA nd on noptimal dDGP m hod pthe opo ddnode. oterm oumixed vPareto bque omp om daba ova o umethod onp DM addition to finding electricity prices in aDG electrical Qua oppo ona ng ea n ng ba ed op m za on QOTLBO me hodo ogy an oaugmented TLBO pDGP. opo ed a MOO ega d ng op ma The DGP problem for size, and type (REG and PEV) is defined as MO integer nonlinear programming (MINLP), in which ✓ [44] -dOPF, -on - fuzzy 2 2014 2--that 1 MO aims at DISCOs contribution that aims in the at competitive DISCOs contribution electricity in market. competitive Moreover electricity modified market. augmented Moreover ε-constrained modified along ε-constrained with method along with [40] MO augmented εo constrained method proposed for planning on short basis for DNWs with ANM ✓ GA w h✓ mm ng m ogy oopo uimproved on wcompetitive omm un nhDNW. MO ncan scenarios) addresses DGP problem of REG and storage integration within [35] 2013 3 2 1 An planning has developed, namely multi-objective harmony (IMOHS), which the ogo u omp on upea nd dvo dgo ong gn DNW y m vnd n w hfuture m ovgon uun on [42]2014 2014 3 1 1 MOPSO algorithm has been used satisfying objectives (minimize DISCO cost and maximize profit) dnog GA p n dohodo oaims oMODGP vng h nhdproblem on n d og mm ng m wo on dfor ng und u nnfunctionalities. dob DG 42 [39] --[38] 2014 2 1 ✓ MCS mb dd GA nn ng m hodo ogy piterative opgphmp ov hng u y obkmu omong hpoubop DG nGA on wDG hevaluate oand olocation) uoin onv locations. The MO probabilistic (mathematical programming) framework has proposed for DER planning (six DG types, size, A go nmp m GoA hfor nnd wh og n ovarious m ob vradial un on oowner ng EA mp oy d onframework osolved vhod MO on dfor bo m vv yd onpower ndop dclustering mma hsearch v m ng ou hn mp MODGP problem is by two stage method including (outer) and EA (inner) optimizations respectively, The multi-criteria planning model toheuristic achieve contrasting objectives (cost and reliability) DGP. further finds aod set of nonoptimal location and sizing of DGP solving multi-objective optimal flow (OPF) problem for DNW. ✓ D SCO [37] 2013 3 1 The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). [43] 2014 4,3 o a on and z ng o DGP o o ng mu ob e ve powe flow OPF p ob em o ad a DNW an NSGA-II is used to obtain compromise (Pareto frontier) solution for a local distribution company (LDC). An improved MOPSO with preference strategy (IMPSO-PS) is proposed for MODGP with the aim to achieve optimal capacity and FDM has utilized for best compromise FDM has utilized solution for among best compromise the set of Pareto solution optimal among solutions. the set of Pareto optimal solutions. ✓ [33] 2012 2 1 [34] 2013 The MOO planning framework based on non-dominated sorting genetic algorithm II (NSGA-II) along with MCS (for uncertain operation Th p opo dPSO NSGA go hm nd m np om wp(QOTLBO) bong don DG gand dud ng om onp on mong on ng ob vwith Quasi-oppositional teaching learning-based optimization methodology, aomp variant ofby proposed as MOO regarding R Y MO PTC Ao MODGP BC D Con bu on MODGP b P nn ng An MO optimization problem for multiple (DG) placement sizes is solved hybrid PSO and SFL algorithms based ✓ ✓ uann ytime mb dd dhuoptimal GA mp oy dowo oelectricity okong vmicro-turbines uuMODGP yop welectricity gh d ob vng un on mp oy ngTLBO uu y hhba oPSO y o MODGP addition to finding generated prices in and competitive electrical market. A hyb d GA p opo d o o v MODGP ob m g d ng DG o on w h GA nd y w p v y [44] ✓ [39] 2014 2 1 that aims at DISCOs contribution in the competitive market. Moreover modified augmented ε-constrained method along un on wh h h o v d by GA n DG w nd p nn NSGA ong w m x m n pp h v ob m d u o d un nd k m n g m n Tab e 5 The on bu on ba ed p ng me hod and e a ed PTC Componen n p ann ng e hn que MODGP MO ed DG p a emen to find varying voltage magnitude and loss sensitivity factor at each node. DGP ✓ dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. [41] 2014 2 1 [42] [36] 2013 3 MOPSO algorithm has been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DG profit) in The DGP problem for location, size, and type (REG and PEV) isoptimal defined as MO mixed integer programming (MINLP), in S yu wv h- OPF, hε constrained m oy mu on GA nd on nstorage d m hod doub d(OPF) oplanning m hod n dDG o types, nd howner b and n which v locations. MO probabilistic (mathematical programming) has for various DER planning (six size, location) ✓ ✓ --- 2--on [40] 2014 2 1 [40] ✓ MO 1mimproved augmented MO method augmented proposed εof constrained for DGP planning method proposed on short term for DGP basis for future on DNWs short term with basis ANM functionalities. future DNWs with ANM functionalities. scenarios) DGP problem REG and integration within optimal location sizing of DGP solving multi-objective power flow problem for An xh haddresses n ES h p n d w h MO ogthe MODGP p ob m und bnonlinear op duachieve ond on framework, followed decision-making tool to select most preferred Pareto optimal solution satisfying competing objectives. MODGP oand mu b dgo on GA nd on ned dframework m pPareto opo dd o o vsolutions. hDNW. omp om dDNW. oon ofor uocapacity on o of DM [43] 4,3 ✓ 2014 MCS mb dd d GA pv- fuzzy n dhm opgogy vstrategy h nhd on nhod dp phof og ng wo on ng u un nn d MOPSO aMO ha been u othe pproposed ob em am ng ob eodfor ve m n m ze SCO ong and [38] 2014 2 1 An MOPSO with preference (IMPSO-PS) is proposed for MODGP with the aim to optimal and A go nm nand m GoA hoaims nund wh go og ng omarket. m mu p ob vradial un ob FDM has for best compromise solution among set optimal EA mp oy d oframework o vhod on dfor ng bo m v y n on nd dn m nd bkb hvou v m ng von ou hn mp omax MODGP putilized ob m oby d by wo h vMODGP m hod nmm ud ng u ng nd EA nn opnD m on pDG vv ym ze DG owne p ofi n ✓ GA w hplanning go mm m hodo oopo uimproved on w omm un ny MO nd The multi-criteria planning model to achieve contrasting objectives (cost DGP. GA further finds aod set non[35] 2013 2 1 An MO has developed, namely multi-objective harmony search (IMOHS), which can the ✓ addition to finding optimal generated electricity prices in ang competitive 43 R 4 3 MO 1 PTC an NSGA-II isp used to obtain compromise (Pareto solution for ahelectrical local distribution company (LDC). [44]2014 Y MCS mb dd d og GA p ng nn ngbased m hodo ogy pfrontier) dsorting obased ov uand yvreliability) o(NSGA-II) omong h DG n onIIas w hevaluate oalong o DGP. uwith on MCS (for uncertain operation [39] 2014 2 1 that aims atplanning DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with ----[34] 2013 2 1 The MOO framework The planning non-dominated framework genetic on non-dominated algorithm IIann sorting genetic along algorithm with MCS (for (NSGA-II) uncertain operation Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, variant of TLBO proposed MOO regarding A B---C D Con bu on omp MODGP bterm da P ng GA owh u d on omp on oMOO uon nd dvo d o g DNW y m n w hfuture ubop m ogwma uMODGP on MOPSO algorithm has been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DG owner profit) in ✓ D SCO [37] 2013 3 1 [41] ✓ The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal (sizing and location). add on o find ng op ma gene a ed e e y p e n ompe ve e e a ke A hyb d GA PSO p opo d o o v MODGP p ob m g d ng DG o on w h GA nd p y h PSO p v y locations. un on h u h o v d by GA o n u op m DG w nd nn ng ✓ [40] 2014 2 1 MO augmented ε constrained method proposed for DGP planning on short basis for DNWs with ANM functionalities. o nd m v y ng vo g m gn ud nd o n v y o h nod DGP ✓ dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. [41] 2014 2 1 Th p opo d MOPSO NSGA go hm mMODGP n framework o the wp nd don DGdoptimal gand dusizes ng omp om onto u achieve on(six mong on oband v ✓ [42] 2014 3 1 An MO optimization problem for multiple micro-turbines (DG) placement is solved by hybrid PSO and SFL algorithms based [43] 4,3 improved with preference strategy (IMPSO-PS) isbof proposed for MODGP theintegration aim optimal capacity The MO probabilistic (mathematical programming) proposed for planning types, size, location) ✓ FDM has utilized for best compromise solution among set Pareto NSGA ong wand hoptimal m xm n DGP ppnd hand ong v ob mhas ng uvarious owith d DER un nd km n ugMODGP mngnoand scenarios) and OPF, addresses scenarios) DGP problem OPF, of REG addresses DGP problem integration of within storage within DNW. optimal sizing of solving optimal power flow (OPF) problem DNW. [36] 2013 3 1 ✓ addition to finding generated electricity prices in ad competitive electrical market. MODGP ow mu on blocation, on GA nd nw duand mgh hod p opo dd oREG o vsolutions. hDNW. b omp om on DM A uw yhylocation d GA oy dofor opg oo von u y d ng ob vng un on ng ufor yvradial hDG y o ✓ [44] 2014 2 The DGP problem for size, and type (REG and PEV) isog defined as MO mixed integer programming in S m hdd hogframework m odydmp mu GA nd on nstorage m hod doub dm ong m hod n dd ooh h b onngothe n which vvv y A go nm n hod GoA hon n dmulti-objective wh go p ng n oand mu p ob un on nnd o(MINLP), ob MODGP pmb ob m oby by wo hnon-dominated und v m hod nvMPSO ud u ou nd EA nn op m ✓ The MOO framework based on sorting genetic algorithm (NSGA-II) along with MCS (for uncertain operation GA pAn mm ng m hodo ogy on w omm nII mong MO on nnd Th mu pm nn mod m hnamely v ng oun nd bmp yoy o DGP GA u non [35] 2013 2 1 MO planning has developed, improved multi-objective harmony (IMOHS), can evaluate An xhmp ugo vDISCOs h nvng ES h p dtool MO ogthe MODGP pnd ob mPareto und bnonlinear opng dnd ond on framework, followed fuzzy decision-making to most preferred optimal solution satisfying competing objectives. ✓ 44 [38] mp oved MOPSO w hn emwoeon enselect ahob egy opo o MODGP w h he aregarding mmp o ap DGP. h eve op asma apa y and o a on ------locations. 2 ✓ [39] that aims atplanning contribution in competitive electricity market. Moreover modified augmented ε-constrained method along [40] MO augmented εGA method proposed for DGP planning on short term basis for future DNWs with ANM functionalities. EA oy d oconstrained o vobtain MO on dthe ng bo hp vh yelearning-based ng n on dPS m nd b(QOTLBO) hvsearch v oed m vwhich ou hn owith ✓ [41]2014 2014 2014 2 2 1 1 1 Quasi-oppositional teaching Quasi-oppositional learning-based optimization teaching (QOTLBO) methodology, optimization aopvariant ofaim TLBO methodology, proposed aas variant MOO of TLBO proposed MOO regarding MOPSO algorithm has been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DG owner profit) in D SCO ✓ An improved MOPSO with preference strategy (IMPSO-PS) is proposed for MODGP with the to achieve optimal capacity and p n d o o v h n on n d p og mm ng m wo k on d ng u u un n o d DG an NSGA-II is used to compromise (Pareto frontier) solution for a local distribution company (LDC). MCS mb dd d n GA nn ng m hodo ogy p opo d o mp ov h u y o h DG n g on w h d o o u un on wh h u h o v d by GA o n u op m DG w nd p nn ng ✓ o nd m v y ng vo g m gn ud nd o n v y o nod scenarios) and OPF, addresses DGP problem of REG and storage integration within DNW. dom n n o u on o w nd b d DG un ng ng nd m n n n h du Th phas opo NSGA go hm ng mdvo n o evolution w ond bofgplacement d DG optimal dsizes ng omp omubop o optimal u ono mong on ngand ob location). v on ✓ - omp - on [42] 2014 1 MO An optimization problem multiple micro-turbines (DG) solved by hybrid PSO and(for SFL algorithms based GA outilized u d-d ofor onfor o nd uon nd d DNW y gand m vis nisproposed w h m uMODGP on [43] 4,3 - 3[37] 2013 3 1 [42] ✓ 2014 The methodology based Pareto frontier differential (PFDE) algorithm for (sizing [44] 2014 2 FDM best compromise solution among the set Pareto solutions. DGP The MOO planning framework based non-dominated sorting genetic algorithm II (NSGA-II) along with MCS uncertain operation optimal location and sizing of optimal DGP for location solving and multi-objective sizing of DGP optimal for solving power multi-objective flow (OPF) problem optimal for power radial flow DNW. (OPF) problem for radial DNW. [36] 2013 3 1 addition to finding optimal generated electricity prices in ad competitive electrical market. ✓ locations. S m y w h h m o mu on GA nd on n m hod doub d o m hod p n d o nd h b n v A hyb d GA PSO p opo d o o v MODGP p ob m g d ng DG o on w h GA nd p y w h PSO p v y The MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) NSGA ong w h m x m n pp o h o v MODGP p ob m on d ng u u o d un n nd k m n g m n MODGP p ob m o v d by wo g h u v m hod n ud ng u ng ou nd EA nn op m on p v y ✓ [41] 2014 2 1 Th mu p nn ng mod m o h v on ng ob v o nd b y o DGP GA u h nd o non Quasi-oppositional teaching optimization (QOTLBO) a variant as oMOO regarding nn ngε dconstrained klearning-based hmethod op m yand ov d mu ob v un mony hngTLBO MOHS whon nomp v(MINLP), u h inDGP An xhy p umb vfollowed hm nwo ymp ESddhvproblem potype nnof d hmp MO MODGP pvMO ob mPbasis und v future bnonlinear ouu don ond framework, by fuzzy decision-making tool to select most p as dhmixed ointeger op mof oDNWs yohng ng ob v ✓ ✓ A uMO dd GA oy oogy vd nd u ywow gh d othe obmethodology, on mp oy yproposed y MODGP The DGP problem for location, size, and (REG PEV) isng defined programming which ------[40] 2014 2 1 MO augmented proposed for DGP planning on short term for with ANM functionalities. scenarios) and addresses DGP REG integration within DNW. ✓ GA w h algorithm go pOPF, mm ng mng hodo uand on w omm d un nsatisfying MO nd onh optimal [42] 2014 3 1 MOPSO has been MOPSO used for algorithm MODGP problem has been satisfying MODGP objectives problem (minimize DISCO objectives cost and maximize (minimize DG DISCO owner cost profit) and maximize in DG owner profit) in An improved with preference (IMPSO-PS) isghfor proposed for MODGP the aim to achieve and [38] 2014 2 1 ✓ 2014 MCS dd p nn hodo ogy pun opo dstorage op mp ov hng u yvnd owith omong h DG ndε-constrained on hmethod d ng ohm o buowith ond [39] that aims DISCOs contribution in the competitive electricity market. modified augmented along A nog m GoA h p strategy n og nflow m pdoubop ob vradial onnw n ocapacity ob v o1go nd m v ng vo gnd m gn nd vgo y nod EA mp oy on oomp MO on ng bo m v nd yused ng on nd dm modu bkmu m ng vgun ou mp dom nmb nop onm u-MOPSO o w bDGP d DG ng ng nfor nMoreover ny h optimal location and sizing of for power (OPF) problem for DNW. An MO m on pvhod m od mu pun monhdmulti-objective owh DG pgenetic m nd ohon vv hyb PSO nd SFL go GA oatplanning u dd o on om nd dvo o gonm DNW n wopo h m uun on ✓ - 4,3-GA -ob [43] 2014 4,3 Th m hodo b dp on P ond on d nfrontier) vo u PFDE go hm p dby oalong m MODGP nd oDG ✓ [44] 2 1 [43] ✓ MCS mb dd dyon GA n odud ouon vsolving h onb nsorting dnd poptimal og mm ng m wo d ng uon uo nhn ong oob doperation an NSGA-II is used to obtain compromise (Pareto solution analgorithm local distribution company (LDC). The MOO framework based non-dominated II (NSGA-II) with MCS (for uncertain Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, aomp variant of TLBO proposed as MOO regarding Th p opo d ogy NSGA go hm ng muMODGP nprices o mthe w bon don DG g ng d ng om onop uelectrical mong on ng v on ✓ addition to finding optimal generated addition toelectricity finding optimal in generated and competitive electricity electrical prices market. in abd competitive market. locations. FDM has utilized for best compromise solution among set of Pareto optimal solutions. NSGA ong w h m x m n pp o h v p ob m d ng u u o un nd k m n g m n un on wh h u h o v d by GA o n op DG w nd p nn DGP ✓ [41] 2014 2 1 Th mu p nn ng mod m o h v on ng ob v o nd y o DGP GA u h nd o non [42] 3 An MO palgorithm nn ng wou mp k dfor vvon op dk n mpoo yframework ov dhng mu ob v(minimize mony hng wh hng nng vowner uNLP hylocation) DGP MOPSO has been used MODGP problem satisfying objectives DISCO cost maximize DG profit) in m wo kand oPSO ow dm by yh DGP d m ng omp mo p ovMO dhmPvarious oGA opDER mnd oMOHS unand on yoPSO omp ng ob v h ✓ ✓ A u y mb dd d GA oy d o o v u y w gh d ob un on mp oy u y h y o MODGP Th DGP p ob m o o on nd yp REG nd PEV d n d x d n g non p og mm M n wh hyb d GA p opo d o o MODGP ob m g d ng DG on w h p y w h p v The MO probabilistic (mathematical programming) has proposed for planning (six DG types, size, and ✓ scenarios) OPF, addresses problem of REG and storage integration within DNW. optimal location of DGP solving optimal power (OPF) radial DNW. An xhmp u v MOPSO hsizing n MO y by ES hforgproposed p nh multi-objective d wwith h MO og proposed MODGP pnd ob mbasis und visproblem ofor dfor ond on [43] 2014 4,3 improved with preference An improved strategy (IMPSO-PS) is strategy for (IMPSO-PS) MODGP with proposed aim to achieve MODGP optimal with theonaim and top achieve optimal capacity and ✓ --- 2-[40] 2014 2 11 [44] ✓ MO augmented εand constrained for DGP on short term DNWs with ANM EA od o on bo m v y n on dflow m nd bkfor vthe oddbby m ng von ou hncapacity mp MODGP p oy o dnd h MOPSO und vpreference m hod u ng EA nn opnSFL m von dom n mb nop oob u-ogy on o w bmethod d ng ng nd m nnmm nud nong h du GA w h✓ go pd og mm ng m ogy oon u nin on w hog omm un nbu mong MO nd An MO m on pvhod m owo mu pun m oowh b DG p PFDE m ndng nd ohou vfuture hyb d PSO nd go b owith dy addition to finding generated electricity prices ang competitive electrical Th m hodo b dpv on P oin on d nu vo u on go hm opo ond op m MODGP nd oDG -optimal -ob ✓ [44] MCS dd dm GA n dhodo odDG oMODGP vng h ndproblem on nplanning dop pu og wo on ng uLDC u un nnfunctionalities. ong ohm dob n1go NSGA o ob n omp om P omethodology, on omp ny [39] 2014 2 1 that aims DISCOs contribution the electricity market. Moreover modified ε-constrained method along A m GoA n go ng ndm omarket. mu pdubop ob vproposed oowner ng ✓ Quasi-oppositional teaching learning-based optimization (QOTLBO) anpvariant ofcost TLBO MOO regarding ✓ 2014 MOPSO algorithm has been used for satisfying objectives (minimize and maximize DG profit) inv GA oatplanning unm duynong omp on oh uponcompetitive nd dvo d o go on DNW y m vm wDISCO h augmented m oun uMCS ononas locations. locations. ✓ DGP o hyb nd vdoPSO vo gk m ud omREG n v y nod The MOO based non-dominated sorting genetic algorithm IId (NSGA-II) along with (for uncertain ✓ [42] 2014 3 1 An MO pm nn ng m wou d vGA op dkmm n m yo ov mu ob v hm mony haim MOHS wh hng vp u hyoperation DGP Th p opo NSGA go hm nd ng npoo w nd bdk dp DG g dd ng omp om u ng on mong onnng ng ob vwh [43] 4,3 m wo kob ow dframework by yh ddgn on mond ng omp hof mo ph Pvw oGA opDER mnd oo u on yPSO omp ng ob vonh improved MOPSO with preference strategy (IMPSO-PS) is proposed for MODGP with the to achieve optimal capacity and Th DGP p ob m o o on nd yp nd PEV d n d MO x n g non n p og mm M NLP n ✓ A d GA p opo d o o v MODGP ob m g d ng DG o on h p y w h v MO p b m h m p og ng m wo h opo d o ou p nn x DG yp nd o FDM has utilized for best compromise solution among the set Pareto optimal solutions. un on wh h u h o v by n u op m DG w nd p nn ng ✓ [41] 2014 2 1 optimal and sizingmp of DGP solving optimal power flow (OPF) problem addition to finding generated electricity prices in aw competitive electrical market. ✓ A uw y location mb dd dpoptimal GA oyhodo dfor oogy o v nd u multi-objective y ow gh d ng ob vounwithin un on ngMO ufornd yradial hDNW. oh yn ondMODGPo non [44] 2014 2 1 ✓ ✓ scenarios) and OPF, DGP problem REG and storage integration DNW. GA pu ng momp on hob omm nbu mong on Th mu nn ng mod m hof v ng nd bmp yoy o DGP GA u MO m on pon m o m uon bu ngo DG pon nd nd ovou vfuture hyb PSO nd go b wvon d An xhmhop ugo v haddresses ES p oh d w h MO MODGP pong ob mobasis und onop dd ond on ✓ Th m hodo b dnvob on P ohmu on d n vo u PFDE go pmu opo dbby ond m MODGP ng ndong locations. NSGA d omm ob n Pn owh on opoyon u ovmshort on omp ny LDC ✓ hngo Dogy SCO on nhom hMODGP omp v m k Mo ov mod d ugm on d hod A nm nog m hod GoA p strategy n do og ng ndhm m pd ob vd un on nm ocapacity nghm ob [40] 2014 2 1 MO augmented εGA constrained method for on term DNWs with ANM MODGP p dd ob o dybu by wo vnplanning m hod nmm ud u ng EA nn opnSFL m poDG ✓ MOPSO algorithm has been used for satisfying objectives (minimize DISCO cost and DG owner profit) invhy An improved with preference (IMPSO-PS) proposed for MODGP with the aim achieve MCS mb dm p n d nd ogproposed vun huoptimization nproblem onDGP dis ogd ng m wo kfor on d ngto umaximize unoptimal nfunctionalities. oon oob dand ✓ dom nh np oob on obby w und do DG ng ng nd m np nop ny g hm ✓ Quasi-oppositional teaching learning-based (QOTLBO) methodology, variant of TLBO proposed as regarding Th p opo douMOPSO NSGA go hm ng m nmong o w nd bpk DG dd ng omp om uu on onMOO [43] 2014 4,3 mnd wo kplanning ow dframework yb ddgn on mond kmm ng oo om hog mo p Povdu o opDER mubop oonn un on ymm ng omp ng ob vonh GA owh uob d ong omp on oom uon nd dvo d o DNW m vuIIad nou w h m o mong u on [44] 2 1 Th DGP p m o o on nd yp REG nd PEV d n d MO x n g non p og ng Mng NLP MO b m h m p og ng m wo h opo d o p ng x DG yp nd on vwh FDM u d o omp o u on h P o m on un on h u h o v by GA n u op DG w nd p nn ng o m v y vo g m ud o n v y o h nod The MOO based non-dominated sorting genetic algorithm (NSGA-II) along with MCS (for uncertain operation ✓ [42] 2014 3 1 addition to finding optimal generated electricity prices in a competitive electrical market. locations. ✓ A hyb d GA PSO p opo d o o v MODGP p ob m g d ng DG o on w h GA nd p y w h PSO p v y ✓ [41] 2014 2 1 optimal location and of DGP solving optimal power flow (OPF) nn ng wo kby v op nudmulti-objective m yon mp mu vovun h mony MOHS wh hyn o ndun v ng u hong xh u vDdd hm ybu ES p d ww h MO od MODGP povng ob mbmod und on dd ond on ✓ ✓ An Th m hodo ogy bon dnv on P odhod on n vo u on PFDE go hm p bmp opo dbho op m MODGP A uMO yp GA mp oy dhfor ooopo vdd gh d ob on oy ngond ufor yradial h oh MODGP u d osizing ob n omp om P oyo onn ung on dm bu on omp ny LDC hn NSGA m SCO on n hogp omp v m kon Mo dproblem ugm on m hod wvonhy ✓ MO ugm nob d n dhm DGP pov ng hofor ovou u DNW w hDNW. ANM on nd MODGP pmb md oon d wo h un v storage m hod novob ud u ng nn op m on poDGP scenarios) and OPF, addresses DGP problem REG and integration within DNW. Th mu p nn ng mod m hof v(IMPSO-PS) on ng o MODGP yu u o and non ✓ An improved MOPSO with preference strategy isy ob proposed the aim toEA achieve optimal A go nm m GoA hfor n dproblem wh go p ogp objectives mm ng n nd o vmnwith mu obDGP vdGA un on SFL nnd ocapacity nghm ob v ✓ MOPSO algorithm been used satisfying (minimize and DG owner profit) inon GA ou uob dbdyn o omp on oom upon dvo d o ygo m w h m u onnd An op m on phod obwo m ogn mudpoMODGP pond m ong u m nng nd oon vg pdubop by hyb PSO go d [44] 2014 2 1 ✓ Th DGP pp m oteaching o on nd yp REG PEV dPgDNW nmm dnh MO xwo nDISCO non n po mm M NLP nb wh MCS mb dd GA p n v h on n(QOTLBO) dnd pk og k d ng u umaximize nngun oregarding ond MO p homp m og mm wo h opo d om DER pcost nn ng xun DG yp oDG FDM d ohas und on mong h ogom o op m ovdu uadou on o MO nd vunn ng vo m ondom nnbndnm vd y Th MOO ng m b non oDG ng np hm NSGA ong w hog MCS oMOO ndop onh dom nh nm oob on obhm w gnd bkdd do ud DG un ng ng nnod hm Quasi-oppositional learning-based optimization methodology, variant of TLBO proposed as ✓ [43] 2014 4,3 1 locations. un on wh h u o v by GA o n u op m DG w nd p nn ng ✓ [42] 2014 3 1 addition to optimal generated electricity prices in pa d competitive electrical m wo k nfinding oPSO owd dopnn by unbu ym d on m k hng oo ond hob mo pMo d Pbmarket. oGA op m o u d on yoh ng omp ng ob v h ✓ ✓ u y mb dd GA mp oy d o o v u y w gh ng ob v un on mp oy ng u y h y o MODGP n NSGA u d ob n omp om P o on o u on o o d bu on omp ny LDC A hyb d GA opo d o o v MODGP p ob m g d ng DG o on w h nd p y w h PSO p v y h m D SCO on on n h omp v y m k ov mod d ugm n on n d m hod ong w ✓ MO ugm d on d hod p opo d o DGP nn ng on ho m o u u DNW w h ANM un on n o nd OPF dd DGP p ob m o REG o g n g on w h n DNW Th mu p ng mod m o v on ng v o nd b y o DGP GA u nd o non ✓ optimal location andm sizing ofhDGP solving power (OPF) problem for radial An MO p nn ng wod kby d for v gop du n multi-objective m y mpv ov doptimal mu n ob v flow h mony h nd MOHS whDNW. h m n v u h p DGP MODGP p ob m o v wo h m hod ud ng u ng ou EA nn op on v y ✓ An improved MOPSO with strategy (IMPSO-PS) is for MODGP to u achieve optimal and m hodo ogy bm d on Ppreference on non vo on PFDE hm opo ong op m MODGP oDG on ✓ MCS mb dd don pw nm ong v h ndom dp pk og ng m wo k onnthe ddaim u opo nn un ong gond Th MO pupob bon m homp poMODGP og mm m h opo d ou pcost nn ng DG FDM dGA ohas o non udd on mong h nwo oproposed o op m up on A n GoA pbon n dproblem wh go og mm noogy oovdu m mu ob vp un onyp ocapacity ob von ✓ Th MOO ng m wo bom d d ouDG ng npmm go hm NSGA ong w hxun MCS oMOO ndd op dom nhoppo nop onm obphod nd bkod DG ng ng nd m nng ngo h Qugo h nohd op m on QOTLBO m vwith TLBO d ng MOPSO algorithm been used for satisfying (minimize DISCO and DG owner profit) An MO munn on obng m mu pun m ong unb n pPgobjectives nhodo nd o DER v pdoby hyb dmaximize PSO nd SFL go ng hm b ind ✓ [44] 2014 2 1 ✓ [24] [25] [26]

--208 D -[23] 2007 2 1 MCS embedded in GA planning methodology isContribution proposed to improve accuracy for stochastic DG integration with tradeoff solution. 16 of 44 Ref. Year MO Aof2017, B-- 10,-C of MODGP-based Planning ✓ [21] 2005 4 1Energies [26] Energies 20082017, 2 208 1 PTC - DGP - with GA with goal programming finds aathe solution with associated uncertainties among MO andofconstraints. [22] 3 Similarly, the same formulation (GA and εmethodology constrained method), double tradeoff method is presented to find the alternative. 5.10, The contribution MODGP-based planning methods and related works. (PTC: Components inDGs planning techniques; MODGP: based DGbest placement). EA isproposed employed to solve IMO considering both time-varying generation and demand behavior aiming atMO various technical impacts ✓ The NSGA algorithm finds arrangements for wind-based regarding compromise solution among contrasting objectives. [27] Table 2008 3 1

Energies 2017, 10, 208

17 of 44

Energies 2017, 10, 208

Table 5. Cont.

16 of 44

Energies 2017,5.10, 208contribution of MODGP-based planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). 16 of 44 Table The

Ref. Energies Year2017, MO208 PTC D Contribution of MODGP-Based Planning 16 of 44 Ref. Year PTC A Aof MODGP-based B BC D C planning Contribution of MODGP-based Planning Table 5.10, TheMO contribution methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). MODGP formulation on second GA and order ε constrained method is proposed to solve the best compromise (tradeoff)issolution for DM Mixedbased integer cone programming (MISOCP) formulation of DGP problem proposed aiming at optimally allocating dispersed ✓ - -- [21]2014 2005 6 4 1 1 [45] Ref. Year TheMO PTC A B C D (DISCO). Contribution of MODGP-based Planning energyand storage systems (DSSs) into ADN followed by AHP for DM among multiple Table related works. (PTC: Components in planning techniques; MODGP: MO objectives. based DG placement). Energies 2017,5.10, 208contribution of MODGP-based planning methods 16 of 44 [22]

2005

3

1

✓

-

-

-

MODGP based on GA and ε constrained methodmethod), is proposed to solve the best compromise (tradeoff) solution for DM Similarly,formulation with the same formulation (GA and ε constrained a double tradeoff method is presented to find the best alternative.

[24] [22]

2008 2005

3 3

1 1

✓ ✓

-

-

-

NSGA-II along with max-min MODGP problem considering future load uncertainties and risk management MODGP based onapproach GA and εsolves constrained method is proposed to solve the best compromise (tradeoff) solution for DM Similarly,formulation with the same formulation (GA and ε constrained method), a double tradeoff method is presented to find the best alternative.

✓ - [21]2014 2005 4 4 1 1 [46] Ref. A supervised BB-BC method evaluates minimization of IMOPlanning for finding optimal location and capacity of one/more voltage-controlled DG(s). Year MO PTC A B- -C D of MODGP-based (DISCO). ✓ [23] 2007 2 1 MCS embedded in GA planning methodology isContribution proposed to improve the accuracy for stochastic DG integration with tradeoff solution. Table 5.10, The contribution of MODGP-basedMODGP planning methods and related works. (PTC: Components in planning techniques; MODGP: MO to based DGbest placement). Energies 2017, 208 16 of for 44 finding the optimal location formulation based on GA and ε constrained method is proposed to solve the best compromise (tradeoff) solution for DM index Chaotic artificial bee colony (CABC) solves MODGP problem regarding the multi-objective performance [22] 2005 3 1 Similarly, with the same formulation (GA and ε constrained method), a double tradeoff method is presented find the alternative. ✓ [24] 2008 5 ✓ - -- - NSGA-II- along with max-min approach solves MODGP problem considering future load uncertainties and risk management 2005 4 1 1 [47] [21] 2014/15 (DISCO). DG units and capacities. ✓ [23] 2007 2 1 MCS inreal GApower planning methodology istime-varying proposed togeneration improve the accuracy forbehavior stochastic DG integration tradeoff solution. EA is embedded employedofto solve IMO considering boththeir and demand aiming at various with technical impacts of Ref. Year MO PTC A B C D Contribution of MODGP-based Planning ✓ [25] 2017, 200810, 2083 6 1 Energies 16 of 44 ✓ ---[22] 2005 1 Similarly, with the formulation (GA and εMODGP constrained method), a doublefuture tradeoff method is presented to find the alternative. [24] Table 2008 NSGA-II along withsame max-min approach problem considering load risk management 5. The contribution of MODGP-based planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DGbest placement). DGP MODGP formulation based on GA and εsolves constrained method is find proposed solve the bestuncertainties compromise (tradeoff) solution forunder DM various operational constraints and An MOPSO based algorithm is presented to best to trade-off solution throughand multiple objectives ✓ -- - --- - MCS 4 [48] [21] - goal programming [23] 2007 2 embedded in solve GA planning methodology istime-varying togeneration improve the accuracy forbehavior stochastic DGand integration with tradeoff solution. EA is employed to IMO considering both and demand aiming atconstraints. variousan technical impacts of ✓ GA with methodology finds aproposed solution with associated uncertainties among MO [26]2015 2005 2008 3 2 1 1 1 (DISCO). demand side management (DSM) is justified in DG planning. FDM is used to identify optimal non-dominated ✓ [25] 2017, 200810, 208 6 1 Energies 16 of 44 solution. Ref. Year PTC Aof MODGP-based B-C D Contribution of MODGP-based Planning ✓ --[24] 2008 3 1 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management 5. TheMO contribution planning methods and related works. (PTC: Components inDGs planning techniques; MODGP: MO based DGbest placement). DGP ✓ The proposed NSGA algorithm finds arrangements for wind-based regarding compromise solution among contrasting objectives. [27] Table 2008 3 1 ✓ -[22] 2005 3 1 Similarly, with the same formulation (GA and ε constrained method), a double tradeoff method is presented to find the alternative. MODGP formulation based on GA and ε constrained method is proposed to solve the best compromise (tradeoff) solution for DM Sequential quadratic programming (SQP) for a set of Pareto front solutions for to find optimal DG sizing and EA is employed to solve IMOmethodology considering both generation and demand behavior aiming atconstraints. various technical impacts of placing. WSM generates a set of ✓ GA with findsistime-varying aproposed solution associated uncertainties among MO [26]2015 2005 2008 2 2 1 1 1 An exhaustive search (ES) has presented with IMOwith MODGP under variable load conditions. ✓ ------ - MCS [21] 4 [49] Ref. - goal programming [23] 2007 2 embedded in GAanalysis planning methodology tofor improve theproblem accuracy for stochastic DGand integration with tradeoff solution. ✓ [25] 2008 6 1 ✓ Year MO PTC Aof MODGP-based B-- -C D Contribution of MODGP-based Planning [28] Table 2009 4 1 (DISCO). acceptable trade-off solutions among contrasting objectives. Fuzzy making (FDM) method provides compromise solutions. 5.10, The contribution planning methods and related works. (PTC: Components inDGs planning techniques; MODGP: MO based DG placement). DGP ✓ The proposed NSGA algorithm finds arrangements for wind-based regarding compromise solution among contrasting objectives. [27] 2017, 2008 3 10, 208 1 Energies Energies 208 2017, 16 of best 44 16 of 44 GA is also used for comparison of results and advocated for large DNW systems even withdecision suboptimal solutions. ✓ with-embedded goal programming methodology finds a with solution associated uncertainties among MO and constraints. [26] 2008 2 1 An exhaustive search analysis (ES)model has presented IMOwith for MODGP problem under variable load conditions. ✓ ✓ ---- - GA [21] 4 [29]2015 2005 2010 3 5 1 1 2 A fuzzy GA isIMO employed to solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. [50] Ref. Atomathematical ofboth MODGP problem solved by the proposed improved NSGA (INSGA-II). The best EA is employed solve considering time-varying generation and demand behavior aiming at various technical impacts of solution obtained with FDM. ✓ Year MO PTC A B-- -C D of MODGP-based Planning [28] 2009 4 1 (DISCO). ✓ ----[23] 2007 2 1 MCS embedded in GA planningof methodology isContribution proposed improve the accuracy forwith stochastic DG integration with tradeoff solution. [25] 2008 6 ✓ -The proposed NSGA algorithm finds arrangements for wind-based regarding compromise solution among contrasting objectives. [27] 3 GA is also planning used for and advocated fortoComponents large DNW systems even suboptimal solutions. ✓ - MODGP-based MCS-embedded GAcomparison is presented toresults solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] Table 2011 5 5. The1 1 contribution 5. Table The contribution of MODGP-based of planning methods methods and related and works. related (PTC: works. Components (PTC: inDGs planning in planning techniques; techniques; MODGP: MODGP: MO based MO DG based placement). DG placement). DGP MODGP formulation based on GA and ε constrained method is proposed to solve the best compromise (tradeoff) solution for DM Multi-objective Shuffled Bat algorithm (MOShBAT), a variant of bat algorithm (BA) proposed for the planning of DGs considering various [22] 2005 3 1 Similarly, with the same formulation (GA and ε constrained method), a double tradeoff method is presented to find the best alternative. ✓ [24] 2008 alongsearch withGA max-min approach solves MODGP problem considering future load uncertainties and risk management An exhaustive analysis (ES) has presented with IMO for MODGP problem under variable load conditions. ✓ ✓ -- - --- - NSGA-II 4 [29] 2010 5 fuzzy is employed to solve fuzzy weighted single objective function employing fuzzy set theory forrespectively. MODGP. [51] [21] 2016 2005 3 1 121 -embedded A hybrid GA-PSO is proposed to solve MODGP regarding DG location (with GA) and MO capacity (with PSO) [31] 2012 3 ✓ GA with goal programming methodology finds aproblem solution with associated uncertainties among constraints. [26] 2008 2 [28] 2009 4 (DISCO). objectives and showed better results than NSGA-II. - BD - CMCS [23] 2007 2 1MO ✓ in solve GA planning istime-varying proposed tolarge improve the accuracy forbehavior stochastic DGand integration tradeoff solution. EA employed to IMO considering both generation demand aiming atsolutions. various with technical impacts of Ref. Year Ref. MO Year PTC APTCB- AC Dembedded Contribution Contribution of MODGP-based ofand MODGP-based Planning Planning GA is is also usedcontrasting for comparison ofmethodology results and advocated for DNW systems even with suboptimal

✓ ---MCS-embedded GA isalgorithm presented to chancewhere constrained programming framework considering future uncertainties loads/DGs. 5 1 A goal attainment method (GoA) hassolve presented, goal programming transforms multiple objective functions into a of single objective ✓ 6 1 The proposed finds arrangements for wind-based DGs regarding compromise solution among contrasting 3 ✓ 3 1 Similarly, withNSGA the same formulation (GA and εMODGP constrained method), a double tradeoff method is presented to find the best objectives. alternative. NSGA-II along with max-min problem considering future load uncertainties and risk management 5 DGP ✓ ✓ MODGP formulation MODGP formulation based onapproach based GA and εsolves GA constrained and εoptimal constrained method is method proposed islocation proposed to solve the to solve best compromise the best compromise (tradeoff) (tradeoff) solution for solution DM algorithm for DM 5 2 fuzzy embedded ismodel employed toonsolve fuzzy weighted single objective function employing fuzzy set theory for MODGP. A for mix penetration (PV and wind-based REGs) and capacitors planning, where MOPSO produces set of potential A hybrid GA-PSO isGA proposed to solve MODGP problem regarding DG (with GA) and capacity (with PSO) respectively. 3 1 function, which isMOO further solved by GA to ensure DG (wind) planning. An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. ✓ ✓ 2005 4 1 4 1 2 2 MCS embedded in GA planning methodology is proposed to improve the accuracy for stochastic DG integration with tradeoff solution. EA is employed to solve IMO considering both time-varying generation and demand behavior at various technical impacts of the best non-dominated solution. ✓ ---GA with goal programming methodology finds apossible solution with associated uncertainties amongaiming MO and constraints. 2 1 4 (DISCO). (DISCO). ✓ MCS-embedded GA is presented to solve chance constrained programming framework considering future uncertainties loads/DGs. [30] 2011 5 1 solutions after considering all trade-offs among distinct objectives and followed by FDM torespectively, identify A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a of single objective ✓ [25] 2008 6 1 MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations GA is also used for comparison of results and advocated for large DNW systems even with suboptimal solutions. ✓ [24] 2008 3 1 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management [32] 2012 5 DGP [33] 2 ✓ 1 -- ✓-The proposed NSGA algorithm arrangements for wind-based DGs regarding compromise solution among contrasting [27] 2008 3 13 ✓ ---function, - time [22] 2005 [22] 2005 3 1 Similarly, Similarly, with the same with the formulation same formulation (GA ε(GA constrained and ε constrained method), method), a double tradeoff a(with double method tradeoff is method presented is presented toPSO) find the tobest findobjectives. alternative. the best alternative. A hybrid GA-PSO is proposed tofinds solve MODGP problem regarding DG location GA) and capacity (with respectively. [31] 2012 which is further solved by GA toand ensure optimal DG (wind) planning. to find varying voltage magnitude and loss sensitivity factor atobjective each node. ✓ [29] 2010 5 2 A fuzzy embedded GA isIMO employed to solve fuzzy weighted single function behavior employing fuzzy theorytechnical for MODGP. EA is employed to solve considering both time-varying generation and demand aiming atset various impacts of Energies 10, 2082007 16 of 44 ✓ 1 --- ✓--GA with goal methodology finds solution with associated uncertainties among MO and constraints. [26] 2017, 2008 2 12 ✓ An exhaustive search analysis (ES) haspresented, presented with IMO MODGP problem under variable load conditions. ✓ ----MCS -embedded [23] 2007 [23] 2 1 MCSprogramming embedded inplanning GA planning in GA planning methodology methodology isawhere proposed is proposed tofor improve to the improve accuracy the for accuracy stochastic for stochastic DG integration DG optimizations integration with with tradeoff solution. A goal attainment method (GoA) has goal programming transforms multiple objective functions intotradeoff aasingle objective [25] 2008 6 MODGP problem is solved by two stage iterative method including clustering (outer) EA (inner) respectively, The multi-criteria model aims toheuristic achieve contrasting objectives (cost and reliability) forand DGP. GA further finds set ofsolution. non✓ [28] 2009 4 1 MCS-embedded GA is presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 ✓ [32] 2012 1 DGP [33] 2 [34] 2013 ✓ 1 -- ✓-The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 3 13 ✓ GA is used for comparison of results advocated for large DNW systems even with suboptimal solutions. ---function, - also [24] 2008 [24] 2008 3 1 NSGA-II along NSGA-II with along max-min with max-min approach approach solves MODGP solves problem MODGP considering problem considering future load future uncertainties load uncertainties and risk management and risk management which is further solved by GA toand ensure optimal DG (wind) planning. to hybrid find time varying voltage magnitude and loss sensitivity factor at each node. dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. Energies 2017, 10, 208 16 of 44 quality). Table 6. The contribution of MOVPQ-based planning methods and related works. (MOVPQ: MOP based on VAR compensation and power ✓ A GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 1 ✓ --GA withembedded goal programming methodology finds a with solution with associated uncertainties among MO and constraints. [26] 2008 2 1 An exhaustive search analysis (ES) has presented IMO for MODGP problem under variable load conditions. ✓of MODGP-based ✓ [29]Table 2010 5 2 A fuzzy GA isIMO employed to solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. EA ismulti-criteria employed EA ismethods employed toframework toby solve considering IMO considering both time-varying both time-varying generation generation and demand and behavior demand aiming behavior at aiming various atcan technical various technical impacts of impacts of 5. The contribution planning and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). MODGP problem issolve solved two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, The planning model aims to achieve contrasting objectives (cost and reliability) for DGP. GA further finds set of non[35] 2013 2 Angoal MO planning has developed, namely improved multi-objective harmony search (IMOHS), which evaluate the DGP. [28] 2009 4 1 attainment method (GoA) has presented, where goal programming transforms multiple objective functions into aasingle objective ✓ 1 -- ✓----A - also [25] 2008 [25] 6 16 ✓ [33] 2012 2 [34] 2017, 2013 The proposed NSGA algorithm finds arrangements for wind-based DGs systems regarding compromise solution among contrasting objectives. [27] 3 Energies 10, 2082008 16 of 44 GA is used for comparison of results and advocated for large DNW even with suboptimal solutions. MCS-embedded GA is presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 ✓ [32] 2012 5 1 DGP DGP to varyingfor voltage magnitude andmicro-turbines loss sensitivity factor at each node. dominant solutions wind-based DG units sizing, siting,(DG) and maintenance schedules. Anfind MOtime optimization problem for multiple placement and sizes is solved by hybrid PSO andMO SFL algorithms function, which is further solved by GA to ensure optimal DG (wind) planning. Table 5. The contribution planning methods related works. (PTC: Components in planning techniques; MODGP: based DGbased placement). Anfuzzy exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. [36] 2013 3 1 --B ✓ [29] 2010 5 2 embedded is and employed tomethodology solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. Ref. Year MO Aof1MODGP-based Contribution of MODGP-based Planning hybrid GA-PSO isGA proposed to solve MODGP problem DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 1 Ref. Year PTC APTC B C-- C ✓ Contribution of MOVPQ-Based Planning ✓ ---- DD-A - planning GA with goal GA with programming goal programming methodology finds atool solution finds aregarding with solution associated with associated uncertainties uncertainties among among and constraints. MO and constraints. [26] 2008 [26] 2008 2 12 ✓ The multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) forMO DGP. GA further finds a set of non[35] MO 2013 An MO framework has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. [28] 2009 4 framework, followed by fuzzy decision-making to select the most preferred Pareto optimal solution satisfying competing objectives. Energies 2017, 10, 208 16 of 44 MODGP problem iscomparison solved by two stage heuristic iterative including clustering (outer) and EA (inner) optimizations respectively, ✓ [34] 2013 2 1 GA is also used for of results and advocated formethod large DNW systems even with suboptimal solutions. --- ✓--MCS-embedded GA isalgorithm presented to solve chance constrained programming framework considering future uncertainties loads/DGs. [30] 2011 5 1 A goal method (GoA) has presented, where goal programming transforms multiple objective functions into a of single objective MODGP formulation based on GA and εsizing, constrained method is proposed toregarding solve the best compromise (tradeoff) solution for DM [33]Table 2012 2 solutions for wind-based DG units siting, and maintenance schedules. ✓ ----- D -dominant - attainment The proposed The proposed NSGA NSGA algorithm finds arrangements finds arrangements for wind-based for wind-based DGs regarding DGs compromise compromise solution among solution contrasting among contrasting objectives. objectives. [27] 2008 [27] 11 3 ✓ 5. The 2008 contribution planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). An MO optimization problem for multiple micro-turbines (DG) placement and sizes isproposed solved by hybrid PSO andsupport SFL algorithms based ✓ [37] 2013 1 The methodology based on Pareto frontier differential evolution (PFDE) algorithm is for optimal MODGP (sizing and location). Ref. Aof1MODGP-based Contribution of MODGP-based Planning ✓ ---B [32] 2012 1 ✓ - ---- C - ✓ [21] 2005 4533533 to find time varying voltage magnitude and loss sensitivity factor atobjective each node. [53] 2005 -PTC Tabu Search (TS) based approach computes non-dominated solutions to provide decision for capacitor location problem. - - - function, [36] 2Year 2013 1 MO 1 ✓ [29] 2010 2 fuzzy embedded GA is employed to solve fuzzy weighted single function employing fuzzy set theory for MODGP. A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 1 which isframework further solved by GApresented to ensure optimal DG (wind) planning. (DISCO). ✓ [35] Energies 2013 2 10, 208 1 MO planning has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the An exhaustive An exhaustive search analysis search (ES) analysis has (ES) presented with IMO with for MODGP IMO problem MODGP under problem variable under load variable conditions. load conditions. framework, followed fuzzy decision-making tool to select the most preferred Pareto optimal solution satisfying competing objectives. The DGP problem forby location, size, andand type (REG and PEV) isobjectives defined as MOand mixed integer nonlinear programming (MINLP), inDGP. which Energies 2017, 2017, 2082009 MODGP formulation based on GA εhas constrained method isfor proposed to solve the best compromise (tradeoff) solution for DM 16 of 44 16 of 44 The multi-criteria planning model aims to achieve contrasting (cost reliability) for DGP. GA further finds set of non✓--B - attainment [28] 2009 [28] 4 1 4 ✓ 5.10, The contribution planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). GA issame presented to solve chance constrained programming framework considering future uncertainties loads/DGs. [30] 2011 5 ✓ [38] 2014 2 1 A goal method (GoA) has presented, where goal programming transforms multiple objective functions into aaof single objective Ref. Year MO PTC Aof1MODGP-based Contribution oflarge MODGP-based Planning MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, ✓ - ------ D - - -MCS-embedded ✓ [21] 2005 4 - - ----- C [34]Table 2013 2 11 [54] 2007 2 1 GA based fuzzy multi-objective approach is presented for optimal capacitor placement. [22] 3 Similarly, with the formulation (GA and ε constrained method), a double tradeoff method is presented to find the best alternative. GA is also GA used is also for comparison used for comparison of results of and results advocated and advocated for large DNW for systems DNW even systems with even suboptimal with suboptimal solutions. solutions. An MO optimization problem for multiple micro-turbines (DG) placement and sizes is solved by hybrid PSO and SFL algorithms based ✓ [37] 2013 3 1 The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). an NSGA-II is used to obtain compromise (Pareto frontier) solution for a local distribution company (LDC). ✓ [32] 2012 5 1 [33] 2 dominant solutions wind-based DG units sizing, siting,regarding and(wind) maintenance schedules. (DISCO). ✓ [36] 2013 3 1 ✓ -- ✓--A hybrid GA-PSO isfor proposed toon solve MODGP problem DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 1 which isembedded further solved by GA to ensure optimal DG planning. to find time varying voltage magnitude and loss sensitivity factor each MODGP formulation based GA and constrained method is proposed to solve the best compromise (tradeoff) solution for DM framework, followed by decision-making tool select the most Pareto optimal solution satisfying competing objectives. ✓ ✓ -B -C -D -function, [29] 2017, 2010 [29] 2010 5 21 5 ✓ A fuzzy embedded A fuzzy GA isfuzzy employed GA is employed to solve fuzzy to solve weighted fuzzy weighted single single function objective employing function employing fuzzy set fuzzy theory set for theory MODGP. for MODGP. ✓ [23] 2007 2 MCS embedded in GA planning methodology isto proposed toatobjective improve the accuracy fornonlinear stochastic DG integration with tradeoff solution. DGP problem for location, size, and typeε (REG and PEV) is multi-objective defined asnode. MO mixed integer programming (MINLP), inDGP. which MO probabilistic (mathematical programming) has proposed for various DER planning (six DG types, size, and location) Energies 208 16 of 44 problem. Ref. Aof2MODGP-based Contribution of preferred MODGP-based Planning ✓ -- -- -of -- -- MODGP-based -- -- - The [21] 2005 4 [35] 2013 1 An MO planning framework has developed, namely improved harmony search (IMOHS), which can evaluate ✓ [55] 2008 4Year 1 MO -PTC Aattainment two-stage immune algorithm (IA) embedding compromise programming solves the multi-objective capacitor [22] 3225. The Similarly, with the same formulation (GA and εframework constrained method), aclustering double tradeoff method is presented to the best alternative. ✓ [38]Table 2014 11contribution 5.10, Table The contribution planning planning methods methods and related and related works. (PTC: works. Components (PTC: Components in planning in planning techniques; techniques; MODGP: MODGP: MO based MO DG based placement). DGplacement placement). A goal method (GoA) has presented, where goal programming transforms multiple objective functions into aafind single objective MODGP problem is solved by two stage heuristic iterative method including (outer) and EA (inner) optimizations respectively, The multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) forfor DGP. GA further finds set of non(DISCO). ✓ [37] 2013 3 1 The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed optimal MODGP (sizing and location). ✓ MCS-embedded MCS-embedded GA is presented GA is presented to solve chance to solve constrained chance constrained programming programming framework framework considering considering future uncertainties future uncertainties of loads/DGs. ofwith loads/DGs. [30] 2011 [30] 2011 5 1 an NSGA-II is used to obtain compromise (Pareto frontier) solution for a local distribution company (LDC). ✓ [24] 2008 3 1 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management ✓ [39] 2014 2 that aims at DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along [32] 2012 5 1 [33] ✓ - - - - - - function, [34] 2013 2 11 MODGP formulation based on GA and ε constrained method is proposed to solve the best compromise (tradeoff) solution for DM An MO optimization problem for multiple micro-turbines (DG) placement and sizes is solved by hybrid PSO and SFL algorithms based ✓ [23] 2007 2 MCS embedded in GA planning methodology is proposed to improve the accuracy for stochastic DG integration with tradeoff solution. which is further solved by GA to ensure optimal DG (wind) planning. to find time varying voltage magnitude and loss sensitivity factor at each node. solutions for wind-based DG units sizing, siting, and maintenance schedules. Energies 2017, 10,2017, 208 16 of 44 16 of 44 ✓ - - --- - - -dominant ✓ 1 -- - ✓ [36] Energies 2013 3 10, 208 11 3 ✓ [21] 2005 4 [56] 2009 2 1 Hybrid SA–PSO and IBVT based approach addresses; fixed and switched type capacitor placement, besides finding [22] 3 Similarly, with the same formulation (GA and ε constrained method), a double tradeoff method is presented to find the best alternative. The DGP problem for location, size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), in which A hybrid A GA-PSO hybrid is GA-PSO proposed is proposed to solve MODGP to solve problem MODGP regarding problem regarding DG location DG (with location GA) (with and capacity GA) and (with capacity PSO) (with respectively. PSO) respectively. [31] 2012 [31] 2012 3 1 MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) FDM has utilized for best compromise solution among the set of Pareto optimal solutions. is employed toby solve IMO considering both time-varying generation and demand behavior aiming at various technical impacts ofa valuable trade-off solution. Ref. Ref. Year Year MO PTC MO A PTC CEA D Contribution Contribution of preferred MODGP-based of MODGP-based Planning Planning framework, followed fuzzy decision-making tool to select the most Pareto optimal solution satisfying competing objectives. (DISCO). ✓ -B [38] 2014 1 ✓ - A- C - B- D - MODGP [24] 2008 3 1 NSGA-II along with max-min approach solves MODGP problem future load uncertainties and risk management Table 5. The contribution of MODGP-based planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). problem is solved by two stageto heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, The multi-criteria planning model aims achieve contrasting objectives (cost and reliability) for DGP. GA further finds set of non[25] 6222 ✓ [35] 2013 1 An MO planning framework has developed, namely improved multi-objective harmony (IMOHS), which can evaluate the an NSGA-II is used to obtain compromise (Pareto frontier) solution forprogramming aconsidering local distribution company (LDC). ✓ [39] 2014 1 that aims at goal DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with objective ✓ ---- - ---- - ---- - A goal attainment A method (GoA) method has (GoA) presented, has presented, where goal where programming goal transforms transforms multiple objective multiple functions objective functions into aasingle into objective aDGP. single [23] 2007 2 MCS embedded in GA planning methodology is proposed toon improve the accuracy for stochastic DG integration with tradeoff solution. [40] MO augmented εattainment constrained method proposed for DGP planning short term basis forsearch future DNWs with ANM functionalities. [33] 2012 ✓ [34] 2013 2 1 DGP ✓ [37] 2013 3 11 The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). MODGP MODGP formulation formulation based on based GA and on GA ε constrained and ε constrained method is method proposed is proposed to solve the to solve best compromise the best compromise (tradeoff) (tradeoff) solution solution for DM for DM ✓ Strength Pareto evolutionary algorithm (SPEA2) incremented by fuzzy logic solves MO combinatorial optimization problem [22] 2005 3 1 Similarly, with the same formulation (GA and ε constrained method), a double tradeoff method is presented to find the best alternative. ✓ ✓ [32] 2012 [32] 2012 5 1 5 1 to find varyingto voltage magnitude andmicro-turbines loss sensitivity factor atgeneration each node. dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. is employed solve IMO considering both time-varying and demand behavior aiming at various technical impacts of16 An MO optimization problem for multiple (DG) placement and sizes is solved by hybrid PSO and SFL algorithms based Energies 2017, 208 of 44 to address Volt/Var ✓ -- MODGP-based - -- - function, - EA - time [21] 2005 4 4 ✓ 1 - -- ✓ The MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) [57] 2010 3[21] 22005 -PTC FDM has utilized for best compromise solution among the set of Pareto optimal solutions. function, which isprogramming further which is solved further by solved GA tonon-dominated by ensure GA to optimal ensure DG optimal (wind) DG (wind) planning. The MOO planning framework based on sorting genetic algorithm IIplanning (NSGA-II) along with MCS (for uncertain operation [24] 335. The NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management Table 5.10, Table The contribution contribution planning planning methods methods and related and related works. (PTC: works. Components (PTC: Components in planning in techniques; techniques; MODGP: MODGP: MO based MO DG based placement). DG placement). ✓ [25] 2008 6 1 -C [36] 2013 11 GA with goal methodology finds aproposed solution with associated uncertainties among MO and constraints. [26] 2 Ref. Year MO Aof MODGP-based B- of Contribution ofplanning. MODGP-based Planning DGP problem for location, size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), inDGP. which (DISCO). (DISCO). control for distribution systems regarding switched and automatic voltage regulators placement. ✓ - --- D - The [23] 2007 222 1 MCS embedded incontribution GA planning methodology iselectricity to capacitors improve the accuracy forfor stochastic DG integration with tradeoff solution. [41] 2014 The multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) DGP. GA further finds a(AVR) set of non✓ [35] 2013 1 An MO planning framework has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP framework, followed by fuzzy decision-making tool to select the most preferred Pareto optimal solution satisfying competing objectives. [38] 2014 1 ✓ [39] that aims at DISCOs in the competitive market. Moreover modified augmented ε-constrained method along with [40] MO augmented ε constrained method for DGP planning on short term basis for future DNWs with ANM functionalities. MODGP problem MODGP is problem solved by isIMO solved two stage byproposed two heuristic stage iterative heuristic method iterative including method including clustering clustering (outer) and (outer) EA (inner) and EA optimizations (inner) optimizations respectively, respectively, ✓ scenarios) and OPF, addresses DGP problem ofconstrained REG and for storage integration within DNW. - - - - an [34] 2013 2 11 employed to solve considering both time-varying generation and demand behavior aiming at(tradeoff) various technical impacts ofalternative. NSGA-II iswith used to obtain compromise (Pareto frontier) solution for aconsidering local distribution company (LDC). ✓ The proposed NSGA algorithm finds arrangements DGs regarding compromise solution among contrasting objectives. [27] 2008 3 MODGP formulation based on GA and εsolves method is associated proposed to solve the best compromise solution for DM -- -- -dominant - is ✓-- C [33] 2017, 2012 [33] 2012 2 11 - EA -with [22] [22] 2005 2005 1 --B--- ✓ Similarly, Similarly, the with same the formulation same formulation (GA and (GA εMODGP and εwind-based constrained method), method), aof double atradeoff double tradeoff method is method presented is presented to find the tolocation). best find alternative. the best solutions for wind-based DG units sizing, siting, and schedules. ✓ [24] 2008 3 NSGA-II along with max-min approach problem future load uncertainties and risk management MO optimization problem for multiple micro-turbines (DG) placement and sizes solved by hybrid PSO and SFL algorithms based [25] 6 ✓ [37] 2013 1 The methodology based on Pareto frontier differential evolution (PFDE) algorithm is for optimal (sizing and ✓ GA goal programming methodology finds aconstrained solution with uncertainties among MO and constraints. [26] 2008 2 132 ✓ Energies 208 16VAR of 44 planning problem. Ref. Year MO MO A PTC C D Contribution Contribution of MODGP-based MODGP-based Planning Planning FDM has utilized for best compromise solution among the set ofmaintenance Pareto optimal solutions. [21] 2005 433 to find time to find varying time voltage varying magnitude voltage magnitude and loss sensitivity and loss sensitivity factor at each factor node. at each node. [58] 2012 2Ref. 1Year -PTC -- -- B-- D - An Elitist NSGA II merged with local search finds best nodes and compensation values ofMODGP the installed capacitors in MO The MOO planning framework based on non-dominated sorting genetic algorithm IIaisproposed (NSGA-II) along with MCS (for uncertain operation Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, variant ofplanning TLBO proposed as MOO regarding ✓ - A Table 5.10, The contribution of1 MODGP-based planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). [36] 2013 1 DGP The MO probabilistic (mathematical programming) framework has proposed for various DER (six DG types, size, and location) An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. (DISCO). [41] 2014 ✓ ✓ ✓ [42] 3 [35] 2013 2 1 An MO planning framework has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. [23] [23] 2007 2007 2 1 2 MCS embedded MCS embedded in GA planning in GA planning methodology methodology is proposed is proposed to improve to improve the accuracy the accuracy for stochastic for stochastic DG integration DG integration with tradeoff with tradeoff solution. solution. framework, followed by fuzzy decision-making tool to select the most preferred Pareto optimal solution satisfying competing objectives. EA is employed to solve IMO considering both time-varying generation and demand behavior aiming at various technical impacts The DGP problem for location, size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), in which ✓ The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 322 ✓ - -- -- -- -- -- The [40] 2014 11 MO augmented εand constrained proposed for DGP planning on short term basis for future DNWs with ANM functionalities. MODGP MODGP formulation formulation based on based GA and on GA εofconstrained and εa constrained method is method proposed isand proposed to solve the to solve best compromise theradial best compromise (tradeoff) (tradeoff) solution solution for for DM [28] 2009 4 1 scenarios) and OPF, addresses DGP problem REG and storage integration within DNW. multi-criteria The multi-criteria planning planning model aims model to achieve aims tocontrasting achieve contrasting objectives objectives (cost (cost reliability) and reliability) for DGP. GA for DGP. further GA finds further a set finds of nona set ofDM non-of optimal location sizing of method DGP for solving multi-objective optimal power flow (OPF) problem for DNW. ✓ [25] 2008 6 ✓ [38] 2014 1 GA with goal programming methodology finds solution with associated uncertainties among MO and constraints. [26] 2 [39] that aims at DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with Energies 2017, 10, 208 16 of 44 ✓ ✓- -- --- -- - -an -- DGP --is optimization [21] 2005 4 1 -- -- ✓ -MILP [34] 2[21] 2013 [34] 22005 2013 11 GA also used for comparison of solve results and advocated for large DNW systems even with suboptimal solutions. [59] 2013 -PTC model is proposed to the and VRC allocation problem and to obtain front satisfying multiple objectives. [22] 332 Similarly, with the same formulation (GA andVR εMODGP constrained method), a distribution double tradeoff method isPareto presented to find the best alternative. An MO problem for multiple micro-turbines (DG) placement and sizes solved by hybrid PSO and SFL algorithms based ✓ ✓ [37] 2013 1 The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). [24] [24] 2008 2008 14 32 ✓ NSGA-II NSGA-II along with along max-min with max-min approach approach solves solves MODGP problem problem considering future load future uncertainties load uncertainties and risk and management risk management NSGA-II is used to obtain compromise (Pareto frontier) solution for aconsidering local company (LDC). Ref. Year MO Aof1 MODGP-based B ✓-- C D dominant Contribution of MODGP-based Planning The MOO planning framework based on non-dominated sorting genetic algorithm IIais (NSGA-II) along with MCS (for uncertain operation dominant solutions solutions for wind-based for wind-based DG units DG sizing, units siting, sizing, and siting, maintenance and maintenance schedules. schedules. An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. (DISCO). (DISCO). Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, variant of TLBO proposed as MOO regarding ✓ MOPSO algorithm has been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DG owner profit) in [36] 2013 3 1 Table 5. The contribution planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). FDM has utilized for best compromise solution among the set of Pareto optimal solutions. ✓ -- -- -- -- -- ✓ ✓ The proposed NSGA algorithm finds for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 3 1 [41] 2014 2 111 [42] [28] 2009 4 [43] 2017, 2014 4,3 framework, followed by fuzzy decision-making tool to select the most preferred Pareto optimal solution satisfying competing objectives. [29] 2010 5 2 A DGP fuzzy embedded GA isIMO employed toarrangements solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. ✓ --- The [23] 2007 2 MCS embedded in GA planning methodology istime-varying proposed to improve the accuracy for stochastic DG integration with tradeoff solution. problem for location, size, and type (REG and PEV) isimproved defined as MO mixed integer nonlinear programming (MINLP), inDGP. which MO probabilistic (mathematical programming) has proposed for various DER planning (six DG types, size, and is employed EA is employed to solve to solve considering IMO considering both both generation generation and demand and demand behavior behavior aiming at aiming various at technical various technical impacts of impacts ✓ 1 -- -- ✓ GA with goal programming methodology finds aconstrained solution with uncertainties among MO and constraints. [26] 2008 2322 132 ✓ Energies 10, 208 16 of 44of to find Pareto optimal scenarios) and OPF, addresses DGP problem of REG and storage integration DNW. optimal location sizing of DGP for solving multi-objective optimal power flow (OPF) problem for radial DNW. -MO [35] 2013 [35] 2013 1 MO planning An MO planning framework framework has developed, has developed, improved namely multi-objective multi-objective harmony harmony search (IMOHS), search (IMOHS), which can which evaluate can evaluate the DGP. MODGP formulation based on GA and εnamely constrained is associated proposed to solve the best compromise (tradeoff) solution for DM GA also used for comparison of results and advocated for large DNW systems even with suboptimal solutions. ✓ addition to finding optimal generated electricity prices in a method competitive electrical market. capacitor placement problem solved with self-adaptive modified honey bee mating optimization (SAMHBMO), proposed -- EA --is [22] 2005 2005 3 Similarly, Similarly, with the with same the formulation same formulation (GA and (GA εframework and εtime-varying constrained method), method), awithin double atradeoff double tradeoff method is method presented is presented to find the tolocation) best find alternative. the best alternative. ✓ [38] 2014 1 ✓ ✓--- --- ---- --- - -An [25] [25] 2008 6 16 ✓ 1 -- -- ✓ ✓ [40] 3[22] 2014 2 11 MO augmented εand constrained method proposed for DGP planning on short term basis for future DNWs with ANM functionalities. ✓ An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. [21] 2005 4 1 ✓ [60] 2013 12008 -PTC [37] 2013 1 The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). NSGA-II isMO used to obtain compromise (Pareto frontier) solution for aconsidering local distribution company (LDC). MCS-embedded GA is presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 ✓ -- --- C -- - ✓ -- an [24] 2008 3 1 NSGA-II along with max-min approach solves MODGP problem future load uncertainties and risk management Table 5. The contribution planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). -B [39] 2014 1 that aims atDGP DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with DGP ✓ Quasi-oppositional teaching learning-based optimization (QOTLBO) afunction variant of TLBO proposed as MOO regarding The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 323 12 ✓ Ref. Year MO Aof1 MODGP-based D Contribution ofmethodology, MODGP-based Planning MOPSO algorithm has been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DG owner profit) in An MO optimization An optimization problem for problem multiple for micro-turbines multiple micro-turbines (DG) placement (DG) placement and sizes and is solved sizes by is solved hybrid by PSO hybrid and SFL PSO algorithms and SFL algorithms based based [28] 2009 4 (DISCO). improved MOPSO with preference strategy (IMPSO-PS) is proposed for MODGP with the aim to achieve optimal capacity and [29] 2010 5 2 A fuzzy embedded GA is employed to solve fuzzy weighted single objective employing fuzzy set theory for MODGP. solutions satisfying maximum objectives. Probabilistic load flow (point estimate method or PEM) address associated uncertainties. ✓ ✓ [23] [23] 2007 2007 2 1 MCS embedded MCS embedded in GA planning in GA planning methodology methodology is proposed is proposed to improve to improve the accuracy the accuracy for stochastic for stochastic DG integration DG integration with tradeoff with tradeoff solution. solution. The MOO planning framework based on non-dominated sorting genetic algorithm II (NSGA-II) along with MCS (for uncertain operation Energies En 2017, g 10,2017 208 10 2082013 16 of 44 16 o 44 [42] 2014 [43] 4,3 --The - has [36] 2013 [36] 1 3 ✓ 1 -- ✓-also used for comparison of results and advocated for large DNW systems even with suboptimal solutions. [44] 2014 DGP problem for location, size, and type (REG and PEV) is defined as MOfor mixed integer nonlinear programming (MINLP), in which MO probabilistic (mathematical programming) framework has proposed various DER planning (six DG types, size, and location) FDM utilized for best compromise solution among the of Pareto optimal solutions. A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 33222 ✓ EA is employed to solve IMO considering both time-varying generation and demand behavior aiming at various technical impacts [41] 2014 11 ✓ -- GA --is GA with GA goal with programming goal programming methodology methodology finds aconstrained solution aoptimal solution with associated with associated uncertainties uncertainties among MO among and MO constraints. and constraints. [26] [26] 2008 2008 2 12 optimal location and sizing of DGP for solving multi-objective power flow (OPF) problem for radial DNW. addition to finding optimal generated electricity prices infinds aset competitive electrical market. framework, framework, followed followed by fuzzy decision-making by fuzzy decision-making tool to select tool the to select most preferred the most preferred Pareto optimal Pareto solution optimal satisfying solution satisfying competing competing objectives. objectives. An exhaustive search analysis (ES) has presented with IMO for problem under variable load conditions. MODGP formulation based on GA and εsolves method isMODGP proposed to solve the best compromise (tradeoff) solution for DM of - --- ✓ [38] 2014 1 ✓ --- --- D --- locations. [22] 2005 3 Similarly, with the same formulation (GA and εMODGP method), awithin double tradeoff method is presented to find the best alternative. MCS-embedded GA is presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 ✓- C [24] [24] 2008 2008 3 ✓ NSGA-II NSGA-II along with along max-min with max-min approach approach solves MODGP problem problem considering considering future load future uncertainties load uncertainties and risk and management risk management Table 5. The contribution of11 MODGP-based planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). scenarios) and OPF, addresses DGP problem ofconstrained REG and storage integration DNW. ✓ [25] 6 1 Ref. Year MO PTC A B Contribution of MODGP-based Planning [28] 2009 ✓ [21] 2005 4 1 an NSGA-II is used to obtain compromise (Pareto frontier) solution for a local distribution company (LDC). ✓ [29] 2010 5 2 A fuzzy embedded GA is employed to solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. [61] 2013 3 1 Accelerated PSO (APSO) based optimization approach solves capacitor planning problem regarding sizing, location, and type. ✓ [39] 2014 2 1 that aims at DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with [40] MO augmented εNSGA constrained method proposed forarrangements DGP planning on wind-based short term basis for future DNWs with ANM functionalities. A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a single objective DGP MOPSO algorithm has been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DG owner profit) in ✓ ✓ The proposed The proposed algorithm NSGA algorithm finds arrangements finds for wind-based for DGs regarding DGs regarding compromise compromise solution solution among contrasting among contrasting objectives. objectives. [27] [27] 2008 2008 3 1 3 1 ✓ ✓ An improved MOPSO with preference strategy (IMPSO-PS) is proposed for MODGP with the aim to achieve optimal capacity and [37] 2013 [37] 2013 3 1 3 The methodology The methodology based on based Pareto on frontier Pareto differential frontier differential evolution evolution (PFDE) algorithm (PFDE) algorithm is proposed is proposed for optimal for MODGP optimal (sizing MODGP and (sizing location). and location). GA is also used for comparison of results and advocated for large DNW systems even with suboptimal solutions. (DISCO). ✓ - - - - - - Quasi-oppositional [23] 2007 2 1 MCS in is GA planning methodology istime-varying proposed togeneration improve the accuracy for stochastic DG integration with tradeoff solution. A hybrid GA-PSO proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 EA is embedded employed EA is employed to solve IMO to solve considering IMO considering both both time-varying generation and and demand behavior behavior aiming at aiming various at technical various technical impacts of impacts of teaching learning-based optimization (QOTLBO) methodology, ademand variant of TLBO proposed as MOO regarding [32] 5 ✓ [43] 2014 4,3 [44] 1 MODGP formulation on GA and εks constrained method is proposed to the best compromise (tradeoff) solution for DM The MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) FDM has utilized for best compromise solution among the set ofin Pareto optimal solutions. MCS-embedded GA isbased presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 523 The MOO planning framework based on non-dominated sorting genetic algorithm IIsolve (NSGA-II) along with MCS (for uncertain operation Table 5. Tab The e contribution 5 The con of bu MODGP-based on o MODGP planning based p ann methods ng me and hods related and works. e a ed (PTC: wo Components PTC Componen planning s n p ann techniques; ng echn MODGP: ques MODGP MO based MO DG based placement). DG p acemen ✓ [42] 2014 1 ✓ ✓ [25] [25] 2008 2008 6 1 6 1 function, which is further solved by GA to ensure optimal DG (wind) planning. addition to finding optimal generated electricity prices in a competitive electrical market. GA with goal programming methodology finds a solution with associated uncertainties among MO and constraints. [26] 2 locations. Ref. Year MO PTC A B C D Contribution of MODGP-based Planning The DGP The problem DGP for problem location, for size, location, and type size, (REG and type and (REG PEV) and is defined PEV) is as defined MO mixed as MO integer mixed nonlinear integer nonlinear programming programming (MINLP), (MINLP), in which in which ✓ An exhaustive An exhaustive search analysis search analysis (ES) has (ES) presented has presented with IMO with for IMO MODGP for MODGP problem problem under variable under variable load conditions. load conditions. [21] 2005 4 1 ✓ [29] 2010 522 A fuzzy embedded GA is employed to solve fuzzy single objective function employing set theory for MODGP. [62] 2013 - 1112 2 ✓ --- - -- - - -that (allocation) planning problem forweighted optimal sizing and location isload solved by ABC based to alternative. maximize the required objectives. [22] 3 Similarly, withand the same formulation (GA and εMODGP constrained method), a flow double tradeoff method isfuzzy presented toapproach find the ✓ -- ✓ [41] 3[28] 2014 ✓ [24] 2008 NSGA-II along with max-min approach solves problem considering future uncertainties and risk management sizing of DGP for solving multi-objective optimal power (OPF) problem for radial DNW. A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into abest single DGP -Capacitor [38] 2014 [38] 12009 2014 ✓ [39] aims atand DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with objective -- - -- optimal - DGP - location [28] 2009 4 1 (DISCO). [40] MO augmented ε constrained method proposed for DGP planning on short term for future DNWs withEA ANM functionalities. scenarios) OPF, addresses DGP problem of REG and storage integration within DNW. ✓ 1 - --- ✓ A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 14 [32] 5 An improved MOPSO with preference strategy (IMPSO-PS) iswind-based proposed for MODGP with the aim to achieve optimal capacity and MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and (inner) optimizations respectively, NSGA-II an is NSGA-II used to is obtain used to compromise obtain compromise (Pareto frontier) (Pareto solution frontier) for solution afor local for distribution abasis local distribution company company (LDC). (LDC). ✓ -- an The proposed NSGA finds arrangements for DGs regarding compromise solution among contrasting objectives. [27] 2008 3 1 formulation based on GA and εof method is proposed to solve the best compromise (tradeoff) solution for DM GA is also GA used isprogramming also for used comparison for comparison of and results advocated and advocated large DNW large systems DNW systems even with even suboptimal with solutions. solutions. MCS-embedded GA isalgorithm presented toresults solve chance constrained programming framework considering future uncertainties ofin loads/DGs. [30] 2011 5 ✓ [23] 2007 2 MCS embedded in GA planning methodology istime-varying proposed toobjectives improve the accuracy forbehavior stochastic DGsuboptimal integration with tradeoff solution. algorithm has been used for MODGP problem satisfying (minimize DISCO cost and maximize DG owner profit) EA is employed to solve IMO considering both generation and demand aiming at various technical impacts ofplanning problem. [44] 3[26] 2014 1PTC 2 11 which is further solved by GA toconstrained ensure optimal DG planning. ✓ -- - MOPSO - function, -with GA GA goal with goal programming methodology methodology finds aconstrained solution finds afor solution with associated with associated uncertainties among MO among and MO constraints. and constraints. [26] 2008 2008 2 1 2 B✓ 1 --B--- D✓ [33] 2012 Ref. 2013 Re Year Yea MO MO A AC Contribution Con of bu MODGP-based on o(wind) MODGP Planning ba P uncertainties ann ng FDM has utilized for best compromise solution among the set of Pareto optimal solutions. [21] 2005 4 ✓ [63] -PTC --- C---- -- D Two-stage modification method with AMHBMO based approach is proposed to solve the multi-objective capacitor The MOO planning framework based on non-dominated sorting genetic algorithm II (NSGA-II) along with MCS (for uncertain operation [22] 3 1 Similarly, with the same formulation (GA and εframework method), aproposed double tradeoff method is presented to find the best alternative. Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, a ed variant ofmultiple TLBO proposed as MOO regarding ✓ [43] 2014 4,3 1 A goal attainment method (GoA) has presented, where goal programming transforms objective functions into a single objective locations. ✓ [25] 2008 6 1 to find time varying voltage magnitude and loss sensitivity factor at each node. The MO probabilistic The MO probabilistic (mathematical (mathematical programming) programming) framework has proposed has for various for DER various planning DER planning (six DG types, (six DG size, types, and location) size, and location) An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. (DISCO). ✓ - - ✓- - - ✓ [41] 2014 2 11 ✓ [29] [29] 2010 2010 5 2 5 2 fuzzy A embedded fuzzy embedded GA is employed GA is employed to solve fuzzy to solve weighted fuzzy weighted single objective single objective function function employing employing fuzzy set fuzzy theory set for theory MODGP. for MODGP. [42] 3 addition to finding optimal generated electricity prices in a competitive electrical market. A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 ✓ [24] 2008 3 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management [32] 5 DGP MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, ✓ ✓ The proposed The proposed NSGA NSGA algorithm finds arrangements finds arrangements for wind-based for wind-based DGs regarding DGs compromise compromise solution solution among contrasting among contrasting objectives. [27] [27] 2008 2008 3 13 ✓ 1 - - - - MODGP [40] 2014 11 MO augmented εand constrained method proposed for DGP planning on short term basis for future DNWs with ANM functionalities. formulation oaims mu based ais on onalgorithm ba GA and onin GA εfor constrained and εto on method aand ned is me proposed hod pMoreover to opo solve ed the o(OPF) obest veregarding compromise he be omp (tradeoff) om e ε-constrained solution adeo for oalong u DM on oobjectives. DM [28] 2009 422 scenarios) and OPF, addresses DGP problem of REG storage integration within DNW. optimal location sizing ofed DGP solving multi-objective optimal power flow problem for radial DNW. ✓ [23] 2007 2 MCS embedded in GA planning methodology is proposed to improve the accuracy for stochastic DG integration with tradeoff solution. function, which further solved by GA ensure optimal DG (wind) planning. ✓ ✓ ✓ [39] 2014 [39] 2014 1 2 1 that aims that at DISCOs at contribution DISCOs contribution the competitive in the competitive electricity electricity market. market. Moreover modified modified augmented augmented ε-constrained method method with along with [33] 2012 2 1 The multi-criteria planning model aims toand achieve contrasting objectives (cost and reliability) for DGP. GA further aof set of objective non✓ 15 - ✓ [21] 2014 2005 21 2005 4 14 GA also for comparison ofto results advocated for large DNW systems even with suboptimal solutions. [64] 4[30] 22011 MOF sused oprogramming mu avoltage o(GoA) sisconsidering mu aneous DG sconstrained and capac o p each acemen and so ved by nves ga ed efinds en evo u y a go hm DEA [22] 3 Similarly, with the same formulation (GA and εtime-varying method), a double tradeoff method is presented todfind the alternative. - EA -is MCS-embedded MCS-embedded GA ised presented GA presented solve chance to solve chance constrained programming programming framework framework considering considering future uncertainties future uncertainties loads/DGs. ofona loads/DGs. [30] 2011 5 1- - -✓ - - -- An MOPSO with preference strategy (IMPSO-PS) is proposed for MODGP with the aim to achieve optimal capacity and A goal attainment method has presented, where goal programming transforms multiple objective functions into abest single is employed to solve IMO both generation and behavior aiming at various technical impacts of ✓ GA with goal methodology finds aconstrained solution with associated uncertainties among MO and constraints. [26] 2008 22 1 to find time varying magnitude and loss sensitivity factor at node. The MOO planning framework based on non-dominated sorting genetic algorithm IIademand (NSGA-II) along with MCS (for uncertain operation [34] 2013 An exhaustive An exhaustive search analysis search analysis (ES) has (ES) presented has presented with IMO for IMO MODGP for MODGP problem problem under variable under variable load conditions. load conditions. Dimproved SCO Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, variant of TLBO as MOO regarding algorithm has been used for MODGP problem satisfying objectives (minimize DISCO cost andproposed maximize DG owner profit) in ✓ - - MOPSO [44] 2014 1 ✓ 1 --- -- ✓--- -- (DISCO). [24] 2008 3 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management [32] 2012 5 [25] 6 FDM has FDM utilized hasfor utilized best compromise for best compromise solution among solution the among set ofwith the Pareto set of optimal Pareto solutions. optimal solutions. MODGP problem is solved byproposed two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, ✓ [41] 2014 2 111 dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. [42] 3 [28] [28] 2009 2009 4 4 ✓ ✓ [43] 2014 4,3 [29] 2010 5 2 fuzzy embedded GA is employed to solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. ✓ locations. [23] 2007 2 1 MCS embedded in GA planning methodology is proposed to improve the accuracy for stochastic DG integration with tradeoff solution. ✓ A hybrid A GA-PSO hybrid GA-PSO is proposed is to solve MODGP to solve MODGP problem problem regarding regarding DG location DG location (with GA) (with and GA) capacity and capacity (with PSO) (with respectively. PSO) respectively. [31] [31] 2012 2012 3 3 1 function, which is further solved by GA to ensure optimal DG (wind) planning. DGP ✓ - - - - Similarly, [33] 2012 2 1 - ✓ scenarios) and OPF, addresses DGP problem ofand REG and storage integration within DNW. The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) for DGP. GA further finds a set of nonoptimal location and sizing of DGP for solving multi-objective optimal power flow (OPF) problem for radial DNW. GA is also GA used is also for used comparison for comparison of results of results advocated and advocated for large for DNW large systems DNW systems even with even suboptimal with suboptimal solutions. solutions. ✓ addition to finding optimal generated electricity prices in a competitive electrical market. [22] 2005 22 2005 3 1 3 1 S with m a the y w same h he formulation ame o mu (GA a and on GA ε constrained and ε on method), a ned me a double hod a tradeoff doub e method adeo is me presented hod p to e en find ed the o best nd alternative. he be a e na ve EAaugmented is employed to solve IMO considering both time-varying generation demand behavior aiming at with various technical impacts of - MO [40] 2014 [40] 2014 2 11 2 ✓ MO augmented ε constrained εpresented constrained method proposed method proposed for DGP planning for DGPfactor planning on short term on and short basis term for future basis for DNWs future with DNWs ANM functionalities. ANM functionalities. to find time varying voltage magnitude and loss sensitivity at each node. ✓ 1 - -- ✓- -- -- -- -MO [34] 2013 2 [35] An planning framework has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. MCS-embedded GA isanalysis to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 ✓ [24] 2008 3 1 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management A goal attainment A goal attainment method (GoA) method has (GoA) presented, has presented, goal where programming goal programming transforms transforms multiple multiple objective objective functions functions into aadeo single into objective ao single [25] 6 MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, ✓ 2- -- -✓ -- MCS - embedded GA with goal programming methodology finds awhere solution with associated among MO and constraints. [26] 2008 2 1 Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, aschedules. variant of TLBO proposed as MOO regarding solutions for wind-based DG units sizing, siting, and maintenance MOPSO algorithm has been used for MODGP problem satisfying (minimize cost and maximize DG owner profit) in An exhaustive search (ES) has presented IMO for MODGP problem under variable load conditions. An MOPSO with strategy (IMPSO-PS) is proposed for MODGP with the aim to achieve optimal capacity ✓ -- dominant [29] [29] 2010 2010 5 2 5 Aimproved fuzzy A embedded fuzzy embedded GA ispreference employed GA isme employed to fuzzy to solve weighted fuzzy weighted single objective single objective employing employing fuzzy set fuzzy theory set for theory MODGP. for MODGP. ✓ 11 ✓ [23] 2007 23 2007 2 14,3 2 MCS embedded in GA planning nplanning GA p methodology ann ng hodo issolve proposed ogy pwith to opo improve ed oobjectives the mp accuracy ove he(cost auncertainties for uand stochastic aDISCO yreliability) oIIfunction oDG ha integration DG nGA eg with aMCS tradeoff on w hand solution. u on objective ✓ 1 -- -- ✓-- -- --- ✓ -multi-criteria [32] [32] 2012 532 5 - ✓ DGP The MOO The planning MOO planning framework framework based on based non-dominated on non-dominated sorting genetic sorting algorithm genetic algorithm IIfunction (NSGA-II) (NSGA-II) along with along MCS with (for uncertain (for uncertain operation operation [33] 2012 2 1 [42] 2014 The model aims to achieve contrasting objectives for DGP. further finds a set of non[43] ✓ [28] 2009 4 [44] 2014 1 An optimization problem for multiple micro-turbines (DG) placement and sizes isbehavior solved by hybrid PSO and SFL algorithms based ✓ 1 - -- ✓- -- -- ---- -optimal A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 12 ✓ ✓ EA is employed to solve IMO considering both time-varying generation and demand aiming atsolutions. various technical impacts of - MO [41] 2014 [41] 2011 2014 2 11 function, function, which is which further is solved further by solved GA toby ensure GA to optimal ensure DG optimal (wind) DG planning. (wind) planning. location and sizing of DGP for solving multi-objective optimal power flow (OPF) problem for radial DNW. to find time varying voltage magnitude and loss sensitivity factor at each node. ✓ addition to finding optimal generated electricity prices in a competitive electrical market. The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 3 1 [34] 2013 2 locations. GA is also used for comparison of results and advocated for large DNW systems even with suboptimal [35] An MO planning framework has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. [36] 2013 3 ✓ ✓ MCS-embedded MCS-embedded GA is presented GA is presented to solve chance to solve constrained chance constrained programming programming framework framework considering considering future uncertainties future uncertainties of loads/DGs. of loads/DGs. [30] [30] 2011 5 1 5 1 ✓ 1 - ✓ - - - - NSGA-II [24] 2008 24 3 16 NSGA along with asolutions ong max-min wand hfor max approach m nmethodology app solves oa hMODGP o finds vesizing, MODGP problem pand considering ob em onintegration future de ng load uwithin uuncertainties e oad un e and a nrisk e and management k managemen scenarios) and OPF, addresses OPF, addresses DGP problem DGP of problem REG and of REG storage integration storage within DNW. DNW. [25] 2008 - scenarios) GA with goal programming atool solution with associated uncertainties among MO and constraints. [26] 2 dominant DG units siting, and maintenance schedules. framework, followed bywind-based fuzzy decision-making to isselect the most preferred Pareto optimal solution satisfying competing objectives. A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into aasingle objective DGP MOPSO algorithm has been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DG owner profit) in respectively. MODGP MODGP problem problem is solved is solved two stage by two stage heuristic iterative iterative method including method including clustering clustering (outer) and (outer) EA (inner) and EA optimizations (inner) optimizations respectively, respectively, An MOPSO strategy (IMPSO-PS) proposed for MODGP with the aim to achieve optimal capacity and The planning model aims toheuristic achieve contrasting objectives (cost and reliability) for DGP. GA further finds set of nonexhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. ✓ [29] 2010 5 2 Aimproved fuzzy embedded GA ispreference employed toarrangements solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. MO optimization problem for multiple micro-turbines (DG) placement and sizes isof solved by hybrid PSO and SFL algorithms based ✓- -- EA-- ✓ - employed - An -multi-criteria hybrid A GA-PSO hybrid GA-PSO iswith proposed isby proposed to solve MODGP to solve MODGP problem problem regarding regarding DG location DG location (with GA) (with and GA) capacity and capacity (with PSO) (with respectively. PSO) [31] [31] 2012 32 12 3 ✓ 1 - -- ✓ is EA emp to solve oyed IMO o o considering ve MO on both de time-varying ng bo h me generation va y ng gene and a demand on and demand behavior behav aiming o at a various m ng a technical va ou e impacts hn a of mpa o Quasi-oppositional Quasi-oppositional teaching learning-based teaching learning-based optimization optimization (QOTLBO) (QOTLBO) methodology, methodology, a variant a variant TLBO proposed of TLBO proposed as MOO regarding as MOO regarding [32] 2012 5 [43] 2014 4,3 [33] [33] 2012 2 1 [44] 1 ✓ The proposed NSGA algorithm finds for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 3 1 [34] 2013 [28] 2009 4 [35] 2 An MO planning framework has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. ✓ -1 - -- -✓- -- -- -- -locations. [36] 2013 33 13 - ✓ [37] The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). - with [42] 2014 [42] ✓ 11 ✓ [25] 2008 25 2008 6 12014 6 function, which is further solved by GA toand ensure optimal DG (wind) planning. addition to finding optimal generated electricity prices in asiting, competitive electrical market. GA goal programming methodology finds atool solution with associated uncertainties among MO and constraints. [26] 2 to find time to find varying time voltage varying magnitude voltage magnitude and loss and sensitivity loss sensitivity factor at factor each node. at each node. dominant solutions for wind-based DG units sizing, and maintenance schedules. GA is also used for comparison ofof results advocated for large DNW systems even with suboptimal solutions. framework, followed fuzzy decision-making to select the most preferred Pareto optimal solution satisfying competing objectives. ✓ - DGP- optimal MCS-embedded GA isby presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 1 location optimal and location sizing and of DGP sizing for solving DGP for multi-objective solving multi-objective optimal power optimal flow power (OPF) flow problem (OPF) for problem radial for DNW. radial DNW. A goal attainment A goal attainment method (GoA) method has (GoA) presented, has presented, where goal where programming goal programming transforms transforms multiple multiple objective objective functions functions into a single into objective a single objective DGP An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. MO optimization problem for multiple micro-turbines (DG) placement and sizes isinteger solved by hybrid PSO and SFL algorithms The problem for location, size, and type (REG and PEV) isobjectives defined as MO mixed nonlinear programming (MINLP), which ✓ -✓ - The -DGP [32] [32] 2012 2012 5 1 improved MOPSO with strategy (IMPSO-PS) iswind-based proposed for MODGP with the aim to achieve optimal capacity and MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, ✓ - An proposed NSGA algorithm finds for DGs regarding compromise solution among contrasting objectives. [27] 2008 3 1 multi-criteria The multi-criteria planning planning model aims model to achieve aims to contrasting achieve contrasting objectives (cost and (cost reliability) and reliability) for DGP. for GA DGP. further GA finds further a set of nonainbased set [28] 2009 4 [35] 2013 22 An MO has developed, namely multi-objective harmony search (IMOHS), which evaluate the DGP. [29] 2010 5 25 fuzzy embedded GA ispreference employed toarrangements solve fuzzy single objective fuzzy set theory for MODGP. [36] ✓ 1- - --- -✓- --- GA [37] 2013 1 The methodology based on Pareto frontier differential evolution algorithm is proposed optimal MODGP (sizing and location). A hybrid GA-PSO iscomparison proposed toof solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 algorithm MOPSO has been used has been for MODGP used for problem MODGP satisfying problem satisfying objectives objectives (minimize DISCO cost DISCO and maximize cost and maximize DG owner DG profit) owner infinds profit) in of non[38] 2014 - - MOPSO [44] 2014 function, which is which further is solved further by solved GA by ensure GA tooptimal ensure DG optimal DG (wind) planning. ✓ 111 ✓ GA goal programming hplanning goa palgorithm ogframework amm methodology ng me hodo finds ogy atosolution nd atool with oweighted uimproved associated onfor w h(wind) a(PFDE) uncertainties o aplanning. ed unfunction e(minimize among an eemploying MO among andfor MO constraints. andsolutions. on a ncan [26] 2008 26 2008 2 12014 2 [33] 2012 - function, -w [34] [34] 2013 2013 2 - ✓ GA is also used for results and advocated large DNW systems even with suboptimal framework, followed by fuzzy decision-making to select the most preferred Pareto optimal solution satisfying competing objectives. ✓ --- -with -locations. [43] 2014 [43] 4,3 4,3 1 - ✓an NSGA-II is used to obtain compromise (Pareto frontier) solution for a local distribution company (LDC). to find time varying voltage magnitude and loss sensitivity factor at each node. dominant dominant solutions solutions for wind-based forfor wind-based DG units DG sizing, units sizing, and siting, maintenance and as maintenance schedules. exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. An MO optimization multiple micro-turbines (DG) placement and sizes isschedules. solved by hybrid PSO and SFL ✓ - - - - The -proposed MCS-embedded GA isproblem presented to solve chance constrained programming framework considering future uncertainties [30] 2011 addition to addition finding tooptimal finding generated optimal generated electricity electricity prices in asiting, prices in aba electrical electrical market. market. The DGP problem for size, and type (REG and PEV) iscompetitive defined MO integer programming (MINLP), inbased which goal attainment method (GoA) has where goal programming transforms multiple objective functions into a of single MODGP MODGP problem problem is solved is by solved two stage by two heuristic stage heuristic iterative iterative method including method including clustering clustering (outer) and (outer) and EA optimizations (inner) optimizations respectively, respectively, ✓1 The p NSGA opo edalgorithm NSGA alocation, go finds hm arrangements nd apresented, angemen for wind-based ocompetitive w nd DGs regarding ed DG algorithm ega compromise dmixed ng is omp solution om nonlinear efor among ofuzzy uEA on(inner) contrasting among on objectives. aalgorithms ngloads/DGs. ob eobjective ve [27] 2008 27 2008 3 15 3 1 - ✓ [28] 2009 4 ✓ ✓ [29] 2010 5 2 A fuzzy embedded GA ison employed to solve fuzzy weighted single objective function employing set theory for MODGP. - -A [36] 2013 3 1 [37] The methodology based Pareto frontier differential evolution (PFDE) proposed optimal MODGP (sizing location). [38] 2014 ✓ [32] 2012 5 1 MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) --is [33] [33] 2012 2 multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) for DGP. GA further finds a can setand of non✓1 ✓ -- -- --- An GA also used for ofby results and advocated for large DNW systems even suboptimal solutions. - The [35] [35] 2013 2013 2 12 2 1 -- ✓ An MO planning An MO planning framework framework has developed, has developed, namely improved namely improved multi-objective harmony harmony search (IMOHS), search (IMOHS), which can which evaluate evaluate the DGP.the DGP. framework, followed by fuzzy decision-making tool to select the most preferred Pareto optimal solution satisfying competing objectives. improved An improved MOPSO with MOPSO preference with preference strategy (IMPSO-PS) strategy (IMPSO-PS) is proposed is proposed for MODGP for with MODGP the with aim with to the achieve aim to(LDC). optimal achieve capacity optimal and capacity and ✓ A hybrid GA-PSO iscomparison proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 1 an NSGA-II isanalysis used to obtain compromise (Pareto frontier) solution for amulti-objective local distribution company is further solved GA towith ensure optimal DG (wind) planning. to find time towhich find varying time voltage varying magnitude voltage magnitude and loss and sensitivity loss sensitivity factor at factor each node. at each node. ✓ 1 - --- ✓- --- An - - function, [34] 2013 22 12 ✓ exhaustive An exhau search ve ea h ana (ES) y has presented ES ha p e en IMO ed w for h MODGP MO o MODGP problem p under ob em variable unde va load ab conditions. e oad ond on ✓ MCS-embedded GA is presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5 1 [44] 2014 [44] 2014 1 The DGP problem for location, size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), in which [39] 2014 2 that aims at DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with based ✓1 dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. [28] 2009 28 2009 4 14 ✓ [29] 2010 5 2 A fuzzy embedded GA ison employed to solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. An MO optimization An MO optimization problem problem for multiple for multiple micro-turbines micro-turbines (DG) placement (DG) placement and sizes and isproposed solved sizes is by solved hybrid by PSO hybrid and PSO SFL and algorithms SFL locations. ✓ - --- - ---- locations. [37] 2013 3 1 - ✓ The methodology based Pareto frontier differential evolution (PFDE) algorithm is for optimal MODGP (sizing and location). [38] 2014 2 A goal attainment method (GoA) presented, where goal programming transforms multiple objective functions into aasingle objective MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) MODGP problem is solved by twohas stage iterative method including clustering (outer) and EAfor (inner) optimizations respectively, The multi-criteria The multi-criteria planning planning model aims model toheuristic achieve aims to contrasting achieve contrasting objectives objectives (cost and (cost reliability) and reliability) for DGP. GA DGP. further GA finds further set finds ofalgorithms nonabased set of non[30] [25] [27] [22] [24] [32]

2011 2008 2005 2008 2012

[36] [31] [32] [33] [34] [35] [30] [39] 2010 29 [38] [40] [37] [31] [32] [35] [36] 2011 30 [39] [33] [34] [40] [41] 2012 31 [38] [32]

[36] 2013 2012 2012 [34] 2013 2011 2014 2010 5 2014 2014 [37] 2013 2012 [35] 2013 2011 5 2014 2012 2013 2014 3 [38] 2014 2012

[29] 2010 [31] 2012 [21] [21] 4 [52] [23] 2016 2005 2007 [26] 2008 [28] 2009

[29] 30 31

2013 3 5 2 2013 2 5 2 25 2 2 2013 3 5 2013 2 3 15 2 12 3 2014 2 5

✓ 1 ✓1 13 2 1 1 ✓1 - ✓ ✓2 ✓ ✓ 1 ✓ 13 ✓1 1 2 ✓1 ✓ ✓1 1 ✓ ✓1 ✓1 12

used a for oGA-PSO ucomparison o isto ompa of results on e uadvocated and advo for large a ed DNW osolution a ge systems DNW even y em with even suboptimal w hcompany ubop solutions. ma o(with u onPSO) respectively. - -is also - GA -- has -- ✓ - GA A hybrid proposed too and solve MODGP problem regarding DGaoptimal location (with GA) and capacity an NSGA-II isedused obtain compromise (Pareto frontier) for local distribution (LDC). FDM utilized for best compromise solution among the setto ofmulti-objective Pareto solutions. An MO planning framework has developed, namely improved harmony search (IMOHS), which can evaluate the framework, framework, followed followed by fuzzy by decision-making fuzzy decision-making tool to select tool the select most the preferred most preferred Pareto optimal Pareto solution optimal solution satisfying satisfying competing competing objectives. objectives. -- ✓ --- - --- - function, MCS-embedded GA is presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. The DGP problem for location, size, and type (REG and PEV) is defined as MO mixed integer nonlinear (MINLP), inDGP. which which further solved by GA ensure optimal DG planning. that aims at DISCOs contribution inDG the competitive electricity market. modified augmented along with to find time varying voltage magnitude and loss sensitivity factor each node. dominant dominant solutions solutions for wind-based for units DG sizing, siting, sizing, and siting, maintenance and maintenance schedules. ✓ A embedded uzzy embedded GA isisemployed GA emp to oyed owind-based ve uzzy opresented, oto ve we uzzy ghunits ed we ng gh eed ob e(wind) ng ve eatob un eMoreover ve on emp un oy on schedules. ng emp uzzy oy ng e uzzy heo ε-constrained yprogramming efunctions oheo MODGP y o method MODGP - ✓ -- A fuzzy -A goal attainment method (GoA) has where goal programming transforms multiple objective into a single objective The MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) MO augmented ε constrained method proposed for DGP planning on short term basis for future DNWs with ANM functionalities. An MO optimization problem for multiple micro-turbines (DG) placement and sizes is solved by hybrid PSO and SFL algorithms based --- ✓ The methodology The methodology based on based Pareto on frontier Pareto differential frontier differential evolution evolution (PFDE) algorithm (PFDE) algorithm is proposed is proposed for optimal for MODGP optimal MODGP (sizing and (sizing location). and location). A hybrid GA-PSO isto proposed to developed, solve MODGP problem regarding DGaoptimal location (with GA) and capacity (with PSO) respectively. an NSGA-II is used obtain compromise (Pareto frontier) solution for localclustering distribution company (LDC). FDM hasplanning utilized for best compromise solution among the set of Pareto solutions. MODGP problem is solved by two stage heuristic iterative method including (outer) and EA (inner) optimizations respectively, The multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) for DGP. GA further finds a set of non✓ An MO An MO planning framework framework has has developed, namely improved namely improved multi-objective multi-objective harmony harmony search (IMOHS), search (IMOHS), which can which evaluate can evaluate the DGP. the MCS embedded MCS embedded GA p GA e en ed p e o en o ed ve o han o ve e on han a e ned on p og a ned amm p og ng amm amewo ng k amewo on de k ng on u de u e ng un u e u a e n un e e o a n oad e o DG oad DG function, is further solved by tonon-dominated ensure optimal DGthe (wind) planning. that aims which atplanning DISCOs contribution in GA the competitive electricity market. Moreover modified augmented ε-constrained method along with DGP. The MOO framework based onsize, sorting genetic algorithm II (NSGA-II) along nonlinear with MCS (for uncertain operation framework, followed by fuzzy decision-making tool to(REG select most preferred Pareto optimal solution satisfying competing objectives. The DGP The problem DGP problem for location, for location, size, and type and (REG type and PEV) and is maintenance defined PEV) is as defined MO mixed as MO integer mixed nonlinear integer programming programming (MINLP), (MINLP), inbased whichinbased which A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a single objective MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) to find time varying voltage magnitude and loss sensitivity factor at each node. -- ✓ -- A -hyb -- -ddominant MO augmented ε constrained method proposed for DGP planning on short term basis for future DNWs with ANM functionalities. solutions for wind-based DG units sizing, siting, and schedules. An MO optimization An MO optimization problem problem for multiple for multiple micro-turbines micro-turbines (DG) placement (DG) placement and sizes and is solved sizes is by solved hybrid by PSO hybrid and PSO SFL and algorithms SFL algorithms A GA hyb d problem GA pPSO opofor pbest opo o compromise o by ed ve two MODGP o ostage ve solution MODGP pheuristic ob emamong p iterative ega ob em dthe ngega DGof d oPareto ngaincluding DG onoptimal ow a hclustering on GA wand h GA apaandyand apa w EA h PSO y(inner) w h e PSO pe vee ype ve y - PSO FDM has utilized set solutions. MODGP ised solved method (outer) optimizations respectively,

Energies 2017, 10, 208

18 of 44

Energies Energies 2017, 10,2017, 208 10, 208

16 of 44 16 of 44

Table 6. Cont. Energies 2017, 208contribution 16 of 44 Table 5.Table The contribution 5.10, The of MODGP-based of MODGP-based planning planning methods methods and related and related works. (PTC: works.Components (PTC: Components in planning in planning techniques; techniques; MODGP: MODGP: MO based MODG based placement). DG placement).

Ref. MO PTC A Contribution of MOVPQ-Based Planning Energies 2017, 208 16 of 44 Ref. Year Ref. YearTable Year MO PTC MO A PTC B B AofCMODGP-based BDC C DD planning methods and related works. Contribution Contribution of MODGP-based of MODGP-based Planning Planning MODGP: 5.10, The contribution (PTC: Components in planning techniques; MO based DG placement). MODGPMODGP formulation formulation based on based GA and on GA ε constrained and ε constrained method is method proposed is proposed to solve the to solve best compromise the best compromise (tradeoff) (tradeoff) solutionbanks solution for DM(SCBs) for DM Fuzzy MOPSO algorithm is proposed applied to find the best solution of DGs and shunt capacitor sizing and placement problem ✓ ✓ - - -- [21] [21] 2005 32005 4 14 1 [65] 2 MO Energies2014 Energies 2017, 2017, 208 208 16FDM. of 44 16 of 44 Ref.10, Year PTC Aof MODGP-based B C(DISCO). D - (DISCO). Contribution of MODGP-based Planning simultaneously uncertainty considered as fuzzy theory. The optimum solution Table 5.10, The contribution planning methodswith and load related works. (PTC: Components in data planning techniques; MODGP: MOextracted based DGwith placement). [22]

[22] 2005

2005 3

13

✓1

- ✓-

--

MODGP based on GAand and constrained methodmethod), proposed to solve the best compromise (tradeoff) solution for DM - Similarly, Similarly, with theformulation with samethe formulation same formulation (GA (GA ε εconstrained and ε constrained method), aisdouble atradeoff double tradeoff method is method presented is presented to find the to best find alternative. the best alternative.

[25]

[25] 2008 [22]

2008 6 2005

13 6

✓1 1

✓- ✓

-- -

MODGP formulation formulation based onbased GA and on GA ε constrained and ε constrained is method proposed is proposed to solve the to solve best compromise the best compromise (tradeoff) (tradeoff) solution solution forthe DM for DM -- MODGP -Similarly, with the same formulation (GA andmethod ε constrained method), a double tradeoff method is presented to find best alternative.

[28] [22] [24]

[25] [28] 2009 [22] 2005 [24] 2008 [27]

4 16 3

✓1 1

✓- ✓

-- -

-- Similarly, -Similarly, with the with samethe formulation same formulation (GA and(GA ε constrained and ε constrained method),method), a doubleatradeoff double tradeoff method is method presented is presented to find the to best find alternative. the best alternative.

✓ -[21] 2005 1 4 1 [66] 2015 5Year -PTC GA based fuzzy multi-objective approach solves withofmaximization ofPlanning fuzzy MOF for optimum sizing and location ofof shunt Energies Energies 2017, 2017, 208 208 16 of 44 16 44 capacitors. Ref. MO Aof- MODGP-based B- - - C D (DISCO). ✓ - ✓ - MCS - embedded [23] Table [23] 2007 2007 25.10, 1contribution 2 MCS embedded in GA planning in related GAand planning methodology methodology is proposed isContribution proposed to improve improve theMODGP-based accuracy the accuracy for stochastic for stochastic DG MODGP: integration DGbased integration with tradeoff with tradeoff solution. solution. 5.10, Table The contribution The of1 MODGP-based planning planning methods methods and related works. (PTC: works. Components (PTC: Components intoplanning in planning techniques; techniques; MODGP: MO MO DG based placement). DG placement). MODGP formulation based onapproach GA and εsolves constrained method is proposed toPSO solve the best compromise (tradeoff) solution for DM An with improved backward/forward sweep (BSFS) load flow based is proposed to mixed integer programming (MIP) problems of both [22] 2005 3 1 - ✓ - - - - NSGA-II Similarly, with the same formulation (GA and εMODGP constrained method), a double tradeoff method issolve presented to find the best alternative. ✓ [24] [24] 2008 2008 3 1 NSGA-II along along max-min with max-min approach solves MODGP problem problem considering considering future load future uncertainties load uncertainties and risk and management risk management ✓ -[21] 5,3 2005 1 4 1 [67] 2016 -PTC Ref. Table Ref. Year Year MO PTC MO A B AofCMODGP-based BD - C D Contribution Contribution of MODGP-based ofand MODGP-based Planning (DISCO). optimal capacitor (delta-connected switched) placement and control toPlanning achieve respective inDG two strategies. - EA - employed [23] 2007 2 10,2017, MCS in related GAconsidering planning methodology istime-varying proposed improve the accuracy for stochastic DGbased integration with tradeoff solution. 5.Table The contribution 5.10, The of1 MODGP-based planning planning methods methods and and related works. (PTC: works. Components (PTC: Components intoplanning in planning techniques; techniques; MODGP: MODGP: MO MO DG based placement). placement). is EA is embedded employed to solve IMO to solve IMO considering both time-varying both generation generation demand and demand behavior behavior aiming ataiming various atobjectives technical various technical impacts of impacts of16 of 44 16 of 44 16 of 44 Energies Energies 2017, Energies 2017, 208contribution 208 10, ✓ 208 ✓ - -- - -- DGP-[24] 2008 3 1 - ✓ NSGA-II along with max-min approach solves MODGP problem considering futurefor load uncertainties and risk management ✓1 [21] [21] 2005 2005 4 14 [68] - DGP Multi-criteria simultaneous planning with ABFO approach is proposed PFF and DG (location and size), considering nonlinear loads. Ref. 2016 Ref. Year 3Year MO 2PTC MO A PTC B ✓ A C BD - C(DISCO). D Contribution Contribution of MODGP-based of MODGP-based Planning Planning (DISCO). --withMCS [23] 2007 embedded in solve GA planning methodology togeneration improve the accuracy forbehavior stochastic DGand integration tradeoff solution. EA is employed to IMOmethodology considering both and demand atconstraints. various with technical impacts of ✓1 GA goal with programming goal programming methodology finds a solution findsistime-varying aproposed solution with associated with associated uncertainties uncertainties among MO among andaiming MO constraints. [26] [26] 2008 2008 2 12 2 1 - ✓ - -- - -- GA Table 5. Table The contribution 5. Table The contribution 5. The of contribution MODGP-based of MODGP-based of MODGP-based planning planning methods planning methods and related methods and related works. and (PTC: related works. Components (PTC: works. Components (PTC: in Components planning in planning techniques; in planning techniques; MODGP: techniques; MODGP: MO based MODGP: MO DG based placement). MO DG based DG placement). ✓ [25] 2008 6 1 MODGP MODGP formulation formulation based on based GA and on GA ε constrained and ε constrained method is method proposed is proposed to solve the to solve best compromise the best compromise (tradeoff) (tradeoff) solution solution for DM for DMplacement). ✓ MO seeker optimization algorithm (MOSOA) iswind-based proposed planning methodology tois integrate distribution automation devices, DGs and - ✓ -- -- -- -- The [22] [22] 2005 2005 3 1 Similarly, Similarly, with the with same the formulation same formulation (GA and(GA ε constrained and constrained method), method), a double atradeoff double tradeoff method method presented presented to and find the to best find alternative. the best alternative. --proposed [24] NSGA-II along with max-min approach solves MODGP problem considering future loadis uncertainties risk management The proposed NSGA algorithm NSGA algorithm finds arrangements finds arrangements for εwind-based for DGs regarding DGs regarding compromise compromise solution solution among contrasting among contrasting objectives. objectives. [27] [27] 2008 2008 3 [21] 2016 [21] 2005 22008 2005 4 [69] 2 1334 ✓ 11 - ✓ - DGP (DISCO). (DISCO). D-STATCOMs in advance power distribution network (APDN). “Max–min” approach is selected most suitable trade-off solution. ✓ ✓ [23] [23] 2007 2007 2 1 2 1 MCS embedded MCS embedded in GA planning in GA planning methodology methodology is proposed is proposed to improve to improve the accuracy the accuracy for stochastic for stochastic DG integration DG integration with tradeoff with tradeoff solution. solution. EA is employed to solve IMO considering both time-varying generation and demand behavior aiming at various technical impacts of ✓ GA with goal programming methodology finds a solution with associated uncertainties among MO and constraints. [26] 2008 2 1 Ref. Ref. Year Ref. Year MO PTC Year MO A PTC MOB A PTC CAn BD A CAn BD C Danalysis Contribution Contribution Contribution MODGP-based of problem MODGP-based Planning MODGP-based Planning Planning exhaustive exhaustive search search analysis (ES) has (ES) presented has presented with IMO with for IMO MODGP forofMODGP problem underofvariable under variable load conditions. load conditions. [21] [70] [23] 2016 [23] 2007 [26]

2008 2009 4 2005 3 2008 2008

2005 2[21] 2007 2 2008

✓1 13 NSGA-II along with along max-min with max-min approach solves MODGP MODGP problem problem considering future future uncertainties load uncertainties and risk and management risk management DGP ✓ - -- - -- NSGA-II The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. 3 1 - ✓ GA --is also GA used is also for used comparison for comparison offor results of and results advocated and advocated for large for DNW large systems DNW systems even with even suboptimal with suboptimal solutions. solutions. MODGP MODGP formulation MODGP formulation based formulation onapproach based GA and on based GA εsolves constrained and on GA εincluding constrained and method εconsidering constrained is method proposed is method proposed toload solve is proposed the to solve best compromise the to solve best compromise the best (tradeoff) compromise (tradeoff) solution (tradeoff) solution for DMfor solution forreliability DM for DM MOSOA proposed planning strategy distribution automation devices like automatic reclosers (RAs) and FACTS devices - is --✓ - MCS - is -with - to - planning 2005 1 1- 4 - -✓ 1-- MCS 3[21] -2005 ✓ 14 2 embedded embedded ingoal GA in solve GAconsidering planning methodology methodology ispresented proposed istime-varying togeneration improve tofor improve the accuracy the accuracy for stochastic forbehavior stochastic DG integration DGand integration with tradeoff with tradeoff solution. solution. employed EA employed solve to considering both both generation and demand and demand behavior aiming various atset technical various technical impacts of impacts of ✓ GA methodology finds aproposed solution associated uncertainties among MO constraints. 2 14 -- ✓ An exhaustive search analysis (ES) has with IMOwith MODGP problem under variable load conditions. ✓1 ✓ -- --- ✓ ✓ (DISCO). (DISCO). (DISCO). - EA 26 5 2 A fuzzy A embedded fuzzy embedded GAprogramming isIMO employed GA isIMO employed to solve fuzzy to time-varying solve weighted fuzzy weighted single objective single objective function function employing employing fuzzyataiming set fuzzy theory for theory MODGP. for MODGP. like DSTATCOM; for VAR compensation in ADN. “Max–min” approach is applied to select the final trade-off solution. ✓ - ✓ 1 ✓

[29] [29] 2010 2010 5 [25] [25] 2008 2008 6 [28] 2009 4 1 - ✓ - -- - -- DGP ✓1 -[24] [24] 2008 2008 3 13 NSGA-II along with along max-min with max-min approachapproach solves MODGP solves MODGP problem problem considering considering future load future uncertainties load uncertainties and risk and management risk management ✓ 1- 3 -- -✓ 1-- NSGA-II The proposed finds arrangements for DGs regarding compromise solution among contrasting objectives. [27] 2008 3 13 - ✓ GA also used for comparison and advocated forand large DNW systems even with suboptimal solutions. - --✓ - DGP - -is - Similarly, -is NSGA ✓2005 [22] [22] 2005 [22] 2005 1 Similarly, with the Similarly, with same the formulation same the formulation same (GA formulation and(GA ε constrained and (GA εwind-based constrained method), ε constrained method), a double method), atradeoff double atradeoff method double tradeoff is method presented is method presented to find isloads/DGs. presented the to best find alternative. the to best find alternative. the best alternative. MCS-embedded MCS-embedded GA presented GA isalgorithm presented towith solveof chance toresults solve constrained chance constrained programming programming framework framework considering considering future uncertainties future uncertainties of of loads/DGs. [30] [30] 2011 2011 5 13 5 EA is employed EA is employed to solve IMO to solve considering IMO considering both time-varying both time-varying generation generation and demand and demand behavior behavior aiming ataiming various atset technical various technical impacts of impacts of ✓ ✓ GA with GA goal with programming goal programming methodology methodology finds a solution finds a solution with associated with associated uncertainties uncertainties among MO among and MO constraints. and constraints. [26] [26] 2008 2008 2 1 2 1 An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. ✓ [29] 2010 5 2 A fuzzy embedded GA is employed to solve fuzzy weighted single objective function employing fuzzy theory for MODGP. ✓ ✓ ✓ ✓ [23] [23] 2007 [23] 2007 2 2007 1 2 1 2 1 MCS embedded MCS embedded in MCS GA embedded planning in GA planning methodology in GA planning methodology is proposed methodology is proposed to improve is proposed to improve the accuracy to improve the accuracy for stochastic the accuracy for stochastic DG for integration stochastic DG integration with DG tradeoff integration with tradeoff solution. with 16 tradeoff solution. GA-PSO hybrid GA-PSO is proposed is proposed to solve MODGP to solve MODGP problemproblem regarding regarding DG location DG location (with GA) (with andGA) capacity and capacity (with PSO) (with respectively. PSO) respectively. [31] [31] 2012 2012 3 Energies 3 2017, 10, - ✓ - - - - A hybrid [25] [25] 2008 2008 6 16 of 44 solution. [28] 2009 4 1 208 DGP--proposed DGP ✓2008 -- -✓ proposed NSGA NSGA finds arrangements finds arrangements foradvocated wind-based for wind-based DGs regarding DGs regarding compromise compromise solution solution among contrasting among contrasting objectives. objectives. [27] [27] 2008 2008 3 The 13 1 GA also used for comparison of results and for large DNW systems even with suboptimal solutions. ✓ 1- 3 of MCS-embedded GA isalgorithm presented to solve chance constrained programming framework considering future uncertainties of loads/DGs. [30] 2011 5contribution ✓ attainment - The -is - NSGA-II - algorithm [24] [24] 2008 [24] 2008 3 13 - ✓ 1-- The NSGA-II along with NSGA-II along max-min with along max-min approach with max-min approach solves MODGP approach solves MODGP problem solves MODGP problem considering problem considering future considering load future uncertainties load future uncertainties load and uncertainties riskaand management risk management risk management Table 7. MOCRU-based planning methods. (MOCRU: MOP based on MOP based on VAR compensation and power quality.). A -goal A -goal attainment method (GoA) method has (GoA) presented, has presented, where goal where programming goal programming transforms transforms multiple multiple objective objective functions functions into single into objective aand single objective [32] [32] 2012 2012 5 5 ✓1 --withGA GA with programming goal programming methodology methodology finds a with solution finds aproblem solution with associated with associated uncertainties uncertainties among MO among and MO constraints. and constraints. [26] [26] 2008 2008 2 12 An✓ exhaustive An exhaustive search search analysis (ES) has (ES) presented has presented IMO with for(wind) IMO MODGP for MODGP problem problem under variable under variable load conditions. load conditions. ✓ - -- - -- function, [29] 2010 5 2 fuzzy embedded GA isby toconsidering solve fuzzy weighted single objective function employing fuzzy set theory forrespectively. MODGP. Agoal hybrid GA-PSO is proposed totoby solve MODGP regarding DG location (with GA) and capacity (with PSO) [31] 2012 3 1 -- ✓ which isEA which further is solved further solved GA ensure GA toIMO optimal ensure DG optimal DG planning. (wind) planning. EA is employed isanalysis employed to EA solve is employed IMO toemployed solve considering IMO to solve both considering time-varying both time-varying both generation time-varying generation and demand generation and demand behavior and demand behavior aiming behavior ataiming various ataiming technical various at technical various impacts technical of impacts of impacts of ✓ ✓ - 6 -- -✓ 1-- of- MODGP-based --✓ - function, [28] [28] 2009 Energies 2009 4 Energies 1Table 4 1 5. The contribution planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). 2017, 10, 2017, 208 10, 208 16 of 44 16 of 44 ✓ [25] [25] 2008 [25] 2008 6 2008 1 6 1 ✓ The proposed The proposed NSGA algorithm NSGA algorithm finds arrangements finds arrangements for wind-based for wind-based DGs regarding DGs regarding compromise compromise solution solution among contrasting among contrasting objectives. objectives. [27] [27] 2008 3 1 3 GA used is also for comparison for comparison of results of and results advocated and advocated for large for DNW large systems DNW systems even with even suboptimal with suboptimal solutions. solutions. ✓ - GA -is also MCS-embedded GA istwo presented to solve chance constrained programming framework considering future uncertainties loads/DGs. [30] 2011 5 1 A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a of single objective DGP DGP DGP MODGP MODGP problem problem isused solved is by solved stage by two heuristic stage heuristic iterative iterative method including method including clustering clustering (outer) and (outer) EA (inner) and EA optimizations (inner) optimizations respectively, respectively, ✓ -- ✓ - D 2012 1 ✓ - -- An [33] [33] 2012 2 1 2 PTC A--MO C Contribution of function MOCRU-Based Planning exhaustive An exhaustive search analysis search analysis (ES) has (ES) presented has presented with IMO with for IMO MODGP for MODGP problem problem under variable under variable load conditions. load conditions. ✓2008 ✓ 1-- 2BPTC ✓ [29] Ref.[32] [29] 2010Year[26] 2010 5 MO 25 5 2 A fuzzy embedded fuzzy embedded GA isisGA employed GA is employed to solve fuzzy to solve weighted fuzzy weighted single objective single objective function employing employing fuzzy set fuzzy theory set for theory MODGP. forrespectively. MODGP. A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) [31] 2012 3 1 Ref. Year A B C D Contribution of MODGP-based Planning function, which further solved by GA to ensure optimal DG (wind) planning. to find time to find varying time voltage varying magnitude voltage magnitude and loss and sensitivity loss sensitivity factor at factor each node. at each node. ✓ ✓ ✓ GA with GA goal with programming goal with programming goal methodology programming methodology finds methodology a solution finds a solution with finds associated a solution with associated uncertainties with associated uncertainties among uncertainties MO among and MO constraints. among and MO constraints. and constraints. [26] 2008 [26] 2008 2 1 2 1 ✓ [28] [28] 2009 2009 4 14 Table 5.1Table The ✓ contribution 5.10, The of MODGP-based of MODGP-based planning planning methods methods and related and related works. (PTC: works. Components (PTC: Components inDNW planning intransforms planning techniques; techniques; MODGP: MODGP: MO based MO DG based placement). DG placement). GA GA used is also for used comparison for comparison of of and results advocated and advocated large for DNW large systems systems even with even suboptimal with suboptimal solutions. solutions. ✓ - contribution -is✓also MCS-embedded GA presented GA is presented toresults solve chance to solve constrained chance constrained programming programming framework framework considering considering future uncertainties future uncertainties loads/DGs. loads/DGs. [30] [30] 2011 2011 5 Energies 5 1 16 of fuzzy 44 MODGP formulation based on GA and εfor constrained method is proposed to solve the best compromise (tradeoff) DM A goal attainment method (GoA) has presented, where goal programming multiple objective functions into anonsingle objective MOO methods like non-dominated sorting genetic algorithm (NSGA) and strength Pareto evolutionary algorithm (SPEA) with c-means problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, The multi-criteria planning planning model aims model to achieve aims to contrasting achieve contrasting objectives objectives (cost and (cost reliability) and reliability) for DGP. for GA DGP. further GA finds further aof set finds ofsolution aof set offor non✓ 1- 3 - ---1-✓ 1--- MCS-embedded -proposed -is The proposed NSGA The proposed algorithm NSGA algorithm NSGA finds arrangements algorithm finds arrangements finds for arrangements wind-based for wind-based DGs for wind-based regarding DGs regarding compromise DGs regarding compromise solution compromise solution among contrasting solution among contrasting among objectives. contrasting objectives. objectives. [27] [27] 2008 [27] 2008 3 2017, 2008 13 208 ✓- ---multi-criteria - - MODGP -- The - --fuzzy [33] 2 12005 ✓1 [34] [71] [32] [34] 201320062012 2013 2 2[21] 2 15 1 --- 4✓ ✓ ✓ -- -- ✓ ✓ [29] [29] 2010 2010 5 5 2 embedded embedded GA isisemployed GA isreinforcement employed to solve fuzzy to solve fuzzy weighted single objective single objective function function employing fuzzy fuzzy theory set for theory MODGP. forrespectively. MODGP. -- -- A fuzzy hybrid A GA-PSO hybrid GA-PSO is(DISCO). proposed isgrid proposed to solve MODGP to solve MODGP problem problem regarding regarding DG location DG location (with GA) (with andemploying GA) capacity andset capacity (with PSO) (with respectively. PSO) [31] [31] 2012 2012 3 1 3 1 (FCM) solve planning problem with high/low-reliability solutions for urban/rural cases respectively. which further solved by GA to weighted ensure optimal DG planning. to find varying voltage magnitude and loss sensitivity factor at each node. dominant solutions solutions for wind-based for wind-based DG units DG sizing, units siting, sizing, and siting, maintenance and maintenance schedules. schedules. An exhaustive An exhaustive search An exhaustive analysis search analysis (ES) search has analysis (ES) presented has (ES) presented with has IMO presented with for IMO MODGP with for IMO MODGP problem for MODGP problem under variable problem under variable load under conditions. variable load conditions. load conditions. Ref. Ref. Year Year MO PTC MO A PTC B dominant A C BDfunction, C Dtime Contribution Contribution of(wind) MODGP-based of MODGP-based Planning Planning ✓ 1- 4 -1-✓ 1- MCS-embedded -✓- attainment - MCS-embedded - - -attainment - GA -is presented [28] [28] 2009 [28] 2009 4 2009 1The 4 - contribution ✓ GA ismethods presented to solve chance to solve constrained chance constrained programming programming framework framework considering considering future uncertainties future uncertainties of loads/DGs. of loads/DGs. [30] [30] 2011 2011 5 1 5 ✓ [22] 2005 Similarly, with the same formulation (GA and εmulti-objective constrained method), aclustering double tradeoff method isGA presented to find the alternative. A A method (GoA) method has (GoA) presented, has presented, where goal programming goal programming transforms transforms multiple multiple objective objective functions functions into asolutions. single into objective a16 single objective 5. of-goal MODGP-based planning and related works. (PTC: Components in planning techniques; MODGP: MO based placement). MODGP problem is solved by two stage iterative method including (outer) and EA (inner) optimizations respectively, The multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) for DGP. further finds aDG set of non✓ -- 3✓ [35] [35] 2013 2013 2 1Table 2 1 MO planning An-goal MO planning framework framework has developed, has developed, namely improved namely improved multi-objective harmony search (IMOHS), search (IMOHS), which can which evaluate can evaluate the DGP. the DGP. Energies 2017, 10, 208 10, ofbest 44 16 of 44 optimal device GA is also GA used is also for GA used comparison is also for used comparison of for results comparison of and results advocated ofconstrained and results advocated for and large advocated for DNW large systems for DNW large systems even DNW with systems even suboptimal with even suboptimal with solutions. suboptimal solutions. MODGP MODGP formulation formulation based on based GA and on GA εheuristic constrained and εwhere method is method proposed isharmony proposed to solve the to solve best compromise the best (tradeoff) (tradeoff) solution solution for DM for DM GA based Optimal electric distribution resource planning (OEDSEP) procedure using acompromise hybrid energy hub concept aims at solving ✓ ✓ -- ✓---1-- - --- An [32] [72] [32] 20122008 2012 5 Energies 5 2017, 1 208 [33] ✓ [34] 2013 ✓ hybrid - --- - - function, -hybrid -time 20051 4 1 - which [23] 2007 embedded inGA GA methodology is proposed toDG improve the accuracy forand stochastic DG(with integration with tradeoff solution. ✓ - - -A A GA-PSO GA-PSO isMCS proposed isfor proposed towind-based solve MODGP totoplanning solve MODGP problem problem regarding regarding DG location location (with GA) (with and GA) capacity capacity (with PSO) respectively. PSO) respectively. [31] [31] 2012 [21] 2012 3 3[21] 12 3 22005 1 -- 142✓ is which further isA solved further by solved by ensure GA to optimal ensure DG optimal (wind) DG planning. (wind) planning. to find varying voltage magnitude and loss sensitivity factor at each node. dominant solutions DG units sizing, siting, and maintenance schedules. (DISCO). (DISCO). An MO optimization An MO optimization problem problem for multiple for multiple micro-turbines micro-turbines (DG) placement (DG) placement and sizes and is solved sizes is by solved hybrid by PSO hybrid and PSO SFL and algorithms SFL algorithms based based ✓ 2- 5 PTC ✓ ✓ ✓ ✓ - ✓ 2- function, allocation and replacement problem (ODARP) by decomposition into three subproblems to attain multiple objectives. [29] [29] 2010 Ref. [29] 2010 5 Year 2010 25 MO A fuzzy A embedded fuzzy embedded fuzzy GA is embedded employed GA is employed GA to solve is employed fuzzy to solve weighted fuzzy to solve weighted single fuzzy objective weighted single objective function single objective function employing function employing fuzzy employing set fuzzy theory set for fuzzy theory MODGP. set for theory MODGP. for MODGP. A B C D Contribution of MODGP-based Planning ✓ 1 - 3✓ - -1- - A [36] [36] 2013 2013 3 [24] 13 2008 ✓ goal- attainment - MO - planning NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties andoptimizations management method (GoA) method has (GoA) presented, has presented, where where programming goal programming transforms transforms multiple multiple objective objective functions functions into arisk single into objective aalternative. single objective MODGP problem problem is solved is by solved two stage by heuristic stage heuristic iterative iterative method including method including (outer) and (outer) EA (inner) and EA (inner) respectively, respectively, The planning model aims to achieve contrasting objectives (cost and reliability) for DGP. GA further finds aDG of non--1-✓- 1-- MODGP --✓ [35] 2013 2 1 An framework has developed, namely improved multi-objective harmony search (IMOHS), can evaluate the DGP. ✓ - -- A --goal -attainment [22] [22] 2005 2005 35. 3✓ Similarly, Similarly, the with same the formulation same formulation (GA and (GA εgoal constrained and εselect constrained method), method), aclustering double aclustering tradeoff double tradeoff method is method presented isoptimizations presented towhich find to best find the best alternative. framework, followed followed by and fuzzy by decision-making fuzzy decision-making tool to select tool to the most the preferred most preferred Pareto optimal Pareto solution optimal solution satisfying satisfying competing competing objectives. objectives. -multi-criteria - MCS-embedded -with MCS-embedded MCS-embedded GA is presented GA istwo presented to GA solve is presented chance to solve chance to solve constrained chance programming constrained programming framework programming framework considering framework considering future considering uncertainties future uncertainties future ofset loads/DGs. uncertainties loads/DGs. [30] [30] 2011 [30] 2011 2011 52017, 1- 5✓ Table 5.52The contribution of MODGP-based of-- MODGP-based methods planning related methods works. and (PTC: related Components works. (PTC: in Components planning techniques; in planning techniques; MO based MODGP: DG placement). MOthe based placement). ✓ ✓ - 1contribution [32] [32] 2012 2012 1Table 5 1 [33] [33] 2 Energies Energies 2017, 10, 208 10, 208 16 of of loads/DGs. 44placement 16 of 44 ✓ - framework, - planning MODGP formulation based on GA and εconstrained constrained method isplanning. proposed toMODGP: solve the best compromise solution for of DM [34] 2013 1The Fuzzy multiobjective model and ant colony optimization (ACO) based approach provides decision support in sectionalizing (SSP) EA is is employed toGA solve IMO considering both time-varying generation and demand behavior aiming at(tradeoff) various technical impacts ofswitch is which further solved further by solved toplanning by ensure GA to optimal ensure DG optimal (wind) DG planning. (wind) to to find varying voltage varying magnitude voltage magnitude and loss and sensitivity loss sensitivity factor at factor each node. at each node. ✓ -- time ---hybrid --time dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. -- 1426✓ 1- 3✓ - -11-✓- 1- function, - which - -methodology - function, [23] [23] 2007 2007 22012 MCS embedded MCS embedded in GA planning in GA methodology methodology isMODGP proposed isMODGP proposed to improve toDG improve thelocation accuracy thesizes accuracy for stochastic for stochastic DG integration DG integration with tradeoff with tradeoff solution. solution. An MO optimization problem for multiple micro-turbines (DG) placement and solved by hybrid PSO and SFL algorithms based ✓-find ✓ [25] 2008 [37] [73] [31] [37] 20132009 2013 3 2[21] 13 12005 The The methodology based on based Pareto on frontier Pareto differential frontier differential evolution evolution (PFDE) algorithm (PFDE) algorithm is proposed is is proposed for optimal for MODGP optimal MODGP (sizing and (sizing location). and location). ✓ A A GA-PSO hybrid A GA-PSO is hybrid proposed GA-PSO is proposed to solve is proposed MODGP to solve to problem solve problem regarding problem regarding regarding DG location (with DG GA) location (with and GA) capacity (with and GA) capacity (with and PSO) capacity (with respectively. PSO) (with respectively. PSO) respectively. [31] 2012 [31] 2012 1 3 (DISCO). problem considering real planning requirements in DNW with DGs. ✓B PTC -C D - MODGP -multi-criteria [36] Year 2013 Ref. 3 PTC 1 AMO DGP Ref. MO Year A B C D Contribution of Contribution MODGP-based of Planning MODGP-based Planning MODGP problem problem is solved is by solved two stage by two heuristic stage heuristic iterative iterative method including method including clustering clustering (outer) and (outer) EA (inner) and EA optimizations (inner) optimizations respectively, respectively, The The multi-criteria planning planning model aims model to achieve aims to contrasting achieve contrasting objectives objectives (cost and (cost reliability) and reliability) for DGP. for GA DGP. further GA finds further a set finds of nona set of non✓ ✓ - --DGP [35] 2013 2 2008 1 - 13 - ✓--1- - -- The MO has developed, namely improved harmony search (IMOHS), which can evaluate the DGP. - - An - NSGA-II - planning [24] [24] 200812 3 1 NSGA-II along with along max-min with max-min approach approach solves MODGP solves MODGP problem problem considering considering future load future uncertainties load uncertainties andmultiple risk and management risk management framework, followed by fuzzy decision-making tool to the most preferred Pareto optimal solution satisfying competing objectives. The problem DGP for location, for location, size, and size, type and (REG type and (REG PEV) and ispresented, defined PEV) is multi-objective as defined MO mixed as MO integer mixed nonlinear integer nonlinear programming programming (MINLP), (MINLP), in which in which Aproblem goal attainment Aframework method goal attainment (GoA) method has (GoA) method presented, has (GoA) presented, where has goal where programming goal where programming goal transforms transforms multiple transforms multiple objective objective functions objective functions into athe single functions into objective aDG single into objective a single objective [33] [33] 2012 2012 2 ✓ - A --goal --attainment 3 Similarly, with the same formulation (GA and εselect method), aprogramming double tradeoff method is presented to find best alternative. ✓ [34] [34] 2013 2013 5. Table 5. -The11contribution of-find MODGP-based of MODGP-based planning planning methods methods and related and related works. (PTC: works. Components (PTC: Components in planning in planning techniques; techniques; MODGP: MODGP: MO MO DG based placement). placement). ✓ GA with goal programming methodology finds aconstrained solution with uncertainties among MO and constraints. [26] 2008 2✓ ✓ [38] [38] 2014 2014 2 [22] 1Table 2 2005 ✓ --✓- time - dominant -find -varying - hybrid -varying MODGP formulation MODGP based on GA and based ε multiple constrained on GAboth and method εsensitivity constrained isevolution proposed method to solve is associated proposed the best compromise to solve thebehavior best (tradeoff) compromise solution for based DM solution forlocation). DM [32] [32] 2012 [32] 2012 5 2012 1The 5 - contribution A method based on strength Pareto evolutionary algorithm (SPEA2) with distribution OPF solves DER placement problem with network to to time voltage magnitude voltage magnitude and loss and sensitivity loss factor at factor each node. at each node. solutions for wind-based for wind-based DG units DG sizing, units siting, sizing, and siting, maintenance and schedules. schedules. EA is employed EA is employed toformulation solve IMO to solve considering IMO considering time-varying both time-varying generation generation and demand and demand behavior aiming ataiming various at(tradeoff) technical various technical impacts impacts of An MO problem for micro-turbines placement and sizes isproposed solved by hybrid PSO and SFL algorithms based ✓ 15 - --11- ✓- 1---dominant [37] 2009 2013 The methodology based on Pareto frontier differential (PFDE) is for optimal MODGP (sizing Energies 10,En 208 16 tradeoff ofand 44of o an is used to isobtain used to compromise obtain compromise (Pareto frontier) (Pareto frontier) solution solution for amaintenance local for distribution a algorithm local distribution company company (LDC). (LDC). ✓ -- - function, -- NSGA-II -- optimization [21] 2005 4 3 16 1 function, which is function, which further issolved which further is by solved further GA to by ensure solved GA toby optimal ensure GA to(DG) DG optimal ensure (wind) DG optimal planning. (wind) DG planning. (wind) planning. [74] 2[21] 22005 - solutions [23] 2007 MCS embedded in GA planning methodology is proposed to improve the accuracy for stochastic DG integration with solution.o ✓ -NSGA-II [25] [25] 2008 2008 - an [36] 2017, 2013 3 En 1 ✓ 14263✓- ✓ The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] (DISCO). (DISCO). reinforcements to assess the trade-off solutions among DER and DSO costs in the long-term grid planning schemes. The multi-criteria The multi-criteria planning planning model aims model todeveloped, achieve aims to contrasting achieve contrasting objectives objectives (cost and (cost reliability) and reliability) for DGP. for GA DGP. further GA finds further atypes, set finds of nonaDGP. setand of nonDGP DGP ✓ 1 -Year ✓ - PTC -MO - - The [35] [35] 2013 2013 2 Ref. 1 2 An MO planning An MO planning framework framework has developed, has namely improved namely improved multi-objective multi-objective harmony harmony search (IMOHS), search (IMOHS), which can which evaluate can evaluate the the DGP. framework, followed by fuzzy decision-making tool to select the most preferred Pareto optimal solution satisfying competing objectives. Ref. Year MO A PTC B A C B D C D Contribution Contribution of MODGP-based of MODGP-based Planning Planning The DGP problem for location, size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), in which MO probabilistic MO probabilistic (mathematical (mathematical programming) programming) framework framework has proposed has proposed for various for DER various planning DER planning (six DG types, (six DG size, and size, location) location) ✓ -- - MODGP problem MODGP problem isalong solved problem isby solved two is stage by solved two heuristic stage by two heuristic iterative stage heuristic iterative method iterative including method including method clustering including clustering (outer) clustering and (outer) EAload (inner) and (outer) EAoptimizations (inner) andmanagement EAoptimizations (inner) respectively, optimizations respectively, respectively, - MODGP 2008 NSGA-II with max-min approach solves MODGP problem considering future loadvariable uncertainties and risk - 3✓-- --11-- ---✓ [34] [22] [34] 2013 2005 2013 2 [24] 12 1 An exhaustive search analysis has presented with for MODGP problem under conditions. [38] 2014 2 1 --- GA --with 3[33] 1✓ Similarly, with the Similarly, same formulation with the same (GA formulation and ε(ES) constrained (GA and method), εmaintenance constrained a IMO double method), tradeoff aschedules. method double tradeoff is presented method to find is presented the best alternative. to find the best alternative. ✓ 12✓--1 ✓- 1-- dominant ✓-- - optimization -methodology - optimization -isproblem [33] [26] [33] 2012 [22] 2012 2 2005 12 ✓ 1324✓ dominant solutions solutions for wind-based for wind-based DG units DG sizing, units siting, sizing, and siting, and maintenance schedules. ✓-- MO GA goal with programming goal programming methodology methodology finds solution finds solution with associated with associated uncertainties uncertainties among MO among and MO constraints. and constraints. [26] 2008 2008 22012 An An MO problem for multiple for multiple micro-turbines micro-turbines (DG) placement (DG) placement and sizes and solved sizes by solved hybrid by PSO hybrid and PSO SFL and algorithms SFL algorithms based based [28] 2009 ---aims [37] 2013 3 1 The based on Pareto frontier differential evolution (PFDE) algorithm is for optimal MODGP (sizing and MODGP MODGP formulation formulation based on GA and on εaand constrained and εak constrained method is method isann proposed to solve the to solve best compromise the best compromise (tradeoff) (tradeoff) solution solution for DM for DM The multi-objective ant colony (MACO) applied for placement of switches and protective devices inwith network addressing an NSGA-II used to obtain compromise (Pareto frontier) solution for aproposed local distribution company (LDC). ✓ ✓ [39] [39] 2014 2014 1Tab 2 [21] 1The that DISCOs aims at DISCOs contribution contribution in the competitive in the competitive electricity electricity market. Moreover market. Moreover modified modified augmented augmented ε-constrained method along method along with EA is employed to solve IMO considering both time-varying generation and demand behavior aiming atsolutions. various technical impacts to time to find varying time to find voltage varying time magnitude voltage varying magnitude voltage loss magnitude sensitivity loss and sensitivity factor loss sensitivity at factor each node. at factor each at each node. Table 5.32The contribution e1✓ 5Tab e4✓ MODGP-based 5on--The on obuMCS MODGP on oat-find MODGP methods ed p ba ann and ed ng p related ann me hod ng works. me and hod (PTC: eoptimization abased and ed Components wo e aGA k ed wo PTC in Componen planning PTC Componen techniques; n pis nnode. ng MODGP: pisproposed ann eDG hn ng que MO eε-constrained hn based MODGP que DG MODGP placement). MO ba MO ed DG ba pdistribution ed a location). DG emen p of a emen ✓ GA also used for comparison of results and advocated for large DNW systems even with suboptimal --2005 --11---4on -- that - planning [36] [36] 2013 2013 3 1 ✓ -- -✓ -ba --proposed - in -is 2005 2[21] -of -bu [23] 2007 [23] 21 embedded MCS GA embedded planning methodology in GA planning is proposed methodology tofor improve is proposed the accuracy to improve for stochastic the accuracy for integration stochastic with DG tradeoff integration solution. with tradeoff solution. ✓ 1- MO - The [25] 2008 The proposed NSGA algorithm NSGA algorithm finds arrangements finds arrangements wind-based for wind-based DGs regarding DGs regarding compromise compromise solution solution among contrasting among contrasting objectives. objectives. [27] [27] 2008 3 1 ✓ ✓ - 1263✓-- ✓ - - - --An [35] [75] [35] 20132009 2013 2 En 12 22007 planning An MO planning framework framework has developed, has developed, namely improved namely improved multi-objective multi-objective harmony harmony search (IMOHS), search (IMOHS), which can which evaluate can evaluate the DGP. the DGP. o framework, framework, followed followed by fuzzy by decision-making fuzzy decision-making tool to select tool to the select most the preferred most preferred Pareto optimal Pareto solution optimal solution satisfying satisfying competing competing objectives. objectives. (DISCO). (DISCO). Pareto optimal based non-dominated multi-objective solutions. The DGP problem for location, size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), in which MO probabilistic (mathematical programming) framework has proposed for various planning (six DG types, size, and location) FDM has FDM utilized has for utilized best compromise for best compromise solution solution among among set ofthe Pareto of optimal Pareto solutions. optimal solutions. DGP multi-criteria The planning multi-criteria planning model planning model to achieve aims model to contrasting achieve aims to set contrasting achieve objectives contrasting objectives (cost and objectives (cost reliability) andDER (cost reliability) for and DGP. reliability) for GA DGP. further for GA DGP. finds further GA aMODGP. set finds further of nona set finds of nona set of non✓ ✓ - The -- along [29] 2010 5 2 A multi-criteria fuzzy embedded GA isaims employed tothe solve fuzzy weighted single objective function employing fuzzy set theory for ✓B-12 PTC -C -✓ ✓ [38] Year 2014 R 2 PTC 12 ✓ - The [24] 2008 [24] 3[34] 2008 1 2013 3✓ 1- ✓ 1D NSGA-II with NSGA-II max-min along approach with max-min solves MODGP approach problem solves MODGP considering problem future considering load uncertainties future load and uncertainties risk management and risk management ---The -1 MO -methodology -isproblem [34] [34] 2013 2013 2 1 An exhaustive exhaustive search analysis search analysis (ES) has (ES) presented has presented with IMO with forbu IMO for MODGP problem problem under variable variable load conditions. load conditions. Ref. MO Y MO A PTC B--- -✓ A C B- D C D Contribution of Con MODGP-based Con on obu MODGP on o distribution MODGP bmodified d Punder b nn d ng P nn ng optimization An MO problem for multiple for multiple micro-turbines micro-turbines (DG) (DG) placement and sizes and is solved sizes by solved hybrid by PSO hybrid and PSO SFL and algorithms SFL algorithms based based ✓ ---NSGA-II ---εoptimization - An -with [22] 2005 2005 1 Similarly, Similarly, with the with same the formulation same formulation (GA and (GA εplacement and εMODGP constrained method), method), aproposed double tradeoff double tradeoff method is method presented is presented to find the to best find the best alternative. ✓ ---methodology [37] [37] 2013 2013 3 The based on based Pareto on frontier Pareto differential frontier differential evolution evolution (PFDE) algorithm (PFDE) algorithm is isaisproposed for optimal for MODGP optimal MODGP (sizing and (sizing location). and alternative. location). an used to obtain compromise (Pareto frontier) solution for aPlanning local company (LDC). ✓ --- ✓----MO [39] 2014 2 R 1 A---Y that at DISCOs contribution in the competitive electricity market. Moreover augmented method ✓ GA goal programming methodology finds aconstrained solution with associated among MO and constraints. [26] 2008 2 1---3 - --- An ✓ [40] [40] 2014 [28] 2 [22] 13 MO MO augmented constrained constrained method proposed method proposed for for planning DGP planning on short on term short basis term for basis future for DNWs future with DNWs ANM functionalities. ANM - EA - dominant [28] 20091 2009 4 1 13 4✓ solutions dominant solutions for wind-based solutions for wind-based DG for units wind-based DG sizing, siting, DG sizing, units and siting, sizing, maintenance and siting, maintenance and schedules. maintenance schedules. schedules. ✓ - is - aims - dominant MCS-embedded GA ised presented toDGP solve chance constrained programming framework considering future uncertainties loads/DGs. [30] 2011 5 1 ✓ ✓ Mu ob eεemployed ve PSO ba pboth ann ng punits opo ed ba ed on wo age nuncertainties he fi age Pasolutions. e ε-constrained owith opat ma yfunctionalities. p n pofealong u with edofo ob a n adeo -augmented [36] [36] 2013 2013 3 1 3 1 employed to EA solve is IMO considering to solve IMO considering time-varying both generation time-varying and demand generation behavior and demand aiming behavior at various aiming technical various impacts technical of impacts GA is also GA used is also for used comparison for comparison of results of and results advocated and advocated for large for DNW large systems DNW systems even with even suboptimal with suboptimal solutions. MODGP formulation MODGP based MODGP o on mu GA o and on mu b ε constrained d on on b GA d nd on method GA on nd is proposed n on d m to hod n solve d m p the hod opo best d p compromise opo o o v d h o o b (tradeoff) v h omp b solution om omp om for d DM o o d u o on o u DM on o DM ✓ ✓ framework, framework, followed followed by fuzzy by decision-making fuzzy decision-making tool to select tool to the select most the preferred most preferred Pareto optimal Pareto solution optimal solution satisfying satisfying competing competing objectives. objectives. [23] [23] 2007 2007 2 1 2 1 MCS embedded MCS embedded in GA planning in GA planning methodology methodology is proposed is proposed to improve to improve the accuracy the accuracy for stochastic for stochastic DG integration DG integration with tradeoff with tradeoff solution. solution.o ✓ 3 The problem problem for location, for location, size, and size, type and (REG type and (REG PEV) defined PEV) is as defined MO mixed as MO integer mixed integer programming programming (MINLP), (MINLP), incan in which [25] 2008 [25] 2008 6 Tab e11 52013 o- MODGP ba ed p ann ng me hod and ehas ato ed wo ksorting PTC Componen p(NSGA-II) ann ng e(with hn que MODGP MO ba ed DG pevaluate aoperation emen MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) ✓ FDM has utilized for best compromise solution among the set of Pareto optimal The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] 3on--12 bu 1- on-The En[31] MOO The MOO framework framework based on based non-dominated on non-dominated genetic sorting algorithm genetic algorithm II II nonlinear (NSGA-II) along with along MCS with MCS uncertain (for uncertain operation [21] 2005 ✓-- -✓ --DGP - MO -DGP - hybrid [35] [35] 2013 2013 2 1The 2✓ An An MO planning An framework MO planning framework developed, framework developed, namely has developed, improved namely improved namely multi-objective improved multi-objective harmony multi-objective harmony search (IMOHS), harmony search (IMOHS), search which (IMOHS), can which evaluate which the DGP. can the ✓ [76] 2010 2, 4 46[35] 32012 ana yplanning among wo onfl ng ob esolve ve oand ais o and CLL nnobjective he esolutions. ond age anonlinear ou ob e(for ve have op mwhich zed Ea hevaluate o ouDGP. ontheepDGP. e en DG --planning --- planning A GA-PSO ishas proposed MODGP problem regarding DG location GA) and capacity (with PSO) respectively. DGP DGP ✓ ✓ -- ✓---12---✓- 1----The - 25✓ [38] [38] 2014 2014 2 2 En ✓ ✓ -- ✓✓ -- NSGA-II 20101 2010 5 1 Ais A embedded embedded GA is employed GA is employed tomax-min solve fuzzy toevolution solve weighted fuzzy weighted single objective single function function employing employing fuzzy set fuzzy theory setfor theory MODGP. forrisk MODGP. ✓ (DISCO). Disfuzzy SCO D SCO ---methodology [41] [41] 2014 2014 2 1 1 ✓ 1-NSGA-II -fuzzy - at - NSGA-II - on [24] 2008 3✓ NSGA-II along with along max-min with approach approach solves MODGP solves MODGP problem problem considering considering future load future uncertainties load uncertainties and risk management management ✓ ✓ - --1-3 -- an [37] [37] 2013 [29] 2013 3 [29] 12 3 [24] 1 --2008 The The methodology based based Pareto on frontier Pareto differential frontier differential evolution (PFDE) algorithm (PFDE) algorithm is proposed is proposed for optimal for MODGP optimal MODGP (sizing and (sizing location). and location). an used to obtain used to compromise obtain compromise (Pareto frontier) (Pareto frontier) solution solution for a local for distribution a local distribution company company (LDC). (LDC). ✓ [39] 2014 2 1 that aims DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. [40] MO augmented ε constrained method proposed for DGP planning on short term basis for future DNWs with ANM functionalities. scenarios) scenarios) and OPF, and addresses OPF, addresses DGP problem DGP problem of REG and of REG storage and integration storage integration within DNW. within DNW. numbe o a on ze and ype b an he ondu o ze e ona z ng w he numbe and o a on and ne wo k opo ogy based AMO goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions intoPSO a single objective An MO optimization An optimization An MO problem optimization problem for multiple problem for multiple micro-turbines forconstrained multiple micro-turbines (DG) micro-turbines placement (DG) placement and (DG) sizes placement and is solved sizes and is by solved sizes hybrid isby solved PSO hybrid and by PSO SFL hybrid and algorithms SFL and algorithms based SFL algorithms based ✓ ✓ GA with goal programming GA with goal methodology programming finds methodology a solution with finds associated a solution uncertainties with associated among uncertainties MO and constraints. among MO and constraints. [26] 2008 [26] 2 2008 1 2 1 [28] 2009 4 R Y MO PTC A B C D Con bu on o MODGP b d P nn ng ✓ MCS-embedded MCS-embedded GA is presented GA is presented to solve chance to solve chance constrained programming programming framework framework considering considering future uncertainties future uncertainties of loads/DGs. of loads/DGs. [30] [30] 2011 2011 5 1 5 1 ✓ ✓ ✓ [22] 2005 3 1 Similarly, with the S m same y formulation S w m h h y w (GA m h h and o mu m ε constrained o on mu GA on nd method), GA on nd a double n on d m tradeoff hod n d m method doub hod is doub presented d o m d hod to o find m p the hod best n p d alternative. o n nd d h o b nd h b n v n v ✓ 13 - ✓ 1- The ✓ - The - -DGP - -DGP - probabilistic - (mathematical EA is employed EAused issize, employed toand solve IMO toand solve considering IMO considering both time-varying both generation generation and demand and demand behavior behavior aiming atsolutions. aiming various at technical various technical impacts of impacts of [36] [36] 2013 [32][36] 2013 3 2012 2013 13 problem location, for location, size, type (REG type and (REG and isadvocated defined PEV) is as defined MO mixed as MO integer mixed nonlinear integer nonlinear programming programming (MINLP), (MINLP), inregarding which in which GA also for comparison of results and for large DNW systems even with suboptimal MO probabilistic MO (mathematical programming) programming) framework framework has proposed has proposed for various for DER various planning DER (six DG types, (six DG size, types, and size, location) and location) FDM has utilized for best compromise solution among the set oftime-varying Pareto optimal solutions. The MOO planning framework based on non-dominated sorting genetic algorithm II (NSGA-II) along with MCS (for uncertain function, which is further solved by GA to ensure optimal DG (wind) Quasi-oppositional Quasi-oppositional teaching teaching learning-based optimization optimization (QOTLBO) methodology, methodology, aplanning. variant aof variant TLBO of proposed TLBO proposed as MOO as regarding MOO ✓ -ba -for -is [25] [25] 1- The Tab e11✓ e1336✓ 5on---The on obu MODGP oproblem MODGP ed p ba ann ed ng p ann me hod ng me hod eMOP adecision-making and ed wo ehodo aPEV) k ed wo PTC k Componen PTC Componen n ann ng p ann hn ng eplanning hn que MODGP MO ba MO ed DG ba aooperation DG emen p aoonemen framework, followed framework, by fuzzy followed by decision-making by fuzzy decision-making tool to tool to the select most tool to the preferred most the preferred most preferred optimal Pareto solution optimal Pareto optimal solution satisfying competing satisfying competing objectives. ✓-2008 ✓ -- -proposed -- A -- framework, NSGA The proposed algorithm NSGA finds arrangements algorithm for arrangements wind-based DGs for regarding compromise regarding solution compromise among contrasting solution among objectives. contrasting objectives. [27] 2008 [27] 3 2008 -11--6on -- on [38] [38] 2014 2014 2 1 MODGP ofollowed mu band d on GA on n(QOTLBO) d mwind-based hod pselect opo d on v ho bque omp om d osolution. o respectively. uo on DM ✓bu The mu ob elearning-based ve pn on ann ng ond p ma yselect DNW ha osingle mu aDGs aPareto an M MO RTS asolution go hm u ed op ve he ob eobjectives. ve p ann ng - --✓ hybrid A GA-PSO hybrid GA-PSO is proposed is proposed to solve MODGP to solve MODGP problem problem regarding DG location DG location (with (with and GA) capacity and capacity (with PSO) (with respectively. PSO) [31] [31] 2012 2012 35Tab ✓ ✓ 2007 21 MCS embedded in MCS GA planning mb MCS dd d mb methodology GA dd d pfuzzy n nn GA ng is pfinds proposed m nn ng m to ogy improve hodo pregarding ogy opo dp accuracy opo ofor mp ov for oped stochastic mp h ov uoGA) heDG yNLP integration u o y h oMODGP with o DG hsatisfying tradeoff n gDG on n w gfor h on ded wobjectives. h ocompeting d umu [41] 2014 2 1The ✓ [42] 77 [42] 20142011 3 En 12 3 En DGP DGP an is used to is obtain used to compromise obtain compromise (Pareto frontier) (Pareto frontier) solution solution forthe amarket. local distribution a dlocal distribution company company (LDC). (LDC). ✓ En[23] o oo u on o - ✓ - NSGA-II A fuzzy embedded GA isin employed to solve fuzzy weighted objective function employing fuzzy set theory MODGP. ✓ ✓ NSGA-II -- 5✓ -- 3 ---2---✓ 1----an [39] [39] 2014 2 1 that that at DISCOs aims atand DISCOs contribution contribution in the competitive the competitive electricity electricity market. Moreover Moreover modified modified augmented augmented ε-constrained ε-constrained method along method with along with 2[29][37] 12010 --aims [40] 2014 2 1 MO augmented εThe constrained method for DGP planning on short term basis for future DNWs with ANM functionalities. scenarios) OPF, addresses DGP problem of REG and storage integration within DNW. MODGP problem is solved byproposed two stage heuristic iterative method including clustering (outer) and EAconditions. (inner) optimizations respectively, optimal location optimal location and sizing and of sizing DGP for of DGP solving for multi-objective solving multi-objective optimal power optimal flow power (OPF) flow problem (OPF) problem for radial for DNW. radial DNW. ✓ ✓ [37] [37] 2013 2013 3 2013 1 3 1 The methodology The methodology based methodology on based Pareto on frontier based Pareto on differential frontier Pareto differential frontier evolution differential evolution (PFDE) evolution algorithm (PFDE) algorithm (PFDE) is proposed algorithm is proposed for optimal is proposed for MODGP optimal for MODGP optimal (sizing and MODGP (sizing location). and (sizing and location). An exhaustive search An exhaustive analysis (ES) search has analysis presented (ES) with has IMO presented for MODGP with IMO problem for MODGP under variable problem load under conditions. variable load D SCO mode o g d e n o emen a ong w h e ona z ng w he SSW p a emen o m n m ze o due o ene gy no upp ed Aalong goal attainment A goal attainment method (GoA) method presented, has presented, where goal where goal programming transforms transforms multiple multiple objective objective functions functions single into am single ✓ ✓ -- PTC -- 2 --✓ [24] 2008 3 1 NSGA-II with NSGA max-min NSGA ong approach wGA hong m solves xwhas m(GoA) hnmethodology m MODGP pp x to mosolve nmethodology hproblem pp ochance vo hMODGP considering osolution vprogramming MODGP pa solution ob future massociated p on ob load dmassociated uncertainties ng on udduP ng oand u d uun risk o management dng nunand nd ninto kam nd n gkobjective m nloads/DGs. n g objective m n location). [33] 2012 GA with GA goal with programming goal programming finds a finds with with uncertainties uncertainties among MO among MO constraints. and constraints. [26] [26] 2008 2008 2 1 1 ✓ ✓ [28] 2009 [28] 4 2009 1 4 R R Y Y MO MO A PTC B A C B D C D Con bu Con on o bu MODGP on o MODGP b b nn d ng P nn MCS-embedded is presented constrained programming framework considering future uncertainties of [30] 2011 5 1 The MO The probabilistic MO probabilistic (mathematical (mathematical programming) programming) framework framework has proposed has proposed for various for DER various planning DER planning (six DG types, (six DG size, types, and size, location) and location) FDM has FDM utilized has for utilized best compromise for best compromise solution solution among the among set of the Pareto set of optimal Pareto solutions. optimal solutions. ✓ ✓ - GA -isMOPSO - MOPSO [32] [32] 2012 2012 5 1 to find time varying voltage magnitude and loss sensitivity factor at each node. The MOO planning framework based on non-dominated sorting genetic algorithm IIa (NSGA-II) along with MCS (for uncertain operation Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, variant of TLBO proposed as MOO regarding algorithm algorithm has been has used been for used MODGP MODGP problem satisfying satisfying objectives objectives (minimize (minimize DISCO cost DISCO and cost maximize and maximize owner profit) owner profit) also used for GA comparison is also used for results comparison and advocated of results for and large advocated DNW systems for large even DNW with systems suboptimal even with solutions. suboptimal solutions. The DGP problem DGP The problem for DGP location, problem for location, size, for and location, type (REG size, type and and PEV) type and is (REG PEV) and is as defined PEV) MO is mixed as defined MO integer mixed as MO nonlinear integer nonlinear integer programming nonlinear programming (MINLP), (MINLP), in which (MINLP), in which ✓ Swhich m y w hNSGA h m o mu on GA nd on d m hod doub d od m hod pv nsolution dng oDG nd hprogramming bin ninomp v in which function, is which further is further by GA to by ensure GA to optimal ensure DG optimal (wind) DG planning. (wind) planning. employed EA solve mp IMO EA oy considering dof mp osolved oy omu vNSGA dfor both MO osolved oproblem vsize, time-varying on MO dand ng on dgeneration h(REG ng m bo vm h and yndefined m ng demand g v nwind-based ng behavior on g n nd d on aiming m nd nd bat m h vmixed nd oab technical hm ng oDG vm impacts ou vof ou mp hn [41] 2014 2 12 --2008 ✓ -- is --- function, -A --- The - to -mu [42] 3 [27] The proposed The proposed algorithm finds arrangements finds arrangements for wind-based for DGs regarding DGs regarding compromise compromise solution among contrasting among contrasting objectives. [27] 2008 3✓ 1-- 2 --1 [43] [43] 2014 4,3 MODGP MODGP ocontribution mu oon b d on on GA d nd on GA on nd nplanning on d hod n d m py basis hod opo dp opo o oand ve(with d h oproblem obque vvarious h omp b om omp om d oPSO) o u oon uemen DM on oobjectives. DM ✓ ---that -1 EA --aims -✓ -aims -contribution [38] 2014 [38] 2014 2 1 1✓ ✓ uzzy ob ve mode palgorithm eed en ed wbo h mod hu fled og eap ng MSFLA go hm a(with op m zdhn ng oo ononp ann ngonumbe and A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location GA) and capacity respectively. [31] 2012 1---3✓ ✓ [39] [39] 2014 2The 11Tab that at DISCOs at DISCOs the competitive the competitive electricity electricity market. Moreover market. modified modified augmented augmented ε-constrained method along method with along with [40] 78[25] [40] MO augmented MO augmented εlocation constrained εed proposed method proposed for DGP for DGP on short on term short term for for DNWs future with DNWs ANM with functionalities. ANM functionalities. ✓ ✓ ✓ 2008 6 Tab e[38] 54,3 bu e12✓ 52014 Tab on The oe35✓ MODGP 5on-Thebu on on ba obu ed MODGP pon ann oto ng MODGP ba me ed hod p ann and ng p emethod ann me aedin ed hod ng wo me and kisin hod PTC ecompromise aband Componen wo ehodo asolve kplanning ed wo PTC Componen pfied PTC ann ng Componen eMoreover hn nobjective p que ann nfuture ng MODGP p ann hn ng MO eε-constrained hn ba MODGP que ed DG MODGP p MO afurther emen ba MO ed DG ba aoo DG The multi-criteria planning model aims to achieve contrasting objectives (cost for DGP. GA finds ap set of scenarios) and OPF, addresses DGP problem of REG and within optimal and sizing DGP for solving multi-objective optimal power flow (OPF) radial DNW. ✓ ✓ addition addition finding to optimal finding generated optimal generated electricity electricity prices in prices aheuristic competitive ink an competitive electrical market. market. --fuzzy -- NSGA-II [29] 2010 5on 2 A embedded A fuzzy GA isconstrained embedded employed GA to solve employed fuzzy weighted to single fuzzy objective weighted function single employing function fuzzy employing set theory for fuzzy MODGP. set theory for MODGP. ✓ ✓ 2011 2[29] 22010 an NSGA-II isba used an NSGA-II to is obtain used to is compromise obtain used to obtain (Pareto compromise frontier) (Pareto frontier) solution (Pareto frontier) solution for aelectrical local solution for distribution a basis local for distribution local company distribution company (LDC). company (LDC). (LDC). MCS mb dd nof GA p nn ng m ogy pstorage opo dintegration ofor mp ov h uDNW. yareliability) o o h DG n gload on w h ded op u aonemen MODGP MODGP problem problem is solved is by solved two stage by two heuristic stage iterative iterative method including method including clustering clustering (outer) and (outer) EAfor (inner) and EA optimizations (inner) optimizations respectively, respectively, DGP DGP ✓ DGP - aan [34] 2013 2 1An exhaustive An exhaustive search analysis search analysis (ES) has (ES) presented has presented with IMO with IMO MODGP for MODGP problem problem under variable under variable load conditions. conditions. D SCO D SCO p emen o e ona z ng w he n d bu on au oma on y em DAS Aframework goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a single objective ✓ 1- MCS-embedded [33] [33] 2012 2012 2 14 2 - ✓1 1- 4 - -The FDM has FDM utilized has for utilized best compromise for best compromise solution solution among the among set of the Pareto set of optimal Pareto solutions. optimal solutions. dominant solutions for wind-based DG units sizing, siting, and maintenance schedules. MOO The MOO framework based on based non-dominated on non-dominated sorting genetic sorting genetic algorithm II (NSGA-II) II (NSGA-II) along with along MCS with (for MCS uncertain (for uncertain operation operation ✓ -- -✓ ---planning -MOPSO --- The -planning -varying 2009 2009 Quasi-oppositional teaching learning-based (QOTLBO) methodology, aproposed variant of TLBO proposed as MOO En[28] En[28] o and location) o MOPSO algorithm has been used MODGP problem satisfying objectives (minimize DISCO cost and maximize DG owner profit) in ✓ MCS-embedded GA is presented to GA solve is presented chance constrained to solve chance programming constrained framework framework future considering uncertainties future of loads/DGs. uncertainties of [30] 2011 [30] 5 2011 1 5 improved An improved MOPSO with preference with preference strategy strategy (IMPSO-PS) proposed is proposed for MODGP for MODGP aim to the aim to optimal achieve capacity optimal capacity and and [32] 2012 1-- D ✓ NSGA ong w h m xfor mprogramming) n ooptimization h(IMPSO-PS) oand vis MODGP palgorithm ob mon on dconsidering ng uMO usystems o achieve d un n nd k m n gregarding m nloads/DGs. to time to find time voltage varying magnitude voltage magnitude and loss and sensitivity loss sensitivity at factor each node. at each node. The MO probabilistic MO probabilistic MO (mathematical probabilistic (mathematical (mathematical programming) framework framework has proposed framework has proposed for has various for DER various planning for DER various planning DER DG planning types, (six DG size, types, (six DG size, location) andvsize, location) ✓ ✓ GA goal programming GA w hyGA methodology w hh og go mm og ng m mm app hodo solution ng m ogy hodo with nd ogy associated ofactor nd u bu on uncertainties w o uprogramming h o w hwith among d un othe d un and constraints. mong MO mong nd on on nnd [26] 2008 21 1✓ ✓B--- PTC -C --An -- B- with [41] [41] 2014 2014 2 2 is The also used is for used comparison for comparison of results of results advocated and advocated for large for DNW large systems DNW even even suboptimal solutions. MO R Y A---YMO A PTC A C Bfind D C D Con bu on oprogramming) Con MODGP Con b on o d bu P MODGP on nn o ng MODGP b dwith Pdn b nn d ng Pdnsuboptimal nn ng [42] 3 [43] 2014 ✓ ✓ Saddresses m Sgo w m hpalso y w m hp h ofinds mu m oon mu on nd GA on nd n on d m hod n d m doub hod doub o with m hod o with m p(six hod nMO p d ondsolutions. nn nd h o band htypes, n v ✓1 [44] R [44] 2014 Y 2 R 14,3 2 PTC function, which is further solved by GA to ensure optimal DG (wind) planning. ✓ ✓ ✓-- --MO -augmented [40] [40] 2014 2014 2 1 2 11 ✓ MO augmented εlocation constrained εand constrained proposed method proposed for DGP for planning planning on short on term short basis term for basis future for DNWs future with DNWs ANM with functionalities. ANM functionalities. [35] 2013 2✓ 1-- -- MO An MO planning framework has developed, namely improved multi-objective harmony search (IMOHS), which candrespectively. evaluate thebnaDGP. scenarios) OPF, and OPF, addresses DGP problem DGP problem of REG and ofGA REG storage and integration storage integration within DNW. within optimal sizing of DGP for solving multi-objective optimal power flow (OPF) problem for radial DNW. ✓ Me hodo ogy omethod aNSGA ve dvMODGP bu on ne wo kDGP ADN dynam expan on pDNW. ann ng ba ed on MOGA Pou opo ed amp egy mob a with mu age --hybrid --and --multi-criteria A GA-PSO A is hybrid proposed GA-PSO to solve is proposed to problem solve MODGP regarding problem DG location regarding (with DG GA) location and capacity (with GA) (with and PSO) capacity respectively. (with PSO) [31] 3 2012 1 3 addition to finding optimal generated electricity prices in ang competitive electrical market. locations. locations. EA mp oy d o o MO on d ng bo h m v y ng g n on nd d m nd b h v o m ng v hn o The The multi-criteria planning planning model aims model to achieve aims to contrasting achieve contrasting objectives objectives (cost and (cost reliability) and reliability) for DGP. for GA DGP. further GA finds further a set finds of nona set of non✓-12 -12-- 5✓ 1---scenarios) ✓ [39] 2012 [39] 2014 [31] 2014 2 [29] 12 ✓2010 that aims that at DISCOs aims that at DISCOs contribution aims at DISCOs contribution in the contribution competitive in the competitive in electricity the competitive electricity market. electricity Moreover market. Moreover market. modified Moreover modified augmented modified augmented ε-constrained augmented ε-constrained method ε-constrained along method with along method along ✓ ✓ The proposed NSGA Th p algorithm opo Th d p opo finds arrangements d NSGA go hm go nd for hm wind-based ng nd m n DGs o m w regarding n nd o b w d nd compromise DG b d g DG d solution ng g omp d among ng om omp contrasting o om u on o mong objectives. u on on mong ng on ob ng v v [27] 2008 3[39] 1 2014 ✓ ✓ [29] 2010 5 2 A fuzzy A embedded fuzzy embedded GA is employed GA is employed to solve fuzzy to solve weighted fuzzy weighted single objective single objective function function employing employing fuzzy set fuzzy theory set for theory MODGP. for MODGP. MODGP o mu MODGP on b MODGP d o on mu GA o on nd mu b on d on on b GA n d d nd on m GA hod on nd p n opo on d m d hod n o d o m v p hod opo h b d p opo o omp o v om d h o o b v d h omp o b om o u omp on om o d DM o o d u o on o u DM on o DM ✓ - MOO ✓ MCS MCS ddbased d mbnisGA dd d p nnn GA ng pnon-dominated mmultiple nn hodo ngsorting mogy hodo psorting ogy opo algorithm dpgenetic opo o mp dalgorithm ov o (NSGA-II) mp h ov u (NSGA-II) hy oisusolved o yalong h MCS oDG hEA nMCS gDG(for on n w g h algorithms on dw operation hrespectively, od u on o u on with MODGP problem solved by two stage heuristic iterative method including clustering (outer) and (inner) optimizations ✓ 1 - Quasi-oppositional -goal -✓ -planning - algorithm [34] 2013 32013 2 An MOmb optimization problem for micro-turbines (DG) placement and sizes by hybrid PSO and SFL based The The MOO planning framework framework on based non-dominated on genetic II II along with with (for uncertain uncertain operation ✓e12 ✓ Quasi-oppositional teaching teaching learning-based learning-based optimization optimization (QOTLBO) (QOTLBO) methodology, aschedules. variant aon of variant TLBO of proposed TLBO proposed as MOO regarding MOO regarding 79 [41] 2011/12 4[34] expan on pattainment ann ng unde gene aES on and demand un eamong aset nmethodology, eoptimal w h op ke DG nMODGP eg amaximize on no as aba aMO on oa single new pobjective oin ea emen on dev e ne wo k MOPSO has been used for MODGP problem satisfying objectives (minimize DISCO cost and DG owner profit) A attainment A method goal (GoA) has method presented, (GoA) where has presented, goal programming where goal transforms programming multiple transforms objective multiple functions objective into aond single functions objective into ✓ - SCO - dominant - FDM [33] 1 2012 2✓ 1-5on An improved MOPSO with preference strategy (IMPSO-PS) is proposed for MODGP with the aim to achieve optimal capacity and DGP dominant solutions solutions for wind-based for wind-based DG units DG sizing, units siting, sizing, and siting, maintenance and maintenance schedules. Tab e 5 Tab The 5 on The bu on o bu MODGP on o MODGP ba ed p ba ann ed ng p ann me hod ng me and hod e a and ed wo e a k ed wo PTC k Componen PTC Componen n p ann n ng p ann e hn ng que e hn que MODGP MO ed DG ba p ed a DG emen p FDM has utilized has FDM for utilized best has compromise for utilized best compromise for best solution compromise solution among the solution among set of the Pareto of the Pareto set of solutions. optimal Pareto solutions. optimal solutions. [36] 2013 3 An exhaustive search An xh analysis u An v xh (ES) u has h v presented n y h n with y h IMO p ES for h n MODGP p d w h n MO problem d w o h MODGP MO under o variable MODGP p ob m load p und ob conditions. m v und b v o d b d on ond on ✓ [41] 2014 2014 2 2 1 ✓ ✓ MCS-embedded MCS-embedded GA is presented GA is presented to solve chance to solve constrained chance constrained programming programming framework framework considering considering future uncertainties future uncertainties of loads/DGs. of loads/DGs. [30] [30] 2011 2011 5 1 1 D D SCO D SCO [42] [28] [42] 3 3 ✓ ✓-- - -[32] 2012 5 4,3 1 1 ✓ ---✓ - ✓ [43] 2014 [32] NSGA NSGA ong w problem hong m xwfuzzy mproblem hof n mREG pp x moand n hREG pp o voloss hMODGP o vto select MODGP p integration ob mp on obeach dm node. ng on DNW. ud Pareto u ngo uoptimal d uun o solution d nun nd n km ndcompeting n gk m m nn gobjectives. m n to find varying voltage magnitude and sensitivity factor at [44] 2009 2 2012 ✓ function, ✓ framework, followed byto decision-making tool the most preferred satisfying 4 1 scenarios) and OPF, and addresses OPF, addresses DGP DGP of storage and integration storage within DNW. within e location onfigu atime on and ew ng The o GA uon on ob ain ned om ed me hod ayvnbasis fie mu pradial efor ob eDNW. ve optimal location optimal and sizing and of sizing DGP for of DGP solving for multi-objective solving multi-objective optimal power optimal flow power (OPF) flow problem (OPF) problem for radial for DNW. is function, solved which is by further GA ensure solved by optimal to DG ensure (wind) planning. DG (wind) planning. addition to finding optimal generated electricity prices aoptimal competitive electrical market. locations. ✓-1- GA GA h pεofh og mm ng m hodo ogy nd ond u on wp hopo o d un mong MO nd on [35] 2013 2013 22014 13 2✓ 12✓-1 13✓- 1- scenarios) MO planning An MO framework framework has developed, has developed, namely improved namely improved multi-objective multi-objective harmony harmony search (IMOHS), search (IMOHS), which can which evaluate can evaluate the DGP. also for comparison GA u ooo results u and oconstrained omp advocated oon uGA for o nd large dvo DNW d dvo systems ong g d even ong with gdoub y DNW suboptimal m n w solutions. h v n o m on o udv on -which -further [40] [35] [40] 2014 [31] [40] 2014 2 [31] 12 ✓2012 MO augmented augmented constrained augmented constrained method proposed method proposed method DGP proposed for planning DGP for planning on short on term basis on term for short future for DNWs basis future with DNWs future ANM with DNWs functionalities. with ANM functionalities. ✓ -- --is✓ ---yAn -used ---hMO -h -w A hybrid A GA-PSO hybrid GA-PSO is proposed ismu proposed solve MODGP to solve MODGP problem regarding regarding location DG location (with GA) (with and GA) capacity and (with PSO) (with respectively. respectively. 2012 S-✓ m w S m mεMO yplanning Soogo w m mu y on w m GA h o m o on n d on nd m hod on nd doub nproblem on d m hod dDNW dplanning oDG m m hod hod doub pnd d od nterm m dv d hod o ofor nd m h hod bncapacity d o nnnd h oPSO) bafunctionalities. nd hthe bnDGP. v n ov EA mp EA oyhdd mp oy oomp vddεh MO ond oon vto on MO dmu ng on bo d huGA ng m bo v hbu y m g vDGP n yn on g nshort nd on m nd bm m h nd oubop bw h mhpoptimal vm ng oubop v mup ng ou vANM hn ou mp hn omp The multi-criteria planning model aims to achieve contrasting objectives (cost and reliability) DGP. GA further finds set of non✓ -- problem -- algorithm [37] 2013 3 1 The methodology based on Pareto frontier differential evolution (PFDE) algorithm isnn proposed for MODGP (sizing and location). R R Y YMO MO A PTC B-- ✓ A C BD Chas D Con Con on omethodology, bu MODGP on ovariant MODGP b dclustering d P b d(outer) ng Pmaximize nn ng Quasi-oppositional Quasi-oppositional teaching teaching learning-based learning-based optimization optimization (QOTLBO) (QOTLBO) methodology, aemploying aun of variant TLBO of proposed TLBO proposed as MOO as regarding MOO regarding MODGP MODGP isGA solved problem by two is stage solved heuristic by two iterative stage heuristic method iterative including method clustering including (outer) and EA (inner) optimizations and EA (inner) respectively, optimizations respectively, ✓ MOPSO MOPSO algorithm been has used been for used MODGP for MODGP problem problem satisfying satisfying objectives objectives (minimize (minimize DISCO cost DISCO and cost and maximize DG owner DG profit) owner in profit) in [34] 2013 2 - PTC 1An improved MOPSO with preference strategy (IMPSO-PS) is proposed for MODGP with the aim to achieve optimal capacity and Th p opo d NSGA go hm nd ng m n o w nd b d DG g d ng omp om o u on mong on ng ob An MO optimization An MO optimization problem problem for multiple for multiple micro-turbines micro-turbines (DG) placement (DG) placement and sizes and is solved sizes is by solved hybrid by PSO hybrid and PSO SFL and algorithms SFL algorithms based based ✓ ✓ ✓ ✓ ✓ ✓ [29] 2010 5 2 A fuzzy embedded A u y is A employed mb u dd y d mb GA to solve dd d mp fuzzy GA oy weighted d mp o oy o v d single u o o y v objective w u gh y d w function ng gh d ob ng v ob fuzzy v on un set mp theory oy on ng mp for u oy MODGP. y ng u h o y y o h MODGP o y o MODGP A wo ep p ann ng a egy p opo ed o op m ze eede a o a on numbe and he ou e e ona z ng w he numbe and o a on The The planning framework MOO planning framework based framework on non-dominated on based non-dominated onwhere non-dominated sorting genetic sorting algorithm genetic II (NSGA-II) algorithm II (NSGA-II) IIobjective with with (for uncertain with uncertain operation operation A goal attainment A goal attainment method (GoA) method has (GoA) presented, has presented, where goal transforms multiple multiple objective functions a MCS single into objective ao single objective MCS mb dd d planning MCS nMOO GA p mb MCS nn ddngd mb mn GA dd hodo d p ogy nnn GA ngbased pp m opo nn hodo ngunits dtype m ogy o hodo mp psorting ov ogy opo hgoal dpgenetic opo oprogramming uismp ydalgorithm ov ooprogramming mp ho hMO ov u mixed hDG y transforms onu go yalong h oon w(NSGA-II) oDG halong nMCS dfunctions gDG o along on noMCS uw ginto on h (for on dwuncertain ooperation h (for d uvwhich oon o u on ✓ ✓- 12 -- ✓--1 -- - ---✓ -- MOO -- The -- -- -✓ 2012 2 14,3 2012 1✓ [42] [42] 2014 [36] 2014 3 [33] 3 1 DGP DGP dominant solutions for wind-based DG sizing, siting, and maintenance schedules. [43] [33] [43] 4,3 ✓ [44] 2014 2 1 ✓ [36] 2013 2013 3 3 1 The DGP problem for location, size, and (REG and PEV) defined as integer nonlinear programming (MINLP), in MODGP MODGP o mu o on mu b d on on b GA d nd on GA on nd n on d m hod n d m p hod opo d p opo o o v d h o o b v h omp b om omp om d o o d u o on o u DM on o DM ✓-1 to -- framework, - An - xhand 2012 5✓ 12 -15✓ 1- optimal -✓--find -✓ - - --varying - location - find [41] 2011 [41] 2014 [32] 2014 2 [32] 12 ✓2012 to voltage time magnitude varying voltage and loss magnitude and factor loss sensitivity at each node. factor each node. location optimal and of sizing DGP for of DGP solving for multi-objective solving multi-objective optimal power optimal flow power (OPF) flow problem (OPF) problem for radial for DNW. DNW. addition totime to optimal finding generated optimal generated electricity electricity prices prices aoby competitive in aoptimal competitive electrical electrical market. market. framework, followed by fuzzy by decision-making fuzzy decision-making tool to select to the select most preferred most preferred Pareto optimal solution optimal solution satisfying satisfying competing competing objectives. objectives. locations. u vfollowed h n yv ES p n w hd MO othe MODGP p ob mPareto und von bkm o dgudm ond -- finding 1 -addition ✓ MCS GA is presented mb MCS dd d mb to GA solve dd p dxw chance GA n p dsensitivity constrained osolved n oin v d oensure hvond programming n o vdtool on hv n nng on framework d og nat do mm p considering og ng mm wo future m uncertainties wo ng on ng loads/DGs. unoon und on nnn oep d 2m oDG DGh 5[41] 1 2014 80[30] 2011/12 2[38] 22014 and esizing nescenarios) numbe n DNW The o on ato egy n ep 1multi-objective on o SPEA2–MOPSO oradial awhich on anoe d w function, function, which which solved further by GA to GA DG optimal (wind) DG planning. planning. ✓ 2 NSGA ong w NSGA hNSGA-II m xhplanning NSGA m ong n pp w ofurther haddresses ong m h ois m haddresses n m MODGP pp xhodo m ohproblem n hu ppp ob m hMODGP on o MODGP p ob uh m u p ob m un ng on um d undng ng okDNW. und dwithin umong un ko d n un n SSW nd n kaum nd nnunevaluate gand km m gthe scenarios) and OPF, and addresses OPF, and DGP OPF, problem DGP of DGP REG and problem of REG storage and of REG integration storage and integration storage within DNW. within DNW. ✓ MCS-embedded ✓ GA w GA go w pused his og go mm p og ng m mm ng m ogy hodo ogy oensure nd uimproved on w o up on o w hd(wind) d un ointegration n un nd MO mong nduof MO on n on - ✓ - scenarios) [35] 2013 2 1An MO framework has developed, namely harmony search (IMOHS), can DGP. an is to obtain compromise (Pareto frontier) solution for a local distribution company (LDC). D SCO D SCO The The planning multi-criteria model aims planning to achieve model aims to achieve objectives contrasting (cost and objectives reliability) (cost for and DGP. reliability) GA further DGP. finds GA set further of nonfinds ainprofit) set ofyp nonMOPSO MOPSO algorithm algorithm has been has used been for used MODGP for MODGP problem satisfying satisfying objectives objectives (minimize (minimize DISCO cost DISCO and cost maximize and maximize DG owner DG profit) owner inomp opreference u d oproblem omp on omcontrasting u nd d ong gn DNW y m vnd w hw m ocapacity u An improved MOPSO MOPSO with with preference strategy (IMPSO-PS) (IMPSO-PS) isgoptimization is proposed for for MODGP with the with aim to achieve aim to achieve optimal capacity and - improved - -multi-criteria -- The --methodology [37] [37] 2013 3 The methodology based based Pareto on Pareto differential frontier differential evolution (PFDE) algorithm (PFDE) proposed isn proposed for optimal for MODGP optimal MODGP (sizing and (sizing and location). ✓ ✓ ✓ A GA-PSO A is hyb proposed doptimization A GA hyb to solve GA p MODGP opo dpstrategy opo o problem o v dproblem MODGP omicro-turbines vm MODGP ob DG m location ob gdMODGP d m (with ng g DG GA) ong and DG ow hthe on (with nd hfor PSO) GA p respectively. ya(inner) w p hon PSO y(inner) w hou PSO plocation). vob y vo [31] 2012 3 4,3 2013 1✓ SPEA2–BMOPSO bdon na y MOPSO o eoregarding ne aheuristic emen MODGP MODGP problem is isteaching by solved two stage two heuristic stage iterative iterative method including method clustering (outer) and (outer) EA and EA optimizations optimizations respectively, respectively, ✓- 13 EA mp oy d GA EA om o vQuasi-oppositional mp EA oyd on dopo mp d oy osolved vPSO ng d bo MO ofrontier oh vfor on MO d vby y ng on ng bo d hproposed npdvo ng bo on voptimization hevolution yn nd g vbmethodology, m yn nd b on g hddoub nalgorithm nd vdis oincluding d on m m nd d bclustering m hv v nd ou oubop bplanning h m hn ng ond vm mp ou vSFL oand hn 2 2013 1 2✓--- ✓-1 1--- - ---An Quasi-oppositional Quasi-oppositional teaching teaching learning-based learning-based optimization learning-based (QOTLBO) (QOTLBO) methodology, aon variant methodology, aGA of variant TLBO aoptimal of proposed variant TLBO of proposed as TLBO MOO proposed as regarding MOO regarding ✓ ✓ Th p opo phPSO NSGA d NSGA go hm go nd hm ng nd npm ng o m w npm nd o w dng nd DG b gand DG dcapacity ng gomp d ng om omp o om on omong on mong ng onmp v vob An MO problem multiple (DG) placement sizes solved hybrid PSO and based - --hybrid [43] [43] 2014 2013 2014 4,3 [44] [34] [44] 2 [34] 12 1 The MO probabilistic (mathematical programming) framework has proposed for various DER DG types, size, and location) ✓ Soptimal yTh SMO w m h yogenerated w m h ho mu m o on mu GA on nd GA on nd on d m hod d(QOTLBO) m hod doub dng ois m d hod oby m pvu hod n(six p dung oon n nd dhn h oalgorithms b nd has bMOO nng v regarding n v --- The -✓ --DGP -DGP [33] 2012 2 ✓ ✓ for wind-based solutions DG for units wind-based sizing, DG units sizing, siting, and schedules. maintenance schedules. -✓-goal -✓ - solutions - todominant -problem [42] [42] 2014 [42] 2014 3 [33] 2014 13 ✓2012 [36] 2013 3✓ 13 -1 12✓ 1- locations. addition addition toattainment finding finding generated optimal electricity electricity prices avhand competitive in competitive electrical electrical market. market. ✓-1 dominant A u ytime mb ddtime dm GA mp oy dsiting, opin oprices u yaw gh d ng ob vat un onnonlinear mp oy ng un poyv ob h o y un on vo MODGP The problem for location, for location, size, and size, type and (REG type (REG PEV) isd defined PEV) isog as defined MO mixed asnode. MO integer mixed integer nonlinear programming programming (MINLP), (MINLP), in ng which in ng which locations. A A method go A (GoA) go nm nu has nm presented, hod nfuzzy m GoA hod where hy GoA goal nand programming p dmaintenance wh nand go wh transforms pothe go mm pomultiple og ng mm n objective ng opund m n mu func oflow m pproblem on mu ob un ng vob on on n oobjectives. ob v ob v to find to find varying voltage varying magnitude voltage magnitude and loss and sensitivity loss sensitivity factor at factor each each node. DGP DGP DGP framework, followed by decision-making tool to select most preferred Pareto optimal solution satisfying competing optimal location optimal location optimal and sizing location and of sizing DGP and for of sizing DGP solving for of multi-objective DGP solving for multi-objective solving optimal multi-objective power optimal flow power optimal (OPF) flow power problem (OPF) for (OPF) radial problem for DNW. radial for DNW. radial DNW. An xh u An v xh h v n y h ES n h p ES h n p d w h n MO d w h MODGP MO MODGP p ob m ob m v und b v o d b ond o d on ond on ✓ ✓ [38] [38] 2014 2014 2 1 2 1 [39] that aims atdd DISCOs contribution innn the electricity market. Moreover augmented method ✓ --✓ An- MO ✓ [32] 2012 5 1 MCS mb MCS d mb nframework GA dd d p nnn GA ngpm hodo ngcompetitive m ogy hodo pfrontier) ogy opo dp solution opo o harmony mp dov o distribution mp hsearch ov udistribution h modified ycompany o u search owhich ycompany h o (IMOHS), oDG hevaluate nε-constrained gDG on n DGP. w g hevaluate on dw o halong od u with oon o u on ✓ 2 -[35] 2013 [35] 2 2013 1 1-planning An framework MO planning has developed, namely has developed, improved namely multi-objective improved multi-objective (IMOHS), harmony can which the can the DGP. ✓ an NSGA-II an NSGA-II is used to is obtain used to compromise obtain compromise (Pareto frontier) (Pareto solution for a local for a local (LDC). (LDC). MCS mb dd d GA p n d o o v h n on n d p og mm ng m wo k on d ng u u un n o o d DG An improved An improved MOPSO MOPSO with preference with preference strategy strategy (IMPSO-PS) (IMPSO-PS) is proposed is proposed for MODGP for MODGP with the with aim to the achieve aim to optimal achieve capacity optimal capacity and and un on wh- ph og unu mm honhGA un wh oum vw on h dhodo by wh uGA hp h oon on u v uhdohodo by op ou GA v m dto oby DG n GA u w nd oo op n pm udon nno DG op nguhobjectives m w nd DG phobjectives wmong nn nd ng pMO nn ng The multi-criteria The multi-criteria planning planning model aims model achieve aims to contrasting achieve contrasting (cost and (cost reliability) and reliability) for DGP. for GA DGP. further GA finds further a set finds of nona set of non✓ ✓ ✓ GA w h go GA w ng go p h og go mm ogy og ng nd m mm ng o m ogy on hodo w nd h ogy o nd u d un w o on n o w d un o d n un nd on mong n MO n mong nd MO on nd n on n o d o o u omp d o on omp on u o nd u dvo nd dvo g d DNW o g y DNW m y v n m w h v ubop n w h m ubop o u m on o u on [37] 2013 3 1 The methodology based Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). MOPSO MOPSO algorithm MOPSO algorithm has been algorithm has used been for has used MODGP been for used MODGP problem for MODGP problem satisfying problem satisfying objectives satisfying objectives (minimize objectives (minimize DISCO (minimize cost DISCO and cost maximize DISCO and cost maximize DG and owner maximize DG profit) owner DG in profit) owner inprofit) in ✓ 1 -2013 ✓ - -1-2 - ✓ 1 An FDM hasoptimization utilized for best compromise solution among the set of Pareto optimal solutions. - - MO [44] [44] 2014 2014 2 [34] 12 [34] ✓ --optimization NSGA NSGA ong w h ong m x w m h n m pp x m o n h pp o v o h MODGP o v MODGP p ob m p on ob d m ng on u d u ng o u d u un o d n un nd n k m nd n g k m m n n g m n 2013 2✓ An MO problem for multiple problem micro-turbines for multiple (DG) micro-turbines placement and (DG) sizes placement is solved and by sizes hybrid is solved PSO and by SFL hybrid algorithms PSO and based SFL algorithms based ✓ ✓ [43] [43] 2014 [43] 2014 4,3 2014 1 4,3 1 4,3 1 The MO The probabilistic MO probabilistic (mathematical (mathematical programming) programming) framework framework has proposed has proposed for various for DER various planning DER planning (six DG types, (six DG size, types, and size, location) and location) ✓ Th A hyb ddominant GA pgo NSGA opo do wo o v MODGP pdhng ob m gsiting, dng DG o on wung h GA nd p y on w h on PSO p on vob ypon locations. MODGP ob mTh MODGP o go v MODGP dpmb by ob wo m pmb ob v h m dump by v do v wo hnd g units nand usiting, ud vnng m hod vng m n ng ud hod ou ng nschedules. ud nd EA ng unn ou ng op nd ou nn nd pynn mh vMODGP y op m v yp v v y solutions solutions for wind-based for wind-based DG units sizing, sizing, maintenance and maintenance schedules. ✓ ✓ ✓ opo NSGA p opo Th hm dpPSO NSGA opo d go ng hm m n go nd oby hm w ng nd bm n DG om w nd gprices od bung w dnd omp DG bshort om g DG du ng obasis gmarket. omp on d ng mong om omp on om om uEAy on ng omong ob uEA on vmong on vwhich ob -locations. -- p -- p ✓ --d addition [36] 2013 [36] 3 2013 1 1 ✓DGP A u yto A u dd GA dd GA oy dd mp og oy oprices vm dudDG u ohod ofor yavbo w u gh yng wcompetitive gh d ob ng vd ob un von un mp oy on mp u oyop o yng o MODGP The problem for size, and type (REG PEV) defined MO mixed nonlinear programming (MINLP), inng addition todominant finding addition optimal finding to generated optimal finding optimal generated electricity electricity in prices competitive inm adand in a on electrical market. electrical [40] 2014 2 1MO augmented εnd constrained method proposed DGP planning on term for DNWs with ANM functionalities. ✓ 3 ✓ framework, ✓ EA mp EA oy dymp od oy ocontribution vlocation, d fuzzy MO odgenerated oelectricity v on MO ng on bo h ng m vpreferred h y g visMoreover nPareto yelectrical ng gcompetitive nas nd don m nd nd dbinteger m h future vnd omarket. bng hm voy ng o ng vmuhng ou vo hn ou mp hn owith mp o framework, by fuzzy followed decision-making decision-making tool to select the most tool to select most preferred optimal solution optimal satisfying solution competing satisfying objectives. competing objectives. -- that --vaims [38] 2014 ✓1 - o -- -nd [39] [39] 2014 2 12 that DISCOs aims DISCOs contribution in the competitive in competitive market. Moreover modified modified augmented augmented ε-constrained method method with along A nm hod GoA h pv nelectricity dp wh go og ng ndistribution oop m mu pon ob vε-constrained un on n along o ng obDGP. vthe DGP. m- vfollowed y ng oatgo vo m goyatdd m v nd gn ynplanning m ng ud vby gng oh m vo gn n goyhud m gn ynd ud on nd nhelectricity onmarket. hv yw ndpothe omulti-objective vog ymm o nod hPareto nod ✓ ✓ - An -nd [35] [35] 2013 2013 2 12 - ✓ planning MO framework framework has developed, developed, improved namely improved multi-objective harmony harmony search search (IMOHS), which can which the xh hMO xhAn nmb u An v xhES u hm h vvo p n ypreference n dES n w hthe MO p ES nnamely w h n MO p dnod ob h m MODGP MO und oMODGP v MODGP p obb m und d ob m v und v oaim d ond oaim d on ond on ocan an NSGA-II is used to obtain compromise (Pareto frontier) solution for ah local company (LDC). ✓ 1 An ✓ MCS MCS d mb GA dd p dynd GA n p d n ohas v d ohho ooMODGP vdstrategy on nsorting on p nfor dmm pis og ng mm m ng wo kond m on wo db k(IMOHS), ng on udbwith u ng un uu nunevaluate n o evaluate doperation oDGcapacity oand d DG and The MOO planning framework based on non-dominated genetic algorithm II (NSGA-II) along MCS (for uncertain An- u improved An improved MOPSO improved MOPSO with preference MOPSO with strategy with preference strategy (IMPSO-PS) (IMPSO-PS) is proposed (IMPSO-PS) is proposed proposed for MODGP with for the MODGP with aim to the achieve with to the optimal achieve to capacity optimal achieve capacity optimal and DGP DGP ✓ ✓ ✓ [37] 2013 [37] 3 2013 1 3 1 The methodology The based methodology on Pareto frontier based on differential Pareto frontier evolution differential (PFDE) evolution algorithm (PFDE) is proposed algorithm for optimal is proposed MODGP for optimal (sizing and MODGP location). (sizing and location). FDM has FDM utilized has for utilized best compromise for best compromise solution solution among the among set of the Pareto set of optimal Pareto solutions. optimal solutions. un onng wh ho optimization umomp hdong omod v dong by GA o multiple nhm u op md DG w nd pbob nn ng 2✓ 12 -1 ✓ 1- ✓ ✓ [44] [44] 2014 [41][44] 2014 2 2014 2014 12 Th mu Th p mu nn Th mod mu p nn p h nn v on mod m o ng ob v o on v h v o ng on nd ob ng v y o v o nd DGP o GA b nd u y h o b DGP nd y o GA DGP u o h non GA nd u h nd o non o non An MO optimization An MO problem problem for multiple for micro-turbines micro-turbines (DG) placement (DG) placement and sizes and is solved sizes is by solved hybrid by PSO hybrid and PSO SFL and algorithms SFL algorithms based based GA o u d GA o omp GA o u on d o o u u o nd on omp dvo on u d o o nd u g dvo DNW nd dvo y o m g d v DNW o n w g h y DNW ubop m y v m n m w o u h v on ubop n w h m ubop o u m on o u on The MO probabilistic framework has proposed for d various DER (six ✓1 - ✓ ✓ A hyb A GA hyb GAaddresses p(mathematical PSO d p DGP opo ohodo o problem vprogramming) dogy MODGP ohodo ond vofogy MODGP po ob m pstorage ob g uhdm ng g DG ong DG on ownun hon GAmong wnnd hplanning GA p nd yMO w p DG hnd PSO ytypes, h size, PSO p n and v yplocation) v y scenarios) and OPF, REG integration within DNW. ✓2013 locations. locations. locations. hdεGA go wPSO psize, hd og mm popo og ng m mm ng m nd uand on w o on ow hdas d un oterm MO mong nd on nw on -- MO - --augmented - GA [36] [36] 2013 problem The for-w DGP location, problem for and location, type (REG size, and and PEV) type is(REG defined and as PEV) MO is mixed defined integer MO nonlinear mixed integer programming nonlinear (MINLP), programming in which (MINLP), in which ✓1 ✓ - The - -DGP [40] [40] 2014 2014 2 13 2 13 - ✓ MO augmented constrained ε go constrained method proposed method proposed for DGP for planning DGP planning on short on term short basis for basis future for DNWs future with DNWs ANM with functionalities. ANM functionalities. [38]

2014

[41] [39]

2014 [42]

[40]

2014

[38] 2 [39] [42] [37] [41] 2014 [40] [38] [39] 2 [43] [42] 2014 [41] [40] 2 [44]

2014 1 2014 [37] 2013 2014 2 2014 [38] 2014 1 2014 2014 3 2014 2014 1 2014

MODGP pmb ob ou by wo gng uDG vntool hod nnd ngn uPareto ng ou nd EA nnuh oyop mhmethod p with v objectives. y dom n y-n mb o✓ dom oaims nw nd ou bundd on ndd DG o o un won w ng dw nd DG b un dd un ng ng nd hng m ng du n nud mnmost h n du nvm h du framework, framework, followed by decision-making decision-making tonud select to the select most the preferred preferred optimal Pareto solution optimal solution satisfying satisfying competing competing objectives. ✓ 2 ✓ ✓ A -u ✓ A dgo GA u yndom A mp oy ym mb GA ofollowed dd ov vdnd dm mp GA uobby oy yfuzzy dmp op gh oy ohv u ong ong yvn ob w gh v ym dwp un gh on d ob mp ng voy obng un on yvariant un mp hoy on ong y mp ouoy MODGP yvproposed ng o MODGP oonyon o MODGP -✓ - uddon 1that at DISCOs contribution in the competitive market. modified augmented ε-constrained along A A go nm n m nm nfuzzy GoA hod h GoA hd nnd p dmtool nelectricity go go mm pmethodology, og ng mm n ng ou n mu oom m pomp mu ob p(LDC). ob un vony un no ng nob o obng v ob v v ✓ an Quasi-oppositional teaching learning-based optimization (QOTLBO) aomp of TLBO ason MOO regarding ✓ ✓ Th p opo Th d p NSGA opo d hod NSGA go hm go nd hm ng nd nwh ng om nwh nd odistribution bog w dnd DG bMoreover dg DG d ng g dalong ng oom u on omong u on mong ng on vngob is ng used an NSGA-II to obtain is compromise used to obtain (Pareto compromise frontier) (Pareto frontier) for avw local solution for ah local company distribution (LDC). company ✓ ✓ The MOO planning framework framework based on based non-dominated non-dominated sorting genetic algorithm genetic algorithm IIv (NSGA-II) II (NSGA-II) with along MCS with MCS uncertain (for uncertain o planning mutilized v yp ng vo g m gn ud nd om n y omu nod 3 13 ✓2013 ✓ ✓mb An---NSGA-II MO nn An m pMCS An kdd nn MO h ng dGA vudd nn op ng wo dcompromise nn m m wo d yfrontier v kPareto mp op ov dsolution d d op m d y nmp m ov vy hm mony ov d mu hob hsolutions. MOHS mony hon wh mony h hkfor MOHS n h u(for MOHS wh DGP wh n voperation hou n u DGP DGP --- pThe -dd ---MOO -nd The methodology The methodology based on Pareto on differential differential evolution evolution (PFDE) algorithm (PFDE) algorithm is proposed isdproposed optimal for MODGP optimal MODGP (sizing and (sizing location). and ✓ ✓ d- MCS GA pwo mb n dmb ofor om vh p d GA hbased p d on o n ovhon vsolving dno od hfrontier pn n oumu og v on hmm nm ng nsorting on dd pmp m og wo n dmm kob ppower on og ng d mm ng ng wo uv kum un wo ng on nfor uvdradial uo ng o un d uhhuDG nun n o dvhoperation oDG o dhlocation). DG FDM has best solution among set of Pareto optimal un MO on un wh on h wh h ok u vhDGP h d by ofor GA d by nvn GA ow op nob uthe DG op w nd DG w nn nd ng pm nn ng ✓1 ✓ 1- MCS 13 2 1 - ✓ optimal location and sizing of multi-objective optimal flow (OPF) problem DNW. An xh u An v xh u h v n y h ES n y h p ES h p d h n MO d w o h MODGP MO o MODGP p ob m p und ob m v und b v o d b ond o d on ond on TheMO MO probabilistic The MO probabilistic programming) framework programming) framework for has proposed DER planning for various DER planning types, size, (six and DG location) types, size, and location) scenarios) scenarios) and OPF, and addresses addresses DGP problem DGP problem of REG and ofhas REG storage integration storage integration within within DNW. Th mu pon ngm mod h v on ng vm oMO nd b(six yDG ointeger DGP uhm h nd o (MINLP), non An m An on MO p(mathematical ob An m MO m oOPF, mu m p(mathematical popo ob on mdpo pmethod o o obumu m ngm p o DG mu po p uproposed m m band nm no o um DG nd bisob n pvarious DG ohod n p vng dDNW. nd by m hyb np nd don PSO ow vw dnd nd by ond SFL hyb v EA dnd by dGA PSO hyb dy bPSO SFL d nd SFL go b which d yypinb which ✓ ✓ The DGP The problem problem for location, for location, size, and size, type and type and (REG PEV) defined PEV) is as defined mixed as MO integer mixed nonlinear nonlinear programming programming (MINLP), inon ✓ 2 ✓ ✓ A -hyb A hyb p opo dop A GA hyb dDGP PSO om dεop o GA vob pnn PSO MODGP opo o p ob obby von d MODGP ooo g om v(REG dgproblem MODGP ng p ob DG pplanning ob gand on d m w ng h g DG GA d o nd DG on ow h y GA h h PSO GA p p ygo w pvm hon ynd PSO w hon PSO pgo vhm yp vvhm -dopGA - PSO 1 MO augmented constrained proposed for DGP on short term basis for future DNWs with ANM functionalities. MODGP MODGP p ob p o v m d by wo v d wo h u h u v hod v m n ud ng n u ud ng ng u ou ng ou nn nd EA op nn m op on m p vd y ✓ MOPSO algorithm has been used for MODGP satisfying objectives (minimize DISCO cost and maximize DG owner profit) in ✓ ✓ GA GA o u d o o u omp d o on omp o u nd u dvo nd d dvo o g d DNW o g y DNW m y v n m w h v ubop n w h m ubop o u o u ✓ 2014 2 1 2 1 ✓ 4,3 -✓ that at DISCOs that contribution aims DISCOs inkthe contribution competitive in electricity the competitive market. market. modified Moreover augmented modified ε-constrained augmented method ε-constrained along with method along with ✓wo Quasi-oppositional teaching teaching learning-based learning-based optimization optimization (QOTLBO) (QOTLBO) methodology, afor variant aof variant of proposed TLBO proposed as ng MOO as regarding MOO regarding nhod nan o on o w nd bby d DG ng ngwh nd m nfor n nmo haong du -go - Quasi-oppositional -o ow 1m-aims k nm m d by wo uoat kplanning m y wo ow d on m ow km d ng yd dgn ooh u oon y dun m kpud ng on hog m oo kelectricity o ng pMoreover oo ohog P mo omethodology, h p op m d p P u on dTLBO op P m omu ng op on ucompany omp m uMCS ob y on ng omp yon ng ng ob ng v vobob v v an NSGA-II NSGA-II isoud used to is obtain used to compromise obtain compromise frontier) (Pareto frontier) solution solution ah local distribution distribution (LDC). ndom A m go A go GoA nm nh moptimal nm poby hod nu n GoA hod wh GoA p go h(Pareto n p d mo wh mm n d ng go p nd o go m mm p mu og ng pmm n ob olocal m v(NSGA-II) mu un o company m pyon ob p(LDC). oon vwith oboun ng vob on un nv v o ng nomp o ob ng The MOO framework based on non-dominated sorting genetic algorithm IIon along (for uncertain operation o nd m v nd y m ng vo v y g ng m vo g ud m gn nd o nd n o v y n o v y o nod h nod ✓1 - ✓ ✓-A 1 3 addition to finding generated electricity prices in a competitive electrical market. ✓ ✓ ✓ A u y A mb u dd y d mb GA dd d mp GA oy d mp o oy o v d u o o y v w u gh y d w ng gh d ob ng v ob un v on un mp oy on ng mp u oy y ng u h o y y o h MODGP o y o MODGP FDM utilized FDM for best has compromise utilized for best solution compromise among the solution set ofmulti-objective among Pareto optimal thego setd of solutions. Pareto optimal ✓ ✓ - ✓ - optimal 1 location optimal location and sizing and of sizing DGP ofh(mathematical DGP solving multi-objective solving optimal power optimal flow power (OPF) flow problem (OPF) problem for dp radial for DNW. DNW. ✓ Th p ng m wo k d ofor vP ddnGA n y mp ov mu v mony hplanning MOHS wh hotypes, vsize, u and hsize, DGP ✓ 2 ✓has Th buMO m d on hodo Th Pnn ogy oby hodo bu on d on dp bonfor P d o on on vo u on PFDE nop don n u on hm vo PFDE ung pvob on opo PFDE go dwithin hm osolutions. op go pm opo hm MODGP oDGP op dradial ng oDGP MODGP op on nd ong nd on o on The MO The probabilistic MO probabilistic (mathematical programming) programming) framework framework has proposed has proposed various for DER various DER planning (six DG (sixnMODGP types, location) and location) un m onhodo wh ogy h An un hon un wh oand vm on h d wh GA hogy h ong u vunn hDGP d nng by op GA v mop doooby DG u wm nd np m uvo nn DG op m w nd DG p w nn nd ng ph nn ng scenarios) OPF, addresses problem of REG and storage integration mu Th mu p nn mod mod m h m v(IMPSO-PS) oo h v ng ob o term vfor nd oDNW. bm nd y ANM ok bopo y uachieve om GA und hm GA nd uDG h ong nd odand non o non An improved MOPSO with preference is proposed for MODGP with the aim optimal capacity ✓ MCS mb MCS dd d mb dd p dob GA n dforfor oDGP n oMODGP vstrategy oh oproblem vo nnsatisfying n on don pob og nobjectives dmm p og mm m ng wo kcost wo ng onto dd umaximize ng un u u SFL nun o oDG ✓ 2 -✓ -- augmented -2 1 εTh MO constrained augmented εGA constrained method proposed for on planning term on for short future DNWs basis for with DNWs functionalities. with ANM functionalities. ✓ MO ✓ -- MOPSO MOPSO algorithm algorithm has been has used been form used MODGP problem satisfying objectives (minimize (minimize DISCO DISCO and cost maximize and DG owner DG profit) owner inhm profit) indd DG An MO op mmethod on pproposed op mu pdplanning mn uhDGP bshort DG pbasis m nng nd oon vfuture ddby hyb PSO nd go n b o 1-

Energies 2017, 10, 208

19 of 44

Energies 2017, 10,Energies 208 2017, 10, 208

Table 7. Cont.

16 of 44

16 of 44

Table 5. The contribution Table 5. The ofcontribution MODGP-based of MODGP-based planning methods planning and related methods works. and (PTC: relatedComponents works. (PTC:inComponents planning techniques; in planning MODGP: techniques; MO based MODGP: DG placement). MO based DG placement). EnergiesYear 2017, 10,MO Energies 208 PTC 2017, 10, A 208 16 of 44 16 of 44 Ref. B C D Contribution of MOCRU-Based Planning Ref.

Year

Ref. MO

PTC Year

AMOB PTC C DA

[24]

2008

[24] 3

2008 1

✓ 3 -

B

C

D

Contribution of Contribution MODGP-based of Planning MODGP-based Planning

Model for MO Integrated Generation and Primary–Secondary Distribution System Expansion Planning (IGDSEP) in the presence of wholesale and Energies 2017,5.10, 208 formulation MODGP based on GA and based ε constrained on GAComponents and method ε constrained is proposed method to solve istechniques; proposed the best compromise toMODGP: solve the best (tradeoff) compromise solution for based DM solution for DM 16 of 44 5. The contribution ofcontribution of MODGP MODGP-based planning methods planning andformulation related works. and (PTC: related works. (PTC: Components planning in planning techniques; based MODGP: DG (tradeoff) placement). MO DG placement). ✓ - - 1-✓ - - [21] Table 2005 4 Table 1 The 4MODGP-based [81] 2012 2[21] 22005 retail markets asmethods MINLP problem is divided into in six sub-problems and solved with MO scenario Driven MINLP method with adaptive GA (AGA) and (DISCO). (DISCO). Energies 2017, 10, 208 16 of 44 integrated AGA (IAGA). Ref. Year Ref. MO PTC Year AMO C D Aof Similarly, BC D ofε Contribution MODGP-based of Planning MODGP-based Planning ✓ -✓ - with - the [22] 2005 [22] 3 Table 2005 1 5. The 3 B- PTC 1Similarly, same formulation with the same (GA formulation and ε Contribution constrained (GA and method), constrained a double method), tradeoff amethod double tradeoff is presented method toMODGP: find is presented the best to find the alternative. contribution MODGP-based planning methods and related works. (PTC: Components in planning techniques; MOalternative. based DGbest placement). MODGP formulation MODGP based formulation on GA and based εplanning constrained onisexpansion GA and method ε constrained isisproposed method to solve is problem proposed the best compromise to solve the best (tradeoff) compromise solution (tradeoff) for DM solution for DM Multistage distribution network planning (MDEP) for DG integration and grid reinforcements is solved ✓ ✓ [23] 2007 Energies [23] 2 Energies 2007 1 2 1 MCS embedded in MCS GA embedded planning methodology in GA proposed methodology to improve proposed the accuracy to improve for stochastic the accuracy DG for integration stochastic with DG tradeoff integration solution. with tradeoff solution. 2017, 208contribution 16with of 44hybrid 16 of 44PSO and ✓ - 2081-✓ MODGP-based [21] 2012 2005 2[21] 4 2017, 2005 1 5.10, 4 10, [82] 2Year The planning methods and algorithm related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). Ref. Table MO -PTC Aof (DISCO). B - C Shuffled D (DISCO). Contribution ofsolutions MODGP-based Planning Frog Leaping (SFL) to find Pareto optimal satisfying multiple objectives. 1-

-✓ NSGA-II - along - with NSGA-II max-min along approach with max-min solves MODGP approachproblem solves MODGP considering problem future considering load uncertainties future load anduncertainties risk management and risk management

MODGP formulation based onεconsidering GA and ε constrained istradeoff proposed to demand solve thebehavior best compromise (tradeoff) solution for DM of ✓ - 208 -✓ - with - the [22] 2005 Energies [22] 3 Energies 2005 1 10,2017, 3 10, 1Similarly, Similarly, same formulation with the same (GA formulation and constrained (GA and method), εtime-varying constrained amethod double method), amethod double tradeoff is Planning presented method to find is presented the at best alternative. to find alternative. employed to EA solve isfeeder employed IMO considering to solve IMO both time-varying both generation and demand generation behavior and aiming at various aiming technical various impacts technical ofthe best impacts 2017, 208 16 of 44 16 of 44 Aof EA B--- is-of C D Contribution of MODGP-based -- MODGP-based -[21] 2005 4 1 [83] 1Year - MO -PTC Optimal routes and conductor size of branches with multi-objective dynamic programming (MODP). The contribution MODGP-based planning planning methods methods and related and related works. (PTC: works. Components (PTC: Components in planning in planning techniques; techniques; MODGP: MODGP: MO based MO DG based placement). DG placement). ✓ ✓ -✓ [25] 2012 2008 2Ref. [25] 6 Table 2008 1 5.Table 65.-The 1-contribution (DISCO). ✓ 2 -✓ MCS - embedded [23] 2007 [23] 2 2007 1 1in MCS GAembedded planning methodology in GA planning is proposed methodology to improve is proposed the accuracy to improve for stochastic the accuracy DGfor integration stochasticwith DG tradeoff integration solution. with tradeoff solution. DGP

DGP MODGP formulation based on GA and ε constrained method is proposed to solve the best compromise (tradeoff) solution for DM

- - -- goal -- programming [21] 2005 4 1 [84] 2Ref. 1Ref. -Year MOgoal tabu search (MOTS) algorithm solves planning problem to check compromise (trade-off) among cost [22] 3 Similarly, with the same formulation (GA and εassociated method), aofamong double tradeoff method is presented tosolutions find the best alternative. ✓ -✓ [24] 2008 Energies [24] 3 2008 1 10, ✓ 1NSGA-II along with NSGA-II max-min along approach with max-min solves MODGP approach problem solves MODGP considering problem future considering load uncertainties future load and uncertainties risk management risk management GA with methodology programming finds methodology a solution with finds aconstrained solution uncertainties with uncertainties MO and constraints. among MO andand constraints. [26] 2013 [26] 2 2017, 2 - -PTC 208 16 and of 44reliability. Year MO A PTC B- with AofCMODGP-based BProposed D C D Contribution Contribution ofassociated MODGP-based MODGP-based Planning Planning (DISCO). Table 5.Table The contribution 5. TheMO contribution of GA MODGP-based planning planning methods methods and related and related works. (PTC: works. Components (PTC: Components in planning in planning techniques; techniques; MODGP: MODGP: MO based MO DG based placement). DG placement). -- isproposed --- NSGA [23] 2007 1 MCS embedded informulation GA planning methodology istime-varying proposed todemand improve the accuracy for stochastic DG integration with tradeoff solution. employed to EA solve isproposed employed IMO considering to solve IMO both considering time-varying both generation and generation behavior and demand aiming behavior at various aiming technical at(tradeoff) various impacts technical ofthe impacts of ✓ 2 ✓ EA -✓ The The algorithm NSGA finds arrangements algorithm finds for arrangements wind-based DGs formethod wind-based regarding compromise DGs regarding solution compromise among contrasting solution among objectives. contrasting objectives. [27] 2008 [27] 3 2008 1 3 1MODGP MODGP formulation based on based GA and on GA ε constrained and ε constrained is method proposed is proposed to solve the to solve best compromise the best compromise (tradeoff) solution solution for DM for DM ✓ Multi-objective seeker optimization algorithm (MOSOA) is proposed for DSP problem with the simultaneous placement of automatic reclosers [22] 2005 3 1 Similarly, with the same formulation (GA and ε constrained method), a double tradeoff method is presented to find best alternative. ✓2005 ✓ 1 -- ✓ -- - -- [25] 2014 2008 4[21] [25] 6 2008 1 6 - - 1 1- 4 -✓ [21] 2005 4 [85] 2Ref. ✓ - CDGP [24] 2008 -Year 3 1 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management DGP Ref. Year MO PTC MO A PTC B-- exhaustive AC D D Contribution ofMODGP MODGP-based of MODGP-based Planning Planning An exhaustive analysis (ES) search has presented (ES) with has IMO presented for MODGP with IMO problem for under variable problem load under conditions. variable load conditions. (DISCO). (DISCO). with solution satisfy maximum conflicting multiple -- B(RAs) -- search [23] 2007 2 1 MCS embedded in GAanalysis planning methodology isContribution proposed toobjectives. improve the accuracy for stochastic DG integration with tradeoff solution. The of An MODGP-based planning methods and related works. (PTC: Components in planning techniques; MODGP: MO based DG placement). Energies 208 2017, 208contribution 16 of 44 16 of 44 ✓ ✓ -✓ [28] 2017, 200910,Energies [28] 4 Table 2009 1 5.10, 4 1[26]

2008

[26] 2 [22]

2008 1 [22] 2005

✓2005 2 3

1- 3 1

EA is toresults solve IMO considering both time-varying generation and demand behavior aiming atsolutions. various technical impacts of -✓ GA goal methodology programming finds methodology aofsolution with finds solution uncertainties with associated uncertainties and constraints. among MO and constraints. is✓also comparison is employed also used of for comparison and advocated results and large advocated DNW for large even DNW with systems suboptimal even with solutions. suboptimal MODGP MODGP formulation formulation based on based GA and on GA εfor constrained and εaconstrained constrained is method proposed proposed to solve the to solve best compromise the best compromise (tradeoff) (tradeoff) solution solution for DM for DM ✓ 1 GA -- with -- goal -used -- programming -for -with Similarly, Similarly, with the with same the formulation same formulation (GA and (GA εassociated andmethod εsystems constrained method), method), aisamong double aMO tradeoff double tradeoff method is method presented is presented to find the to best find alternative. the best alternative.

[24] 2008 -2005 3 NSGA-II along with approach solves problemplacement consideringoffuture load SSWs uncertainties and risk management [25] 6 ✓ 1 -- ✓ -- - MOSOA -- - DGP - is proposed [21] 2005 4 [86] 2Ref. 2[21] - 111- 4 -✓ formax-min DSP problem with theMODGP simultaneous RAs and to achieve maximum reliability and lower cost. ✓2007 ✓ proposed The algorithm NSGA finds algorithm finds for wind-based DGs for is wind-based regarding compromise DGs regarding solution compromise among contrasting solution among objectives. contrasting objectives. [27] 2015 2008 [27] 3 2008 1 3 Year MO A 1 The D Contribution of MODGP-based Planning (DISCO). (DISCO). [29] 2010 [29] 5 2010 2 5 - PTC 22 ✓✓ A B--fuzzy A fuzzy GA isMCS embedded employed GA toarrangements solve isIMO employed fuzzy weighted toarrangements solve single fuzzy objective weighted single objective employing function fuzzy employing set theory for fuzzy theory for MODGP. ✓C -- embedded -✓ -- NSGA - MCS -proposed [23] [23] 2007 2 1 embedded embedded in solve GA planning in GA planning methodology methodology istime-varying proposed proposed tofunction improve to improve the accuracy the accuracy forbehavior stochastic foraiming stochastic DGMODGP. integration DG integration with tradeoff with tradeoff solution. EA is employed to considering both generation and demand atset various technical impacts of solution. 5.10, The contribution Table 5.10, The of contribution MODGP-based of1 An MODGP-based planning methods planning and related methods works. and (PTC: related Components works. (PTC: in Components planning techniques; in planning MODGP: techniques; MO based MODGP: DG placement). MO based DG placement). - exhaustive - search [25] 2008 6 1- 3 -✓ GA with goal programming methodology finds a solution with associated uncertainties among MO and constraints. [26] 2 Energies 2017, Energies 208 2017, 208 16 of 44 of 44 An exhaustive analysis (ES) search has analysis presented (ES) with has IMO presented for MODGP with IMO problem for MODGP under variable problem load under conditions. variable load conditions. MODGP formulation based on GA and ε constrained method is proposed to solve the best compromise (tradeoff) solution for DM MO planning for optimal placement of switching devices has solved by NSGA-II with two approaches with the aim of satisfying objectives namely ✓ [22] [22] 2005 2005 3 1 Similarly, Similarly, with the with same the formulation same formulation (GA and (GA ε constrained and ε constrained method), method), a double a tradeoff double tradeoff method is method presented is presented to find the to best find alternative. the best alternative. MCS-embedded MCS-embedded GA is presented to GA solve is presented chance constrained to solve chance programming constrained framework programming considering framework future considering uncertainties future of loads/DGs. uncertainties of loads/DGs. [30] Table 2011 [30] 5 2011 1 5 ✓ ✓ [24] 2008 NSGA-IINSGA-II along with along max-min with max-min approachapproach solves MODGP solves MODGP problemproblem considering considering future load future uncertainties load uncertainties and risk and management risk management16 DGP ✓ -✓ [28] 2015 2009 3[21] [28] 4 2009 1 - - 2005 4 - - 1[87] 1[24] -2008 ✓ 1 GA The proposed NSGA algorithm finds arrangements for cost, wind-based DGs regarding compromise solution among objectives. [27] 2008 3 1- 2 -✓ is✓also for GA comparison also used of for results comparison and advocated of results for and large advocated DNW systems for even DNW with systems suboptimal even with solutions. suboptimal (DISCO). DG network reliability and equipment with no island network operation. --- GA-PSO -used --- -unavailability, -is ✓ [23] [23] 2007 2007 2 1 MCS embedded MCS in GA planning in GA planning methodology methodology istime-varying proposed is proposed tolarge improve to improve the accuracy the accuracy for stochastic for stochastic DG integration DG integration with tradeoff with tradeoff solution. solution. A B---hybrid A is hybrid proposed GA-PSO to solve is proposed MODGP to problem solve MODGP regarding DG location regarding (with DG GA) location and capacity (with GA) (with and PSO) capacity respectively. (with PSO) respectively. [31] 2012 [31] 3 2012 1 3 EA is employed EA is embedded employed to solve IMO to solve considering IMO considering both both time-varying generation generation and demand and demand behavior behavior aiming atsolutions. aiming various atcontrasting technical various technical impacts of impacts of ✓ GA with goal programming methodology finds aproblem solution with associated uncertainties among MO and constraints. [26] 2008 2 B PTC 1 Ref. Year Ref. MO PTC Year A MO C D A C D Contribution of Contribution MODGP-based of Planning MODGP-based Planning ✓ 1 A --fuzzy ✓ -- embedded - -- - An - isembedded [25] [25] 2008 2008 6 1 6 ✓✓ exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. ✓ [29] Table 2010 [29] 5 Table 2010 2 5. The 5 2- 3 A fuzzy GA employed GA tomax-min solve isformulation employed fuzzy weighted to solve single fuzzy objective weighted function single employing function fuzzy employing set theory for fuzzy MODGP. set theory for MODGP. [22] 2005 3 1 Similarly, with the same (GA and ε constrained method), a double tradeoff method is presented to find the alternative. ✓ attainment -✓ -proposed [24] [24] 2008 2008 NSGA-II NSGA-II along with along max-min approach approach solves MODGP problem problem considering considering future load future uncertainties load uncertainties risk and management risk management A method goal attainment (GoA) has method presented, (GoA) where has presented, goal programming where goal transforms programming multiple transforms objective multiple functions objective into aand single functions objective into abest single objective 5. The contribution of contribution MODGP-based of1 A MODGP-based planning methods planning and related methods works. and (PTC: related Components works. (PTC: in Components planning techniques; inobjective planning MODGP: techniques; MO based MODGP: DG placement). MO based DG placement). DGP DGP ✓ --goal -- formulation -- -fault The NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] 2008 3 [28] 2009 4 MODGP MODGP based formulation on GA and based εwith constrained on GA and method εsolves constrained is MODGP proposed method to solve is proposed the best compromise to solve the best (tradeoff) compromise solution (tradeoff) for DM solution for DM The current limiter (FCL) allocation problem with DGs is formulated as multi-objective constrained nonlinear programming (NLP) problem ✓ -✓ [32] 2012 [32] 5 2012 1 5 -- - 1 1-GA is also used for comparison of results and advocated for large DNW systems even with suboptimal solutions. ✓ ✓ [21] 2005 4 2005 1 4 1 [88] 2 MCS-embedded MCS-embedded presented to GA solve is presented chance constrained to solve chance programming constrained framework programming considering framework future considering uncertainties future of loads/DGs. uncertainties of loads/DGs. [30] 2015/16 2011 3[21] [30] 5 2011 5 [23] 2007 2 MCS embedded in GA planning methodology is proposed to improve the accuracy for stochastic DG integration with tradeoff solution. function, further solved which is by further GAanalysis toIMO ensure solved by optimal GA toDG ensure (wind) optimal planning. DG (wind) planning. EA is employed EA is employed to solve to solve considering IMO considering both time-varying both time-varying generation generation and demand and demand behavior behavior aiming ataiming various atconstraints. technical various technical impacts of impacts of ✓ 1 function, - ✓ - which - - -is -with GA GA goal with programming goal programming methodology methodology finds a with solution finds aobjectives. solution with associated with associated uncertainties uncertainties among MO among and MO constraints. and [26] [26] 2008 2008 2 12 An exhaustive search (ES) has presented IMO for MODGP problem under variable load conditions. (DISCO). (DISCO). solved using PSO to achieve multiple conflicting -- - and -- -is [25] [25] 2008 6 1 ✓ Ref. Year Ref. MO PTC Year MO C A 1 B-- ✓C D Contribution of Contribution MODGP-based of Planning MODGP-based Planning [28] 2009 A2008 4 B PTC 16 D✓

fuzzy embedded GA is employed toarrangements solve fuzzy weighted single objective function employing fuzzy set theory forrespectively. MODGP. -✓ A is hybrid proposed GA-PSO to solve iscomparison proposed MODGP to problem solve MODGP regarding problem DG location regarding (with DG GA) location and capacity (with GA) (with and PSO) capacity respectively. (with PSO) --hybrid - GA-PSO -- - DGP NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management DGP MODGP MODGP is solved problem byproposed two is stage solved heuristic byεalgorithm two iterative stage heuristic method iterative including method clustering including (outer) clustering and EA (inner) (outer) and EA best (inner) respectively, respectively, ✓ 1A -✓ -is The proposed The NSGA algorithm NSGA finds finds arrangements for wind-based for wind-based DGs regarding DGs regarding compromise compromise solution solution among contrasting among contrasting objectives. objectives. GA also used for of results and advocated for large DNW systems with suboptimal solutions. --✓ -- ✓ --- problem -- the with Similarly, same formulation with the same (GA formulation and constrained (GA and method), ε constrained a double method), tradeoff amethod double tradeoff iseven presented method tooptimizations find is presented the alternative. tooptimizations find the best alternative. ✓ Similarly, MODGP MODGP based formulation on GA and εpresented constrained on GA and method εand constrained isisproposed method to solve is proposed the best compromise to solve the best (tradeoff) compromise solution (tradeoff) for DM solution for DM Optimal coordinated voltage control (OCVC) to solve MOP using Pareto optimization to find the optimal values ✓ --goal -- formulation --varying MCS-embedded GA isbased to solve chance constrained programming framework considering future uncertainties loads/DGs. A attainment A method goal attainment (GoA) has method presented, (GoA) where has presented, goal programming where goal transforms programming multiple transforms objective multiple functions objective into aatset single functions objective into a of single objective EA is employed to solve IMO considering both time-varying generation and demand behavior aiming various technical impacts of of voltage of the to find time to find voltage time magnitude varying voltage and loss magnitude sensitivity factor loss sensitivity at each factor atfor each node. ✓ 1 MCS ✓ -✓ - in -with GA GA goal with programming goal programming methodology methodology finds aproposed solution finds anode. solution with associated with associated uncertainties uncertainties among MO among and MO constraints. and constraints. An exhaustive An exhaustive search analysis search analysis (ES) has (ES) presented has presented with IMO with for IMO MODGP for MODGP problem problem under variable under variable load conditions. load conditions. --✓ A fuzzy embedded GA is employed to solve fuzzy weighted single objective function employing fuzzy theory for MODGP. embedded MCS GA embedded planning methodology in GA planning is proposed methodology to improve is proposed the accuracy to improve stochastic the accuracy DG for integration stochastic with DG tradeoff integration solution. with tradeoff solution. ✓ 1 (DISCO). - ✓ -- - generators - - (DISCO). ✓ and OLTC, addressing various technical aspects. ✓ A hybrid GA-PSO is proposed to solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] which further solved which is by further GA tofor ensure solved by optimal GA to DG ensure (wind) optimal planning. DG (wind) planning. DGP The planning multi-criteria model aims planning to achieve model contrasting aims to achieve objectives contrasting (cost and objectives reliability) (cost for and DGP. reliability) GA further for DGP. finds GA asolutions. setfurther of nonfinds aof set of non- objectives. ✓ 1 function, ✓ -- along - -- with -isfunction, -max-min The proposed The proposed NSGA NSGA algorithm finds arrangements finds arrangements for wind-based for wind-based DGs regarding DGs regarding compromise compromise solution solution among contrasting among contrasting objectives. [27] [27] 2008 2008 3 1- 3 -✓ GA is also GA used is also for used comparison comparison of results of and results advocated and advocated for large for DNW large systems DNW systems even with even suboptimal with suboptimal solutions. MCS-embedded GA isalgorithm presented to solve chance constrained programming framework considering future uncertainties loads/DGs. [30] 2011 5 ✓ --- multi-criteria [24] 2008 [24] 3 2008 1 1 NSGA-II along approach with max-min solves MODGP approach problem solves MODGP considering problem future considering load uncertainties future load and uncertainties risk management and risk management ✓ 3 ✓ NSGA-II [34] 2013 [34] 2 2013 1 2 --1-✓ --✓ ---- problem --- the [22] 2005 [22] 3 2005 1 3 1 Similarly, with Similarly, formulation with theis same (GA formulation and εanalysis constrained (GA and method), εmaintenance a IMO double method), tradeoff amethod double tradeoff is problem presented method tooptimizations find isvariable presented the best alternative. toconditions. findinto theabest alternative. A goal attainment method (GoA) has where goal programming transforms multiple objective functions single objective MODGP MODGP issame solved problem by two stage solved heuristic by two stage heuristic method iterative including method clustering including (outer) clustering and EAfunction (inner) (outer) and EA (inner) respectively, optimizations respectively, ✓ 2 dominant GA with goal programming methodology finds aconstrained solution with associated uncertainties among MO and constraints. [26] 2008 2 1 dominant for wind-based solutions DG for units wind-based sizing, siting, DG units and sizing, siting, and schedules. maintenance schedules. An exhaustive An exhaustive search analysis search (ES) has (ES) presented has presented with with for IMO MODGP for MODGP problem under variable under load conditions. load ✓ -- solutions -✓ [29] [29] 2010 5 2 fuzzy A embedded fuzzy embedded GA is employed GA employed topresented, solve fuzzy to solve weighted fuzzy weighted single objective single objective function employing employing fuzzy fuzzy theory set for MODGP. for MODGP. -- -- to A hybrid GA-PSO is proposed toisiterative solve MODGP problem regarding DG location (with GA) and capacity (with PSO) respectively. [31] 2012 3 1 ✓ 1 EA--- is✓ [32] 5 employed EAsolve is employed IMO considering to solve IMO both considering time-varying both generation time-varying and demand generation behavior and demand aiming behavior at various aiming technical atset various impacts technical oftheory impacts of ✓2010 -5 -✓ [33] 2012 [33] 2 2012 1 2 -✓ ✓ [28] [28] 2009 2009 4 1 4 ✓ ✓ [23] 2007 [23] 2 2007 1 2 1 MCS embedded in MCS GA embedded planning methodology in GA planning is proposed methodology to improve is proposed the accuracy to improve for stochastic the accuracy DG for integration stochastic with DG tradeoff integration solution. with tradeoff solution. which is further solved by GA toand optimal DG (wind) planning. [25] 2008 [25] 6 2008 6 to to find voltage time magnitude varying voltage and loss magnitude sensitivity and factor loss sensitivity at each node. factor at each node. -- planning The proposed NSGA finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. [27] 3 1 [35] 2013 [35] 22017, 10, 2013 1208 ✓2011 2 An--find MO An framework MO planning has developed, framework namely has developed, improved namely multi-objective improved harmony multi-objective search (IMOHS), harmony search which (IMOHS), can evaluate which thesolutions. DGP. can evaluate the DGP. GA also GA used is also for used comparison for comparison of results of results advocated and advocated for large for DNW large systems DNW systems even with even suboptimal with suboptimal solutions. ✓ 1 DGP ✓ time - --varying - function, -is MCS-embedded MCS-embedded GA isalgorithm presented GA is presented to solve chance toensure solve constrained chance constrained programming programming framework framework considering considering future uncertainties future uncertainties of loads/DGs. [30] [30] 2011 5 1- 5 -✓ Energies 16 of 44loads/DGs. A goal attainment method (GoA) has presented, where goal programming transforms multiple objective functions into a of single objective DGP ✓ ✓ [24] 2008 [24] 3 2008 1 The 3 1 NSGA-II along with NSGA-II max-min along approach with max-min solves MODGP approach problem solves MODGP considering problem future considering load uncertainties future load and uncertainties risk management and risk management [32] 2012 5contribution MODGP problem is solved by two stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, The multi-criteria The planning multi-criteria model aims planning to achieve model contrasting aims to achieve objectives contrasting (cost and objectives reliability) (cost for and DGP. reliability) GA further for DGP. finds GA a set further of nonfinds a set of nonTable 8. of MONTR-based planning methods. (MONTR: MOP based on distribution network reconfigration.). An exhaustive search analysis (ES) has presented with IMO for MODGP problem under variable load conditions. An MO optimization MO problem optimization for multiple problem micro-turbines for multiple (DG) micro-turbines placement and (DG) sizes placement is solved and by sizes hybrid is solved PSO and by SFL hybrid algorithms PSO and based SFL algorithms based ✓ ✓ ✓ ✓ [29] [29] 2010 2010 5 2 5 2 fuzzy embedded fuzzy embedded GA is employed GA is employed to solve fuzzy to solve weighted fuzzy weighted single objective single objective function function employing employing fuzzy set fuzzy theory set for theory MODGP. for MODGP. A hybrid A GA-PSO hybrid GA-PSO is proposed is proposed to solve MODGP to solve MODGP problem problem regarding regarding DG location DG location (with GA) (with and GA) capacity and capacity (with PSO) (with respectively. PSO) respectively. [31] [31] 2012 2012 3 1 3 1 function, which is further finds solved by GA towith ensure optimaluncertainties DG planning. - with - goal - programming [33] 2012 ✓ GA with goal methodology programming methodology a solution finds associated a solution with(wind) associated among uncertainties MO and constraints. among MO and constraints. [26] 2008 [26] 2 2008 1 1 [34] 2013 [34] 2013 [28] 2009 4 ✓ 2 --✓ GA [36] [36] 3 1 3 -1-EA EA solve is employed IMO considering to solve IMO both considering time-varying both generation time-varying and demand generation behavior and demand aiming behavior at various aiming technical atsolutions. various impacts technical of of to find time varying voltage magnitude and loss sensitivity factor at each node. dominant for wind-based solutions DG for units wind-based sizing, siting, DG units and sizing, maintenance siting, and schedules. maintenance schedules. GA also used for comparison ofto results and advocated for large DNW systems even with suboptimal framework, framework, by fuzzy followed decision-making by fuzzy decision-making tool to select the most tool topreferred select the Pareto most preferred optimal solution Pareto optimal satisfying solution competing satisfying objectives. competing objectives. ✓employed -- isproposed -- solutions - followed -- - to -is MCS-embedded MCS-embedded GA isalgorithm presented GA presented solve chance to solve constrained chance constrained programming programming framework framework considering considering future uncertainties future uncertainties of impacts loads/DGs. loads/DGs. [30] 2011 5 1 5 MODGP-based A goal attainment Aproblem goal attainment method (GoA) method has (GoA) presented, has presented, where goal where programming goal programming transforms transforms multiple multiple objective objective functions functions into into objective a of single objective ✓2011 ✓ 1 dominant -✓ [25] 2008 [25] 6 2008 1 contribution 6 1-of MODGP is solved byistwo stage heuristic iterative method including clustering (outer) and EA (inner) optimizations respectively, The planning methods and related works. (PTC: Components planning techniques; MODGP: MO based DG placement). The The proposed algorithm NSGA finds arrangements finds for arrangements wind-based DGs forinwind-based regarding compromise DGs regarding solution compromise among contrasting solution among objectives. contrasting objectives. [27] [27] 3 5.[30] 3 En Table oaasingle ✓ -- planning -- MO - -- NSGA - DGP -methodology [32] [32] 2012 2012 5 1 5 -✓ 1 DGP [33] 2 The planning aims to achieve contrasting objectives (costDG and reliability) for DGP. GA further finds setand of non✓2012 [35] 2013 [35] 2 2013 1 2 1- 3 An An framework MO planning has is developed, framework namely has developed, improved namely multi-objective improved harmony multi-objective search (IMOHS), harmony search which (IMOHS), can evaluate which the DGP. can evaluate the DGP. ✓ [29] 2010 5 2 fuzzy embedded ison toevolution solve fuzzy weighted single objective function employing fuzzy set theory for MODGP. [37] [37] 3 The methodology The based on Pareto frontier based differential Pareto frontier differential (PFDE) evolution algorithm (PFDE) isDG proposed algorithm for optimal is proposed MODGP for optimal (sizing and MODGP location). (sizing location). ✓ - -✓ -- search - function, -multi-criteria A hybrid A GA-PSO hybrid GA-PSO isGA proposed ismodel proposed towith solve MODGP to solve MODGP problem problem regarding regarding DG location location (with GA) (with and GA) capacity and capacity (with PSO) (with respectively. PSO) respectively. [31] [31] 2012 3 1 function, which which further isemployed solved further by solved GA toby ensure GA to optimal ensure DG optimal (wind) planning. (wind) planning. to find time varying voltage magnitude and loss sensitivity factor at each node. ✓ 1 An - exhaustive [34] 2013 2 1 An exhaustive analysis (ES) search has analysis presented (ES) has IMO presented for MODGP with IMO problem for MODGP under variable problem load under conditions. variable load conditions. Ref. Year [26] MO PTC A B C D Contribution of MONTR-Based Planning -✓ -- programming with goal GA with goal methodology programming finds methodology a solution with finds associated awhere solution uncertainties with associated among uncertainties MO and constraints. among MO and constraints. 2008 2 1 MO✓ 2 1 dominant solutions for wind-based DG units siting, and maintenance schedules. B C -- MO D -- optimization Contribution ofand MODGP-based [28] 2009Ref. [26] [28] 4 Year 2008 2009 4 An An MO problem optimization for multiple micro-turbines for multiple (DG) micro-turbines placement and (DG) sizes placement isPlanning solved and by sizes hybrid isclustering solved PSO and by SFL hybrid algorithms PSO and based SFL based ✓ GA MCS-embedded GA isproblem presented to solve chance constrained programming framework considering future uncertainties loads/DGs. [30] 2011 5PTC 1A The DGP problem The for DGP location, problem size, for and location, type (REG size, and and PEV) type issizing, (REG defined as PEV) MO isobjectives mixed defined integer as MO nonlinear mixed integer programming nonlinear (MINLP), programming in which (MINLP), in which A goal attainment A goal attainment method (GoA) method (GoA) presented, has presented, goal where programming goal programming transforms transforms multiple multiple objective objective functions functions intoalgorithms aaof single into anonsingle objective MODGP MODGP problem problem is solved is by solved two stage by two heuristic stage heuristic iterative iterative method including method including clustering (outer) and (outer) EA (inner) and EA optimizations (inner) optimizations respectively, respectively, The multi-criteria planning model aims to achieve contrasting (cost and reliability) for DGP. GA further finds set of objective is also GA comparison also used of for results and advocated ofhas results for and large advocated DNW systems for large even DNW with systems suboptimal even with solutions. suboptimal solutions. ✓2012 ✓ 1 GA -✓ [36] 2013 [36] 3 2013 1 3 1-o5 [38] [38] 2 2014 2 ✓ The on MODGP p annapp ng me hod and ecomparison aalgorithm edfuzzy wo PTC Componen p ann ng eDGs hn que MODGP ed DG pwhich a objectives. emen --- ba --- eMODGP -used ---ve -for -is [32] 2012 [33] [33] 2 [34] 2013 En Tab oo o om objectives. ✓ 5 -✓ The proposed The proposed algorithm NSGA finds arrangements finds for arrangements wind-based DGs for wind-based regarding compromise regarding solution compromise among contrasting solution among objectives. contrasting [27]En 2014 2008 [27] 3 e 5 [32] 2008 1 3 1 formulation based GA and εkdecision-making constrained method is proposed to solve best compromise (tradeoff) solution for DM [35] 2bu- on1 An MO planning has developed, namely improved multi-objective harmony search (IMOHS), can evaluate the A mu obed oa htime ba ed on aon m gene aDG go hm mGA aanselect m a the ade othe among enode. abGA) yMO ndba ecompeting and powe erespectively. Pa e DGP. o on o u on o ob a n ad a framework, followed framework, by fuzzy followed decision-making by tool to select the most tool to preferred Pareto most preferred optimal solution Pareto optimal satisfying solution satisfying competing objectives. -is NSGA NSGA-II used an NSGA-II to obtain is compromise used to obtain (Pareto compromise frontier) solution (Pareto frontier) for local solution distribution for alocation local company distribution (LDC). company (LDC). A hybrid GA-PSO is proposed to solve MODGP problem regarding DG (with and capacity (with PSO) 2012 3 1- 2 ✓ function, function, which isframework which further iso solved further by solved GA toby ensure GA to optimal ensure DG optimal (wind) planning. (wind) planning. to find to find varying time voltage varying magnitude voltage magnitude and loss and sensitivity loss sensitivity factor atDG factor each node. at each dominant solutions for wind-based units sizing, siting, and maintenance schedules. ✓ -✓ - --fuzzy - -- embedded [90] 2009 [29] 5 20101[21] [31] ✓ an ✓ [29] 5-2005 2010 2 - 4 ✓ -5 12A A fuzzy GA is embedded employed GA to solve is employed fuzzy weighted to solve single fuzzy objective weighted function single objective employing function fuzzy employing set theory for fuzzy MODGP. set theory for MODGP. (DISCO). opo e onfigu aMO onattainment om(ES) p ann ng pe pe ve An An exhaustive analysis search has analysis presented (ES) with has IMO presented for MODGP with IMO problem for MODGP under variable problem load under conditions. variable load conditions. optimization problem for multiple micro-turbines (DG) placement and sizes solved by hybrid PSO and SFL algorithms -✓ - exhaustive - search [37] 2013R [37] 3 Y 2013 1 MO✓ 3 1A methodology based methodology on Pareto frontier based on differential Pareto frontier evolution differential (PFDE) evolution algorithm (PFDE) is proposed algorithm for optimal is isproposed MODGP for optimal (sizing and MODGP location). (sizing and location). The The MO (mathematical probabilistic programming) (mathematical framework programming) has proposed framework for has various proposed DER planning for various (six DER DG planning types, size, (six DG location) types, size, location) goal method (GoA) has presented, where goal programming transforms multiple objective functions into aasingle objective MODGP MODGP problem problem is solved is by solved two stage by two stage heuristic iterative iterative method method including clustering clustering (outer) and (outer) EA (inner) and EA optimizations (inner) optimizations respectively, PTC C og De -- probabilistic Con bu on ocontrasting MODGP b dincluding P objectives nn ng The multi-criteria The multi-criteria planning planning model model toheuristic achieve aims to achieve contrasting objectives (cost and (cost reliability) and reliability) for DGP. for GA DGP. further GA finds further finds of nonabased set respectively, of non[35] 2013 2 An MO planning framework has developed, namely improved multi-objective harmony search (IMOHS), which can evaluate the DGP. ✓2012 -2 B -✓ - MO -- - A [28] 2009 4 2009 1 4 1 [36] 3 MCS-embedded MCS-embedded GA is presented to GA solve is presented chance constrained toaims chance programming constrained framework programming considering framework considering uncertainties future of loads/DGs. uncertainties ofset loads/DGs. [30] 2011[22] [28] [30] 5 2005 [33] 2011 5 ✓ -- ng [32] 2012 1 ✓ -used -is [33] 2obu 1ed ✓ --ann [34] [34] 2013 2013 1 - ---pba - also with the same formulation (GA and εsolve constrained method), aDG double tradeoff method isfuture presented to find the best GA is GA comparison also used of for comparison and advocated of results for and large advocated DNW systems for even DNW with systems suboptimal even with solutions. suboptimal solutions. framework, followed by fuzzy decision-making tool to the most preferred Pareto optimal satisfying competing Tab The The on on✓ ann ed p me afor hod ng me and hod eDISCOs aband ed wo eMCDM acontribution ked wo PTC k Componen PTC Componen n p ann nselect ng ann edlarge hn que efactor hn MODGP MODGP MO ba MO ed ba ped aomethod DG emen p aoalternative. emen The DGP problem The DGP location, problem size, for and location, type (REG size, and and type issizing, (REG defined and as PEV) MO is mixed defined integer as MO nonlinear mixed integer programming nonlinear (MINLP), programming indSFL which inbased which ✓ ✓ --ba - Similarly, [39] 2014 [39] 2 e 5on 2014 1 bu3on 2 1-MODGP 1-o MODGP that aims at DISCOs that contribution aims at the competitive in electricity the competitive market. electricity Moreover modified Moreover augmented modified ε-constrained augmented ε-constrained along with method with a on om h p p function, isin further solved by GA to ensure (wind) planning. En e 5Tab o wo along MODGP ofind mu on dresults on GA nda on nPEV) dp m hod p omarket. oo vma h bque omp om dDG osolution u on DM to time towhich find varying time voltage varying magnitude voltage magnitude and loss and sensitivity loss sensitivity factor at each node. at node. Fuzzy mu dfor on mak ng go hm opo doptimal o popo op nng on o each ou ava ab yhybrid arespectively. DNW ga ng n(MINLP), kobjectives. onfigu v o dominant dominant solutions solutions for wind-based for wind-based DG units DG units siting, sizing, and siting, maintenance and maintenance schedules. schedules. An MO optimization problem for multiple micro-turbines (DG) placement and sizes is solved by PSO and algorithms ✓ ✓ A hybrid GA-PSO A is hybrid proposed GA-PSO to solve is proposed MODGP to problem solve MODGP regarding problem DG location regarding (with DG GA) location and capacity (with GA) (with and PSO) capacity (with PSO) respectively. [31] 2012 [31] 3 2012 1 3 1 [38] 2014 [38] 2 2014 2 ✓ 91 2009 [29] 2 20101[23] [29] -✓ an - -fuzzy - - embedded MCS embedded in GA planning methodology is proposed to improve the accuracy for stochastic DG integration with tradeoff solution. [36] 2013 3 11✓ ✓ 5 2007 2010 2 2✓ 5 2 A A fuzzy GA is embedded employed GA to solve is employed fuzzy weighted to solve single fuzzy objective weighted function single objective employing function fuzzy employing set theory for fuzzy MODGP. set theory for MODGP. [37] The methodology based on Pareto frontier differential evolution (PFDE) algorithm is proposed for optimal MODGP (sizing and location). NSGA-II is used an NSGA-II to obtain is compromise used obtain (Pareto compromise frontier) solution (Pareto frontier) for a local solution distribution for a local company distribution (LDC). company (LDC). FDM has FDM for best compromise utilized for best solution compromise among the solution set of among Pareto optimal the set of solutions. Pareto optimal solutions. D problem is solved by two stage iterative including clustering (outer) EAGA (inner) optimizations respectively, ope a- ona bu onMO phas ann ng The The multi-criteria planning planning model aims model toheuristic achieve to achieve contrasting objectives objectives (cost and (cost reliability) and reliability) forand DGP. DGP. further GA finds further set finds of nonaDGP. set of non✓ - dutilized - - CO - MODGP -multi-criteria 2013 2013 2PTC B 1A2C B✓ An planning An MO planning framework framework has developed, has developed, namely improved namely improved multi-objective multi-objective harmony harmony search (IMOHS), search (IMOHS), which can which evaluate can evaluate the the DGP. framework, followed by fuzzy decision-making tool to most preferred Pareto optimal solution satisfying competing objectives. R R Y [35] Y MO[35] PTC MO A D1 C D Con bu Con on oaims bu MODGP on ocontrasting MODGP bselect dmethod Pthe b nn dng Pmultiple nn ng method goal attainment (GoA) has method presented, (GoA) where has presented, goal programming where goal transforms programming transforms objective multiple functions objective intofor a single functions objective into aasingle objective [33] 2012 ✓ ✓ ✓ - --goal - attainment [24] 2008 3 ✓2013 NSGA-II along with max-min approach solves MODGP problem considering future load uncertainties and risk management - probabilistic -m-- - yA -is [34] [34] 2013 2 11- 2 --✓ 1A MCS-embedded MCS-embedded GA presented to GA solve is presented chance constrained to solve chance programming constrained framework programming considering framework future considering uncertainties future of loads/DGs. uncertainties of loads/DGs. [30] 2011 [30] 5 2011 1 5 1 The DGP problem for location, size, and type (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), in which ✓ [32] 2012 [32] 2012 The MO MO (mathematical probabilistic programming) (mathematical framework programming) has proposed framework for has various proposed DER planning for various (six DER DG planning types, size, (six and DG location) types, size, and location) w h h m o mu on GA nd on n d m hod doub d o m hod p n d o nd h b n v ✓ 3 -✓ MO [40]En 2014 [40] 2 e 5 2013 2014 1 2bu- on1 1-o MODGP augmented εdominant MO constrained augmented method εoptimization proposed for DGP proposed planning for on DGP short term basis for short future term DNWs for future DNWs functionalities. with ANM functionalities. to find varying voltage magnitude and loss sensitivity factor node. dominant solutions solutions for wind-based for DG units DG units siting, sizing, and siting, maintenance and maintenance schedules. schedules. The on ed po --ann ng me hod eGA aproblem ed wo kmethod PTC Componen ann eeach hn que MODGP MO ba ed DG p ao and emen An MO optimization An MO problem multiple multiple micro-turbines (DG) placement placement and sizes and is solved sizes solved hybrid PSO hybrid SFL and algorithms -- ba -- add En En Tab o PSO o SFL [37] The methodology based on Pareto frontier differential evolution (PFDE) algorithm proposed optimal MODGP (sizing location). MODGP MODGP mu oon mu b time d on on band GA dconstrained nd on GA nd nwind-based on d m hod nfor dDG m psizing, hod opo dp opo on p oplanning v d h ong obat(DG) von h omp bh om om d owith disuby ofor on o on uby DM on DM [38] 2014 2 Th1 function, pap mu ob vIMO d op afor on op m za on pmicro-turbines ob m on d ng aomp ubasis ais on o noANM up o and mpa oandalgorithms a based ona mbased va a on n annua which further solved which is by further toon ensure solved by optimal GA to ensure (wind) optimal planning. DG (wind) planning. EA employed to solve considering both time-varying generation and demand behavior aiming at company various technical impacts of ✓ ✓ - DISCOs -is -isfunction, -contribution [36] 2013 3 1- 3 --✓ ✓ ✓ ✓ ---hybrid --- MC an NSGA-II is to to obtain compromise (Pareto frontier) solution aomulti-objective local distribution (LDC). 92 2009 [31] A GA-PSO is hybrid solve isthe proposed MODGP to problem solve regarding DG location regarding (with DG GA) location and capacity (with GA) (with and PSO) capacity (with PSO) respectively. 3 2012 1 6 ✓2013 3 1 mb dd d An nGA-PSO pused nn ng mnon-dominated hodo ogy pMODGP opo d oproblem mp ov h to u ythe o hand DG nsearch gaugmented on wrespectively. hwith dwhich oalong ocan ufinds on [39] 3 2012 20142[25] [31] [39] 2 2008 [36] 2014 2 1that aims at that aims atGA DISCOs in contribution competitive in electricity the competitive market. electricity Moreover market. modified Moreover augmented modified ε-constrained method ε-constrained with method along with ✓ - bu -CO The MOO planning framework MOO planning based framework on based on sorting non-dominated genetic algorithm genetic IIfor (NSGA-II) algorithm along II (NSGA-II) with MCS along (for uncertain MCS operation (for uncertain operation The multi-criteria planning model aims achieve contrasting (cost reliability) DGP. GA a can set of non✓ ✓ - -- CO -A -proposed [35] [35] 2013 2013 2 12 An planning framework framework has developed, has developed, namely improved namely improved multi-objective harmony harmony (IMOHS), search (IMOHS), which evaluate evaluate the the DGP. framework, framework, followed followed by fuzzy by decision-making fuzzy decision-making tool to tool the select most preferred preferred Pareto optimal Pareto solution optimal solution satisfying satisfying competing competing objectives. objectives. D DGP problem for location, size, and type (REG and PEV) isbobjectives defined as MO mixed nonlinear programming (MINLP), inDGP. which d 1 MODGP on eede ope a planning on pMO ann ng The pby ob em oto ved won hiterative b naselect ysorting PSO BPSO omost find eede w hfor ng hedu efurther MODGP is MO solved problem by two is stage solved heuristic two iterative stage heuristic method including method clustering including (outer) clustering and EA integer (inner) (outer) optimizations and EA (inner) respectively, optimizations respectively, DGP -✓ --D -- problem --GA The [41] 2014R [34] [41] 2 Y 2013 2014 1 MO✓ 2 1 PTC A B C D Con bu o MODGP d P nn ng ✓ [38] 2014 2 1 ✓ N ong w h m m n pp o h o v MODGP p ob m on d ng u u o d un n nd k m n g m n The MO probabilistic (mathematical programming) framework has proposed for various DER planning (six DG types, size, and location) A goal attainment A method goal attainment (GoA) has method presented, (GoA) where has presented, goal programming where goal transforms programming multiple transforms objective multiple functions objective into a single functions objective into a single objective FDM FDM for best has utilized for best solution compromise among solution among Pareto optimal thestorage set of solutions. Pareto optimal solutions. ✓2013 - m✓has - utilized - hOPF, [33] 2012 [33] 2 2012 1 2 1- 3 -✓ scenarios) and scenarios) addresses and DGP OPF, problem addresses ofcompromise DGP REG and problem storage REG integration and integration DNW. DNW. dominant solutions for wind-based DG sizing, siting, and maintenance schedules. Enbu o SFL An MO optimization An MO optimization problem for multiple micro-turbines micro-turbines placement (DG) placement and sizes and solved solved PSO hybrid PSO algorithms algorithms based based ✓ ✓ -w -methodology me h y w m hand h ocompromise mu m o on mu GA on nd GA on nd nathe on dunits mset hod nof d m doub hod doub dhn o m dhn hod m pwithin hod nMO osizes nnd hby ofor bhybrid nd hby b np nSFL v and [37] 2013 3 1 1ed The The methodology based on based Pareto on frontier Pareto differential frontier differential evolution evolution (PFDE) algorithm (PFDE) algorithm is isdis proposed optimal for MODGP optimal MODGP (sizing and (sizing location). and location). an NSGA-II is used to (Pareto frontier) solution aoMODGP local distribution company ✓ ✓ -o ng --ba ---find --- ayng ---varying 2012 [32] 5 o2008 2012 1 bu2on 5 1 Tab e 5[32] The on Tab e 5Tab on The e MODGP 5on[37] The on ba✓ obu ed pon ann MODGP me pba ann ed p ann me a --ed hod ng me ktime hod PTC emagnitude aband ed Componen wo emethodology aproblem kobtain ed wo PTC kfor nsensitivity Componen pmultiple PTC ann ng Componen eBPSO hn n p que ann nng MODGP p (DG) ann ewithin MO eeach ba que ed DG MODGP p MO apdisproposed emen ba ed ba p ed ao(LDC). DG aovand emen ✓ -hod - and GA with goal programming finds with associated uncertainties among and constraints. [26] 1-MODGP time to find voltage varying voltage and magnitude and factor loss sensitivity at each node. factor at node. MODGP owo on dconstrained on GA nd on nsolution d hod pplanning. opo dnd omarket. vfor hMoreover omp om dDG o DNWs uemen on DM ✓ -on -mu [36] [36] 2013 3 1 1 to Re onfigu ob em an MO amewo kloss ha ohDGP ved by ba ed ea hng aoque go hm abasis find ngMO he op ma au o he w he along o a with y ob e ve e ab y and mp oy dnplanning osolved olearning-based vmethod on densure ng bo m vm ynamely ngfor goptimal nDGP on d m nd bbmost ha vm o m ng vaugmented ou hn mp o ✓ -- augmented - EA -- pdmb [39] 2014 2 1- 3 -✓ that aims at DISCOs contribution in the competitive electricity modified ε-constrained method which is further which by further GA toogy solved by optimal GA to DG ensure (wind) DG (wind) planning. ✓2013 [40] 2 1 MO εfunction, MO augmented εMO proposed method for planning on short term basis for short future term DNWs for with future with ANM functionalities. ✓ function, ✓ Quasi-oppositional Quasi-oppositional teaching teaching optimization learning-based (QOTLBO) optimization methodology, (QOTLBO) aon variant of TLBO a proposed variant of as TLBO proposed regarding as MOO regarding 2013 2 1 An MO framework has developed, improved multi-objective harmony (IMOHS), can evaluate the 93 2012 [40]En 2 2014 1[27] [35] framework, framework, followed followed by fuzzy by decision-making fuzzy decision-making to select tool to the select most the preferred preferred Pareto optimal Pareto solution optimal solution satisfying satisfying competing competing objectives. objectives. ✓ MC dd nconstrained GA dd d p The n nn GA ng pmis nn hodo ng m hodo p ogy opo dpproposed opo otype mp dovtool otype mp hand ov uplanning hand y ohas u o ymethodology, h oas oDG h n gMO DG on nsearch w g ANM hnonlinear on dMOO wfunctionalities. o hnonlinear o(six d uwhich oDG on otypes, ufinds on The DGP problem DGP problem for location, for location, size, and size, and (REG (REG PEV) isobjectives defined PEV) is defined MO mixed as integer mixed integer programming (MINLP), (MINLP), inDGP. which in which ✓ MO probabilistic (mathematical programming) framework proposed for various DER planning size, and location) - MC - --mb The proposed NSGA algorithm finds arrangements for wind-based DGs regarding compromise solution among contrasting objectives. En En o o o The multi-criteria The planning multi-criteria model aims planning to achieve model contrasting aims to achieve objectives contrasting (cost and reliability) (cost for and DGP. reliability) GA further for DGP. finds GA aprogramming set further of nona set of non✓ D CO [42] 2014 [42] 3 2008 [38] 2014 1 3 ✓2014 3 11- 2 --✓ powe o ✓ [38] 2014 2 1 1 DGP FDM has utilized best compromise among set of Pareto optimal solutions. problem MODGP ishNSGA-II solved problem by two is stage solved heuristic two iterative stage heuristic method iterative method clustering including (outer) clustering and EA (inner) (outer) optimizations and (inner) optimizations respectively, R Y [34] RMO R Y [34] B2013 MO C on DA PTC B D MODGP CN D ed Con on oby Con MODGP bu Con bsolution on odifferential d bu P MODGP on nn oincluding ng MODGP b d P boptimal nn dng P ✓2013 ✓ - B -o3CMODGP -✓ 2013PTC 2AYMO 1 PTC 2 1 The MOO planning The framework MOO planning based framework on non-dominated on sorting non-dominated genetic algorithm genetic II (NSGA-II) along IInisproposed (NSGA-II) with MCS along (for uncertain with MCS operation (for uncertain operation optimal optimal and sizing location of DGP and for sizing solving of DGP for solving multi-objective power flow power problem flow for (OPF) radial problem DNW. for radial DNW. An MO optimization problem for multiple micro-turbines placement and sizes solved by hybrid PSO and SFL algorithms based GA N ong w m h n m pp m ofor nbu his pp owind-based vohas hMODGP obased von MODGP pcompromise ob m poptimal on ob d m ng on u(DG) d uschedules. ng o(PFDE) u(OPF) d oalgorithm dng nun nd kconditions. m nd n proposed gk m m nEA n(LDC). g m n ✓ -mong -wydominant -m [37] 2013 3 1A The methodology The methodology on based Pareto frontier Pareto frontier differential evolution evolution (PFDE) is is optimal for optimal (sizing and (sizing location). and location). The p ann ng me hod and eDG acontribution ed wo kmulti-objective PTC Componen nthe psorting ann ng euun hn que MODGP ed DG arespectively, emen an an NSGA-II is used to obtain used to compromise obtain (Pareto (Pareto frontier) solution solution for ann local for distribution a algorithm local company company (LDC). ---- ba ---- location ----GA [39] 2014 2 1 that aims at DISCOs in the competitive electricity market. Moreover modified augmented along with An exhaustive analysis (ES) presented with IMO MODGP problem under variable load ✓✓ - on✓ -✓ 1 dominant [33] 2012 [33] 2 e 5 [37] 2012 1 solutions wind-based solutions for units sizing, siting, DG units maintenance siting, and maintenance schedules. [41] 2014[28] Tab [41] 2014 2bu ✓ wfor hpsearch h m o based mu on GA nd on nsizing, d for mfrontier) hod doub dn oalgorithm m hod p distribution nMO d ba ofornnd h ε-constrained bp MODGP n MODGP vmethod [36] 3 1 ✓ GA wOPF, hto go og mm m hodo ogy ostorage uand oo node. un mong MO nd on -gh - btime ✓ -find -me -varying [40] 2009 2013 2014 4 2 1 MODGP 1 MO augmented εMO method proposed for DGP planning short term basis for future DNWs with ANM functionalities. o-✓ muto MODGP onng MODGP d ooy on mu GA nd mu b on dproblem on on b GA nng dddconstrained nd on m GA hod on nd png n opo on dnd m dyMODGP hod n osize, do vn p hod opo hw bhed dp opo o omp o v(minimize om ddnd h ohe ointegration bop von h omp om o u omp on om odng DM ooptimal od uou osolution on o unonlinear DM on oprofit) DM find voltage time magnitude varying voltage and loss magnitude sensitivity and factor loss sensitivity at each factor at each node. We hod mu ecomparison eva ua on ed GA pon opo ma up g ada on nhn ud ng add ona eede owhich hang ng no ma o ed oop scenarios) and scenarios) and DGP OPF, problem addresses of DGP REG and problem of REG integration and storage within DNW. within DNW. MOPSO algorithm MOPSO has been algorithm for has MODGP been used problem for satisfying problem objectives satisfying objectives DISCO (minimize cost and maximize DISCO cost DG and owner maximize DG owner profit) in framework, followed by fuzzy decision-making tool to select the most preferred Pareto satisfying competing objectives. EA mp EA mp ooon oy oaddresses vhas d MO o oused vaon ng on bo dcompromise hba m bo von hadvocated m ng g vm yprogramming) ng on g nframework nd dfind on m nd d b m h vd nd oooptimal bb h m vng omixed vm ou vheme hn mp omp oin DGP DGP problem for location, for location, size, and type and (REG type and (REG PEV) is defined PEV) is as defined MO as MO integer mixed nonlinear integer programming programming (MINLP), (MINLP), inDGP. in which En En ootypes, o The MO The probabilistic MO probabilistic (mathematical (mathematical programming) framework has proposed has for various for DER various planning DER planning (six (six DG size, types, and size, location) and location) GA isad also used for of results and for DNW systems even with suboptimal solutions. FDM utilized for best among the set of Pareto solutions. ✓ ✓ 94 Tab 2012 3 on 3Rbu MC mb dd d n GA p nn ng m hodo ogy psolution opo dhn olarge mp ov heand ud yhn oMODGP oproposed h DG nsearch gmong on w h d oDG uv on ✓ ✓ 2013 [35] 2✓ 2013 1 bu 2 1 An MO planning An framework MO planning has developed, framework namely has developed, improved namely multi-objective improved harmony multi-objective search (IMOHS), harmony which (IMOHS), can evaluate which the DGP. can evaluate the e 5[35] The Tab e 5Tab on The oYe MODGP 5on[38] The on on ba obu ed-MODGP p on ann o ng MODGP ba me ed hod p ba ann and ed ng p e ann me ed hod ng wo me and k hod PTC e a and ed Componen wo e a k ed wo PTC k n Componen p PTC ann ng Componen e n p que ann n ng MODGP p ann hn ng que MO e ba que ed DG MODGP p MO a emen ba MO ed DG ba p ed a DG emen p a emen ✓ ✓ [43] 2014 [43] 4,3 4,3 Th p opo d N GA go hm nd ng m n o w nd b d DG g ng omp om o u on on ng ob ✓ ✓ [38] 2014 2014 2PTC 1 2 1 MO A B C D Con bu on o MODGP b d P nn ng D CO D multi-criteria CO Dfinding COnThe opo ogy o to me h p ma yto d bu on w hthe DG The we gh an be ad u ed by for he de on make o find o u nonon a(sizing ng equ edalong ob e with ve MOO planning framework based on sorting genetic algorithm II (NSGA-II) along with MCS (for uncertain operation The planning multi-criteria model aims planning toeede achieve model contrasting aims to achieve objectives contrasting (cost and objectives reliability) (cost for and reliability) GA further forfor DGP. finds GA aaregarding setfurther of finds ayregarding set of nonQuasi-oppositional Quasi-oppositional teaching learning-based teaching optimization learning-based (QOTLBO) optimization methodology, (QOTLBO) aon variant methodology, of TLBO a proposed variant of as TLBO MOO proposed as MOO DGP addition addition optimal generated finding optimal electricity generated prices electricity in anon-dominated competitive prices electrical in ad competitive market. electrical market. ✓ [37] 2013 3 1 The based on Pareto frontier differential evolution (PFDE) is optimal MODGP and location). NSGA-II an NSGA-II is used is obtain used to compromise obtain compromise (Pareto frontier) (Pareto frontier) solution solution for amarket. local distribution aDGP. local distribution company company (LDC). ✓ - DGP ✓ --- optimization A- fuzzy embedded is εm employed to fuzzy weighted single objective function employing fuzzy set theory for --- MO ---GA - an -methodology 2014 2 1 that aims that at DISCOs aims at DISCOs contribution contribution competitive in the competitive electricity electricity market. Moreover Moreover modified modified augmented augmented ε-constrained ε-constrained method along method with [40] MO augmented constrained method proposed for DGP planning short term basis for future DNWs with ANM functionalities. ✓ 1 An ✓ --✓ [41] 2014 [34] 2013 [34] 2 2013 1 5 ✓2014 2 1- 2 y w N ong woptimization hGA m nto pp ohmicro-turbines hsolve oon vin MODGP phod ob m on ng um u oalgorithm dand un n nd km n gMODGP. m no(LDC). [42] 2014[29] [39] [42] 3✓2010 [39] 3 2- m ✓ An MO problem for multiple problem for multiple (DG) micro-turbines placement and (DG) sizes placement is solved by sizes hybrid is solved PSO and by SFL hybrid algorithms PSO and based SFL algorithms based An h u v h n y E p n d w h MO o MODGP p ob und v b o d ond on MODGP o mu on b d on GA nd n d m p opo d o o v h b omp om d o o u on DM ✓ ✓ h- m m- location yGA ogo w m mu hp hh y on wfor m GA hpwith hog ond mu mDGP on oon mu GA nstrategy don nd m GA hod on doub n on d m hod nof d d omo doub hod hod doub dn o nd m ddnhod onn o for nd m pwithin h hod bnfor p dradial ondnproblem doptimal vhaim o bnnd hradial bn DNW. voptimal n v capacity and scenarios) and OPF, addresses DGP problem REG and storage integration DNW. dominant solutions dominant wind-based solutions DG for units wind-based sizing, siting, DG sizing, maintenance siting, and schedules. maintenance schedules. ✓ optimal optimal and sizing location of and for solving multi-objective DGP for solving optimal multi-objective power flow optimal power problem flow (OPF) DNW. for w og go mm ng m mm hodo ng m ogy hodo nd ogy ond nd ucompromise on w otype hdbu o w hm d un othe d un MO nd MO on nnd on ✓ ✓ -oCMODGP --ba -ann 2013 [36] 3AY 1 PTC 3PTC 1A R Y [36] 2RMO R Ye 5Tab MO B2013 MO C5on D✓ A B B Dhed C D bu oof Con MODGP Con bsolution on P on nn ng MODGP b d Ppnd b nn ng Pwith ng An MOPSO An improved MOPSO with preference (IMPSO-PS) strategy isaon proposed (IMPSO-PS) for MODGP is proposed the MODGP to with the toahn capacity achieve and The DGP problem for location, size, and (REG and PEV) is defined as MO mixed integer nonlinear programming (MINLP), in which - GA -w ed MCS-embedded GA is presented tofuzzy solve chance constrained framework considering future uncertainties of loads/DGs. [30] 2011 1-MODGP MO The MO probabilistic (mathematical (mathematical programming) programming) framework has has proposed for various for DER various planning DER planning (six DG types, (six size, types, and size, location) and location) 95 2013 1PTC NSGA ohng ve Pa eng ouoy op ma eCon onfigu aon on pbest ob em ounits avuoand nong epgprogramming onfigu aann on o(OPF) umong on among y ng wo ob eDG ve FDM has FDM utilized has for utilized best compromise for solution among among set of the set of optimal Pareto solutions. optimal solutions. Tab The e 5on The bu on obu on✓ pimproved ba ann p me hod me and epreference avand ed wo ehodo asizing k ed PTC k Componen PTC Componen n ann n ng p eDG hn ng que eproposed hn MODGP que MODGP MO ba MO ed ba p ed DG emen p amp emen The MOO planning framework based on non-dominated sorting genetic IIachieve (NSGA-II) along with MCS (for uncertain operation EA mp d hod oprobabilistic oomp MO on dwo ng bo hbu m yMODGP nframework on dPareto m nd balgorithm h vaim oubop m ng v ou o DG framework, followed framework, by fuzzy followed decision-making by decision-making tool to select the most tool to preferred select the Pareto most preferred optimal solution Pareto optimal satisfying solution competing satisfying objectives. competing objectives. En En o o GA o d o on o u nd dvo d g DNW y m v n w h m o u on ✓ [44] 2014 [44] 2 1 ✓ D CO [38] 2014 2 1 ✓ ✓ ✓ MC mb dd d MC n GA p mb MC nn dd ng d mb m n GA dd hodo d p ogy n nn GA ng p p m opo nn ng d m ogy o hodo mp p ov ogy opo h d p opo o u mp y d ov o o mp h o h ov u h y o n u g o y h o on w o DG h n d g DG o on n o u w g on h on d w o h o d u o on o u on [41] 2014 ✓ ✓ ✓ Quasi-oppositional teaching learning-based (QOTLBO) methodology, among TLBO as regarding [35] 2013[31] [39] [35] 2 2012 [39] 2013 1 3 ✓2014 2 1-MODGP 1- 2 o--✓ An MO planning An framework MO has has namely multi-objective harmony multi-objective (IMOHS), harmony search which (IMOHS), can which the DGP. can evaluate the DGP. Th dp N opo GA daims goplanning GA hm go hm ng nd mnamely nmethod ng omimproved w nproblem nd w dregarding nd DG dg DG d gh omp ng om omp o om uand on oom uvariant on mong ng on ob ng voprofit) ob vMOO algorithm MOPSO has been algorithm used for has MODGP been used problem for satisfying problem objectives satisfying (minimize DISCO (minimize cost and maximize DISCO DG and owner maximize DG in owner profit) inalong with mu MODGP MODGP d on GA oon nd bN on don on bGA nnd ddframework nd on m GA hod on nd pcontribution nopo on ddeveloped, dMODGP hod n ocompetitive dob(Pareto m voptimization p opo hproposed bbfrontier) d pimproved opo o omp o vng om dDGP omarket. ointegration bdobjectives von dshort h omp osearch bGA) om o u omp on od DM oon odof ucost o on oevaluate uproposed DM on DM locations. locations. NSGA-II is used to obtain compromise solution for amarket. local distribution company (LDC). ✓ - Th -pb--opo A GA-PSO is proposed to solve MODGP DG location (with capacity (with PSO) respectively. ✓ --- on -ohybrid ---umu - an -mu 2014 2 1 1 MOPSO that that DISCOs aims at DISCOs contribution in the the competitive electricity electricity Moreover Moreover modified modified augmented ε-constrained method along method [40] [40] MO augmented MO augmented εdeveloped, constrained εddifferential constrained proposed method for DGP for planning planning on term short basis future for DNWs future with DNWs with functionalities. functionalities. DGP scenarios) and OPF, addresses problem ofhod REG and storage within [42] 2014 ✓ ✓ ✓ [43]5 2 2013 20143 [43] 4,3 A y mb dat GA oy oPareto opDGP vopo y woin gh d ng ob un mp oy ng uterm hon om ynd o MODGP ✓ 4,3 -NBGA - C B -✓ [37] 3✓ 1 PTC 3PTC 1A The methodology The based methodology frontier based on frontier differential (PFDE) evolution algorithm (PFDE) proposed algorithm for optimal isnDNW. proposed MODGP for optimal and MODGP location). (sizing location). 96 2013 [37] MO mh- za on app oa hohon ba ed on NSGA ha ed mu yea MO pvng ann ng n ud ng DG afor oond aaugmented and wo kANM eε-constrained a and on a owith u u e e n o emen mong y m mu on GA on n d m hod doub don o hod nbasis ok hg b nonfigu v ANM ✓YMO2013 ✓ ong wop N GA m N m GA nvCO w hoptimal ong hhdd m hPareto n m MODGP pp ohmp noptimal hsizing p pp oob vohmicro-turbines hMODGP on o vnd dubu MODGP pevolution ng ob m u p on ob oprices un ng on dd unP omarket. uis nd d u un kelectrical o m dm un nby m nd ky m nd nd gon m n(sizing n ne m n DNW. R R Y MO ✓ A D An C D Con Con onao(DG) omicro-turbines bu MODGP on MODGP b btransforms nn d ng Pund nn ng An u An hpp u ow h vm nw yvutilized hfinding E n om ymultiple p E nm p d wprices h n MO d w hu MODGP MO oodm MODGP p ob m pund ob m vproposed bmultiple vn o dgsizes bsolutions. ond opisdon optimal location and of DGP for solving multi-objective optimal power flow (OPF) problem for radial D CO DhMO COoptimization Dgoal An MO problem optimization for problem for multiple placement and (DG) sizes placement is solved and hybrid solved PSO and by SFL hybrid algorithms PSO and based SFL algorithms based operation addition to finding addition to generated electricity generated electricity in electrical in au competitive market. A attainment method (GoA) has presented, where goal programming objective functions single MO probabilistic (mathematical programming) framework has for various DER planning (six DG types, size, and location) FDM has FDM has for utilized best compromise for best compromise solution solution among the among set of the Pareto set of optimal Pareto optimal solutions. GA w mb hThe go p og mm ng m hodo ogy nd obased ucompetitive on w hmp o dng un n mong MO nd on nwinto The MOO The planning MOO planning framework framework based on non-dominated on non-dominated sorting genetic sorting algorithm genetic algorithm II (NSGA-II) IIun (NSGA-II) along with along MCS with (for MCS uncertain (for uncertain operation En Tab En e 5 En on En[32] o oaprogramming oobjective o ✓ ✓ Quasi-oppositional teaching learning-based optimization (QOTLBO) methodology, aa variant of TLBO proposed as MOO regarding ✓ ✓ [36] 2013 [36] 3 2013 1 3 1 d GA p n d o o v h n on n d p og mm m wo k on d ng u u n o d DG ✓ MC dd n GA nn ng m hodo ogy p opo d o ov h u y o o h DG n g on h d o o u on The DGP problem The for DGP location, problem size, for and location, type (REG size, and and PEV) type is (REG defined and as PEV) MO is mixed defined integer as MO nonlinear mixed integer programming nonlinear (MINLP), in which (MINLP), in which 2012 5 1 The bu Tab on o e MODGP 5 The on ba bu ed p on ann o ng MODGP me hod ba and ed p e ann a ed ng wo me k hod PTC and Componen e a ed wo k n p PTC ann ng Componen e hn que n MODGP p ann ng MO e hn ba que ed DG MODGP p emen MO ba ed DG p a emen EA mp oy d EA o o v mp EA MO oy on d mp o d oy o v ng d bo MO o o h v on m MO d v y ng on ng bo d g h n ng m bo on v h y nd m ng d g v m n y nd ng b on g h n nd v o d on m m nd nd ng d b m h v v nd ou o b h m v hn ng o v m mp ng ou v o hn ou mp hn o mp o ✓ ✓ -m -o yv -ow -mu -omp [41] [41] 2014 2014 2 1- 2 y w 1 An GA GA u d- h oo uimproved dwith ook on omp oonoon on uGA owith nd upreference dvo nd d dvo o gd DNW oproblem gdoub ypreferred DNW m ythe v n m w hwith v ubop n w h m ubop onom up m on othe u on MODGP MODGP o on mu b d on on b GA d nd on GA on nd n on d m hod n d m p hod opo p opo o o v d h o o b v h omp b om omp d o o d u o on o u DM on o DM [42] 3 ✓ ✓ ✓ NSGA MO n wo onfigu a on p ann ng p ob m ga d ng h ad o b w n m m z ng pow o max m z ab y and m n m z nv m n w h m h h m o m mu h y on w m GA h h nd mu m mu n d nd m GA hod on nd doub n on d m hod n d d o m m hod hod doub p d o n m d d hod o o nd m p h hod b n d o n nd d v h o b nd h b n v n v MOPSO algorithm has been used for MODGP satisfying objectives (minimize DISCO cost and maximize DG owner profit) in framework, followed framework, by fuzzy followed decision-making by fuzzy decision-making tool to select the most tool to select Pareto most preferred optimal solution Pareto optimal satisfying solution competing satisfying objectives. competing objectives. improved MOPSO An preference MOPSO strategy (IMPSO-PS) strategy is proposed (IMPSO-PS) for MODGP is proposed for the MODGP aim to achieve with optimal aim to capacity achieve optimal and capacity and function, which is further solved by GA to ensure optimal DG (wind) planning. ✓ [38] 2014 [38] 2 2014 1 2 1 ✓ ✓ ✓ ✓ [39] that aims at DISCOs contribution in the competitive electricity market. Moreover modified augmented ε-constrained method along with ✓- N - hyb - dan -daugmented [40] [40] 1 -NSGA-II MO MO augmented εm constrained εoo(Pareto constrained method method proposed for planning DGP planning on short term short for future for DNWs future with DNWs ANM ANM functionalities. Th opo NwO GA go hm nd mfor npproposed w bfor don DG goptimal dstorage ng om o(LDC). u on mong on ngnvDNW. ob v functionalities. scenarios) scenarios) OPF, and addresses OPF, DGP problem DGP problem ofnd REG and of REG storage and integration integration within within DNW. ✓ optimal location and sizing DGP solving multi-objective power flow (OPF) problem for radial 97 2013 [44] 3 20142 [43] -u A GA P pis opo oobtain vof MODGP obosolution m gfor dDGP ng DG odistribution on holocal GA ndbasis pterm ybasis w hk Pm O pm y with 2014 4,3 ✓ ✓2014 GA ong hand m nod pp haddresses ong vfrontier) MODGP p ob m dsolution ng u uwoy don nDNW. n(LDC). g [44] 2✓ 1 2 -DGP✓ 1- 2 -✓ isp used NSGA-II to obtain compromise used to compromise local for aomp company distribution company DGP ✓ ✓eCO CO A u y on AD mb dd ym d mb GA dd dnn mp GA oy dmp oy od vMCS dstage u oPareto vogy w uh gh ysolution dw ng gh d ob ng vhaevolution ob un un mp on ng mp uoy ysolutions. ng uh oyneo(inner) ynd he oe ydw o anddan pDGP ebased A non equen aE eopyheuristic ma eevolution ab yofrontier) and p eDG on y em eun pon oEA v o(sizing on up eam o ob a n a u a e e u MC n mb MC nn dd ng mb GA dd hodo d p on ogy nto GA ng pmopo hodo ng m ogy o hodo mp ov opo dp opo oeuh(Pareto mp ydov o oprices hMODGP oMODGP ov u hcompetitive yvon on(PFDE) u go yh oon w h deis gDG on u w goon hMODGP on oa hMODGP od uand oon o eam/down u on addition optimal generated electricity in aoset electrical market. ✓D✓ ✓ -MC ✓ - mb B -dd -DoGA -p -dev [37] MO 2013 [37] 3AY B2013 1 MO 3PTC 1A The methodology The Pareto frontier based on differential frontier differential (PFDE) algorithm is proposed algorithm for optimal MODGP for optimal MODGP location). (sizing and location). MODGP is solved by two iterative method including clustering (outer) and optimizations respectively, locations. locations. FDM has utilized for best compromise among the of Pareto optimal EnR TabYe 5 En o o as PTC R C2MODGP C D Con bu on o MODGP Con bw dann bu P on nn omp ng bob d Pund nn ng An hd unproblem vmethodology hfinding np ynn h p nPTC d MO pnd m voDG bnalgorithm demen ond on The MOO The planning MOO planning framework framework based on based non-dominated on non-dominated sorting genetic sorting algorithm genetic IIaoaoproposed (NSGA-II) II (NSGA-II) along with along MCS with (for MCS uncertain (for uncertain operation operation Quasi-oppositional Quasi-oppositional teaching teaching learning-based learning-based optimization optimization (QOTLBO) (QOTLBO) methodology, methodology, variant of variant TLBO of proposed TLBO proposed as MOO regarding MOO regarding MOPSO algorithm has been used for MODGP problem satisfying objectives (minimize DISCO cost and maximize DG owner profit) in A go nm n m hod GoA h p n d wh go p og mm ng n o m mu p ob v un on n o ng ob v Tab Thee 5on Tab The bu e 5on Tab on The o bu e MODGP 5 on on The o bu on on ba o bu ed MODGP p on ba ann o ed ng MODGP p ba ann me ed hod ng p ba ann me and ed hod ng p e ann me a and ed hod ng wo e me a and k ed hod wo PTC e a and k ed Componen wo PTC e a k ed Componen wo PTC k n Componen p ann n ng Componen p e hn n ng p que ann e hn n ng MODGP p que ann e hn ng MODGP que MO e hn ba MODGP que MO ed DG ba MODGP p MO ed DG ba p MO ed a DG emen ba p ed a DG emen p a emen [33] 2012 1 EA mp oy d o o v MO on d ng bo h m v y ng g n on d m nd b h v o m ng v ou hn mp o ✓ ✓ ✓ GA w h go p og GA mm w h GA ng go m w p hodo h og go mm ogy p og ng nd m mm hodo ng o u m ogy on hodo w nd h ogy o o nd u on d un w o u h on n o w h mong d un o MO d n un nd on mong n MO n mong nd MO on nd n on n The MO probabilistic The MO (mathematical probabilistic programming) (mathematical framework programming) has proposed framework for has various proposed DER planning for various (six DER DG planning types, size, (six and DG location) types, size, and location) ✓ ✓ ✓ ✓ -- m --mb -w --GA -mb -varying [41] [41] 2014 2014 2 112 1 The yN m h y wdd m om mu m oon GA on GA on npsensitivity d mp hod np(REG d mpdm doub hod dis odefined d hod o pnonlinear hod p d nnd nd hgaim ok m bond bo nDG v capacity MC MC dd d GA p dw GA p d oomu o d ohvnd n om vand on hvnd n and non on dtype og non d mm og ng mm ng wo kund m on wo d kmdm ng on umixed dm unnd ng un uoprogramming un n oh d innDG [42] [42] 3 ✓ GA ong wN hDGP GA mbon m ong n hGA pp w ofor ong m hhlocation, oho vproblem ha nomp m MODGP pp n hvp pp oob o hMODGP on osize, d MODGP ng ob m u ob o un ng on um ddoub u nis ng om dm uun ko n un n gubop n k um nnd nn gd programming m noovdoptimal [43] 2014 4,3 ✓ to time voltage magnitude loss factor at each node. An improved MOPSO with preference (IMPSO-PS) proposed for MODGP with the to problem The DGP size, for and location, type (REG and PEV) is and as PEV) MO mixed integer as integer (MINLP), which (MINLP), which GA owh d on ohod uof nd dvo ohdefined DNW y v n w h ofuture u on MODGP on MODGP d find on ohuonfigu nd mu on on bon nεm ddoconstrained on m GA nd psizing on d n ostrategy da oum vd hod bg p opo omp om d ong ointegration von d hpower o b owithin omp on om om DM dnonlinear om o achieve u on DM ✓ -opo [40] 2014 2 -N 1- 3opoo-✓ MO augmented method proposed for DGP planning short term basis for DNWs with ANM functionalities. Dmu DFR p ob m ou mu d n MO o am wo kMO ou nng o ab yng and ong vvfor w h adap and v inmod fi d ba a go hm SAMBA o scenarios) scenarios) and OPF, and OPF, addresses problem DGP problem of REG and of REG storage and storage integration DNW. within DNW. optimal location optimal location and sizing and DGP for of DGP for multi-objective solving multi-objective optimal flow power (OPF) flow problem (OPF) problem for radial DNW. radial DNW. un h hn vaddresses d by GA oopo nin op m DG w nd p addition to finding optimal generated electricity in ahahcompetitive electrical [44] ✓✓ ✓ Th p dN Th p -go Th hm N opo nd GA d N goO ng GA hm m go nd o hm w nd ng nd b m np dDGP ng DG o w np nd gsolving o d bmp w dng nd omp DG om go DG d ng oonn u goptimal omp d mong om omp on o om on oDG mong ob u on on v on --bu --ddpPon [38] [38] 2✓ 2014 1 2 1 [39] 4 20142 [39] that aims at DISCOs that aims atu DISCOs in the contribution competitive electricity the competitive market. Moreover market. Moreover augmented modified ε-constrained augmented along method along with ✓ 98R 2014 MC mb MC dd dd p nn GA ng p m nn hodo ng m ogy hodo p ogy opo d p opo odg d oprices hnelectricity y ohon o y hou oh oDG hdistribution nymarket. gw on n hhou on w omethod h ε-constrained oad uset oon o uvwith A hyb dDGP A GA hyb d O GA p opo opo oused o v d MODGP oobtain o vo MODGP ob m ob g dng m ng g DG o ng DG on w on GA w nd GA p p hou Py w O P pd O vob yp vob mpBoy dan EA oGA o EA MO oy on dmb on dGA oy ocontribution vP ng d bo MO on o hdp vCon on m MO d vmodel y ng on ng bo d gve h nng m bo on vom h y nd m ng v n yMODGP nd bon gbfrontier) hd nd vd du m m nd nd dmodified bu m vng vnd o blocal h m vu hn ng ond vDGP. m mp ng vgw omong hn mp hn omp o on The multi-criteria aims to achieve contrasting objectives (cost and reliability) for further finds ofynonlocations. Ye 5 The R YMO R Y MOTab A PTC MO A PTC MO C on Bba D✓ A PTC CEA D AoC Chod D bu Con on Con MODGP bu Con MODGP b on om dbu P bov on nn omp d P MODGP nn d ng P b nn dh ng ng NSGA-II an NSGA-II to obtain is to (Pareto compromise solution (Pareto for abov local solution distribution for ann company (LDC). company (LDC). ✓ ✓ D CO D CO A ymp dd dplanning GA mp oy d obu ok v uann y w gh d ng ob vng un on mp oy ngMO yGA oDISCO yned o MODGP aD ea bba evand omp u on among on acompromise ob eed The MOO planning framework based on non-dominated genetic algorithm II (NSGA-II) along with (for operation TabR on PTC bu on✓ oY e MODGP 5 BThe bu ed pB on ann ng MODGP me ed p euis ann aused ed ng wo me km hod PTC Componen ehas aMO wo non pnd PTC ng Componen eoptimization hn que n MODGP psorting ann ng MO ePund hn ba que ed DG MODGP pnd emen MO ba DG p aand emen ✓ ✓ Quasi-oppositional Quasi-oppositional teaching teaching learning-based learning-based optimization (QOTLBO) (QOTLBO) methodology, methodology, aoau(minimize variant ah of variant TLBO of proposed TLBO proposed as MOO regarding MOO regarding MODGP pmb ob oand vwng d by wo g h ufrontier) vof m hod n ud ng u ng ou EA nn op m onMCS p vuncertain yas MOPSO MOPSO algorithm algorithm been has used been for used MODGP for MODGP problem problem satisfying satisfying objectives objectives (minimize DISCO cost and cost maximize maximize DG owner DG profit) owner inprofit) in GA w h go og mm ng m hodo ogy o u on w h o d un n mong nd on [34] 2013 2 1 An improved MOPSO with preference strategy (IMPSO-PS) is proposed for MODGP with the aim to achieve optimal capacity and An h u v An h h n u An y v h E u h h v p n y n h d E n y h h p E h o n MODGP p d w h n MO p d w ob o h m MODGP MO und o v MODGP p ob b m o p und d ob ond m v on b v o d b ond d on ond on FDM has utilized FDM for best has compromise utilized for best solution compromise among the solution set among Pareto optimal the set of solutions. Pareto optimal solutions. ✓ ✓ ✓ N GA N ong w hong mbhod w m hGoA n mbhod pp m oGoA n pp ohon vd oDG o vd m pwh obd m phod on ob m ng on udvng u ng oschedules. udvu ob d om nun nd noduDM kvon m nd n kon m m ng nn m ng no vDM A go A go nm n dom nm non m hon phGA n p dhMODGP wh n dMODGP go po og go mm pobdog ng mm n o m n mu ounm p mu ob p von ob un un ndugoDM n ugoDM ob ob v [41]✓ 2 DGP DGP DGP -- oon -- probabilistic -om --GA -mb -mu [42] [42] 2014 2014 3 1 3 yMODGP 1 The ✓ ✓ dominant solutions for wind-based units sizing, siting, and maintenance [43] [43] 4,3 4,3 [44] 2014 2 1 ✓ MODGP o mu MODGP mu b MODGP d on on mu b GA on nd on GA d on nd on GA n d d nd on m hod n nd m p hod n opo on d p hod n opo o m v d p opo h o v d p opo h o omp o b om d h o omp o b om h omp o d u omp o om o on o o on o on ✓ m w h h m o mu y on w GA h h nd m on o mu n d m GA hod nd doub on n d d o m m hod hod doub p n d d o o nd m h hod b p n d v o nd h b n v MO The MO (mathematical probabilistic programming) (mathematical framework programming) has proposed framework for has various proposed DER planning for various (six DER DG planning types, size, (six and DG location) types, size, and location) MC dd d GA p n d o o v h n on n d p og mm ng m wo k on d ng u u un n o d DG ✓ GA ✓ o u d GA scenarios) and addresses DGP problem ofDNW REG integration within DNW. nd vd d ylocation ng vo glocation gn uduof nd omfor nm v y ohand h nod optimal optimal and sizing and sizing DGP of DGP solving for multi-objective solving multi-objective power optimal flow power (OPF) flow (OPF) problem for radial for DNW. radial DNW. Th pd opo N hm nd ng o w nd b d DG goptimal d ng omp om ou ung onproblem mong on ngo ob vofunctionalities. addition addition toou finding to optimal optimal generated electricity electricity prices prices am competitive in competitive electrical market. market. o-unaugmented omp GA oo u omp om omp u nd on omp dvo oOPF, on ufinding d oooby nd gogy DNW nd dvo ynaneou o g d o nfor w g y DNW ubop y vm n w o u hv vnd on nelectrical h ovm m on uANM on locations. 99R 2015 A puzzy ACO ba ed au hm ha em en ed odvo mu aoin ostorage adnn on oaharmony PV and DSTATCOM un -✓ - mp -on [40] 3 MO 20144R [40] 2✓ 2014 1✓C2 ✓D✓ 2 1-GA 1- h D MO εog MO constrained augmented εMO proposed method for DGP proposed planning on short planning term basis for short future term DNWs basis for with future functionalities. with ANM on un wh on h wh u v hmethod dgo by oconstrained v d u oo op n m ud DG op m w nd DG pand w nd ng p ng EA EA oy d oogo oy ogo vhph d o MO oGA om v on d ng on bo dnGA hgenerated ng m bo v h y m ng g vimproved n yvNR ng on g nDGP nd dm on nd nd d bm m oubop bw h m vm ng oubop ou von hn ou DNWs mp hn mp ✓ ✓ go GA w GA ng go w p hodo h mm ogy ng nd mm ng op uGA m ogy on w hmp o nd unamely on d un w o hd on n o w hmp dnn un MO un nd mong n MO naugmented nd MO on nd no -og -mm [35] [39] 2013B2014 An MO planning has developed, search (IMOHS), which the Y [39] AY B C D Con bu on MODGP Con bu P on nn ng MODGP b dhon P nn ng D w CO CO CO D CO ✓ dthat n GA nn ng m dd hodo ogy nframework GA pphodo opo nn ng m hodo ogy hthe opo uu yMODGP d og ooo omong h ov hDG n u gon ymethodology, oon w omong dDGP DG o(minimize noGA ugw on on w hevaluate ovalong oasDGP. uDG on A hyb d GA Pd og O p opo d odhodo ohnd vo MODGP ppb ob m dmulti-objective ng DG on on w h GA nd yond hon P ncan Oand p d and ynon ✓✓✓ ✓ -MC - mb --dd -D -pyhMC -um 2014PTC 2✓ 1 MO 2PTC 1A aims at DISCOs that aims DISCOs in the contribution competitive in competitive market. electricity Moreover market. modified Moreover modified ε-constrained augmented method ε-constrained with method along ✓ Th mu nn ng mod oyov huelectricity v on ng ob v oproposed nd b yhMODGP u h nd o Quasi-oppositional learning-based optimization (QOTLBO) aop variant of TLBO proposed MOO regarding An hmb up v huimproved n y E p n dMODGP wm hfor MO MODGP p ob m und vou bo d MOPSO algorithm algorithm has been has used been for used problem problem satisfying satisfying objectives objectives (minimize DISCO cost cost maximize owner owner inprofit) ✓ ✓ ✓ ✓ ✓ A up y mb AGA GA u A mb mp dd oy ym d mb GA ocontribution o vMOPSO dpat mp GA oy y w d mp oteaching gh oy oMOPSO vdnon-dominated dubm u ong od v ob w gh v ynd d w un ng gh on d ob ng voy ob ng un uu vng on y un mp hoy on o ng y mp o u y ng uMODGP h oy ynn o h MODGP oDISCO yon o MODGP An improved An MOPSO with preference with preference strategy (IMPSO-PS) is is proposed for MODGP for with the with aim to the achieve aim to optimal achieve capacity optimal capacity and with and in DGP DGP MODGP MODGP p ob ob ohdd vm d by o wo v d by g wo hnd g hopo um v hod v m n ud hod ng n(IMPSO-PS) u ud ng u ou ng nd EA nn nd EA op m op m p on vhm ymaximize pav y DGprofit) The MOO planning The framework MOO planning based framework on based on non-dominated genetic algorithm sorting genetic II (NSGA-II) algorithm II (NSGA-II) with MCS along (for uncertain with MCS operation (for ✓ ✓ ✓ Th opo dh1dd N Th p opo Th hm dp N opo nd GA d N go ng GA hm m n go nd oGoA hm w ng nd m n ng DG om w nwh gostrategy bum w d nd omp bmp om d g DG dd ng oom g omp on du ng mong om omp on ooun om uoy on ng oom mong ob u on v mong ng on ob ng vng ob --dw --go --GA [42] 2014 3 1 An MO optimization problem for mu p m o usorting bd nng DG pp m n nd o v doalong by hyb dbnnd Pon O nd FL go bv avvuncertain d ✓ ✓ MODGP ow mu on b MODGP d on GA o nd mu on on b n d d on m GA hod nd p on d n d o v hod h b opo omp d o o v h b o u omp on o DM d o o u on o DM [43] [43] 2014 4,3 4,3 ✓ ✓ ✓ ✓ A u MOO am wo k ha p opo d o n wo k onfigu a on n h p n o m u b n ba d DG Enhan d g aoperation h a go hm EGSA a go hm o v m y m h y m m h h y o w m mu m h h y o on w mu m GA h o nd mu m GA on o on nd mu GA n on d on nd m GA hod n on d nd doub hod n on d m doub hod n d d o m m doub hod d hod o m doub p d hod o n m p d d hod o o n nd m p d h hod n nd p d h o nd d v h o b n nd v h b n v n [44] [44] 2014 2 1 2 N GA ong w h m N m n pp o ong h w o v h m MODGP m n p pp ob o m h on o v d MODGP ng u p ob o d un on n ng nd u k o m d n g m n k m n g m n A go nm n m hod h p n d go p og mm ng n o m mu p ob v un on n o ob v ona FDM has utilized FDM for best has compromise utilized for best solution compromise among the solution set of among Pareto optimal the set of solutions. Pareto optimal solutions. ✓ ✓ dom ndmb noptimal oo uvdp on o p won nd bv dod DG un ng ng nd m ndmm ncompetitive nywo h on du -o -nd-pmb -nvnd 2✓2013 2014 1 3✓✓ 2 1-MC ✓ 1- mb --✓ - MC location and sizing of DGP for solving multi-objective optimal power flow (OPF) problem for radial DNW. GA o u o omp on o u nd dvo d o g DNW m v n w h ubop m o u on addition addition to finding to optimal finding generated optimal generated electricity electricity prices in prices a in a competitive electrical electrical market. market. 100 2016 [41] 3 20142[36] [41] dd d GA MC dd GA dd d GA h n d o n o d n h p n o og v on h mm n ng n on d p m og wo n d mm k p on og ng m ng ng u k u m un wo d k ng on n u d u o ng o un d u u DG n un o n o d o DG o d DG locations. locations. m o y m ng vo v y g ng m vo gn g ud m gn nd ud o nd n o v y n o v y h o nod h nod ✓ ✓ GA hu GA go p hm og og ng hodo ng ogy nd ogy o nd u on w oopo u hmp on ow d un oob nun mong noon mong MO on nd n on n scenarios) OPF, scenarios) addresses and DGP OPF, problem addresses of DGP REG and problem storage of REG integration storage within integration DNW. An hmb u vhedmb An hwvex hn An yand vw h Ed ungo h h vmm npow y nd hhmm dE npm w ynn hovhde MO p Edphodo h o n MODGP p dm w h nng MO p dnd w ob ov ho m MODGP MO und ogh vn MODGP pand ob od p und d ob ond m v und bn v owithin dond bnond oDNW. don ond on m wo k op by u yng d on kn oo hmp mo pm dm P o MO op m o ung on yon ng omp ng obo uo on v D CO D CO omp MOP o owed by uzzy on mak ng oo op m zed Pa e o on u on ✓ ✓ ✓ ✓ MC MC dd MC n GA dd d p mb MC n nn GA dd ng p mb nn GA dd hodo ng m ogy n nn GA hodo ng p m ogy opo hodo m ogy opo o hodo mp d p ov ogy opo o mp h d p ov o u h y d ov o o u h o y h ov u h o DG y h o n u g o DG y h o on w g o DG h on d w g DG o h on n o d u w g o on h o d u w o on h o d u o on un on wh h u h v d by GA o u op DG w nd p nn ng EA mp oy o o EA MO mp d oy ng d bo o o v MO y on ng d g n ng bo on h m d m y nd ng b h v o on m nd ng d v nd ou b h v hn m mp v o ou hn mp ✓✓ -✓ PMO - ✓ -opo -uMO [40] 2014 [40] 2✓ 2014 1 2 -A hyb 1- d GA augmented MO augmented method constrained proposed method DGP proposed for DGP short planning for short future term DNWs basis for with ANM functionalities. with ANM functionalities. An pconstrained nn ng m woεob kvhas dg op d nu m y mp ov don mu obterm vbasis hon mony h MOH wh hDNWs v u tovoptimal hy DGP ✓ ✓ yεd dd dpmod GA oy dvhv od op vfor y w gh d ng ob vogproposed un on ng uMODGP y hwith oh yO onvob MODGP MOPSO algorithm used for problem satisfying objectives (minimize DISCO cost and DG owner profit) ✓ O A p dTh A GA hyb d Pd o O omb GA v p P MODGP opo O d pnn opo o p omp dhd mMODGP oo o MODGP ng ob DG m o ob gplanning on d m w ng h g DG GA d ong nd DG on p ow on y GA wu w hproposed nd hPofor GA O p nd p yfor w p vm hon PGA y ofuture O w haim P p ymaximize p An improved An improved MOPSO MOPSO preference with preference strategy strategy (IMPSO-PS) (IMPSO-PS) is is MODGP with the to the achieve aim capacity optimal capacity and in and ✓ mu mu phuteaching nn ng mod owith v on vMODGP ng on ng ob og vm nd b nd bmp DGP yoy o GA DGP h nd u nd o non oachieve non ✓ p dm p N opo GA d N go GA go nd hm nd m ohm w nd obob w nd DG b d DG d ng omp d om omp onun om u on on uproposed on on ng vn ob GA ow u GA o- hhyb omp GA om o omp u d o nd omp dvo oon u obeen nd u g dvo nd d dvo ynpon o m gd DNW o n w gumethodology, h yPFDE DNW ubop nhom w oang h v ubop h m ubop ond u u Th hodo bhhm dpp on Pm on d n vo u on go hm pw opo m MODGP ng nd on vMOO ✓ ✓ Quasi-oppositional learning-based teaching optimization learning-based (QOTLBO) optimization (QOTLBO) variant of TLBO among variant ofgon TLBO MOO regarding as mGA yN h1dQuasi-oppositional hTh m-opo om mu y on GA m od mu nppp d on GA hod nd doub npv dd d o m m hod hod doub pyvm nnd dmethodology, dy on on o nd m h hod bm nnd v okas nd hgproposed b vony regarding -GA [43] 2014 4,3 ✓ ✓ N ong GA w N ong GA m wDGP N hu ong non m pp w ong n-m m hhogy pp w oh m vnd o non m hMODGP ong m voby non hMODGP oong ob vomm m hMODGP pnoDNW ong ob v m MODGP ng on ob m ng on ob ov ud on ud u nun ng u d unun kng o m nd d n kd m nd npop gkon m nmong nd m mng nm n m ✓ ✓ -m -omb [44] [44] 2 1 MODGP pow ob m o g hon ud v m hod no ud ng u EA op p ✓ ✓ An MO op p ob m owo mu p m opwh u budog nu DG punng m n nd oou v pg dd by hyb dm Ov nd FL go b vuncertain d The MOO planning framework MOO planning based framework on non-dominated based on sorting non-dominated genetic algorithm sorting genetic II algorithm along II (NSGA-II) with MCS along (for with MCS (for operation ✓ ✓2014 MC pv noptimal d hd onp oog h on nmmm d pnog og mm ng m wo kmu uvob u Pnn un nnuncertain oonng ohm -m [42] 2014 [42] 3 2014 1 3 -DGP 1- 2 -✓ addition to generated electricity prices in aob competitive market. nmdd ndom A A go GoA nm nThe hd mdd nm p hod nvGA non m GoA wh hdw GoA go p dvp mm n ng go p n o go p mu ng p mm n ng o m n mu un o(NSGA-II) m p oelectrical ob obng un ng on un n ng o ob vdoperation obDG v locations. locations. ✓ nhod nu dom o--u undd on o o udd won nd od b hod w nd DG b un DG ng ng ng nd m ng n nd m n h nm du nund h du An h--dp An vnd hdmb u hob v nod y hfinding E n y h p E nhodo dun w h nonh MO dnd w o hm MODGP MO o MODGP pnnd ob p ob m v und bon von ougnvhn do bnon ond omp dou on on ✓ ✓ ✓ A go uw ymp mb A GA u ymp A mb mp oy ymp mb GA ov d mp GA u oy d mp op gh oy o vd dh u ond ng owh yng vob w ud gh v d w un ng gh on d ob mp ng voy ob un u vng on yv un mp hoy on ng y mp oy MODGP yv ng u h oyn yp omp ho MODGP oon ynw omp MODGP ✓✓✓ ✓ -dd -dgo [41] 2014 [41] 2✓ 2014 1 2 -A 1- mb Th DGP pddd m oof o on yp REG nd PEV d ndbng dMO MO m doon n og mm ng MdNLP h ✓ MC n GA MC nn ng m hodo ogy n GA p py ng opo nn ng d m o mp ov ogy h pyun opo u yoptimal d oodb mp o h ov hm DG nbu gon y w omong hm dv DG ounhn oond uvgon on on hradial oDNW. o nuowh on optimal location optimal and sizing location DGP and for sizing solving of multi-objective DGP for solving multi-objective power flow optimal (OPF) power problem flow for (OPF) radial problem DNW. for ✓ EA EA oy d mp EA o oy o v EA MO o oy o v on MO o oy o v ng on d bo MO o d o h v ng on m bo MO d v h y ng on m ng bo d g v h n y ng m bo on g v n y m ng d on g v n y nd nd ng on g m h n v nd o on m h nd v nd o d m h v m v nd ng ou o b h v m ng ou o m ng o hn ou o hn o mp o m y ng vo g m gn ud nd n v y o h nod ✓ ✓ GA h go p og mm GA ng m w hodo h go ogy p og nd mm o u on hodo w h ogy o nd d o u on n w h mong o d un nd n n MO nd m wo k o ow d by u y d on m k ng oo o h mo p d P o op o on y ng omp ng ob v ✓on wh hscenarios) and OPF, scenarios) addresses and DGP OPF, problem of DGP and problem storage REG and storage within integration within DNW. A d GA Pd p opo dhaddresses oowith o GA vnuu MODGP poov ob mof gintegration d ng DG onn on hhfor GA nd oMOH pm w hon pu voptimal y DGPcapacity and un ✓ unuMO hon wh ohyb vd on h d by uGA h h oO onk u v uhd by op ov v m dd by DG w nd oREG op npnmp ud nn DG op ng m w nd DG p w nn ng pm ng ✓ ✓ ✓ An improved MOPSO preference strategy (IMPSO-PS) proposed MODGP with aim tonn achieve ✓ ✓ An pDGP An nn ng po nn ng wo m h wo d kGA op d d op m dy m y d mu ov d mu v ob h(minimize v hDNW. MOH hy wh vkPhm uO DGP GA GA oun owh u omp o omp o u nd dvo nd dvo g d DNW oproblem g ynd DNW y vmony m w hw vmony ubop n h u on othe un ✓ MC mb dd d MC mb MC d mb GA dd om p d GA hv non p d on ofor oon vp no o d honv p n om og v on h mm nm ng nng on don p m og n do mm kob pud on og ng d mm ng ng wo un ku m on un wo d kw ng on nm udhom uong ng o un d u uDG nun oDG n ond oDG dhnon DG nudd N GA u d o ob n omp om P o on on ovm o du bu on omp ny LDC ✓ N GA ong wDGP h m -p N m GA nMO pp ohas ong hvogy w ong hm m MODGP m n pp ob h on ong v d MODGP ump u pwo ob m un on d nis ng nd umong ko m don un n gubop nwh nd gdvhmaximize m noprofit) DGP MOPSO algorithm MOPSO been algorithm used has MODGP been used problem for MODGP satisfying objectives satisfying objectives DISCO (minimize cost and maximize DISCO cost and owner DG inon owner profit) in ✓ mu p nn ng mod m o h v ng ob o nd b y DGP GA u h o [44]✓ 2014 2 DGP 1 opo ✓ ✓ Th p dN GA go Th hm p opo nd d N GA n go o hm w nd nd b d DG m n g o d ng w nd omp b om d DG o u g on d ng omp om o ob u on v mong on ng ob v ✓ Th m hodo b d on P o on d n vo u on PFDE go hm p opo d o op m MODGP ng nd o A go nm ndduon m GoA h pglearning-based n do wh go pmud og mm ng n ov m mu pdTLBO ob un on nhm o v y regarding Quasi-oppositional teaching learning-based teaching optimization (QOTLBO) optimization methodology, (QOTLBO) variant methodology, of ap proposed variant of TLBO MOO proposed regarding asb MOO obPQuasi-oppositional m MODGP v MODGP dpmb by ob wo m p ob o vob h m by o v do by gmv hv m u vu hod hu nDG ub u ud vp ng ung hod vw m n ng hod ou ng non u ud nd ng EA ng uoa nn ou ng op nd ou m EA on nn nd EA op nn m vhMODGP y op m on yp vd ✓ 4,3 -MODGP -✓ - u -u [43] 2014 [43] 4,3 2014 1 1- d pGA ✓ locations. An MO op An MO m op on m pP m pvhod owo ob mu m p owo mu m o png m um DG b n pd DG n p nd m n nd d by vvhn by d hyb dp FL Oh nd FL b ng d yob ✓ ✓ ✓ A yhd A dd d mb GA GA oy dmp od oy ohodo d opy vprices w yun d w gh d ob ob un vng on un mp oy on mp u oy yp ng uO oyvnd o oas ynon MODGP ✓ A hyb O povp--n opo hyb d oymp O o GA v pgogy MODGP opo O pogy opo o p ob o MODGP ob m o ob gdaon d m ng h g DG ong nd p ow h on y GA w nd h P GA p y w vhng hng P yynPO pgo Oo vhn yppgo vvhm Th MO pgo ob b m m pod og mm ng m wo kng h pDG d o vdu ou DER pmarket. nn DG yp nd o o on o-gohyb o EA MO on doptimal oy ng d bo otomp om hd m MO v yMODGP on ng do g n ng bo on nd m v m y nd ng b hdvn vvnun o dmarket. vw nd ou bng hvhyb ond vw oog ou mp ✓ -oy -m [42] 2014 [42] 3✓ 2014 1✓ ✓ ✓ 3 -EA 1- mp ✓ ✓ GA w hGA go wpdaddition hog GA mm w og GA ng go mm w p hodo hd og ng m mm hodo og ng nd mm hodo o u m on w og nd uh ogy on ow o nd uhhngh on d ow o u on nun ow h mong d o mong d nun nd MO on mong nh noO on mong nd MO nPm on on dom nP n o udd on o w nd bogy d DG un ng ng nd m n nm nnd hm to finding addition generated finding optimal electricity generated electricity in competitive prices electrical in aopo competitive electrical An hvo An yGA hEgn um vvo p n d wng yhond hvm Egn h p ob hh m MO und ogGA d ob ond m und on bnon omp onP ✓ ✓ Th DGP ph ob m oof ongn on nd yp REG PEV d nb p dMO MO m dnd nMO g flow ndndond p mm ng M NLP n wh h un wh hp hDGP by GA ond op DG w nd nn ng ✓ ✓h um vv optimal o nd yMC ng om ndwo m go m v nd yoptimal ng ud vsizing ynd gngu oh m vo n gand ud vMO ynd o oMODGP nhknno onu h vpnod yw nd nd ovog ymulti-objective h oMODGP nod hPop nod location and location for sizing of multi-objective DGP for solving optimal power flow optimal (OPF) power for (OPF) problem DNW. for radial DNW. ✓ ✓ m oon wo ow km d oby u d ydby dp on ddm ng on m oo ng oo ohm mo pog mo dp dm op P m oproblem op ong um on ong uradial y on ng omp yon ng ng ng ob v h ✓ mb MC dd d mb GA dd dow GA n d n ovy v dsolving hkpud n oog vn on n on n dmm ph og ng mm m ng wo k on wo don kmu on ud un u u nun ong n o dvoboDG ov dhong ✓ hkgo m D CO on bu on h omp m Mo ov mod d ugm nachieve d on non d m hod A gop nTh A m goomp A GoA nm nAn h nm p nng nm GoA hod wh huo GoA p go hon p d wh mm nobn go pm nod og o go m pymu ng nob ng vom nmu un omong m pon on ob n po vhuob un ng vob on un noptimal vo nomp oob v obDG DGP DGP ✓ ✓ ✓ ✓ Th opo Th dp d N opo dphod N go opo Th GA hm dpMOPSO N go opo nd GA hm d hod N go nd GA hm m n go ng nd o hm m w n nd ng nd obm w d nd ng DG mdng w nwh dv nd gstrategy DG bpmp ng w d nd gmm omp DG d bd ng om d gpkmm omp DG dob ng oois om um gproposed omp on d ng o om u omp on o mong om uaim on ng on o mong ob uwith on ng on v mong ob vng ob ng vng ob vw vcapacity An MO p nn ng m wo k h d v op d m y ov mu v h mony MOH wh h n u DGP GA o unm oGA GA on o o u u d o nd omp dvo on d o o u g DNW nd dvo y d v o n w g h DNW ubop y m o u v n w h ubop m o u on An improved improved with preference MOPSO strategy with preference (IMPSO-PS) is proposed (IMPSO-PS) for MODGP with for the MODGP to the aim to capacity achieve optimal and and n N GA u d o ob n omp om P o on o u on o o d bu on omp ny LDC ✓ ✓ ✓ MODGP pGA ob ovdopo vng dfor by wo gused uvfor vPFDE m hod nDG ud ng u ng ou nd nn op m on profit) po DG y mu Th pu m mu nn ng Th mod mu p m om ng p mod hv on mod m ng h m ob v oh hm vMODGP on nd ob ng v boob ynd o satisfying v oon nd DGP om b ndowu yuDISCO ho bopo DGP nd ycost GA DGP uEA ohmaximize h GA nd uhcost h o ymaximize non o ynon MOPSO been used has problem satisfying objectives (minimize objectives (minimize DISCO DG owner invon owner profit) in ✓✓ -✓ -un - algorithm -hm [44] 2014 [44] 2✓ 2014 1 2 -Th 1- on Th hodo Th m ogy hodo bhas ogy dopo on bon P o on P on oby d on no don vo ung on vo u on PFDE go hm go pGA opo hm dp op dom oand MODGP op m MODGP nd ong nd on ✓ A A GA hyb P O pnn P Oalgorithm d p v dMODGP MODGP ooom ohn v MODGP ob m p ob gon dw m g DG dw onn onn w hb on GA nd h GA nd yond w Pnon y PO w P png Oand vnd p FDM h u bphp omp oopnd u on mong h Png oMO op on wh hMOPSO hon un o vw on h d by wh h od u unn d by op o GA v m d obeen DG GA u w o op n p uMODGP DG op ng m nd DG ng p ng ✓ go og GA ng hd ogy og nd mm ng oo u m on hodo w h ogy nd d o on nproblem w h mong oob nd ndon noovmong nd onO ✓ An An v ph u An hvmm hn du An ywh v h E nyhodo um y huGA hv ph E n y nGA hu d E noob yhh h n MO p d Egn w o h n MODGP MO p d w o h n MO p w ob oun hm MODGP MO png und ob opmp v p und ob bng m v p und d ob ond von ovund ond op dd on bby op on MO op m on m o mu m onn un bu n DG m nod nd voy hyb go vhm d ✓✓4,3 ✓ ✓ ✓ -GA -✓ -dhyb [43] 2014 [43] 4,3 2014 1 1-nhyhnu Aunuw mb GA A mp d mb o vng dh yw mp gh oy d vn ob v yun wnd gh on d ng u vun yun hon ob y mp MODGP ng u don ydond hDG ondynypoFLMODGP locations. locations. Th MO pgo ob bow m m pong og mm m wo knm hp opo d ohm ou DER pMO nn ng nd ob on o nd v yon vo gw m ud nd onuREG v y oMODGP h o udd on dom oDGP n-wnto dom nd o-u b unoy on n d DG onm oudd un nd ob hod w ng d nd DG b ng un dd nd DG m un ng ng nng ng hog m ng du n nd m noy np h nm du nnod du ✓ ✓ ✓ dom ✓ addition finding addition optimal to generated finding optimal electricity generated prices electricity in a competitive prices electrical in a competitive market. electrical market. Th Th p ob DGP m p o ob o m o o on nd yp nd REG yp nd PEV nd d PEV n d d MO n d MO d n m g d non n g n non p og n mm p og ng mm M NLP ng M n NLP wh h n wh h A go A go nm n m hod n m GoA h GoA p h n p d wh n d go wh p go mm p og ng mm n ng o m n mu o m p mu ob p v ob un v on un n o on ng n o ob ng v ob v ✓ MO ugm n d on n d m hod p opo d o DGP p nn ng on ho m b o u u DNW w h ANM un on MODGP p ob m MODGP o v MODGP d p by ob wo m p ob g o v h m d u by o wo v d by g v wo h m u hod g h n u ud v ng m u hod v m n ng ud hod ou ng n u ud nd ng EA ng u nn ou ng op nd ou m EA on nn nd EA op p nn m v y op on m p on v y p v y ✓ Th opo N goMC Th hm pd opo nd N ng GA m n go ohm nd d ng DG m gonm odmgng omp bphm om DG ohng ugubop on ddm omp om ng ou u on on mong ngng vwv h GA p GA o udd ooGA uomp dp GA ooh um omp o ookudd omp udovd oow o nd omp uby dvo o on u dvo nd dvo DNW o nd ynoo DNW ov dwywo vnDNW ond ndm gon h yvd DNW n m wpMo yvm m w ong u hmod vmon ubop nokwop oon h on ubop oong m ovon uunon wo und ydw ddonond on m ng mo Pmong m y ng ob ✓ ✓ d GA non d mb oon d GA hpkd non p non o d p ougd og v hk mm ngddvo ng kyw og d mm ng un u un wo on nm d u uDG n oob dhong DG ✓ m D CO on on nvbd h omp kubop ov du ugm nouob don nonn domp m ✓ ✓ ng m hn v ondo ng oproposed nd bthe yaim u aim nd o non ✓ ✓ MC ✓mb An MO pm dd nn ng An MO m wo pTh An MO hmu ng p dm vunn m ng d n m m wo dn ybu v kmod mp op ov dd n mu op m dy mp ov vyo mp hm mony ov duob hvob hMODGP v h wh mony hh MOH n vdhooto uDGP MOH wh hhGA DGP wh noptimal vhh u vh ohod u DGP DGP nimproved N GA nknn N GA ugn dhud o op ob uwo dn ohnn omp ob om n omp Po om ooo Pon oy on ooob on ondnv d omony bu don omp bu on omp LDC ny LDC An MOPSO An improved with preference MOPSO strategy with preference (IMPSO-PS) strategy is proposed (IMPSO-PS) for is with for MODGP with to achieve optimal and capacity and un on un wh on h wh h h uGA v h d by o GA vh d o GA u op nob m u DG op w nd DG p w nn ng pmp nn ng o u nd yO o nd m go m v nd v nd g ng o m vo gn ud m v gn y nd ud ou oon nd n hvm n omu v y omu nod hMOH nod Th MOO pv nn m wo kMO b d non dom non d o ng g go hm N GA ong hoMODGP MC oo capacity un n op on ✓ An u vv P hvo An y h E ung v p h d n y h Eo MODGP p nou p dnod w ob hum MO und v MODGP bong o p d ob ond m und on vny dMODGP on ✓ ✓ ✓ Amb u dd y Ang GA u y dom d mb mp GA oy yyd d mb mp GA ovo dd oy oyon vng dbo mp u o oy o yw vg w domp mp u o gh oy y vd w dhn ong gh vd ob w ng gh v yn d ob w un ng gh v on ob un mp ng voy on ob un u on yng un mp hoy on oyp ng y mp o uhoy MODGP o ybp ng yo op uachieve oyw y the hhMODGP oP yO MODGP Th m hodo ogy d on P oby on d vo ud on PFDE go hm opo d op m ng vnd o on ✓ ✓ GA ppdd A hyb d o GA MODGP opo p ob d m g o v d MODGP ng DG p on m h gnd GA dh DG o on y wou w h P GA nd vhhm yond w pgo FDM h uon dO onm om ooyu mong hw oo onop m uh on -✓ -dmb -nopo -uMO [44] 2014 [44] 2✓ 2014 1✓ ✓✓✓ 2 -A 1- hydop ndd nop o uP opbpom w nd b d DG un ng ng m n nvoy h du ✓ Anhyb MO m locations. An onMO ob op An m m oboom mu m p p ob m on owo ob umu bby m n powo DG mu m pog o png um m m bon nm no o um DG nd bknd nm p DG m n pvPng dnd nd by m hyb np nd dGA P oh v O dDNW nd by oO hyb FL vnd dng by d go Pnn hyb O nd dGA bP FL O d nd go FL hm b vhm d yynon b von d Th MO Th p ob MO p ob m b h m h p m og mm p mm ng wo h wo p opo k h d p opo o v d ou o DER v ou p nn DER p nn DG ng yp DG yp nd o nd on o MODGP MODGP p ob p ob o v m d by o v d g h u g h u v hod v m n ud hod ng n u ud ng ng u ou ng nd ou EA nd EA op nn m op on m p on p locations. n nd OPF dd DGP p ob o REG nd o g n g on w n Th mu Th p mu nn ng Th mod mu p m nn o ng p mod h nn v ng on mod m o ng h m ob v o on v h o ng on ob ng v b ob y o v o nd DGP o b nd u y h o b DGP y o GA DGP u o h non nd u h nd o non o GA o u d o omp GA on o o u u d o nd omp dvo on d o o u g DNW nd dvo y m d v o n w g h DNW ubop y m m o u v on n w h ubop m o u on ✓ ✓ ✓ MC ✓mb MCdd d mb MC GAdd p d mb MC nDGP p dmb GA n oobvp dm GA oh on o vop d on oh on n o vdn on od h pn ond og v nyp on d hmm pnREG og ng n on dmm pmog ng wo n dmm kpmon og wod kdmng on ng wo udkum on un wou on n uo ng n o un u uoDG n o und mm oDGn ongd MoDG o d DG Thdd pdd nd PEV dng nmm MO mng ddkun ng gunudnon nd p og NLP n wh y h [31] [22] [33]

2012 2005 2012

[21] 2005 [89] [23] 2007 [32] 2016 2012

[29] [31] 3 [24] [27] [22] 3 [33] 2 [30]

[26] 4 [29] 3[21] [23] 2 [32] 5 [25] [28]

✓ ✓ ✓

2010 2012 1 2008 [27] 2008 2005 1 2012 1

5 ✓2008 3 3 -✓ ✓ 3 2 -

2 1 1-- 3 1 1-

2011 5 1 2008 2 1 2 2005 1 4 2010 5 2 2[26] ✓2008 2007 2 2012 1 5 1-2008 6 -- - 1 [28] 2009 2009 4 1 2012 3 14

✓

n mm hod A GoA nm pby nun m dby hod wh GoA pp mm dng nd o go m p mu og pmm ob ngu v vond obmony m poon ng vob on ng nun oob ng obDGP ✓ A go MO ugm n d on nk dyhgo m opo dmwh o yDGP p ng onob ho m oo n um u oboDNW wun hvANM ✓ An MO ph nn ng m wo dhhod vng op dn n mp ov hun MOH wh vong uon hong m✓ k onm ou ow dnd wo uo kgo m y wo ow kDG d on m ow kbu d ng y dgn ood u oon dun m kpud on hhog mo m oo ko ng oo P mo onn h p op m d p o on op P m yon omu ng op omp ng u ob y on ng omp ng vv ✓ h GA m h D m on on nbu hb on omp n omp v vnwnd yoh m kn ydmp Mo m kn ov mod ov dmod ugm do nun ugm dohngTLBO on n n on dv myodhng hod n ndomp m hod wv ob h m v nd m vo v yCO gng m vo g ud gn nd nd opdv ov y omu nod hPp nwo nAwh dom opby n wd nO dom nd oN bounPd on nby dng o opou un nd op bon w ng nd DG ng d DG m un ng nng ng hm m ng du n m nmo h nMo du nnod h Qu oppo on hyop ng n ng bw d m on QOTLBO m hodo ogy vyGA ng w h ✓✓ Aunhyb u✓ yd mb dd A mp uv oy yyCO d mb oD dd ow vng d GA u w mp oy d dnd opng vn ob v yy un gh on dh ng oy ob u yon un oP mp uv hd yp hh MODGP ✓ dom GA hyb P dO AodGA hyb Popo GA hyb d p opo o d O o GA v d P MODGP opo O o d MODGP opo o p ob oo dgh m MODGP p ob g o voong d m MODGP ng g DG onn ob g DG on d m ng g DG on GA dh oon w ng nd hDG on GA o w nd h y GA wpdu whon hnd h Ow p oy hMODGP nd P py Ow p P p y opo Ow yP pyOooMOO v un yp gv ndyop n GA dnuv o nvm om Pnop op on ond u ong ovgo d bu on omp ny LDC h on u✓ h un o on d wh hu oon u hob v m do DG GA nd onon pnudom um op ng DG w p nn ng Th MOO pGA nn m wo komp b by d on n dow o ng gnd n hm Ny GA ong w h vMC on ✓on

Energies 2017, 10, 208

20 of 44

4.1. Numerical Methods (1)

The ε-constrained method: In this technique, a specific objective function selects as master and others as slaves. Moreover, the slave objective functions considered as new constraints

• • • • (2)

Monte Carlo Simulation (MCS): The iterative techniques based on use of random numbers.

• • • • (3)

•

Benefits: Better computation efficiency, high precision and use as inner optimizations. Demerits: Rigid problem formulation and few variants can be inclusion in the calculations. Preference: Several applications in literature (for high precision optimization problems). MOPT Ref : DGP (A) [41]; CRU (C) [74].

Cone programming (Cone P): The technique addresses nonlinear convex problems. It aims at minimization of linear objective function over intersection of affine linear manifold intersection and product of (second order quadratic) cones.

• • • • (7)

Benefits: Simple to implement and determine optimal global solution (of small problems). Demerits: Requires very high computation and not efficient for large distribution systems. Preference: ES based optimum solutions compared with the other algorithm and performance is found by relative deviation among two solutions. MOPT Ref : DGP (A) [25,28,33].

Optimum power flow (OPF): The conventional method is used to solve complex planning problems, aiming at optimal performance of power systems.

• • • • (6)

Benefits: Simple to implement and aims at a tradeoff solution. Demerits: Requires very high computation and practically time consuming. Preference: Several applications in literature. MOPT Ref : DGP (A) [26,32].

Exhaustive Search (ES): Simple problem solving method aims at finding all possible solutions.

• • •

(5)

Benefits: Efficient for problems with less iteration, resulting in efficient processing time. Demerits: More the complex problem, more iteration and more processing time. Preference: Mainly deterministic and probabilistic types by behavior and outcome of the random process. Also used as inner optimizations in MO problems. MOPT Ref : DGP (A) [23,41].

Goal programming (Goal P): Suitable for MO formulations aiming at a tradeoff solutions.

• • • • (4)

Benefits: The technique efficiently generates a set of Pareto optimal or non-inferior solutions in MO-based problems. Demerits: High computation required for greater number of objective functions. Preference: Enables decision maker select a solution on preference basis. MOPT Ref : DGP (A) [21,22,39,40].

Benefits: Efficient computation efficiency and better precision for convex optimizations. Demerits: May cause inaccurate solutions for real time problem. Preference: Several applications in literature (for convex optimization problems). MOPT Ref : DGP (A) [45].

Sequential quadratic programming (SQP): The iterative method is proficient at solving nonlinear formulation with inequality constraints. The procedure involves.

Energies 2017, 10, 208

• • • • (8)

Benefits: Considered as the fastest method to solve nonlinear programming (NLP) problems. Demerits: Complexity increases with multiple quadratic subproblems at each iteration. Preference: Few applications in literature (mostly used for DGP optimization problems). MOPT Ref : DGP (A) [49].

Mixed integer linear programming (MILP): The conventional technique is appropriate to solve problems with linear objective functions, abiding associated constraints.

• • • • (9)

21 of 44

Benefits: The approach shows excellent convergence properties. Demerits: The more realistic model experiences more difficulty in finding solutions. Preference: Applicable to discrete and continuous variables (linear optimization problems). MOPT Ref: VPQ (B) [59].

Mixed integer nonlinear programming (MINLP): The conventional numerical technique is a combination of linear programming (LP), mixed integer programming (MIP) and nonlinear programming (NLP), respectively.

• • • •

Benefits: This approach demonstrates better computation efficiency and accuracy. Demerits: Difficult to implement nonlinear objective functions and with much iteration. Preference: Appropriate to address discrete, continuous variables, nonlinear power flow. Also, provides accurate, reliable and efficient solution for MOP formulation. MOPT Ref : DGP (A) [38–40]; CRU (C) [72,77,81].

(10) Dynamic Programming: (DynP): This is is sequential optimization technique based on multiple stages. The technique is capable to address main problems by breaking into sub-problems.

• • • •

Benefits: The approach addresses concerned problem in efficient and reliable manner. Demerits: Difficult to implement for large number of objectives. Preference: Capable to address RT complex problems with less time for optimum solution. MOPT Ref : CRU (C) [83].

4.2. Meta-Heuristics (MH) Methods (1)

Genetic algorithms (GA): A kind of adaptive heuristic search algorithm, based on the concept of natural selection and genetics.

• • • • (2)

Particle swarm optimization (PSO): In this method, a set of arbitrarily (randomly) activated solutions moves in the search process, aiming at best solution over some iterations.

• • • • (3)

Benefits: Simple, easy to understand, does not depend on the initial solution. Demerits: More computational time is needed and can converge at local optima due to intensification of parameters in search process. Preference: Numerous applications, since it does not require complex mathematical knowledge in the implementation of required solution. MOPT Ref : DGP (A) [21–23]; VPQ (B) [66]; CRU (C) [79,81,88],[89]]; NTR (D) [90,94].

Benefits: Easy to code, efficient computation time and better convergence than GA. Demerits: When the problem dimensions increase, the algorithm loses robustness. Preference: Considerable preference is given in the literature, aiming at large-scale DNs, with modifications in code and efficient tuning with the controller parameters. MOPT Ref : DGP (A) [43,44,48,52]; VPQ (B) [61,65],[67]]; CRU (C) [76,88]; NTR (D) [92,93].

Harmony search algorithm (HS): The metaheuristic is based on the concept of decision variable (musician) generates (plays) a value (note) for searching a best global optimum (harmony).

Energies 2017, 10, 208

• • • • (4)

•

• •

•

Benefits: The approach uses a lesser number of control parameters, better capability to deal with complex multidimensional optimization problems and exhibits good convergence properties. Demerits: The convergence rate is poor for constrained optimization problems. Preference: Modified versions of ABC algorithms are proposed for solving real world optimization problems. However, the approach has still not been considered in depth by the researchers. MOPT Ref : DGP (A) [47]; VPQ (B) [62].

Tabu search (TS): This meta-heuristic optimization technique utilizes adaptive memory to produce the most flexible search behavior. The algorithm operates in sequential way, starting from searching from an initial point and then selecting a new point in the search space as the next current point.

• • • • (8)

Benefits: Excellent convergence property with less computation time. Demerits: The improper tuning of system parameters may increase the complexity. Preference: Applied to nonlinear multidimensional functions, power flow problems (with continuous/discrete control variables) and DGP with good convergence speed. MOPT Ref : DGP (A) [46].

Artificial bee colony (ABC): An algorithm based on the concept of foraging behavior of honey bees and considers the fact that the food source represents a feasible explanation of the problem. great amount of a food source resembles the characteristics of a converged solution.

•

(7)

Benefits: Short simulation times and no parameters are required for the algorithm to work. Demerits: Nonlinearities and increased iterations for global optima of large scale real DN. Preference: Capable of finding global or near global optimum solution. MOPT Ref : DGP (A) [42].

Big bang big crunch (BB-BC): The nature inspired standard BB-BC algorithm consists of two steps. First, the formation of initial candidate solutions spread randomly across the search space. Second, a concurrence operator groups all solutions at only one solution, called “center of mass.”

• • •

(6)

Benefits: Easy to implement and offers near optimum solutions in some problems. Demerits: The standard HS suffers drawbacks (inaccuracy) in finding optimum solutions. Preference: The drawbacks have been addressed with improved variants, namely improved HS (IHS) and novel global HS (NGHS) respectively. MOPT Ref : DGP (A) [35].

Teacher learning algorithm (TLA): The method is motivated by the concept of teacher’s impact on the results achieved by students in a class and aims a student towards the best qualification.

• • • • (5)

22 of 44

Benefits: Fast convergence properties and easy tuning of the controller parameters. Demerits: It requires a suitable initial solution. Also, finding an optimal global solution in complex and multi-dimensional problems is not guaranteed. Preference: The approach is suitable for simple and comparatively less complex problems. MOPT Ref : VPQ (B) [53]; CRU (C) [77,84].

Immune algorithm (IA): The immune (system) algorithm belongs to the class of AI based meta-heuristics and is based on the concept of human body’s defense process against viruses. It starts with a randomly generated population with solutions reproduced at different rates. Later, suitable solutions are duplicated at high rates, followed by mutation at various rates. Finally, a selection operator is applied to produce suitable solutions.

Energies 2017, 10, 208

• • • • (9)

23 of 44

Benefits: The method does not depend on initial solution and is capable of obtaining a global optimum in complex problems with more accuracy than GA. Demerits: It requires more computation time than GA and PSO. The tuning of the system parameters is a tedious task for real world (MOP) problems. Preference: The approach utilized in various planning problems in literature. MOPT Ref : VPQ (B) [55].

Simulated annealing (SA): A stochastic search algorithm motivated by the similarity between the solid annealing procedure and optimization problems. The first metropolis process refers to jumping property that deals with a worse solution having the probability to be accepted as a new solution. Later, a cooling schedule slows down the probability of the worst solution in the search space.

•

•

• •

Benefits: Robustness is comparatively high. It does not depend on the initial solution. The algorithm is capable of finding a global optimum solution in combinational and complex (large scale) problems. Demerits: It requires more computation time than TS. The appropriate tuning of the system parameters is difficult for real world problems, particularly; it is not suitable for multiple planning candidates in large power systems. Preference: The approach utilized in various asset planning problems in literature. MOPT Ref : VPQ (B) [56].

(10) Honey bee mating (HBMO): A metaheuristic optimization algorithm inspired by the mating process of honey bees. The method aims at reproducing the mating process between a queen and drones, until new broods are generated, to find the most suitable solution. If the best brood among the brood population is better than the queen, it replaces the queen.

• • • •

Benefits: It shows suitable performance to solve various complex planning problems. Demerits: High dependence on adjusting initial parameters and premature convergence resulting in tracking local optimal solution. Preference: Modified versions of HBMO are utilized to address abovementioned demerits. MOPT Ref : VPQ (B) [60,63].

(11) Bacterial foraging (BFO): The algorithm is inspired by E. coli bacterias’ foraging properties, by an activity called “chemotaxis”. Simply, the algorithm deals with mimicking the chemotactic movement of (virtual) bacteria in the search space of the problem (food search).

• • • •

Benefits: Improved BFO variants are developed to improve optimization performance. Demerits: Poor convergence and decrease in search performance, with increase in search space and problem complexity respectively. Preference: A few models have developed for solving practical planning problems. MOPT Ref : VPQ (B) [68].

(12) Ant colony optimization (ACO): This AI method is based on the probabilistic searching behavior of real ants in pursuit of food to find the shortest paths from their nest to a food source (solution).

• • •

Benefits: It is easy to understand, simple to code and needs less computation time. Demerits: Poor convergence, difficult tuning of controller parameters and uncertainty in achieving global optima for simple/complex DN planning problems. Preference: The technique is utilized in several DN (expansion) planning problems. The deficiency in standard ACO is met with efficient variants, such as mix-max ACO and ant colony search algorithms (ACS), respectively. These variants have better convergence, however, the computation time increases.

Energies 2017, 10, 208

•

24 of 44

MOPT Ref : CRU (C) [73,75]; NTR (D) [99].

(13) Bat algorithm (BA): This population-based meta-heuristic algorithm mimics a group of bats, searching for the location exhibiting maximum availability of prey. The echo solutions of micro bats represent the most feasible solution.

• • • •

Benefits: High robustness and better convergence in comparison with GA, PSO, and HS. Demerits: Difficult tuning of controller parameters and may give inaccurate solutions for real time problems. Preference: The different variants of technique, notably shuffled-bat algorithm (ShBAT), are used by researchers to solve various planning problems. MOPT Ref : DGP (A) [51]; NTR (D) [98].

(14) Evolutionary algorithm (EA): This metaheuristic (AI) algorithm is inspired by the concept of biological evolution with operators (reproduction, mutation, recombination, and selection) and implemented by assigning fitnesses to each possible solution.

• • •

•

Benefits: Efficient and robust process to find near-optimal overall solution. Demerits: Premature convergence, low precision, and possibility of finding local optima. Preference: The technique has many applications in literature, to solve various planning optimization problems involving integer variables. The variants of this technique predominately include DE, TLA, SOA, GSA, SPEA, SPEA 2, NSGA, NSGA-II, and SFL. MOPT Ref : VPQ (B) [64].

(15) Differential evolution (DE): An artificial intelligence (AI)-based meta-heuristic based on a population technique (mutation, crossover, and selection) for finding global optimal solutions.

• • • •

Benefits: Its main features include the ability to solve complex optimization problems. Demerits: Prone to premature convergence and possibility of falling into local optima. Preference: Few applications in literature. MOPT Ref: DGP (A) [37]; VPQ (B) [64].

(16) Seeker optimization algorithm (SOA): This (meta-heuristic) EA is based on the notion of simulating an individual human’s (seeker) searching behavior among a population (seekers), for intelligent solution they search with their memory, experience (learning) and uncertainty reasoning.

• • • •

Benefits: High robustness and better convergence in comparison with GA and PSO. Demerits: Requires initial solution and high computation for complex problems. Preference: The technique has been employed by researchers to solve various planning problems. MOPT Ref : VPQ (B) [69,70]; CRU (C) [85,86].

(17) Gravitational search algorithm (GSA): This algorithm is based on the law of gravity, where agents’ (objects’) performance is measured by their masses. Thus, heavier masses show better solutions.

• • • •

Benefits: Promising results for complex and high dimensional search space problems regarding robustness and convergence, in comparison with GA and PSO. Demerits: Requires efficient initialization of G-parameter to control search accuracy. Preference: The technique is not treated in depth to solve various planning problems. MOPT Ref : NTR (D) [100].

(18) Strength Pareto EA (SPEA): This aims at implementation of elitism, by maintaining an external set of non-dominated (higher fitness) populations (set of solutions), found during the whole iteration loop. The fitness of solutions within a population depends on the best solution in the external set.

Energies 2017, 10, 208

• • • •

25 of 44

Benefits: Well developed to address complex nature and stochastic planning problems. Demerits: High computation requirements; more time required and need to reduce external number of solutions to a specific population size. Preference: The method has been considered in various planning problems of complex nature. MOPT Ref : CRU (C) [71].

(19) Strength Pareto EA 2 (SPEA 2): The algorithm (second generation) addresses the limitations in SPEA 2 (first generation).

• • • •

Benefits: Improved fitness assignment scheme, precise guidance of the search process and preservation of boundary solutions, better than SPEA. Demerits: Convergence performance reduces as the search space increases. Preference: The technique is used to find non-dominated solutions for complex and stochastic-based planning problems. MOPT Ref : VPQ (B) [57].

(20) Non-dominated GA (NSGA): The algorithm (first generation) aims at employing the notion of selection method, based on classes of dominance of all solutions. The algorithm, in each generation (iteration), indicates individuals (non-dominated solutions) within the population, to form non-dominated solution sets (Pareto front).

• • • •

Benefits: Well developed to address multi-objective based planning problems. Demerits: High computational complexity, lack of elitism, need to specify sharing parameters and slow non-dominated sorting procedure. Preference: To find non-dominated solutions in design, test and planning problems. MOPT Ref: DGP (A) [27]; NTR (D) [96].

(21) Non-dominated GA II (NSGA-II): The algorithm (second generation) addresses the limitations in NSGA (first generation), to allow efficient application to constrained planning problems.

• • • •

Benefits: Efficient constraint management, elitism, parameter-less approach and fast non-dominated sorting procedure, batter than NSGA. Demerits: Performance of convergence reduces with as the search space increases. Preference: The technique is employed by researchers to solve complex nature planning problems, aiming at accurate, diverse and well-spread Pareto fronts. MOPT Ref : DGP (A) [24,38,41]; VPQ (B) [58]; CRU (C) [87]; NTR (D) [95].

(22) Shuffled frog leaping (SFL): This population-based meta-heuristic method is inspired by the concept of mimetic evolution of a group, aiming at the quality of the meme (of an individual) and improves the performance (individual frog) towards a goal (highest food availability).

• • • •

Benefits: Efficient and better computing performance with global search ability. Demerits: More computation and premature convergence due to the DN size complexity. Preference: Several applications in literature and needs to be investigated as a potential candidate for combinational problems, aiming at future distribution concepts. MOPT Ref : DGP (A) [36]; CRU (C) [78,82].

Energies 2017, 10, 208

26 of 44

4.3. Hybrid Methods The hybrid techniques are developed to address the limitations in the previous techniques. The efficient development of hybrid methods helps to find global optimization solutions. Moreover, these techniques can manage complex (stochastic- and uncertainty-based) optimization problems. However, such methods are comparatively hard to code and have relatively limited example is literature. The prominent MOP-based hybrid methods are the following: (1)

Hybrid GA: These methods aim at solving combinational type problems, accommodating inner optimizations and implementing decision making processes to sort out feasible solutions.

• • • •

(2)

Hybrid PSO: These methods aim at modifications that address the limitations in simple PSO.

• • • • (3)

Benefits: Better convergence, more precision, and improved performance since high fitness/ value chromosomes are used to produce the next generation. Demerits: Requires more computation time (global optima cannot be reached in a limited time). The linear change of decision variables can result in suboptimal solutions. Preference: The techniques have employed for optimal solution with maximum objectives. MOPT Ref : DGP (A): The ε-constrained method [21,22]; MCS [23]; goal programming [26,32]; fuzzy, WSM [29]; MCS, AHP [30]; PSO [31] and multi-criteria stochastic programming model (MSPM) [34]; VPQ (B): GA based fuzzy multi-objective method [54,60].

Benefits: Better convergence, more precision, and improved optimization than PSO. Demerits: Requires efficient tuning of several control parameters and the method can suffer from partial optimism. Preference: These methods results in better performance than standard PSO. MOPT Ref: DGP (A): SFL [36]; VPQ (B): SA [56] and mixed integer programming (MIP) [67]; CRU (C): SPEA 2 [80], SFL [82] and nonlinear programming (NLP) [87]; NTR (D): MCDA [94]

Hybrid EA: These techniques are employed in the literature to address optimal solutions with decision making in multi-objective planning problems.

• • • •

Benefits: Highly efficient, robust and quick convergence to find optimal solutions. More suitable for multi-objective planning and decision making problems. Demerits: The inertia weights are randomly adjusted in the Pareto front. The weighted methods require more knowledge for decision making. Preference: These methods show promising aspects to address active DP, including storage, REGs, EV, DSM, and DR, stochastic generation, uncertain demands and controllable loads. MOPT Ref : DGP (A): Two-stage heuristic (ES + clustering techniques) [33] NSGA-II + MINLP [38]; CRU (C): NSGA + SPEA + FDM [71] and SPEA 2 + OPF [74]; NTR (D): NSGA II + MCS-OPF based approach [97].

4.4. Decision Making and Other Methods The decision-making methods are crucial for finding trade-off solution among a set of solutions in MOP formulations. Notable features are as follows:

• • • •

Benefits: The method enables decision maker to sort out best possible solution. Demerits: The weights (weight methods) require extensive knowledge. Also, the results can be specifically weight dependent (in case of IMO). Preference: FDM is among one of the widely used approaches. MOPT Ref : DGP (A): WSM [29,49]; AHP [45]; FDM [36,39,49,52]; VPQ (B): Max-min Approach [69,70]; CRU (C): FDM in [71,73,78,86]; NTR (D): FDM in [100] and fuzzy multi-criteria decision making (MCDM) algorithm implemented in “proof of concept” in [91].

Energies 2017, 10, 208

27 of 44

4.5. Impact of Load Flow Method Load flow (LF) or power flow is an important and vital tool for analysis, planning and operation of DNW in steady-state conditions. LF indicates the system variables, which exceed the respective constraints limits. The necessary action must be followed to bring back the system in stable operation zone. In addition, LF and AC based OPF (ACOPF) constitutes an integral part of MOPT problem, as inner optimization. The associated LF (and ACOPF) based solution techniques can be defined by formulation type, solver and initialization. On the basis of literature review, the load flows can be safely classified on the basis of formulation (F), solver (S), and initialization (I), as indicated by (F,S,I). Moreover, interaction among (F,S,I) and impacts of LF (and ACOPF) methods as inner optimization [101] in MOP, has also presented in this sub-section as follows. 4.5.1. Load Flow Formulation (F) The branch LF models (BLFM) constitute the largest portion of LF formulations. Being inner optimization, LF methods formulates constraints in the main optimization (MOP) problem. The LF methods with usually flat initialization values, solve BLFM formulations, which indicate equality constraints or power/load balance. The prominent technical inequality constraints include voltage limits (between maximum and minimum allowable limits) and branch thermal limits (must be less than the maximum admissible apparent power of the line). LF methods are responsible for operating and retaining the test DNW system within technical inequality constraints. The ACOPF tool (to solve complex and nonlinear power flow problems) formulates any set of constraints through three types of major formulations, namely polar power–voltage power flow (PSV), rectangular power-voltage power flow (RSV) and rectangular current injection (RIV), respectively [101]. These formulations (which indicate equality constraints) and associated technical constraints are usually solved with MCS, modified LF methods and commercial solvers. Ample details regarding inequality constraints have provided in Section 2.7.4. 4.5.2. Load Flow Solution Methods (S) The LF solution methods or simple solvers depend on the application of planning problem and type of loads addressed. For load profiles, refer to Tables 1–4, respectively. The solvers can be broadly classified as follows: (1)

Traditional LF solvers: The traditional LF solution methods, aiming at solving equality and inequality constraints, predominately include:

• • • • (2)

Newton Raphson (NR) as in [48,52,68,79,94]. Gauss-Seidel (GS) [82]. Fast decoupling (FDLF) [31]. Backward/forward sweep (BSFS) LF as in [24,59,63–65,69–71,73,74,77,83–86,91].

Modified LF solvers: The modified LF solution methods, aiming at solving equality and inequality constraints, mainly include:

• • • • •

3-Phase (Φ)-4-wire BSFS LF as in [23,25,27–29,46,47,72,96]. Simple (iterative and algebraic) LF as in [50,54,57,58,93,99]. Numerical based LF as goal attainment method (GoA) in [32], SQP in [49] and MCS in [23], [30,38,39,41,81,97]. Meta-heuristics (MH) based LF as in [35,55,67,75,76,95]. Probabilistic LF (PLF) as in simple PLF in [21,22] and 2 m point estimation method (2 m-PEM) as in [60,98].

Energies 2017, 10, 208

•

(3)

28 of 44

Other LF methods and frameworks as in [33,42–44,51,53,66,80,90,92]; in addition to LF frameworks as multi-criteria stochastic planning model (MCSPM) with central limit theorem (CLT) [34] and IBVT in [56].

Software-based commercial solver packages: The commercial solvers, aiming at solving nonlinear equality and inequality constraints, notably include:

• • • •

• •

Load model synthesis (LOADSYN) [26]. DIgSILENT® for balanced/unbalanced LF (B/U LF) [36] and static/dynamic modelling [100]; Voltage stability and optimization (VS&OP) Tool [61,62]. General algebraic modeling system (GAMS) for modeling mathematical programming models as in [37]. CONOPT solver of GAMS solves the nonlinear optimization problem (NLP) as in [39]. The DICOPT solver of GAMS solves MINLP problem as in [40]. GUROBI commercial optimization solver; solves linear (LP), quadratic (QP) and quadratic constrained programming (QCP) respectively, as in [45]. OpenDSS (open electrical power distribution system simulator) software [89].

4.5.3. Initialization (I) The convergence of solutions depends on an effective initialization method. The initialization methods can be broadly classified on the basis of three prominent types of initializations (or starts) reported in the literature [21–101]. Further details are given later in this section.

•

• •

Flat/Normal/Base/Random Start: These methods constitute the largest part of initialization methods in the reported literature. Either the load flow initiates flat or starts in BLFM, where the starting voltage is equal to the root (standard/substation/slack bus) voltage. Also, candidate solutions are generated randomly and loads are normally distributed (in the case of PLF). However, these methods have high computation costs. Btheta (Bθ) Start: This method is a natural extension of the current injection method (CIM) normally considered for optimal dispatch. This method, like flat starts, can require high computation cost. Hot Start: The initial optimal solution is determined for effective and efficient generation of candidate solutions. The time computation cost can be significantly reduced. Also, more realistic solutions including uncertainty in problems, are addressed in much reduced time.

At each start point, power flow constraints are solved to initialize BLFM, PSV, RSV and RIV formulations. The initialization methods are important in a way that they remain feasible for both equality (formulations) and inequality constraints. 4.5.4. LF Impact with Interaction in MOP Problems The general LF technique is the arrangement of test problem (DNW size), formulation (F), solver (S) and initialization (I). Moreover, interaction among (F,S,I) and considered test problem (T) as inner optimization (T,F,S,I) [101] needs to be considered in the context of MOP problems. MOP problems mostly consist of inner and outer optimization methods. In MOPT planning problems, efficient LF solutions (with interaction of T,F,S,I) support the main algorithm (outer algorithm) and are followed by a decision-making method. A comprehensive comparison of the impact of LF, on the basis of interactions in MOP problems; is presented in Table 9. The interaction his arranged according to problem types (DGP, VPQ, CRU and NTR), DNW size (T), formulation (F), solver (S) as inner optimization, initialization I (for both inner and outer optimization), main algorithm and DM.

Energies 2017, 10, 208

29 of 44

Table 9. Power/Load Flow method impact by interaction in overall MOPT planning problem. Problem Type

Test Problem (DNW Size)

Formulation (Form)

Solver/Inner Algorithm Method/(LF Method)

Initialization Inner/Main

Main (Outer) Algorithm

Decision Making

[21]

Real Italian

BLFM

Simple PLF (Heuristic)

Normalized

GA + ε constrained

-

[22]

18 Bus

BLFM

Simple PLF (Heuristic)

Normalized

GA + ε constrained

-

[23]

34 Bus

BLFM

3-Phase-4 wire LF/

Flat/Random

GA + MCS

-

[24]

9 Bus

BLFM

BSFS LF with FLd

Flat/Random

NSGA-II

Max-min

[25]

34 Bus

BLFM

3-Phase-4 wire LF (BSFS)

Flat Start

ES

Weight

[26]

34 Bus

BLFM

LOADSYN / FGP

Flat/Random

GA

Weight

[27]

34 Bus

BLFM


Flat/Random

NSGA

-

[28]

16; 37 Bus

BLFM


Flat/Random

ES, GA

Weight

[29]

33; 69 Bus

BLFM


Flat/Random

Fuzzy GA

WSM

MCS-GA

AHP

[30]

37 Bus

PSV

MCS/CCP

Hot(MCS)/ R

[31]

33; 69 Bus

BLFM

Fast decoupled LF (FDLF)

Hot/Random

GA + PSO

Penalty

[32]

25; 33 Bus

BLFM

Goal attainment method

Base/R

GA + Goal P

Weight

[33]

24 Bus

BLFM

Classical LF and LSF

Hot(LSF)/R

2-Stage (MH +ES)

-

[34]

37 Bus

MCPM

MCSPM with CLT

Bθ/Hot

GA + MCOM

-

[35]

33; 69 Bus

BLFM

IMOHS Framework

Random (R)

NGHS-II

Max-min

[36]

33 Bus

BLFM

B/U LF in DIgSILENT®

Base/Hot

PSO + SFL

FDM

[37]

33; 69 Bus

BLFM

GAMS Model as MINLP

Bθ/Random

PFDE Algorithm

-

[38]

38 Bus

PSV

MCS

Random (R)

NSGA-II (MINLP)

-

[39]

9 Bus

RSV/GAMS

CONOPT/MCS

Normalized

AεCM

FDM

[40]

69 Bus

PSV/GAMS

DICOPT

Bθ /Normal

AεCM

-

[41]

13 Bus

PSV

MCS

Bθ/Normal

Fast NSGA-II

-

[42]

69 Bus

BLFM

BIBC and BCBV based LF

Base/Hot

QOTLBO

-

[43]

33 Bus

PSV

Modified branch based LF

Hot (weight)

MOPSO

FDM

[44]

33 Bus

PSV

B/F updated DistFlow LF

Bθ/Hot

IMOPSO-PS

-

[45]

34 Bus

BLFM

GUROBI

Base/Linear

MISOCP

AHP

[46]

69; 123 Bus

BLFM

3-Phase(Φ)-BSFS LF

Flat/Random

S-BB-BC

Weight

[47]

38; 69 Bus

BLFM

3-Φ B/F propagation LF

Flat/Random

CABC

Weight

[48]

28 Bus

BLFM

Newton Raphson (NR)

Flat/Hot

MOPSO

FDM

[49]

15 Bus

BLFM

SQP (Taylor expansion)

Bθ

SQP + WSM

FDM

[50]

33; 292; 588 Bus

PSV

L-Index based LF

Bθ/Random

INSGA-II

FDM

[51]

33 Bus

RIV

3-Φ Current injection LF

Flat/Hot

MOShBAT

-

[52]

28 Bus

BLFM

Newton Raphson (NR)

Flat/Hot

MOPSO

FDM

[53]

94 Bus

BLFM

Forward sweeping LF

Flat/Hot

TS

-

[54]

69 Bus

BLFM

Simple algebraic LF

Flat/Random

Fuzzy-GA

Weight

[55]

69 Bus

BLFM

SA based LF

Base/R

IA

-

[56]

12 Bus

BLFM

IBVT

Random

SA + PSO

-

[57]

69 Bus

BLFM

Algebraic iterative LF

Flat/Hot

SPEA 2

-

[58]

94 Bus

BLFM

Iterative branch based LF

Flat/Random

ENSGA-II

-

[59]

69; 136 Bus

BLFM

BSFS LF

Flat/Linear

MILP Model

-

[60]

9; 34 Bus

BLFM

2 m-PEM

Bθ/Hot

SAMHBMO

-

[61]

34; 118 Bus

BLFM

VS&OP

Flat/Hot

APSO

Weight

[62]

34; 94 Bus

BLFM

VS&OP

Flat/Hot

ABC

Weight

[63]

9; 34; 69 Bus

BLFM

BSFS LF

Bθ/Random

AMHBMO

-

[64]

115 Bus

BLFM

BSFS LF

Flat/Random

IDEA

-

[65]

33; 94 Bus

BLFM

BSFS LF (Harmonics)

Flat/Random

Fuzzy MOPSO

FDM

[66]

51; 69 Bus

BLFM

Compensated line LF

Flat/Hot

Fuzzy-GA

Weight

[67]

60 Bus

BLFM

Improved BSFS (PSO)

Flat/Hot

PSO

WSM

[68]

34 Bus

BLFM

NR

Bθ/Uniform

ABFO

Weight

[69]

54 Bus

PSV

BSFS LF

Flat/Uniform

MOSOA

Max-min

[70]

54 Bus

PSV

BSFS LF

Flat/Hot

MOSOA

Max-min

Energies 2017, 10, 208

30 of 44

Table 9. Cont. Problem Type

Test Problem (DNW Size)

Formulation (Form)

Solver/Inner Algorithm Method/(LF Method)

Initialization Inner/Main

Main (Outer) Algorithm

[71]

41 Bus

[72]

Urban DNW

[73] [74]

BLFM

BSFS LF

Flat/Uniform

NSGA + SPEA

FCM

BLFM

3-Φ-4 wire LF (BSFS)

Flat/Random

GA + MINLP

Weight

Rural; Urban

BLFM

Compensated BSFS

Flat/Random

ACO

FDM

355 Bus

BLFM

BSFS LF

Flat/Random

SPEA 2

-

[75]

18; 51 Bus

TGM

Simple GA based LF

Flat/Random

MACO

-

[76]

21; 100 Bus

BLFM

Penalty factor (GA)

Flat/Random

MOPSO

-

[77]

180 Bus

BLFM

Compensated BSFS

Flat/Random

MORTS

-

[79]

33; 177 Bus

BLFM

NR (Branch exchange)

Flat/Uniform

MOGA

-

[80]

21; 54; 100 Bus

BLFM

Conventional LF

Flat/Random

SPEA2-B/MOPSO

-

[81]

Urban DNW

PSV

MCS

Random (R)

IGDSEP (AGA; IAGA)

Weight

[82]

Actual DNW

BLFM

Gauss Approach

Random (R)

PSO + SFL

-

[83]

21; 54; 100 Bus

BLFM

BSFS LF

Bθ/Hot

DynP

Weight

[84]

54 Bus

BLFM

Compensated BSFS

Bθ/Hot

MOTS

-

[85]

54; 100 Bus

BLFM

BSFS LF

Flat/Hot

MOSOA

Weight

[86]

54 Bus

BLFM

BSFS LF

Flat/Hot

MOSOA

FDM

[89]

13; 34 Bus

BLFM

OpenDSS

Flat/Uniform

MOGA

Weight

[90]

17; 33; 172 Bus

BLFM

Power summation LF

Flat/Uniform

Micro-GA

-

[91]

Actual DNW

BLFM

BSFS LF

Flat/Uniform

MCDM

-

[92]

16 Bus

BLFM

DistFlow B/F update

Flat/Uniform

BPSO

Weight

[93]

33; 123 Bus

BLFM

Simple iterative LF

Flat/Random

BPSO

-

[94]

Sample DNW

BLFM

NR

Flat/Uniform

GA

Weight

[95]

33; 67 Bus

BLFM

Branch exchange MH

Flat/Hot

NSGA-II

-

[96]

38; 119 Bus

BLFM

3-Phase-4 wire (BSFS)

Flat/Uniform

NSGA

-

[97]

69 Bus

BLFM

Non-dominated MCS

Random

NSGA-II

-

[98]

32 Bus

BLFM

2m-PEM

Random

SAMBA

-

[99]

33; 67 Bus

BLFM

Simple iterative LF

Flat/Hot

Fuzzy-ACO

Weight

[100]

33 Bus

BLFM

DIgSILENT®

Random

ESGA

FDM

(DPL)

Decision Making

5. Contribution of the Review Work 5.1. Reviewed Work Contributions The contributions of reviewed work are shown chronologically in Tables 5–8; addressing each PT category under the MOP framework as MODGP, MOVPQ, MOCRU, and MONTR, respectively. 5.2. Assessments of Multi-objective Planning Methods Among numerical methods, OPF is commonly exploited as an inner optimization algorithm to facilitate high precision, stochastic, efficient time computation and uncertain operation situations. However, the problem formulation is very rigid, and few variations can be incorporated. Dynamic programming (DynP) and exhaustive search (ES) although they promise to find a global optimum solution, however they have not advocated for large distribution systems. Likewise, a major drawback of MCS (being inner optimization) is the high computation time. In comparison, MINLP, SQP, and dynamic programming are the most efficient numerical methods. Metaheuristic (MH) methods like evolutionary algorithms have been considered as a natural way of addressing complex MOP problems. However, the high computational requirements and possibility of premature convergence towards local optimal solutions need to be addressed with more efficient methods. The simple MH methods require more computation time. Also, more number of functions is required to achieve nearly high quality results. Also, constrained and unconstrained multi-objective problems are difficult to optimize. Similarly, GA and PSO provide near optimal solutions for large distribution systems.

En

En En

En

o

o

o

DM LENT DPL RBund Bu BL5.MContribution D BuBu LENT DPL BuBL BL R nd RBL DM EACO GA Bu MReview BL M mp mp Hnd Mm uE mp ACO LW DM u AMBA ACO W BL mp L LDPLE GA H uGA Wh W h hDM mm PEM mLLENT PEM Bu mR nd m AMBA muACO PEM AMBA R ndHm BuBu mDPEM RH5. nd m AMBA 5. 5.The Contribution ofunder the of Review the Work Work Contribution of the Review Work ofBu the Review Work The contributions The contributions of reviewed ofM reviewed work work shown are shown chronologically chronologically in Tables inMONTR, Tables 5–8; addressing 5–8; addressing each each contributions The contributions ofContributions reviewed of work reviewed are shown chronologically are shown chronologically in Tables 5–8; in Tables addressing 5–8; addressing each each PTContribution category PT category under the MOP the MOP framework framework asare MODGP, as MODGP, MOVPQ, MOVPQ, MOCRU, MOCRU, and and MONTR, respectively. respectively. 5.1. Reviewed Work Contributions PT category under the MOP framework aswork MODGP, MOVPQ, MOCRU, and MONTR, respectively. 5.1. Reviewed Work Contributions 5.1. Reviewed Work En En En ouDE En oh DMhoR ndHm D mp LENT DPL RBu nd nd LENT E GA DPL E GA DLENT LENT DPL RH nd E Bu MReview BL M mp LPEM L BL HmM Work mp ACO uACO ACO LW DM W ACO WD BL M mp L DPL HReview uGA WhDM BuBu mDPEM m Bu mR AMBA mGA PEM AMBA AMBA BuBu m5. PEM m AMBA 5. Contribution 5. Contribution ofcategory the of Review the Work Work Contribution of the 5. Contribution ofBu the Review Work The contributions The contributions ofBL reviewed of reviewed work work are shown are The shown chronologically contributions chronologically of in reviewed Tables in Tables 5–8; work addressing 5–8; are addressing shown each chronologically each in Tables u5–8; addressing The contributions of reviewed work are shown chronologically in Tables 5–8; addressing each PT category PT category under under the MOP the framework MOP framework as MODGP, as MODGP, MOVPQ, MOVPQ, MOCRU, MOCRU, and MONTR, and MONTR, respectively. respectively. 5.1. Reviewed Work Contributions 5.1. Reviewed Work Contributions PT category PT under the MOP under framework the MOP framework as MODGP, as MOVPQ, MODGP, MOCRU, MOVPQ, and MOCRU, MONTR, and respectively. MONTR, respectively. 5.1. Reviewed Work Contributions 5.1. Reviewed Work Contributions En98 En omLPEM Bu LENT D LENT DPL RBund mR nd DEmp LENT E GA DPL GA DBLFM DPL RH nd E Bu Bu BL M BL MDkew mp mp L BL HmM ACO uACO ACO LW DM W uE ACO WD BL M mp L DPL Hm uGA WhDM h DMhoR ndHm Bu Rev BL MCon Bu m PEM RLENT m Bu AMBA mGA PEM AMBA AMBA BuBu mnd PEM AMBA 98 32 Random Bus BLFM 2m PEM 32 Bus Random 2m PEM SAMBA Random SAMBA 2m En PEM SAMBA 5 Con on ounder he Rev ew Wo 5 Con on o he ew Wo 5bu k bu on oMOP he Rev Wo k The contributions of reviewed work are shown chronologically in Tables addressing each The bu contributions of reviewed work The contributions are shown chronologically offramework reviewed work in Tables areMOVPQ, shown 5–8; addressing chronologically each inu5–8; Tables 5–8; addressing each PT category PT category under the MOP the MOP framework as MODGP, PT as category MODGP, under MOVPQ, MOCRU, the MOP MOCRU, and framework MONTR, and MONTR, as respectively. MODGP, respectively. MOVPQ, MOCRU, and MONTR, respectiv 5.1. Reviewed 5.1. Reviewed Work Contributions Work Contributions 5.1. Reviewed Work Contributions PT category under the framework as MODGP, MOVPQ, MOCRU, and MONTR, respectively. 5.1. Reviewed Work Contributions En5.2. EnD o WD 5.2. Assessments 5.2. Assessments of Multi-objective ofm Multi-objective Planning Planning Methods Methods En om Assessments of Multi-objective Planning Methods Bu Bu BLhe MRev Dk Wo LENT DPL DPL Bu RBu nd m R nd Dk Emp LENT E AMBA GA DPL R PEM ndHm Bu DkLENT LENT R he nd m E GA BL M bu mp L ew BL M mp H L u ACO BL Hmm M WWo hBu uACO ACO L DM W uRE nd ACO Bu BL M mp L HR uGA W DM h DMhom BuBu 5 Bu PEM BL M R mDPL PEM AMBA nd BL M m mnd PEM AMBA Con 5 Con bu on bu oas on he ohe Rev ew Wo ew k 5 Con bu on oaddressing Rev ew 5 Con on o Rev Wo The contributions of reviewed work are shown The chronologically contributions of in reviewed Tables 5–8; work addressing are shown each chronologically inFuzzy TablesACO 5–8;GA addressing The contributions of reviewed The contributions work are shown of reviewed chronologically work are in shown Tables chronologically 5–8; in each Tables 5–8; addressing each 33 BL 67 Bus BLFM S mp 99 e e a ve 33 LF 67 Bus BLFM F a Ho S mp e e a Fuzzy ve LF ACO We F a gh Ho We PT category under the MOP framework as MODGP, MOVPQ, MOCRU, and MONTR, respectively. 5.1. Reviewed 5.1. Reviewed Work Contributions Work Contributions 5.1. Reviewed Work Contributions e e a ve LF99 Bu FPT aM Ho Fuzzy ACO We gh category under the MOP framework PT category MODGP, under the MOVPQ, MOP framework MOCRU, as and MODGP, MONTR, MOVPQ, respectively. MOCRU, and MONTR, respectively. 5.1. Reviewed Work Contributions En En M EnD oh DBuu ELENT o 5.2. Assessments 5.2. of Multi-objective of Multi-objective Planning Planning Methods 5.2. Assessments 5.2. Assessments of Multi-objective of Multi-objective Planning Methods Planning Bu LENT DPL BL MM R nd LENT m Buu ER GA R nd BLH mm M WWo DM GA DM E GA Bu BL M Dk LENT DPL nd m BL mp Bu LAMBA BL mp H L Methods ACO uE BLDPL ACO M WDMhW mp ohR nd m L H Bu mp L DPL HR ACO Bu mMethods PEM nd mGA PEM AMBA AMBA m PEM BuBu 5 RAssessments nd m 5 Con on oDkw on he ohe Rev he ew Rev Wo ew k Wo k 5PT Con bu on o he Rev ew k 5–8; 5PT Con bu on o ew Wo The contributions contributions of reviewed of reviewed work are work shown are shown chronologically chronologically The in contributions in Tables 5–8; addressing 5–8; of addressing reviewed each work each are shown and chronologically in TabD The contributions ofCon reviewed work are shown chronologically in Tables addressing each PT category under the MOP framework as MODGP, category MOVPQ, under MOCRU, the MOP and MONTR, as respectively. MODGP, MOVPQ, MOCRU, MONTR, respectiv 5 1Con RESGA vBL w dnd Wo Con bu on PT under the MOP framework under as MODGP, the MOP MOVPQ, framework MOCRU, as MODGP, and MONTR, MOVPQ, respectively. MOCRU, and MONTR, respectively. 5 1 Rcategory v wmpd Wo k Con bu on 5bu 1The Rnumerical vbu dDPL Wo kRev bu on Among Among numerical methods, methods, OPF OPF is commonly is commonly exploited exploited as an as inner an inner optimization optimization algorithm to to ® ®R Among numerical methods, OPF is commonly exploited asTables an inner optimization algorithm En oBu 5.2. Assessments 5.2. Assessments of Multi-objective of Multi-objective Planning Planning Methods 5.2. Methods Assessments of Multi-objective Planning Methods En En oframework o 5.2. Assessments of Multi-objective Planning Methods M LENT DPL R D m LENT DPL nd DM E BL GA M D DM LENT m 33BLRandom Bus BLFM D5category gS 100 LENT 33 Bus BLFM Random D LENT DPL ESGA FDM Random ESGA FD BL M Dk LENT DPL R GA nd m E GA DM LENT® DPL100 BuBu FDM Bu BL M mp L gS Bu BL H M mp u ACO Lalgorithm hm toH DPL uR ndACO W M En L 5 H u ACO Wnd h Bu m PEM Bu BL M m m PEM AMBA RWnd AMBA m PEM Bu RThe m AMBA Con Con bu on bu oDkBu on he ohe Rev he ew Rev Wo ew k Wo k 5 Con bu on okEas he Rev ew Wo k 5 Con bu on o Rev ew Wo The contributions contributions of reviewed of reviewed work are work shown are The shown chronologically contributions chronologically of in reviewed Tables in Tables 5–8; work addressing 5–8; are addressing shown each chronologically each in Tables 5–8;MOCRU, addressing The contributions of reviewed work are shown chronologically in Tables 5–8; addressing each PT category PT category under under the MOP the framework MOP framework as MODGP, as MODGP, MOVPQ, MOVPQ, MOCRU, PT category MOCRU, and under MONTR, and the MONTR, MOP respectively. framework respectively. as MODGP, MOVPQ, and 5 1 R 5 v 1 w R d v Wo w d Con Wo k bu Con on bu on 5 1 R v w d Wo Con bu on PT category under the MOP framework as MODGP, MOVPQ, MOCRU, and MONTR, respectively. 5 1 R v w d Wo k Con bu on Among Among numerical numerical methods, methods, OPF is OPF commonly is commonly exploited exploited an inner as an optimization inner optimization algorithm algorithm to to Among numerical Among methods, numerical OPF methods, is commonly OPF is exploited commonly as exploited an inner as optimization an inner optimization algorithm to algorithm to En En o o facilitate facilitate high high precision, precision, stochastic, stochastic, efficient efficient time time computation computation and uncertain and uncertain operation operation situations. situations. 5.2. Assessments of Multi-objective Planning Methods En o facilitate high precision, stochastic, efficient time computation and uncertain operation situations. 5.2. Assessments of Multi-objective 5.2. Planning Assessments Methods of Multi-objective Planning Methods LENT Buo AMBA BLHm M Wo D LENT DPL DM E GA BuBu DPL R nd Eew GA Bu BL M Wo mp LBuon BL R M nd uLE ACO HhR u AMBA ACO W h D BL M 5 Con mp LENT Lo he Hm u ACO Wnd h Bu BLCon M m Bu PEM R m m PEMDPL Bu m PEM AMBA RWnd m nd m bu Rev 5 ew Wo bu on ohe he Rev ew k 5 Con bu he Rev ew kTab 5 Con bu on o Rev Wo k The con on o ev ewed wo kas aD eyDM hown ono og ca y n Tab eoptimization 5–8 add eGA ng each TheDcon bu on on o ev ewed wo The kkk con aunder eCon hown bu on ch o ono ev og ewed ca wo n Tab kMODGP, av exploited eMOVPQ, ech hown 5–8 add ch ono eas ng og each ca ymp nand ecommonly 5–8 add eMOVPQ, ng each PT category PT category under the MOP the framework MOP framework MODGP, PT as category under MOVPQ, MOCRU, the MOP MOCRU, and framework MONTR, MONTR, as respectively. MODGP, respectively. MOCRU, and MONTR, respectiv 5 1 R 5 v 1 w R d v Wo w d Con Wo k bu Con on bu on 5 1 R w d Wo k Con bu on PT category under the MOP framework as MODGP, MOVPQ, MOCRU, and MONTR, respectively. 5 1 R v w d Wo k bu on Among Among numerical numerical methods, methods, OPF OPF is commonly is commonly Among exploited numerical as an as inner methods, an inner optimization OPF optimization is algorithm algorithm exploited to to as an inner optimization algorith Among numerical methods, OPF is commonly exploited an inner algorithm to En En o facilitate facilitate high precision, high precision, stochastic, stochastic, efficient efficient time computation time computation and uncertain and uncertain operation operation situations. situations. 5.2. Assessments of Multi-objective Planning Methods 5.2. Assessments of Multi-objective Planning Methods o facilitate high facilitate precision, high stochastic, precision, efficient stochastic, time efficient computation time computation and uncertain and operation uncertain situations. operation situations. 5.2. of Multi-objective 5.2. Assessments Planning of Methods Multi-objective Planning Methods Bu BL M D LENT DPL Bu BL R M nd m D LENT DPL E GA R nd DM m E GA DM Bu BL En M Assessments D LENT DPL R nd m E GA DM BL M mp Bu LRev mp Hm L BuAMBA u AMBA ACO W mp hbe uRev L BLACO W Hh u BuBu However, mhe PEM R mvery nd Buoincorporated. M Wo m PEM R nd m W h Bu on BL M mk Bu PEM BL Mformulation R nd m PEM R m AMBA However, the problem the problem formulation formulation is very is rigid, rigid, and few and variations few variations can be can incorporated. Dynamic Dynamic the problem very rigid, and few variations can be incorporated. Dynamic 5 Con unde bu he 5The ew Con Wo bu he ew Wo 5end Con bu on he k 5ca Con on o Rev ew Wo k The con bu con on bu o on oev ewed ev wo ewed kacommonly awo e5inner hown eoptimization hown ch con ono ch bu og ono on ca og y oinner n ca ev Tab y ewed n einner Tab 5–8 eoptimization add k5–8 aadd add hown ngve eng each ch ng ono each og n Tab e ACO 5–8 add e ng con bu on oev ewed wo k ais ek hown ono og ca yH n Tab ewo 5–8 eew each PT ego unde he MOP amewo k MODGP MOVPQ MOCRU and MONTR eeBL pec yto 5 1 En RM 5 vBL 1However, w RyThe dbu vnumerical Wo w dprecision, Con Wo ko bu Con on bu on 1km R vMODGP w dch Wo knd Con bu on PT caAmong heoMOP amewo PT ca ego k aDkcommonly y MODGP unde he MOVPQ MOP amewo MOCRU aThe and MONTR MOVPQ pec ve MOCRU y and MONTR ePlanning pec ve ytoca 5facilitate 1Rev R vhigh w dBu Wo kon Con bu on Among methods, OPF is exploited an optimization algorithm numerical methods, OPF Among is numerical exploited methods, as OPF isaMethods commonly exploited as an to algorithm 208 31 of y44uncertain Enego y10, En En ohLENT omDynamic facilitate facilitate high precision, precision, stochastic, stochastic, efficient efficient facilitate time time computation high computation precision, and uncertain and stochastic, uncertain operation efficient operation situations. time situations. computation and operation DM situat 5.2. 5.2. Assessments of Multi-objective of Planning Planning 5.2. Assessments of Multi-objective Methods oh high stochastic, efficient computation and uncertain operation situations. 5.2. Assessments of Multi-objective Methods LENT Bu DPL BL MPlanning RMethods D nd LENT DPL Bu Easalgorithm GA BL M nd m Dcan DM DPL E GA Rmp nd DM E GA M mp Lformulation u ACO W Bu Bu BuBuWork BLHowever, mp Bu LMulti-objective BL M mp H u ACO HR W hm ACO W BL M man PEM Rthey m AMBA AMBA Bu BLew M mThe PEM Rfew nd AMBA However, However, the problem the problem formulation is very is rigid, very and rigid, few and variations few variations can be incorporated. be Dynamic Dynamic the However, problem the problem formulation is very rigid, and very few rigid, and can incorporated. can be incorporated. Dynamic 5The Con bu he Rev Wo 5k Con bu on okvariations he Rev ew Wo k 5 Contribution of2017, 5Assessments Contribution of the Review Work 5 Con bu on o heEnergies Rev ew Wo programming (DynP) (DynP) and exhaustive and exhaustive search search (ES) although although they promise promise to find au5–8 find global aMODGP optimum con bu con on bu o on ev oev ewed ev wo ewed kacommonly aPEM wo e5PT hown atime eLis hown ch con ono ch bu og ono on ca og y oan n ca ev Tab y ewed nebe Tab 5–8 wo eincorporated. add ka5–8 aadd eglobal add ng eng each ch ng each ogas caanLyinner n Tab eMONTR 5–8 Haddalgorith (DynP) and exhaustive search (ES) although they promise to find aoptimum global optimum con bu on oformulation ewed wo k aHinner em hown ono og ca y nvariations Tab eto eM each PT ca PT ego ca yThe ego unde y unde he MOP he amewo amewo k MODGP ca MODGP ego MOVPQ yexploited unde MOVPQ MOCRU he MOP MOCRU and amewo MONTR and k MONTR eoperation pec eve pec MOVPQ yto ve y MOCRU and ee pecngv 5the 1 k R Review v numerical w d Wo kprogramming Con 5 bu 1programming R on vefficient w dhigh Wo ko Con bu on 1kam R v(ES) w dch Wo Con bu on PT ca ego ydBu unde he MOP amewo kis and MODGP MOVPQ MOCRU and MONTR ehown pec ve yono 5 1The R vhigh w Wo kon Con buMOP on Among numerical methods, OPF is Among exploited numerical as an inner methods, optimization OPF is commonly algorithm exploited optimization Among methods, Among OPF numerical is commonly methods, exploited OPF as an commonly optimization algorithm as inner to optimization algorithm to En En En o o facilitate precision, stochastic, efficient time computation and uncertain operation situations. Assessments 5.2. Assessments of Multi-objective of Multi-objective Planning Planning Methods 5.2. Methods Assessments of Multi-objective Planning Methods En o facilitate high precision, stochastic, facilitate time precision, computation stochastic, and efficient uncertain time operation computation situations. and uncertain situations. 5.2. Assessments of Multi-objective Planning Methods Bu 5.2. BL M D LENT DPL R E GA DM Bu BL M D LENT DPL R nd m Bu BL M D LENT Bu DPL BL M R D nd m LENT DPL E GA R nd m DM E GA DM Bu Bu BL M BL M mp mp L L H H u ACO u ACO W h W h Bu BL M mp L H u ACO W h m PEM Bu R nd m BL M AMBA m PEM R nd m AMBA Bu BL M m PEM R nd m AMBA However, However, the problem the problem formulation formulation is very is very rigid, However, rigid, and few and the variations problem few variations formulation can be can incorporated. be is incorporated. very rigid, Dynamic and Dynamic few variations can be incorporated. Dyn However, the problem formulation is very rigid, and few variations can be incorporated. Dynamic 5 Con bu on o he Rev ew Wo 5 Con bu on o he Rev ew Wo k 5 Con bu on o he Rev ew Wo k programming programming (DynP) (DynP) and exhaustive and exhaustive search search (ES) although (ES) although they promise they promise to find to a find global a global optimum optimum The con The bu con on bu o on ev o ewed ev wo ewed k a wo e k hown a The e hown ch con ono ch bu og ono on ca og y o n ca ev Tab y ewed n e Tab 5–8 wo e add k 5–8 a e add hown ng e each ch ng ono each og ca y n Tab e 5–8 add e programming programming (DynP) and (DynP) exhaustive and exhaustive search (ES) search although (ES) they although promise they to promise find a global to find optimum a global optimum The con bu on o ev ewed wo k a e hown ch ono og ca y n Tab e 5–8 add e ng each solution, solution, however however they they have have not advocated not advocated for large for large distribution distribution systems. systems. Likewise, aLon PT ego ca yAmong ego unde unde he MOP he MOP amewo kis aH MODGP PT k aLENT ca MODGP ego MOVPQ yexploited unde MOVPQ MOCRU he MOP MOCRU and amewo MONTR and kACO MONTR a Likewise, MODGP emethods, pec emajor ve pec y MOCRU and MONTR e situat pecng v however they have not advocated for large systems. Likewise, ayve major Rm vn wprecision, Wo k5PT Con bu 5Among 1ego R vefficient wyBu dnumerical koBL Con bu on 5and 1distribution R vinner woptimization dand Wo koperation Con bu PT ca y unde he MOP amewo k and MODGP MOVPQ MOCRU MONTR eGA pec ve 5ca vhigh w dBu kWo Con bu on Among numerical methods, methods, OPF OPF commonly is commonly exploited exploited as an inner as Among an inner numerical optimization algorithm algorithm OPF toomajor is commonly exploited as an inner numerical methods, OPF is commonly as an optimization algorithm to En ohaMOVPQ oyto facilitate precision, stochastic, efficient facilitate time computation high precision, uncertain stochastic, situations. time computation uncertain operation 2BL A1Rev m n oWo Mu ob vamewo P ann M hod En high facilitate stochastic, high precision, time stochastic, computation efficient and uncertain time computation and situations. uncertain situations. 5facilitate 2 A5 1Con odBuMu ob vR 5solution, P 2on ann A ng M m n hod Mu ob vng P ann ng M hod D LENT D DPL DPL RBu nd m R nd Eefficient GA E DM DM Bu BL M D LENT DPL R EGA DM Bu M mpWo L HEn uoperation mp W u Wop mp L ACO W h Buo En BL M m PEM BL M R m m PEM Bu AMBA BL m Mm mincorporated. AMBA Rthey ndHand mpromise AMBA BL M m PEM R nd m AMBA However, the problem formulation is very rigid, and few variations can be incorporated. Dynamic However, the problem formulation However, is very the rigid, problem and few formulation variations can very be rigid, incorporated. and few Dynamic variations can be Dynamic bu on he 5The ew Con Wo bu k on o he Rev ew 5OPF Con bu on ouoperation he Rev ew Wo k programming programming (DynP) (DynP) and exhaustive and exhaustive search search programming (ES) although (ES) although (DynP) they they promise and promise exhaustive to find a5–8 find search global aLikewise, (ES) optimum optimum toeMONTR find a ACO global optim The con bu PT on oMPT ev con wo bu kBu a(being on ehe hown oev ev ch ewed ono wo og ca athe y The eadvocated n hown con Tab edistribution ch bu 5–8 ono on add og oan ca end ev y ng ewed n each Tab wo ekis kaIn aPEM add eaalthough ch each og ca yinner n Tab 5–8 addalgorith e pec ng programming (DynP) and exhaustive search (ES) although they promise to find a5–8 global optimum con bu on o ewed wo k ais hown ono og ca y n Tab eto add eaMINLP, ng each solution, solution, however however they have they not have advocated not advocated for large for large distribution systems. systems. Likewise, ang major ca ego ca yThe ego unde unde he MOP MOP amewo amewo kis acommonly MODGP PT kH aecommonly ca MODGP ego MOVPQ yexploited unde MOVPQ MOCRU he MOP MOCRU and amewo MONTR and MONTR MODGP eeglobal pec emajor ve pec MOVPQ yto ve y MOCRU and euncert v solution, however they however have they not advocated have not for large distribution for large distribution systems. Likewise, systems. Likewise, major and major 5 1ego Rewed vsolution, wyLENT dBu Wo kBL Con bu on 5 1k R vistime w dch Wo knd Con bu on PT ca y unde he MOP amewo kis and MODGP MOVPQ MOCRU and MONTR ehown pec ve yono 5 1 R v w d Wo k Con bu on 5 Among Among numerical numerical methods, methods, OPF is Among exploited numerical exploited as an inner methods, as an optimization inner OPF optimization commonly algorithm algorithm exploited to as an optimization Among numerical methods, OPF commonly as inner optimization algorithm to drawback drawback of MCS of MCS (being inner inner optimization) optimization) is high the high computation computation time. time. In comparison, In comparison, MINLP, drawback of MCS (being inner optimization) is the high computation time. comparison, MINLP, En En En o o facilitate facilitate high precision, high precision, stochastic, stochastic, efficient efficient time computation computation facilitate and uncertain and high uncertain precision, operation operation stochastic, situations. situations. efficient time computation and 5 2 A 5 2 m A n o m Mu n o ob Mu v ob P ann v ng P ann M 5 ng hod 2 M A hod m n o Mu ob v P ann ng M hod facilitate high precision, stochastic, efficient time computation and uncertain operation situations. 5 2 A m n o Mu ob v P ann ng M hod Bu M D LENT DPL Bu R m D E GA LENT DPL DM R m E GA Bu BL M D DPL R nd m E GA DM Bu BL M mp Bu L BL M mp L Bu u ACO BL H M W h mp u ACO L W h H u ACO WD BL M mp L H u ACO W h Bu m PEM Bu R m BL M AMBA m PEM Bu R nd BL m M m AMBA PEM R nd m Bu m PEM R nd m AMBA However, the formulation is very rigid, However, and few the variations problem formulation can be incorporated. is very rigid, Dynamic and few variations can beono incorporated. Dyn the problem formulation However, the isproblem very problem rigid, formulation and few variations is very rigid, can be and incorporated. few variations Dynamic can be incorporated. Dynamic 5 Con bu on o he Rev ew Wo k 5 Con bu on o he Rev ew Wo k 5However, Con bu on o he Rev 5 ew Con Wo bu k on o he Rev ew Wo k programming (DynP) and exhaustive search (ES) although they promise to find a global optimum The con bu on o ev ewed The con wo k bu a e on hown o ev ch ewed ono og wo ca k y a e n Tab hown e ch 5–8 ono add og The e ca con ng y each n bu Tab on e 5–8 o ev add ewed e ng wo each k a e hown ch og ca y n Tab (DynP) and exhaustive programming search (ES) (DynP) although and exhaustive they promise search to find (ES) a although global optimum they promise to find a global optimum The con bu on o ev ewed wo k a e hown ch ono og ca y n Tab e 5–8 add e ng each solution, solution, however however they they have have not advocated not advocated solution, for large for however large distribution distribution they have systems. not systems. advocated Likewise, Likewise, a for major large a major distribution systems. Likewise, a m PT ca ego y unde he MOP PT ca amewo ego y unde k a MODGP he MOP MOVPQ amewo PT MOCRU k a ca MODGP ego and y unde MONTR MOVPQ he MOP MOCRU e pec amewo ve and y k MONTR a MODGP e pec MOVPQ ve y MOCRU and MONTR e pec v Contr but ons 5 1 Rev ewed Work Contr but ons solution, however they have not advocated for large distribution systems. Likewise, a major 5 1 R v w d Wo k Con bu on 5 1 R v w d Wo k Con bu on PT ca ego y unde he MOP amewo k a MODGP MOVPQ MOCRU and MONTR e pec ve y 5 1 R5v1w Rev d Wo ewed k Con Work buprogramming on Among nume ca me hod OPF common y exp ocomputation ed athe an nne op m za on go hm oohis Among nume Bu ca me hod OPF Among common nume yDPL exp ca me o ed hod ang an OPF nne common op m za y exp on aand omethods. go ed hm acomputation an ocan nne op maEAMBA za on aDynamic go hm origid, drawback drawback of MCS of (being MCS (being inner optimization) inner optimization) is the high is the computation high computation time. In time. comparison, In comparison, MINLP, MINLP, drawback of drawback MCS (being of MCS inner (being optimization) inner optimization) is the high computation is high time. In comparison, time. In comparison, MINLP, MINLP, En En En o om facilitate facilitate high precision, high precision, stochastic, stochastic, efficient efficient facilitate time computation time high precision, uncertain stochastic, and uncertain operation efficient operation situations. time computation situations. and uncertain operation situat 2 A 5 2 m A n o m Mu n o ob Mu v ob P ann v P ann M 5 ng hod 2 M A hod m n o Mu ob v P ann ng M hod En facilitate high precision, stochastic, efficient time computation and uncertain operation situations. 5 2 A m n o Mu ob v P ann ng M hod D LENT R D nd m LENT DPL Bu E GA R nd BL M DM D LENT GA DPL DM R nd m E GA D Bu BL M D LENT DPL R nd m E GA DM Bu 5 BL M mp L Bu H BL M u mp ACO L W h Bu H BL M mp u ACO L W h H Bu mp L H u ACO W Bu Bu BL M BL M m PEM m PEM R nd m m Bu AMBA PEM R nd m SQP, SQP, and dynamic and dynamic programming programming are the are most the most efficient efficient numerical numerical methods. m PEM AMBA SQP, and dynamic programming are the most efficient numerical methods. However, However, the problem the problem formulation formulation is very is rigid, very and rigid, few and variations few However, variations can be the incorporated. problem be incorporated. formulation Dynamic very and few variations can be However, the problem formulation is very rigid, and few variations can be incorporated. Dynamic Con 5 bu Con on bu ocon he on Rev o he ew Rev Wo ew k k 5solution, Con on oand he Rev ew Wo k programming (DynP) and exhaustive search programming (ES) although (DynP) they promise and exhaustive to find a search (ES) optimum although they promise toIn aMOCRU global optim The oinner ev ewed kasystems. aThe con hown ch on ono oan og ca ev ewed ynne n Tab wo eglobal kacommon 5–8 aglobal eetc.) hown ehigh ch ng ono ca ynne ntime. Tab efind 5–8 add ngh programming and programming exhaustive search (DynP) (ES) although exhaustive they promise (ES) to find abu global they promise to find optimum The con bu on o solution, ev ewed wo eunde hown chMOP ono og ybu n Tab eon 5–8 add eWo ng each Artificial (AI)-based metaheuristics (for example ACS, and hybrid solution, however they have not advocated for large distribution systems. Likewise, aMINLP, major PT ego yintelligence he PT ca amewo ego y unde k aobu MODGP he MOP MOVPQ amewo MOCRU k MODGP and MONTR MOVPQ PT ca eMOCRU ego pec y ve unde yABC, and he MONTR MOP amewo ego pec keach aon y MOVPQ and however they have not advocated however for large they have distribution not advocated for Likewise, large distribution aoptimum major systems. Likewise, ay major 5 1eca R v kwpa(DynP) dec Wo k5 Con bu 5 1ego on R vca w dBu Wo kBL Con bu on 5OPF 1Rksearch R v2m w dethe Wo Con bu on PT ca y unde he MOP amewo awo MODGP MOVPQ MOCRU and MONTR eadd pec ve Among Among nume nume ca me ca hod me OPF hod common common Among y exp o ykhigh nume exp ed aMu o ca an me aua nne an hod op m OPF op za m on za aAMBA go on ahm y exp o hm ocomputation ed oog aMODGP an op mcomparison, za on aego Among nume ca me hod OPF common yalthough exp o ed aed nne op m za on aACO go hm ove drawback drawback of MCS of MCS (being (being inner optimization) optimization) drawback is the is high the of computation MCS computation (being inner time. time. optimization) In comparison, In comparison, is the MINLP, MIN drawback of MCS (being inner optimization) is the high computation time. In comparison, MINLP, En En En o o ac a e h gh p ec on ocha c e c en me compu a on and unce a n ope a on ua on 5 2 A 5 2 m A n o m Mu n ob Mu v ob P ann v ng P ann M 5 ng hod M A hod m n o ob v P ann ng M hod En o ac a h gh on ocha ac c e a e c en h gh me p ec compu on a ocha on and c e unce c en a n me ope compu a on a on on and unce n ope a on ua 5 2 A m n o Mu ob v P ann ng M hod Bu BL M D LENT DPL Bu nd BL M D E GA LENT DPL DM Bu R nd m D LENT E GA DPL DM R nd m Bu D LENT DPL R nd m E GA DM Bu Bu M BL M mp mp L L H H u Bu ACO u BL M W h W mp h L H BL M mp L H u ACO W h Bu BL M m Bu PEM R nd m m PEM Bu AMBA R nd BL m M m PEM R nd m AMBA SQP, and SQP, dynamic and dynamic programming programming are the are most the efficient most efficient numerical numerical methods. methods. SQP, and dynamic SQP, and programming dynamic programming are the most are efficient most numerical efficient methods. numerical methods. However, However, the problem the problem formulation formulation is very is rigid, However, very and rigid, few the and variations problem few variations formulation can be can incorporated. be is incorporated. very rigid, Dynamic and Dynamic few variations can be incorporated. Dyn However, the problem formulation is very rigid, and few variations can be incorporated. Dynamic Metaheuristic Metaheuristic (MH) (MH) methods methods like evolutionary like evolutionary algorithms algorithms have have been been considered considered as aadd natural as aeono natural 5programming Con bu on ocon he Rev ew Wo k 5PT Con bu on odistribution he Rev ew Wo k Metaheuristic (MH) methods like evolutionary algorithms been considered as ang 5solution, Con bu on he Rev ew Wo k programming (DynP) (DynP) and exhaustive and exhaustive search (ES) (ES) although they promise they promise to find (DynP) to aCon and abu global optimum exhaustive optimum search (ES) although they to The bu o ev ewed wo The kayca con aMODGP ehigh hown bu on ch othey ono ev og ca y wo n Tab kand eglobal hown 5–8 ch og each cay n Tab eand 5–8MONTR add e promise ng each and exhaustive search (ES) although promise to find global optimum The bu on o drawback wo kaw aMODGP hown ch ono og ca y(DynP) n Tab eon 5–8 add esearch ng each solution, however they have not advocated solution, for large however distribution they have systems. not advocated Likewise, aMINLP, for major distribution systems. Likewise, aua m PT ca ego ymp unde he MOP amewo k ego y unde he MOP MOCRU amewo kEfind MONTR aaGA MODGP ego pec MOVPQ MOCRU pec v however they solution, have however they for have large not distribution advocated systems. for large Likewise, aewed major systems. ayua major R dkeonWo Con bu on 5programming 1computation R vand whave Wo kaathe on egocon y unde he MOP amewo MOVPQ and MONTR eann ve 5ev 1 ewed R5 v12 w dvkMCS Wo Con bu 5programming on R vnot w dP Wo kca Con bu on Among Among nume nume me ca hod me OPF hod OPF common common Among yalthough exp o ythe nume exp ed aMu o an ed me anne an hod op nne m OPF op on za acomparison, go on ahm exp o hm onatural ed ay an nne op za ae go Among ca me hod common y exp oRMOVPQ ed aon an nne op za on aACO go hm ove drawback of MCS inner optimization) the computation time. In comparison, of (being inner drawback of MCS is the (being high computation inner time. In is comparison, MINLP, time. In MINLP, En En olarge ac a5programming ac eMetaheuristic h gh abu eadvocated p hnume ec gh pMu on ec ocha on ocha cpromising evOPF cpec coptimization) en ac cis me en aways compu ehod me h gh compu apo p on and aca on on unce unce adcWo n ope em n aLikewise, cm ope en on ahod me ua on compu on ua on aoLENT on and unce a find nmaddress ope aon A m ookes Mu ob 2optimization) A vhMOCRU m ann noMetaheuristic ng o(MH) M Mu hod P ann 5 ng 2can M A m n ohigh ob vaddress P ann ng M En ac aHoweve eand gh p ec on ocha cof eaLENT cem en me compu aec on and unce ahccomplex n ope ang on on 5 21Metaheuristic A m n ob vexhaustive P ng M hod Bu BL M D D LENT DPL nd m R nd Bu Eza GA E GA M DM DM DPL RAMBA ndACO m on D LENT DPL DM BL M Bu Las BL M mp H Ln Bu u ACO H W mp u L W hRo u Wh BuBu m Bu R nd m m Bu AMBA BL m Mbeen m PEM AMBA nd SQP, SQP, and dynamic dynamic programming are the are most the most efficient and dynamic numerical numerical programming methods. methods. are most efficient numerical methods. Bu BL M PEM R nd m AMBA and dynamic programming are the most efficient numerical methods. Thechrono contr but of rev ewed work The are contr shown but chrono ons rev og ca ewed yefficient work Tab are es 5–8 shown chrono og each ca y n Tab es 5–8 ng Howeve he p ob em o mu aob on ve y gDPL d ew va aodeal can be nco acommon ed Dynam coDng Howeve he obon em ohe mu aEn on ve y(being he p dprogramming ob and em ew o mu va aadvocated on on ve be yalthough g nco dalgorithms and athe ew ed va Dynam aocha on can be nco po aas ed Dynam rk PT arecashown og ca yThe npcon Tab 5–8 address ng each methods can consider to with MOP problems Metaheuristic (MH) (MH) methods methods like evolutionary like evolutionary algorithms algorithms have been have considered been considered as aas natural ahm natural methods (MH) like methods evolutionary like evolutionary algorithms have considered have considered natural as 55ons Con 5PT Con Wo bu on o he Rev ew Wo k 5SQP, Con bu on o he Rev ew 5SQP, Con on o he ew Wo k programming (DynP) (DynP) and exhaustive and search programming search (ES) (ES) (DynP) they promise they and exhaustive promise to find to arequirements search find aBL global (ES) optimum although optimum they promise to aeglobal optim bu on o ev ewed The con wo kg bu aPEM ethe on hown o ev ch ewed ono og wo The ca kayfor con y aPEM eHowever, n Tab hown bu edistribution on ch 5–8 ono add ev og ewed etime. ca ng ynne wo each n Tab kand apo eglobal hown 5–8 add ch eono og each ca ycHm n Tab eMONTR 5–8 add eon ng each programming (DynP) and exhaustive search (ES) although they promise to find abeen global optimum way way of addressing of addressing complex complex MOP MOP problems. problems. However, the high high computational computational requirements and way of addressing complex MOP problems. However, the high computational requirements and solution, however however they have they not have advocated not advocated large for large solution, distribution systems. however systems. Likewise, they Likewise, have aMINLP, major not aua advocated major for large distribution syste ca ego ymp unde he MOP amewo PT ca k ego MODGP y unde he MOVPQ amewo MOCRU k ak MODGP MONTR MOVPQ ego pec MOCRU y and pec ve yhD solution, however they have not for large distribution systems. Likewise, ay major 1MOVPQ RRev vBL 5 w 12ew R dbe Wo vethe w kk Con dBu Wo kbu Con on bu on PT ca ego y unde heoptimization MOP amewo k abu MOCRU and MONTR eann pec ve 5SQP, 1eof R vn w dca Wo kRev Con bu on Among Among nume nume me ca hod me OPF hod OPF common common Among exp o yalthough nume exp ed aMOP o ca an ed me anne an hod op OPF op za m on common za aEAMBA go on ahm exp o hm oand oacon anatural an nne op mcomparison, a go Among nume ca me hod OPF common y exp o ed aon an nne op m za on aACO go ove drawback MCS (being inner optimization) drawback is the high of computation MCS (being inner optimization) In comparison, is the high computation time. In MIN drawback of MCS (being drawback inner optimization) of MCS (being is inner high optimization) computation time. is the In high comparison, computation MINLP, time. In comparison, MINLP, En En En om oed ac a ac e h gh a p h ec gh p on ec ocha on ocha c e c c en e ac c me en a compu e me h gh compu a p on ec and a on on unce and ocha unce a n c ope e a n a c ope en on a me ua on compu on on a and unce a n ope a on ua 5and 2Con A mMODGP non oo 5solution, Mu ob 5 A v P ann m n ng o M Mu hod ob v P ann ng M hod 5 2 A m n o Mu ob v P ann ng M hod ac a h gh p ec on ocha c e c en me compu a on and unce a n ope a on ua on 5 2 A m o Mu ob v P ng M hod D LENT DPL R D nd LENT DPL E GA R nd DM D LENT GA DPL DM Eza GA M Bu L BL M mp H L Bu u ACO BL H M W h mp u L W h H u ACO W BL M mp L H u ACO W h BuBu m Bu PEM m m PEM Bu AMBA m m PEM R nd m AMBA SQP, and dynamic programming are the most efficient numerical methods. Bu m PEM R nd m AMBA SQP, dynamic programming are and most dynamic efficient programming numerical methods. are the most efficient numerical methods. Howeve Howeve he p ob he em p ob o em mu a o mu on a ve on y Howeve ve g d y and g d ew he and va p ob ew a em va o a can mu on be a can nco on be po ve nco a y ed po g Dynam a d ed and Dynam c ew va a on can be nco po a ed Dyn Howeve he p ob em o mu a on ve y g d and ew va a on can be nco po a ed Dynam c Metaheuristic Metaheuristic (MH) (MH) methods methods like evolutionary like evolutionary Metaheuristic algorithms algorithms (MH) have have methods been been considered like considered evolutionary as a natural as a algorithms natural have been considered as a na Metaheuristic (MH) methods like evolutionary algorithms have been considered as a natural 5 bu he Rev ew Wo k 5 Con bu on o he Rev ew Wo k 5 Con bu on o he Rev ew Wo k 5ev Con bu on o he ew Wo kwo og amm ng DynP and exhau ve ea ch ES afor hough hey p om ecomputational ohigh nd arequirements gLikewise, oba op mum The con buand on o ev wo kon aDynP eRev hown ch ono og n Tab edistribution 5–8 add The etime. con ng each bu on orequirements ev wo ky aa e an hown chopono og ca ya go nveTab p og amm ng DynP exhau p og ve amm ea ch ng ES a on hough and exhau hey padvocated om ea ey ch o nd ES aoptimal aono g oba op mum pof om ecomparison, o nd gMOVPQ oba op mum The con bu on o The con wo k bu a(being hown oca ch ev ewed ono og ca k y aca nan ecommon Tab hown elocal 5–8 add og eaed ca ng y each n Tab eyand 5–8 add eaMINLP, each way of way addressing of addressing complex complex MOP problems. MOP problems. However, However, the high the computational high and and way of addressing way of complex MOP complex problems. MOP However, problems. the high computational the computational requirements requirements and solution, solution, however however they have they not have advocated not solution, large however large they distribution have systems. not systems. advocated Likewise, Likewise, for major aMINLP, major distribution systems. Likewise, aua PT ca ego y and unde he ca amewo ego unde kepLENT aaddressing MODGP he MOP MOVPQ ca k ego afor MODGP y unde and MONTR he MOVPQ MOP ehey amewo MOCRU pec ve k aop MODGP MONTR eng pec MOCRU and MONTR eACO pec ym solution, however they have not advocated large distribution ay major 5p 1ca RMOP vewed w dewed Wo km Con bu 5 12ng R vthe w dch Wo khough Con on 1MOCRU R vann wdynamic d Wo kfrom Con bu on Among nume me Among OPF nume common me y hod ove OPF ed ave Among nne op ylocal nume m za o on me amethods. abu go an hod hm OPF o za on aewed go exp hm oDM ed nne muope za ca me hod common yd exp o ed an nne op za on aaACO go hm ove possibility possibility of of premature convergence convergence towards towards local optimal solutions solutions need need to be to addressed be addressed with with possibility of premature convergence towards optimal solutions need be addressed drawback drawback MCS of MCS (being inner optimization) inner optimization) the is the computation high drawback In time. MCS comparison, In (being comparison, inner optimization) isand the high computation drawback of MCS (being inner optimization) is high computation time. In MINLP, under the MOP framework PT category MODGP under MOVPQ the MOP MOCRU framework and as MONTR MODGP respect MOVPQ ve MOCRU y MONTR respect v En olarge as5 MODGP MONTR respect ve y ac aamm ac eAmong h gh aamm epremature pthe hnume ec gh on ec ocha on ocha cexp eOPF cPT cthe en ac cis me en aamm compu me hfor gh aexp p on ec and aca on unce and ocha unce acan n ope em athe n aLENT cm ope en on me ua on on ua ava unce nnco aon on 5PT 2hod A n oefficient Mu ob vamewo P ann 5MOCRU A M hod m n oHowever, Mu ob vnne P ann ng M hod efficiently aac future perspective. The hybrid metaheuristic techniques show robustness En En osystems. aamm eof has gh pyDynP ec on ocha emp cech en me aew on and unce ahcnd n ope ato on ua on 2 APT category m n MOVPQ o Mumore ob 5 vMetaheuristic Pand ng M hod Bu D R D nd m LENT DPL E GA R DM D E GA Ro nd E GA D D LENT DPL R nd E GA Bu En BL M mp Bu L BL M mp H L Bu u ACO BL H M W mp u L W hwith Hand Wh Bu BL M L H ACO W Bu M m m R nd m m AMBA m PEM AMBA m AMBA SQP, and dynamic programming the most SQP, efficient and dynamic numerical programming methods. are most efficient numerical Bu m PEM m AMBA SQP, programming SQP, and are dynamic most programming numerical are methods. efficient numerical Howeve he pBu ob he em p ob oBL em mu ao mu on acewed on y Howeve ve gch d gehigh ew he and va pBu ob acomputation em va on o and can mu on be nco on be po ve nco acommon y ed po gaDPL ad ed and Dynam coand ew con aand on can po aoba ed time. Dyn Howeve he p ob em oDPL mu aare on ve y g d and ew va aon on can be nco po ed Dynam con (MH) methods like evolutionary algorithms have been considered as aDynam natural 5Howeve Con 5 bu Con on bu o he on Rev he ew Rev Wo ew k Wo kawo 5PT Con bu on o he Rev ew Wo k (MH) methods like Metaheuristic evolutionary (MH) algorithms methods have like been evolutionary considered algorithms as ahey natural have been considered as ahng natural 5PT Con bu on o he Rev ew Wo ktowards p og p og ng DynP ng DynP and exhau and exhau ve ea ve p ea ES og ch ES hough ng acompu hough DynP hey p and om exhau p eon om ecomparison, ve o aurequirements ea nd oba ch aaeoba op ES mum acompu op hough mum hey pmethods. om ebeoao nd aarequirements g op m The con The con on oo on ev ewed o ev wo k aPEM emost hown ky aPEM eHowever, ch hown ono ch og ono ca yon og ca Tab yaoem eve n 5–8 Tab eg add 5–8 add ng each eand p og ng and exhau ve ea ES acompu hey p om eand o nd arequirements g op The con bu on o ev ewed wo k anne eaed hown ono ca y n Tab ecomparison, 5–8 add eDM ng each way way of addressing of addressing complex complex MOP problems. problems. way However, of high the complex high computational computational MOP problems. requirements and the high computational way of addressing complex MOP problems. However, the high computational ohe u on howeve hey have no advoca o ge dMONTR bu y L kew eon aoba ma oMINLP, ca ydgh unde MOP amewo abu MODGP MOVPQ MOCRU and ca eavn ego pec y unde y he MOP amewo kaeach amost MOVPQ MOCRU and o on hey have no o u advoca on ed oprogramming aMOP hey ge have d on advoca y em orequire L kew aanumerical ge eaMOCRU d asolutions ma bu on em Lg kew emum ay ma oMODGP Rhoweve Wo kec Con bu 5 12ego on R vemethods. w dBu Wo kBL Con bu on 5 1ng R vlocal w dhough Wo kch Con bu ego yavego unde he MOP ca amewo y kunde aD MODGP he MOP MOVPQ amewo MOCRU MODGP and MONTR MOVPQ eand pec ve yo MONTR eHowever, pec ve 5 1Metaheuristic R vM w dbu Wo kk Con bu on Among nume ca me hod Among OPF nume common ca me y exp hod o OPF ed aed an common op yMu exp m za o on ed Among aog go an hm nne nume oAlso, op ca m me za hod on aMINLP, go OPF hm ond ytime. exp ed nne op Among nume ca me hod OPF common y exp o ed asolutions an nne op m za on aACO go hm ocon possibility possibility of premature of premature convergence convergence towards local optimal local optimal solutions need to need be to be addressed with with possibility of possibility premature of premature convergence towards towards optimal local optimal need solutions to be need addressed to be with addressed with drawback drawback of MCS of (being MCS (being inner optimization) inner drawback the is the of computation MCS high (being computation inner time. optimization) In time. In is MINLP, the computation In comparison, MIN drawback of MCS (being inner optimization) is the high computation time. In comparison, En En ohigh ac ew hM p(MH) on ac ocha aamm hcBu gh ehoweve pLENT cm en ec on me compu ocha abu cno eon ac cis and en aamm unce ehigh me h gh compu athe nBu p ope ec on on on and ocha ua unce on cnd ey aM nnco caddressed ope en aDPL me on compu ua on unce aconsidered nnco aan on ua 5Metaheuristic A m n oPEM Mu ob voptimization) 5convergence P 2Rk ann A M m naddressing hod ohe ob P ann ng hod Enuca En o ohew ac a e h gh p ec on ocha c e c en me compu on and unce a n a ua on 52 A m n o Mu ob PT v 55PT P1Con ann ng hod more more efficient efficient methods. The simple The simple MH methods MH methods require require more more computation computation time. time. Also, more number Bu BL M DPL R D nd m LENT DPL E GA R nd DM D LENT E GA DM m E GA D more efficient methods. The simple MH methods more computation time. Also, more number Bu D LENT DPL R m E GA DM Bu M BL M mp mp L L Bu H BL H M u mp ACO umore Lnumber W h W hRcommon Hand uope ACO W BL M mp L H uope ACO W Bu BL Bu nd m m PEM AMBA m m PEM AMBA AMBA SQP, and SQP, dynamic and dynamic programming are the are most the efficient most efficient numerical SQP, methods. and methods. dynamic programming are the efficient numerical methods. SQP, and dynamic programming are the most efficient numerical methods. Howeve Howeve he p ob he em p ob o em mu a o mu on a ve on y Howeve ve g d y and g d ew and va p ob ew a em va on o a can mu on be a can nco on be po ve a y ed po g Dynam a d ed and Dynam c va c a on can be po a ed Dyn Howeve he p ob em o mu a on ve y g d and ew va a on can be nco po a ed Dynam Metaheuristic (MH) methods like evolutionary Metaheuristic algorithms (MH) have methods been considered like evolutionary as a natural algorithms have been as a na bu on o he Rev 5 ew Con Wo bu k on o he Rev ew Wo k 5 Con bu on o he Rev ew Wo k Metaheuristic methods like evolutionary (MH) methods algorithms like evolutionary have been considered algorithms as have a natural been considered as a natural p og amm p og ng DynP ng DynP and exhau and exhau ve ea ve ch p ea ES og ch a ES hough ng a hough DynP hey p and hey om exhau p e om o e ve o a g ea nd oba ch a g op ES oba mum a op hough mum hey p om e o nd a g oba op m The con bu on ohey ev ewed wo kop aadvoca etowards hown The ch con ono bu og on ca y oR n ev Tab ewed en 5–8 wo add kam agaddressed eed hown ng each ch ono og caany n Tab em 5–8 add eegoangh p og amm ng DynP and exhau ve ea ES an hough hey p om edon oOPF nd oba op mum The con bu onkng o awback ev ewed wo aunde eCon hown ch ono og ca yoch n Tab ehoweve 5–8 add ebu ng each way of addressing complex MOP problems. However, the high computational requirements and of addressing complex MOP way problems. of addressing However, complex the high MOP computational problems. However, requirements the high and computational requirements and o u on o u howeve on howeve hey have hey no have advoca no ed u o on ed a ge o d a ge bu d hey on bu have y on em no y advoca L em kew L e kew a o ma a o a ge ma d o bu on y em L kew m PT ca ego PT ca y unde ego y he MOP he MOP amewo amewo k a MODGP k a MODGP MOVPQ MOVPQ MOCRU MOCRU MONTR and MONTR e pec ve e pec y ve y o u on howeve have no advoca ed o a ge d on y em L kew e a ma o 5 1 R v w d Wo Con bu on 5 1 R v w d Wo k Con bu on 5 1 v w Wo k Con bu on PT ca ego y unde he MOP amewo k a MODGP MOVPQ MOCRU and MONTR e pec ve y 5 1 R v w d Wo k bu on quality solutions), promises powerful global optimization methods and have ability to address Among nume ca me hod OPF common Among nume y exp o ca ed me a hod an nne op common za on y go exp o hm ed o a nne op za on a Among nume ca(high meway OPF common y exp o ed a an nne op m za on a go hm possibility possibility of premature of premature convergence convergence towards possibility local local optimal of premature optimal solutions solutions convergence need need to be to towards addressed be addressed local with optimal with solutions need to be addressed possibility of premature convergence towards local optimal solutions need to be with d o MCS be ng nne op m za on he h gh compu a on me compa on M NLP dhod awback o MCS be nne op d awback m za on o MCS he be h gh ng compu nne a on m za me on compa he h gh on compu M NLP a me n compa on M NLP En oa ac hm gh pBu ecEn on ac ocha aWo eon cdynamic hmethods. egh cDynP p en ec me on compu ocha aP clike on emp and chigh en unce me aon compu nBu ope ac aamethods. on avbeen eand h ua gh unce on pconsidered ec ave nnd on ope ocha aad on cnumber ua emore on chRcen me compu and on and unce 5Howeve 2Con Aa e bu non ooob Mu ob 5functions 2M A vahis P ann m n ng orequired M Mu hod ob vD 5are 2MH ann A ng M m n hod ohe Mu ob P ann ng M hod En 5 En o olike acRev aamm eMetaheuristic gh pmp ec on ocha cewed eLENT cch en me compu aew on and unce atime. n ope aAlso, on ua on more efficient more efficient methods. methods. The simple The simple MH methods methods require more computation more computation time. Also, time. more number more Bu D D DPL LENT DPL R nd R nd m D E GA LENT Eas GA DPL DM DM nd GA more efficient more efficient The methods. simple The MH simple methods MH require methods more require computation more computation Also, more time. number Also, number LENT DPL R nd E GA DM BL Bu L BL M H Lrequire Bu uo ACO H W h mp u ACO L W Hm uE ACO WD Bu BL M m PEM m BL M AMBA m PEM AMBA SQP, and SQP, dynamic and programming programming are the most the SQP, efficient most and efficient dynamic numerical numerical programming methods. are the most efficient numerical methods. Bu BL M m Bu PEM BL M R nd m m PEM AMBA m AMBA SQP, and dynamic programming are the most efficient numerical methods. he p em Howeve o mu he ve p ob y em g d o and mu a ew on va Howeve a ve on y can g d be and p nco ob po em va a o ed a mu on Dynam a can on be c nco y po g ed and Dynam ew va a on can be nco po a ed Dyn Howeve he p ob em o mu a on ve y g d and ew va on can be nco po a ed Dynam c of functions of required is required to achieve to achieve nearly nearly high quality quality results. results. Also, Also, constrained constrained and unconstrained and unconstrained of functions is to achieve nearly high quality results. Also, constrained and unconstrained Metaheuristic (MH) (MH) methods methods like evolutionary evolutionary algorithms algorithms have Metaheuristic have considered been considered (MH) methods a natural as natural evolutionary algorithms have been Metaheuristic (MH) methods like evolutionary algorithms have been as a natural he 5 ew Con bu k on o he Rev ew Wo 5 Con bu he Rev ew Wo k 5 Con bu on o he Rev ew Wo k p og amm p og ng amm DynP ng and exhau and exhau ve ea ve p ea ES og ch amm a ES hough ng a hough DynP hey p and hey om exhau p e om o nd e ve o a g ea oba ch a g op ES oba mum a op hough mum hey p om e o a g oba op m The con bu on o ev ewed The con wo k bu a e on hown o ev ch ono og wo ca k y The a e n Tab con hown e bu ch 5–8 on ono add o og e ca ev ng ewed y each n Tab wo e k a 5–8 e add hown e ch ng ono each og ca y n Tab e 5–8 add e ng p og ng DynP and exhau ve ea ch ES a hough hey p om e o nd a g oba op mum The con bu on o ev ewed wo k a e hown ch ono og ca y n Tab e 5–8 add e ng each way of addressing complex MOP problems. way However, of addressing the high complex computational MOP problems. requirements However, and the high computational requirements way of addressing complex way MOP of addressing problems. complex However, MOP the problems. high computational However, requirements the high computational and requirements and o u on o u howeve on howeve hey have hey no have advoca no advoca ed o u o on ed a howeve ge o d a ge bu d hey on bu have y on em no y advoca L em kew L ed e kew a o ma e a o a ge ma d o bu on y em L kew e a m PT ca ego y unde he MOP amewo k a MODGP PT ca ego MOVPQ y unde MOCRU he MOP and amewo MONTR k a MODGP e pec ve MOVPQ y MOCRU and MONTR e pec v o u on howeve hey have no advoca ed o a ge d bu on y em L kew e a ma o 5 1 R v 5 w 1 R d Wo v w k Con d Wo k bu Con on bu on 5 1 R v w d Wo k Con bu on PT ca ego y unde he MOP amewo k a MODGP MOVPQ MOCRU and MONTR e pec ve 5 1 R v w d Wo k Con bu on Among nume ca me hod OPF Among common nume y exp ca me o ed hod a an OPF nne common op m za y on exp o go ed hm a an o nne op m za on a go hm o Among nume ca me hod OPF common y exp o ed a an nne op m za on a go hm possibility of premature convergence towards local optimal solutions need to be addressed with possibility of premature convergence possibility towards of premature local optimal convergence solutions towards need to local be addressed optimal solutions with need to be addressed with d awback d awback o MCS o MCS be ng be nne ng op nne m op za m on d za awback on he h gh o he compu MCS h gh compu be a ng on nne a me on op n me compa m za n compa on on M he on NLP h M gh NLP compu a on me n compa on M N d awback o MCS be ng nne op m za on he h gh compu a on me n compa on M NLP En o ac a e h gh p ec on ocha c e ac c en a e h me gh compu p ec a on on ocha and unce c e a c n en ope a me on compu ua a on on and unce a ope a on ua 5 2 A m n o Mu ob v P ann ng M hod 5 2 A m n o Mu ob v P ann ng M hod En En En o o o ac 5a e2 hAssessments gh p ec on 5SQP ocha cog edynam en me compu amu on unce ae(MH) n ope on ua on m ncbu ohe ob 5Howeve 2Rev vA P ann m n ng M omethods. Mu hod ob vexhau ng M hod more more efficient methods. The simple MH methods MH more require efficient require more more computation computation The simple time. MH time. methods Also, more more require number more time. more num Bu BL M D LENT DPL BL M R D m LENT DPL E R nd m DM D LENT Eas GA DPL DM EAlso, D more efficient methods. The simple MH methods require more computation time. Also, more Bu BL M mp Lhigh Bu H BL M u mp ACO Lnumber W ha Hm u AMBA ACO W Bu BL M mp Bu L BL M mp H LP uand ACO Ha W hme uconsidered ACO W hunconstrained Bu m PEM Bu R nd m AMBA m PEM R hey and dynam cng p og amm ng amu eP he mo eog camm en ca me hod Bu m Bu PEM R nd m m R nd m AMBA cMu ptime og amm ng SQP aefficient and he mo dynam eay co co psimple en og nume amm ca ng aog me e5p hod he mo enume cmethods. en nume ca hod Howeve pbu obSQP em o Howeve aeis on he ve pBu ob em g d o and ew aann on va amethods ve on y can g d be nco ew po Howeve va ed on Dynam he can pemethods ob cAlso, be em nco po mu ay ed aanumber on Dynam vek y ccomputation g don ew va akew on can be he p ob em o mu aFurthermore, on ve y gR d and ew va aAlso, on can be po aounconstrained ed cand of functions of functions required is required to achieve to achieve nearly nearly high quality quality results. results. Also, Also, constrained and unconstrained and unconstrained of functions of is functions required is to required achieve nearly to achieve high nearly quality high results. quality constrained Also, constrained and and o2 AMu tamm -ob ect P 5oMOP ann 2ca Assessments ng Methods Mu tm -ob ect ng Methods ng Methods Metaheuristic Metaheuristic (MH) methods methods like evolutionary like evolutionary Metaheuristic algorithms algorithms (MH) have been have considered been evolutionary aas natural as algorithms natural have been considered aTab na Metaheuristic (MH) methods like evolutionary algorithms have been considered aDynam natural 5 Con on ove he 5 ew Con Wo bu k on oThe he Rev ew Wo kand Con bu on o he Rev ew Wo knco 5p Con bu on o he Rev ew Wo kwo pand ng DynP and p og exhau amm ve ea DynP ch ES and anne hough ve hey ea p ch om ES ehoweve ng aunde othe hough DynP nd aGA and hey oba exhau pme op om mum ve o ea nd ch arequirements gy ES oba op hough mum pHowever, om eand oem nd aGA gaddressed oba op m The con on o ev ewed wo k hown The con ch ono bu on ca o y n ev Tab ewed eAMBA 5–8 wo add kg ahigh The eearesults. eamethods hown con ng each ch bu ono on og o ca ev ewed n Tab wo eve 5–8 aond enne add hown eand ch ng ono each og ca yeas npec og amm ng DynP and exhau ve ea ch ES aann hough hey pconstrained om eand o nd arequirements g oba op mum multi-objective multi-objective problems problems are difficult are difficult to optimize. to optimize. Similarly, Similarly, GA and GA PSO and PSO provide provide near near optimal optimal The con bu on ev ewed aPEM eve hown ch ono og ca y n Tab elike 5–8 add eon ng each multi-objective problems are difficult to optimize. Similarly, GA and PSO provide near optimal constrained real planning problems. decision-making also need to be way of way addressing of addressing complex complex MOP problems. MOP problems. However, However, high the way computational of addressing computational requirements complex MOP and problems. the high computa way of addressing complex MOP problems. However, the high computational and o u on o u howeve on howeve hey have hey no have advoca no advoca ed o u o on ed a ge o d a ge bu d hey on bu have y on em no y advoca L em kew L ed e kew a o ma a o a ge ma d o bu y L e a m PT ca ego y unde he PT ca amewo ego y unde k a MODGP he MOP MOVPQ amewo MOCRU PT k a ca MODGP ego and y MONTR MOVPQ he MOP MOCRU pec amewo ve y and k MONTR a MODGP e pec MOVPQ y MOCRU MONTR v u on howeve hey have no advoca ed o a ge d bu on y em L kew e a ma o 5 1 R v w d Wo k Con bu 5 1 on R v w d Wo k Con bu on 5 1 v w d Wo k Con bu on PT ego y unde he MOP amewo k MODGP MOVPQ MOCRU MONTR e pec ve y Among nume ca me hod Among OPF nume common ca me y exp hod o OPF ed a Among an common nne nume op y exp m ca za me o on ed hod a go an OPF hm nne o common op m za exp o go ed hm a an op m za on a go hm o possibility of premature convergence towards possibility local optimal of premature solutions convergence need to be towards addressed local with optimal solutions need to be possibility of premature possibility convergence of towards premature local convergence optimal solutions towards need local to optimal be addressed solutions with need to be addressed with d awback d awback o MCS o MCS be ng be nne ng op op za m on d za awback on he h gh o he compu MCS h gh compu be a ng on nne a on op n me compa m za n compa on on M he on NLP h M gh NLP compu a on me n compa on M d awback o MCS be ng nne op m za on he h gh compu a on me n compa on M NLP En En o ac a e h gh p ec on ocha ac c e a c e en h gh me p ec compu on a ocha on and c e unce c en a n me ope compu a on a ua on on and unce a n ope a on ua on 5 2 A 5 2 m A n o m Mu n o ob Mu v ob P ann v ng P M ann hod ng M hod En En En o o o ac a e h gh p ec on ocha c e c en me compu a on and unce a n ope a on ua on 5 2 A m n o Mu ob v P ann ng M hod more efficient methods. The simple MH methods require more computation time. Also, more number Bu BL M D LENT DPL Bu R nd m BL M D E GA LENT DPL DM R nd m E GA D efficient methods. The simple more MH efficient methods methods. require The more simple computation MH methods time. Also, require more more number computation time. Also, more number D LENT DPL Ra D nd LENT DPL E GA R nd DM E GA Bu mp Levo Bu H ubeen mp ACO LW hDM H on u AMBA ACO WN Bu BL M mp Bu L BL M mp H Ly uis ACO HRev W ha uhe ACO W h m PEM AMBA m PEM SQP and SQP dynam and dynam c p og c amm p og ng amm e ng he a mo e he SQP e mo c and en e dynam nume c en nume c ca p og me ca amm hod me ng hod a e mo e c en nume ca me hod Bu m Bu PEM m m PEM AMBA m AMBA SQP and dynam c og amm ng a e he mo e c en nume ca me hod Howeve he p ob em o mu on Howeve ve y g d he and p ob ew em va o a mu on can on be ve nco y po g d a and ed Dynam ew va c a can be nco po a ed Dyn Howeve he p ob em o more mu aMe on ve y g d and ew va a on can be nco po a ed Dynam c of functions of functions is required is required to achieve to achieve nearly nearly high of functions high quality quality results. required results. Also, Also, to constrained achieve constrained nearly and unconstrained and high unconstrained quality results. Also, constrained and unconstra of functions is required to achieve nearly high quality results. Also, constrained and unconstrained Me aheu c MH me hod ke evo u ona a go hm have been con de ed a a na u a 5 Con 5 bu Con on bu o he on Rev o he ew Rev Wo ew k Wo k 5 Con bu on o he ew Wo k aheu c MH me hod ke Me evo aheu u ona c y MH a go me hm hod have ke been u con ona de y ed a go a a hm na u have con de ed a a na u a 5 Con bu on o he Rev ew Wo k p og amm ng DynP and p exhau og amm ve ng ea DynP ch ES and a exhau hough hey ve ea p om ch ES e o a nd hough a p g og oba hey amm op p om ng mum DynP e o nd and a g exhau oba op ve mum ea ch ES a hough hey p om e o The con The bu con on bu o on ev ewed o ev wo ewed k a wo e hown k a e ch hown ono ch og ono ca y og The n ca Tab con y e n 5–8 bu Tab on e add 5–8 o e ev add ng ewed each e ng wo each k a e hown ch ono og ca y n Tab p og amm ng DynP and exhau ve ea ch ES a hough hey p om e o nd a g oba op mum multi-objective multi-objective problems problems are difficult are difficult to optimize. to optimize. Similarly, Similarly, GA and GA PSO and provide PSO provide near optimal near optimal The con bu on o ev ewed wo a e hown ch ono og ca y n Tab e 5–8 add e ng each multi-objective multi-objective problems are problems difficult are to difficult optimize. to Similarly, optimize. GA Similarly, and PSO GA provide and PSO near provide optimal near optimal way of way of addressing complex complex MOP problems. MOP problems. way of However, high the computational high computational requirements requirements and the high computational way of complex MOP problems. However, the high requirements and oefficient on howeve hey o5have on no howeve advoca ed hey have aMOVPQ ge no d advoca o on on ed y howeve o em anBu LCon dnd hey bu have enne aMOP on ma no yneed oproblems. advoca em L ed kew orequire ewith ge ma dtowards oRacommon bu on y em Lon kew e on aM m PT ca ego ygh he MOP amewo kpremature apbe PT ca ego ysystems. unde he MOP and MONTR k PT ca MODGP eacomputational ego pec yme ve unde MOVPQ yann he MOP MOCRU and MONTR k MODGP ehod MOVPQ ve yed MOCRU and o ud on howeve hey have no advoca orequire aamewo ge d bu on y em L kew eM aNLP ma 5ac 1u RAmong venume w dBu Wo k5d Con bu 1u on R vmethods. w dBu Wo kMODGP Con bu 5d 12However, R vgh w dcPEM Wo knd bu on solutions for large large distribution distribution systems. PT ca ego unde he MOP amewo aecommon MODGP MOVPQ and MONTR pec ve yo 5d 1solutions R vhod wann dfor bu on solutions for large distribution systems. nume ca me OPF common y exp o ed aMOCRU an nne op za on Among hm nume o ca me hod OPF yapec exp on arequirements anDynam nne op me hod Among OPF nume common ca me yem exp ed OPF nne op y m exp za om on acomplex asolutions go an hm nne o op m za on aeHowever, go oand possibility possibility of premature of convergence convergence towards optimal local optimal solutions possibility solutions need of to be addressed to convergence addressed with local optimal solutions nee possibility of premature convergence towards local optimal need to be addressed with awback awback oaddressing MCS omethods. MCS be ng be nne ng op nne m op za m za on he haddressing gh o he compu MCS h gh compu be akew ng on ago me on op n me compa m n compa on on he on hhaahm gh NLP compu acan on me compa N awback oy MCS ng nne op m za on he h gh compu ahave on n M am h pca on ac ocha aamm erequired cBu h eMu gh cCon en ec me on compu ocha aon ac ckan on abu and ced eu en unce me p ec athe ope on aaMOCRU ocha on and cua econ unce cD en aza ned me ope compu aon on aaM ua on on and n ope aEaAlso, ua 2ob A m n oWo ob vohod Po ann ng M 5 hod A m n oge Mu ob Pon ng M hod 5more 2aAmong nbu ounde vaddressing P ng M hod more efficient simple MH methods more require efficient more methods. computation The simple time. MH Also, methods more more num Bu D LENT DPL R m E LENT DPL DM nd m methods. The more simple efficient MH methods require The simple more MH computation methods time. more more computation number time. more number BL M D LENT DPL BL M Ra D nd LENT m DPL E GA R m DM E GA DM BL M Llocal Bu H BL M u mp ACO Lnumber W Hunce u ACO W mp Lp mp H L u ACO H W hv uGA ACO W h m PEM AMBA SQP and SQP dynam and dynam cBu p og amm og ng amm ag eon ng he mo eon he SQP eawback mo cand en dynam nume en nume ced ca p og me ca amm hod me ng hod apremature ek mo eg camewo nume ca PEM Bu BL M m PEM m AMBA R nd m AMBA AMBA SQP and dynam og amm ng amp eatowards he mo eona ceona en nume ca me hod Howeve he p ob o mu Howeve ve y he g p d and ew o va mu ame on can ve be ycompa nco g dbe and ana ed ew va on ccomputation be nco po ed Howeve ob em o of mu on ve yBu gMu doec and ew va aeWo can be nco po aachieve ed Dynam ch of functions is to achieve nearly high quality results. Also, constrained and unconstrained aiming finding optimal weights, for FDNs. functions is required to achieve of functions high is required quality results. to Also, nearly constrained high quality and results. Also, constrained aheu aheu cm MH cme me MH hod hod ke evo ke u evo ona Me u aheu anner go y aem hm go cunconstrained MH have hm been hod been ke de con evo de aza u ona aunconstrained y u aNLP aDynam na go uakhm hm have con de aon ay nacDm 5A Con on he Rev ew Con bu k on o he Rev ew Wo k 5p Con bu on o he Rev ew Wo Me aheu ccproblems MH me hod ke evo u y aob go hm been con de ahm na u ahm og ng and exhau ve og ea ch amm ES acompu DynP hough and hey exhau p om ohe ea nd ch aza ES oba ato hough op mum hey pme e ge otime. nd aGA g oba The con bu o ev ewed The con wo knumer bu acThe eed on hown o ev ch ewed ono og ca k y aThe eyed n Tab con hown eAlso, bu ch 5–8 on ono add o og eon ca ev ng yon each n Tab wo eAlso, k 5–8 amore epo add hown een ch ng ono each ca yom nbeen Tab en 5–8 add e op ng p og amm he ng DynP and exhau ve ea ch ES aon hough hey pfor om eDynP o nd ame oba mum multi-objective multi-objective problems are difficult are difficult to optimize. multi-objective to optimize. Similarly, Similarly, problems GA and GA are PSO difficult PSO provide provide to optimize. near near optimal Similarly, optimal and PSO provide near opt multi-objective problems are difficult to optimize. Similarly, GA PSO provide near optimal way add ng comp ex MOP p ob em Howeve he h gh compu aewed ona emen way add ng comp ex MOP p ob add em ecnume Howeve ng comp ex h MOP gh compu p em ona Howeve equ emen he h gh and compu aequ ona equ emen oan uA on howeve hey have oawback u on no howeve advoca hey o have ahe ge no d advoca bu on y o em aoptimal ge L kew o d u eand bu aM howeve ma y oen em hey L have kew no ehhgo aMH advoca ma ed orequire aneed d bu on e PT ca ego PT ca yMe unde ego y unde he MOP he MOP amewo k aop MODGP k alocal MODGP MOVPQ MOVPQ MOCRU PT ca MOCRU ego y unde MONTR and he MONTR eetc.) pec amewo ve enumber pec y ve aaand yoog MODGP MOVPQ MOCRU and ong u on howeve hey have no advoca ed orequire aof ge d bu y em L kew esimple aM ma o 5daac R vnner dBu kat Con bu 5p 12ego on R vomethods. w dBu Wo kBL Con bu on 5d 12ob R vaBL w dng Wo kLnd Con bu on solutions solutions for distribution large distribution systems. systems. PT ca y unde he MOP kRm awo MODGP MOVPQ and MONTR eaed pec ve 5d 12Me R5 vM w dBu Wo kmethods. Con bu on solutions for solutions large distribution for large distribution systems. systems. Among Among nume nume ca hod ca me OPF hod common OPF common y exp o y ed exp acommon o an ed nne aon an op nne m za op on m aed go on oon ond Among ca me OPF exp ahave an nne op m on aand go oGA Among numer methods OPF snearly common y exp ca ohod methods ted as scompu opt m zat on y exp ave gor o ted thm as an nner opt m zat on aed gor th possibility possibility of premature of premature convergence convergence towards possibility towards optimal local premature solutions solutions convergence need to need be towards addressed to be addressed local with optimal with solutions be addressed possibility of premature convergence local optimal solutions need to be addressed with s common y pexp oconsidered, ted as opt m zat aon gor to orequired MCS be d nne op oLENT m MCS za on be ng he nne h gh op za awback on on me o he MCS h ny gh compu be ng on nne on op NLP me m za n compa on he on gh NLP compu acan me compa on M N awback o MCS be ng nne op m za on he h gh aed on me compa on NLP Artificial Artificial intelligence intelligence (AI)-based (AI)-based metaheuristics metaheuristics (for example (for example ACS, ACS, ABC, ABC, etc.) and and hybrid hybrid Artificial intelligence (AI)-based metaheuristics (for example ACS, ABC, etc.) and hybrid agh eeisw h gh phe ec on ocha cBu ehe en me compu aP on and aand ope ac aMOCRU on ahod econstrained h ua gh on p ec on cM eyhey cRhm en me compu ato on and unce 5o 21awback m nWo oon Mu ob 5 A vefficient P ann m n othm M Mu hod ob vamewo ann 5 ng A M hod m n ocompa Mu ob P ann ng M hod ac eca hbu p ec ocha ac away eon cAmong h elarge gh co p en ec me on compu ocha ach cewed on eLENT and cke en unce me aOPF compu n ope ahough ago on on and ua unce on ave aMOP on ua 5 A m n op Mu ob vamewo ann ng M hod more efficient more The simple The simple MH methods MH methods require require more computation more more computation efficient time. methods. Also, time. more Also, The number more number methods more computation D DPL R m D GA LENT DPL DM E GA D more efficient methods. The simple MH methods more computation time. Also, more Bu D DPL R D nd LENT m DPL E GA R nd m DM E GA DM M Lan H u ACO W BL M mp Lp BL Bu M BL mp H M Lunce mp u ACO H W hv H uope ACO u W ACO h W hon m nd m Bu BL M m PEM m AMBA SQP and SQP dynam and dynam cBu og cPEM amm og ng amm ag eP ng he mo etowards he SQP eaCon mo cand en dynam nume cch en nume cAMBA ca p og me ca amm me ng hod eEawo he mo eocha cemen en nume ca me hod m PEM Bu m PEM M AMBA m PEM AMBA R nd m AMBA SQP and dynam cng pbu og amm ng amp ecompu he mo ecommon ceona en nume ca me hod Howeve he pBu ob em oBL Howeve mu a on ve p y ob em g d o and mu ew a Howeve on va a ve on y he can g p d be ob nco em ew po o va mu a ed a on Dynam can c be y nco g d po and a ed ew Dynam va a on c be nco po a ed Dynam c of functions is required to achieve nearly high of functions quality results. is required Also, to constrained achieve nearly and unconstrained high quality results. Also, constrained and unconstra of functions to of achieve functions nearly is required high quality to achieve results. nearly Also, high constrained quality results. and unconstrained Also, and unconstrained Me aheu Me aheu c MH c me MH hod me hod ke evo u evo ona Me u y aheu a go y a hm c MH have hm me been have hod con been ke de con evo ed de a u a ed ona na a y u a a a na go u a have been con de ed a a na Me aheu c MH me hod ke evo u ona y a go hm have been con de ed a na u a 5 Con bu on he Rev ew Wo k 5 bu on o he Rev ew Wo k 5 Con on o Rev 5 Con ew Wo bu k on o he Rev ew Wo k p og amm ng DynP and exhau p og ve amm ea ch ng ES DynP a and exhau hey p om ve ea e o ch nd ES a a g hough oba op mum p om e o nd a g oba op mum The con bu on o ev ewed The con wo k a e on hown o ev ono og wo ca k y The a e n con Tab hown e bu ch 5–8 on ono add o og e ca ev ng ewed y each n Tab e k a 5–8 e add hown e ch ng ono each og ca y n Tab e 5–8 add e ng poog amm ng DynP and exhau ve ea ch ES a hough hey p om e o nd a oba op mum multi-objective problems are difficult to optimize. Similarly, GA and PSO provide near optimal The con bu on o ev ewed wo k a e hown ono og ca y n Tab e 5–8 add e ng each multi-objective problems are difficult multi-objective to optimize. problems Similarly, are difficult GA and to PSO optimize. provide Similarly, near optimal GA and PSO provide near optimal way way add o e add ng e comp ng comp ex MOP ex p MOP ob em p ob way Howeve em o add Howeve e he ng h gh he comp compu h gh ex compu MOP a ona p ob equ ona em emen equ Howeve and he and h gh compu a ona equ emen way o add e ng comp ex MOP p ob em Howeve he h gh compu a ona equ emen and uca on hey have no advoca oza uame on ed howeve o aDPL ge hey d have bu on no advoca em L ed kew orequire eGA aDM age dng bu on em L ACO kew anum m PT ca egonume yema unde he MOP PT amewo ego y unde k aybe MODGP MOP MOVPQ amewo PT MOCRU k ca MODGP ego and yme unde MONTR MOVPQ he MOP e(for MOCRU pec amewo ve yMH and kEbe MONTR aGA MODGP enumber pec MOVPQ ve yuoon MOCRU MONTR pec v u on howeve hey have ed ocMu a5SQP d bu on em L kew eop au ma ocowa 5more 1conve R vamm 5d w 1yvac R dand Wo w khoweve Con dBu Wo kBL bu Con bu on 5SQP 1nne R vBL w dche Wo knd Con bu on solutions solutions for large large distribution distribution systems. systems. solutions for large distribution systems. 1ge R w dfor kmethods. Con bu on solutions for large distribution Among nume ca me hod OPF common Among y exp o nume ed aop ca an me nne hod op m za on common aoMOP hm y exp oann oma ed aca an nne op m on ae ecan go h Among ca me hod OPF common y exp o ed op za on adeal hm o po b o p ema u ey conve owa d oca op ma op u on need o be add ego ed w h poThe b yadvoca o ppm u eac po gence owa yintelligence o d p ema oca u op egence conve ma ova gence need d o oca be add ma ed o w u on need be add eNLP ed w hhybrid d awback o MCS be ng nne awback op m o za MCS on be ng he nne h gh m aona on me n hmore compa gh compu awback on aM M o on NLP MCS me n compa ng nne on op M m NLP za he hyand gh compu aed on me awback ovbmethods. MCS ng nne op m za on he h compu ahave on me n compa on M Artificial Artificial intelligence intelligence (AI)-based (AI)-based metaheuristics metaheuristics (for example (for example ACS, ACS, ABC, ABC, and etc.) hybrid and hybrid intelligence (AI)-based (AI)-based metaheuristics metaheuristics (for example ACS, example ABC, ACS, etc.) ABC, and hybrid etc.) and aefficient eArtificial ac h gh am p eon hec gh on p ec ocha on csystems. ocha eke cp cen eu me cas en compu me compu apo on and ae2aog on unce and ah unce n ope aetc.) aem on ope aona ua on on ua on 5no 2og A n oBu ob vefficient P ann M 5p 2he hod A m n oMH Mu ob vm P ann 5d ng A hod m n oeEnco Mu ob P M hod amp eWo h gh p ec on ocha can ehigh en compu ago on and unce n ope amore on ua on modified LF methods and software-based LF platforms provide better performance aiming 5d 2Me A oArtificial Mu ob ven P ng M hod more simple The simple MH methods more methods require efficient require more methods. more The computation simple time. time. methods more Also, number number more computation time. Also, more Bu M D LENT DPL R m DM more efficient methods. The simple MH methods require computation time. Also, more Bu D LENT DPL BL Bu M Rann D BL nd M LENT m DPL D LENT E GA R nd m DM R nd m M mp L H Bu u ACO BL h mp L HDM u za W Bu BL M L mp H Bu L M ugh ACO mp H W L hW uGA ACO H W h ACO W hva SQP and dynam p og SQP an dynam ecBu he mo cThe og eon amm cand nume aastochast eog he ca mo and hod dynam en nume ccomputation ca amm me ng hod aOPF he mo eg cvemen en nume me hod and dynam cng og amm ng ang ecompu he mo eo ceand en nume ca me hod optimization optimization methods methods can be can consider be consider as promising as promising ways ways to deal to with with complex complex MOP MOP problems problems more optimization methods can be consider promising ways to deal with complex problems more Howeve he p ob em oBL mu ang ve g d ew aon on can be nco po Howeve aM ed Dynam he p ob cAlso, o mu aEmore on ve y g d and ew aAlso, on be Howeve he ob em o Howeve mu a on he ve p y ob g em d and o mu ew a va on a ve on y can g d be nco ew va a ed Dynam on can c be po a ed Dynam c of functions of functions is required is required to achieve to achieve nearly nearly high quality high quality results. results. Also, of functions constrained Also, constrained is required and unconstrained and to achieve unconstrained nearly high quality results. constra of functions is required to achieve nearly quality results. Also, constrained and unconstrained aheu Me aheu MH c me MH hod me hod evo ke evo Me u y ona aheu a go y a hm go c MH have hm me been have hod con been ke de con evo ed de a u a ed na a y u a a a na go u hm a have been con de a a na Me aheu c MH me hod ke evo u ona y a go hm been con de ed a na u a 5 Con bu on o he Rev ew Wo k 5 Con bu on o he Rev ew Wo k 5 Con bu on o he Rev 5 Con ew Wo bu k on o he Rev ew Wo k p amm ng DynP and p exhau og amm ve ng ea DynP ch ES and a exhau hough hey ve amm ea p om ch ng ES e DynP o a nd hough and a g exhau oba hey op p om ve mum ea e o ch nd ES a a hough oba op hey mum p om e o nd a g oba op mum The con bu on o ev ewed The con wo k bu a e on hown o ev ch ewed ono og wo ca k y The a e n con Tab hown e bu ch 5–8 on ono add o og e ca ev ng ewed y each n Tab wo e k a 5–8 e add hown e ch ng ono each og ca y n Tab e 5–8 add e ng multi-objective problems are difficult to optimize. multi-objective Similarly, problems GA and are PSO difficult provide to optimize. near optimal Similarly, GA and PSO provide near opt The con bu on o ev ewed wo k a e hown ch ono og ca y n Tab e 5–8 add e ng each multi-objective problems multi-objective are difficult to problems optimize. are Similarly, difficult GA to optimize. and PSO provide Similarly, near GA optimal and PSO provide near optimal fac tate h gh prec s on stochast c fac eff tate c ent h t gh me prec computat s on on and c uncerta eff c ent n operat t me computat on s tuat on ons and uncerta n operat on s tuat way way add o e add ng e comp ng comp ex MOP ex p MOP ob em p ob way Howeve em add Howeve e he ng h gh he comp compu h gh ex compu MOP a ona p a ob equ ona em emen equ Howeve and he and h gh compu a ona equ emen c ent t me computat on and uncerta n operat on s tuat ons way o add e ng comp ex MOP p ob em Howeve he h gh compu a ona equ emen and u on howeve hey have no o u advoca on howeve ed o a hey ge have d bu no on advoca y em ed o L kew a ge e d a ma bu o on y em L kew e a ma o PT yded unde MOP PT ca amewo ego yMu unde k aycommon he MOP MOVPQ PT MOCRU k aca MODGP ego and y unde MONTR MOVPQ eaua MOCRU pec amewo ve and kEpza MONTR ao MODGP emo pec MOVPQ ymo MOCRU and MONTR pec u on howeve hey no advoca okc5po aPT ge d bu on L kew eop ma ocommon vego w Wo Con bu 5 12y on R veme w den Wo kMODGP Con bu on 5d 1ng R vBL w dhe Wo kEed Con on solutions for large distribution systems. ca ego y unde he MOP amewo kRaand ahod MODGP MOVPQ MOCRU and MONTR pec ve solutions for large distribution solutions for large distribution systems. Among nume hod Among OPF nume ca me yamm exp hod o OPF acommon an Among nne op nume y m exp za o ca on ed me aBL go hod an hm nne OPF oA op m common za on y aed go exp o hm ed ond an nne op on aeand hDv Among nume ca me hod y exp o au an nne op m on aadd go hm ove bme o b p y ema o p u ema eThe conve u eng conve gence gence po d owa oca bequ dhod op oca ma pen op ema oeew ma eon o conve need on gence oMOP be owa add be epo d ed oca w ed hann w ma h u on need be ego ed po b y o p u ehod conve gence owa d oca op o uto on o be add eua w h d awback o MCS be m za awback on oew MCS h gh be compu ng nne aM on op me za n compa he on h gh compu NLP ahave on me nmocompa on do awback o MCS be ng have nne m za on h gh compu arequired on me n compa on M NLP Artificial intelligence intelligence (AI)-based (AI)-based metaheuristics metaheuristics Artificial (for intelligence example (for example (AI)-based ACS, ABC, ABC, etc.) metaheuristics etc.) and and hybrid hybrid (for ABC, etc.) hy Artificial intelligence (AI)-based metaheuristics (for example ACS, ABC, etc.) and hybrid ac aepo esystems. h gh ec on ocha cperspective. eOPF cowa en ac me aquality compu ehod gh aequ p on and on unce ocha aneed nnco ope eke aLENT co en on me compu on ang on and unce aDM ope aadd on ua 2ca A 5SQP m A n oon m n ofuture ob Mu vto ob P ann ng Ped ann hod M 5SQP 2u A m n om Mu ob P M hod ac eca h gh p ec ocha cMH cve me compu awo on ahod n ope aec on on 5SQP 2Artificial A m oema Mu ven P ng M hod mo eand cWo hod mp enne me eresults. mo ema compu aACS, on me mo eeed numbe Bu he BL M D LENT DPL nd m Bu GA DM D DPL R m Eza GA mo e5op e1aog cR en me hod The mp mo een eand me hod me equ eco mo eRvamewo mp eoptimize. MH aunce me on me A o mo mo eMOP ebu compu acneed on A onear eM Bu BL M D LENT DPL BL M D nd Bu LENT m DPL M E GA D R nd LENT m DPL DM GA R nd DM EaAlso, SQP and dynam phe og amm ng an dynam eBu he mo caem p eem og nume ng apromising ehybrid ca he mo eh cand nume and me dynam hod cpo og amm ng aaop eynumbe he ecompu cACS, en ca and pbu og amm ng acan ecompu he eo cona en nume ca me hod optimization optimization methods methods can be can consider be consider as as promising ways to ways deal to with deal complex with complex problems MOP problems more more optimization optimization methods can methods be consider be as consider promising as ways promising to deal ways with to complex deal with MOP complex problems MOP more problems more Howeve Howeve he pp ob he em p ob o mu em aperspective. on mu aann ve y g ve d y and g d va ew ahm on va can ame can be nco ame ed Dynam ed Dynam cand coptimize. Howeve p ob o mu ato on ve yego g d va au on can be nco po ed Dynam coGA of functions of functions is is to achieve achieve nearly nearly high of functions quality high quality is required results. Also, constrained Also, achieve constrained nearly and unconstrained high and quality unconstrained results. constrained and unconstra of functions is required to achieve nearly results. Also, constrained and Me aheu c5 MH me Me hod aheu ke evo u MH ona me y aMH hod go hm ke evo have u been aheu con ao go ced de MH hm ed aca have aon hod na been u ak con evo de u ona avemen y ahm na go u hm aexample been con de ed aM na Con bu on o he Rev ew Wo k 5 Con bu on o Rev ew Wo Me ccrequired MH me hod ke evo u ona yeand aono go have been con de ed am anne na u ama efficiently efficiently from from acon future adistribution perspective. The The hybrid metaheuristic metaheuristic techniques techniques show show robustness robustness (high (high 5 bu on ounde heon Rev 5 Con ew bu k o he Rev ew Wo k efficiently from future The hybrid metaheuristic techniques show robustness (high pamm amm ng DynP exhau ea ch ES aThe hough hey p om oem nd aog p g og oba amm op ng mum DynP and exhau ve ea ch ES aem hough pme om eng o con The bu o on ev ewed o ev ewed kon aawback emo hown kame The aon eed ch hown bu ch og on ono ca y o og n ca ev Tab y eve 5–8 Tab eadd k a5–8 eunconstrained add hown ng each eand ch ng each og yyon n Tab eMONTR 5–8 add eehod p og ng DynP and p exhau og amm ve ng ea DynP ch ES and aob hough exhau hey ve p ea om ch eMe ES o acon nd hough aand g oba hey op pnumbe om mum esystems. o nd ao g oba op mum multi-objective multi-objective problems are difficult are difficult to optimize. Similarly, Similarly, GA multi-objective and PSO and provide PSO problems provide near are optimal difficult optimal to Similarly, GA and PSO The con bu on o ewed wo k ahigh eem hown ono ca yex n Tab ey 5–8 add ehod ng each multi-objective problems are difficult to optimize. Similarly, GA PSO provide near optimal o way add o edynam add ng eon comp ng comp ex MOP ex p MOP ob em way Howeve add Howeve ey he ng h gh he comp compu h gh compu MOP aewed ona p awo ob equ ona em emen equ he and hza gh anume equ way o add ng comp ex MOP p ob Howeve he hhe gh compu an ona equ emen o2vuca on howeve hey have u on no ed hey o have aMOVPQ ge no o u advoca on bu howeve y o aop hey L kew have d eand bu no abe ma o em ed o L kew aamo ge ethey d aM bu on kew eadd aM ma oop ego ygh he MOP ca ego y unde k aproblems MODGP he MOP MOCRU k ca MODGP y MONTR MOVPQ MOP eGA MOCRU pec amewo and k MONTR aABC, MODGP enumbe pec MOVPQ ve y MOCRU and eaemen pec v 5way 1ca R vR w daheu Wo khoweve Con bu on 5PT 12ACS, R vd w dch Wo kge Con on solutions for large distribution systems. solutions for large distribution ca ego yme unde he MOP amewo kp awo MODGP MOVPQ MOCRU and MONTR pec ve yono 5solutions 1Con R wafor Wo kload Con bu 5PT 1The vawback dP Wo kme Con bu on large distribution solutions systems. for large systems. me hod common Among yamm exp o nume ed aob ca an me nne hod op OPF m za on common Among abu go hm yadvoca nume exp ca ed ame an OPF op m common on aunce y go exp hm ed ohey apo an nne at uncertainty of and in comparison to traditional models. Also, provide Among nume ca me OPF common exp au an nne op za on aeeHoweve go ohybrid po b po o b p y ema op p ema ecM conve u eev conve gence gence owa po d owa oca b op oca ma p ema o ma u eon o conve u need on gence need oMOP be add be eon d add oca w op ed h w ma h oca u need oMOP be ed po b y o p ema u ebe conve gence owa dmetaheuristics oca op o u on need o be add eed ed w h d o MCS be ng op d m za on o MCS he be h ng compu nne op aLF on m za me on n compa he h gh on compu NLP adeal on me n compa on NLP d awback oand MCScfew ngvar nne op m za on he h gh on me n compa on M NLP Artificial (AI)-based metaheuristics (for example ABC, etc.) and hybrid Artificial intelligence (AI)-based Artificial metaheuristics (for (AI)-based example ABC, etc.) (for example hybrid ACS, etc.) ac edcan h p on ac aamm cegeneration, h eadvoca gh cu p en ec me on compu ocha aperspective. cas on eem ac and ccan en aech unce eways h me gh aunde n p ope aon ancorporated on on on ocha and ua on em ahe cowa en ope aeme on compu ua aacan on on and aoLona nnco ope ao ua 5PT m nnume oec Mu ob 5PT 2y A vamewo ann m n ng ointelligence Mu hod ob vamewo P ann 5Howeve ng A M hod m n oto Mu ob von P ann M ac aocha h gh ec on ocha cewed eke cevo en me compu aec on and unce n ope aA ua on mo emu eon mo cWo een ew cOPF en hod The hod mp The MH mp me ed MH hod mo me eoptimize. hod equ cy en enco equ mo me eama egh hod mo The compu aACS, on mp aunce me eacnd on MH A me me o mo hod o numbe equ eand numbe eng mo eo compu ancorporated on me A mo eDyn num mo ecompu eREG cbu en me hod The mp edifficult MH me hod equ ecompu mo eon compu aned on me A o mo eat SQP and dynam cd p og ng apromising ehybrid SQP he mo eona dynam cehybrid en ccompu nume p og amm ca me hod eo mo cand en nume ca me hod SQP and pbe og amm ng a on eAAmong he mo ebu en nume ca me hod optimization optimization methods methods can can consider be consider as as optimization promising ways to methods deal to with deal can with complex be consider complex as MOP problems problems more ways more to with complex problems m optimization methods can be consider promising ways deal with complex MOP problems more Howeve he peaintelligence ob em o mu ahod on ve y g d and ew he va p ob ahm em o can mu be ang nco on po ve apromising y ed ghod Dynam d cand ew va agh on can be aon ed Howeve he pequ em ohave ano on ve y g and ew va on be po ed Dynam cned o unc on equ ed ofrom ach eve nea yaaob h gh qua y eoptimization u A o con and uncon ave ned oo unc ed o1comp ach eve o nea unc yfrom on h gh qua equ ed y eare oenne u ach A eve o nea con yy h an gh ned qua and y uncon eand ueu aca A o con ang ned and uncon ned the prob em formu at on However s very r the g d prob and em few formu var at at ons on can s very be r g d and few Dynam var c ons be Dyn very r gHowever d dynam at ons be ncorporated Dynam c Me aheu cc5 MH me hod Me aheu ke evo c u MH ona me y a hod go hm ke have evo u been con y go de ed hm Me a have a aheu na u been a c con MH de me ed hod a a na ke u evo a u ona y a go hm have been Con bu on o he Rev ew Wo k Me aheu c MH me hod u ona y a go have been con de ed a a na u a efficiently efficiently from a future a future perspective. perspective. The The hybrid metaheuristic metaheuristic techniques techniques show show robustness robustness (high (high 5 Con bu on oob he Rev 5 Con ew k 5 on Con o he bu Rev on ew he Wo Rev k ew Wo k efficiently from efficiently a future perspective. a future The hybrid The metaheuristic metaheuristic techniques techniques show robustness show robustness (high (high p og amm p og ng DynP ng and DynP exhau and exhau ve ea ch ve ES ea a hough ES a hough hey p om hey e p om o e a o g oba nd a op g oba mum op mum The con on o ev ewed The con wo k bu a e on hown o ev ch ono og wo ca k y a The e Tab con hown e bu ch 5–8 ono add o og e ev ng ewed y each n Tab wo e k 5–8 a e add hown e ch ono each og ca y n Tab e 5–8 add e ng p og amm ng DynP and exhau ve ea ch ES a hough hey p om e o nd a g oba op mum multi-objective multi-objective problems problems are difficult to optimize. multi-objective to Similarly, Similarly, problems GA and GA are PSO and difficult provide PSO to provide optimize. near optimal near Similarly, optimal GA and PSO provide near opt multi-objective problems are difficult to optimize. Similarly, GA and PSO provide near optimal way o add e ng way ex MOP o add p e ob ng em comp Howeve ex MOP he p h ob way gh em compu o add Howeve a ona ng he comp equ h gh emen ex compu MOP and p a ob ona em equ Howeve emen he and h compu a ona equ emen way o add e ng comp ex MOP p em Howeve he h gh compu a ona equ emen quality quality solutions), solutions), promises promises powerful powerful global global optimization optimization methods methods and have and have ability ability to address to address quality solutions), promises powerful global methods and have ability to address o u on howeve hey have no advoca ed o a ge d bu on y em L kew o on e a howeve ma o hey have no advoca ed o a ge d bu on y PT ca ego PT ca y unde ego y unde he MOP he MOP amewo amewo k MODGP PT k a ca MODGP ego MOVPQ y unde MOVPQ MOCRU MOP MOCRU and amewo MONTR and k MONTR a MODGP e pec e pec MOVPQ y ve y MOCRU and MONTR e pec ve u on howeve hey o u on advoca howeve ed hey o have a ge no d advoca bu on ed o em a ge L kew d e bu a on ma o y em L kew e a ma o 5 R v w d Wo k Con bu on 5 1 R v w d Wo k Con bu on solutions solutions for large for distribution large distribution systems. systems. solutions for large distribution systems. PT ca ego y unde he MOP amewo k a MODGP MOVPQ MOCRU and MONTR e pec ve y 5 1 R v w d Wo k Con bu 5 1 on R v w d Wo k Con bu on solutions for large distribution systems. Among Among nume nume ca me hod ca me OPF hod common OPF common y exp o y ed exp ahod o an nne Among aaca an op nne nume m za op on ca aMOP me za go hod on hm go OPF oon hm common occompu ycompu exp o ed aetc.) an nne op Among nume ca me hod OPF common ycompu exp o ed au an nne op m za on aeocha go hm ocNLP po b po yawback o bcey p yn ema op p up ema eon conve u eng conve gence gence owa po d owa oca b d op oca o ma p op ema o ma u eon o conve u need on gence need oMOP be owa add om be epo d add ed oca w op ed h w ma h o ahod u on need o on be add eNLP ed po b o p ema u ebe conve gence owa d oca op ma oed u on need oope be add ed w h dac awback ointelligence MCS be ng d nne op m o za MCS be he nne h gh compu op d awback m za a on o MCS me he be h compa gh ng compu nne on op a M on m NLP za me on n compa he h gh on compu M a on me n compa on M Artificial intelligence (AI)-based metaheuristics Artificial (for intelligence example ACS, (AI)-based ABC, etc.) metaheuristics and hybrid (for example ACS, ABC, and hy Artificial (AI)-based Artificial metaheuristics intelligence (AI)-based (for example metaheuristics ACS, ABC, (for etc.) example and hybrid ACS, ABC, etc.) and hybrid a e h gh p ec on ocha c e c en ac me a compu e h gh a p ec on and on unce ocha a n c ope e ac c a en on e me h ua gh compu on p ec a on on and unce c e a n c ope en me on ua on a and unce 5 2 A m n o Mu ob 5 2 A v P ann m n ng o M Mu hod ob v P ann 5 2 ng A M hod m n o Mu ob v P ann ng M hod ac a h gh ec on ocha c e c en me a on and unce a n a on ua 5 2 A m o Mu ob v P ann ng M hod mo e e mo c e en me c en hod me The hod mp The e MH mp me e MH hod mo me e hod equ e c en e equ mo me e e compu mo e The compu on mp a me e on MH A me me o mo A hod o numbe mo equ e numbe e mo e a on me A o mo e num mo e e en me hod The mp e MH me hod equ e mo e compu a on me A o mo e numbe SQP and dynam c p og amm ng SQP a e and he mo dynam e c en og nume amm ng me a e hod he mo e c en nume ca me SQP Me andaheu dynam cc pMH og amm ng a e he mo e c en nume ca me hod optimization methods can consider as promising ways to deal with complex problems more optimization methods can be consider optimization as promising methods ways can to be deal consider with complex as promising MOP ways problems to deal more with complex problems more Howeve he p ob em o Howeve mu a on he ve y ob g em d and o mu ew a on va a Howeve ve on y can g d be he and p nco ob ew po em va a ed o a mu Dynam on can on c be ve nco y g d a and ed Dynam ew va a on can be nco po a ed Dyn Howeve he p ob em o mu a on ve y g d and ew va a on can be nco po a ed Dynam o unc o on unc equ ed equ o ach ed eve o ach nea eve y nea h gh o y qua unc h gh on y qua e u y equ e A u ed o con A ach o a con eve ned nea a and ned y uncon h and gh qua uncon a ned y e a u ned A o con a ned and uncon a o unc on equ ed o ach eve nea y h gh qua y e u A o con a ned and uncon a ned Me aheu c MH me hod ke evo Me u aheu ona y a c go MH hm me have hod been ke con evo u de ona ed y a a go na u hm a have been con de ed a a na me hod ke evo u ona y a go hm have been con de ed a na u a 5way Con bu on o he Rev ew Wo k 5 Con bu on o he Rev ew Wo k efficiently efficiently from from a future a future perspective. perspective. The hybrid The efficiently hybrid metaheuristic from metaheuristic a future techniques perspective. techniques show show The robustness hybrid robustness (high metaheuristic (high techniques show robustness ( 5 Con bu on o he Rev 5 Con ew Wo bu k on o he Rev 5 Con ew Wo bu k on o he Rev ew Wo k efficiently from a future perspective. The hybrid metaheuristic techniques show robustness (high p og amm ng DynP and exhau ve ea ch p ES og amm a hough ng DynP hey p and om exhau e o nd ve a g ea oba ch op ES mum a hough hey p om e o nd a g oba op m The con bu on o ev ewed wo k a e hown The ch con ono bu og on ca y o n ev Tab ewed e 5–8 wo add k a e hown ng each ch ono og ca y n Tab e 5–8 add e ng p og amm ng DynP and exhau ve ea ch ES a hough hey p om e o nd a g oba op mum mu ob ec ve p ob em a e d cu o op m ze S m a y GA and PSO p ov de nea op ma The con bu on o ev ewed The con wo bu e on hown o ev ch ewed ono og wo ca k a y e n Tab hown e ch 5–8 ono add og e ca ng y n each Tab e 5–8 add e ng each mu ob ec ve p ob em a e d mu cu ob o ec op ve m p ze ob S em m a a e y d GA cu and PSO op p m ov ze de S m nea a op y GA ma and PSO p ov de nea op ma generation of various scenarios under all load models, which were limited in conventional methods. o add e ng comp way ex MOP o add p ob e em ng comp Howeve ex MOP he p h gh ob em compu Howeve a ona way equ he h o emen gh add compu e and ng a comp ona equ ex MOP emen p ob and em Howeve he h gh compu a way o add e ng comp ex MOP p ob em Howeve he h gh compu a ona equ emen and quality quality solutions), solutions), promises promises powerful powerful global global optimization optimization methods methods and have and ability have ability to address to address quality solutions), quality solutions), promises powerful promises global powerful optimization global optimization methods and methods have ability and have to address ability to address o u on howeve u on howeve hey have hey no have advoca no advoca ed o ed a ge o d a ge bu d on bu y on em y L em kew L e kew a ma e o a ma o PT ca ego y unde he MOP PT ca amewo ego y unde k a MODGP he MOP MOVPQ amewo MOCRU PT k a ca MODGP ego and y unde MONTR MOVPQ he MOP e MOCRU pec amewo ve y and k MONTR a MODGP e pec MOVPQ ve y MOCRU and MONTR e pec v o u on howeve hey have no advoca ed o a ge d bu on y em L kew e a ma o 5 1 R v w d Wo k Con bu on 5 1 R v w d Wo k Con bu on solutions solutions for large for distribution large distribution systems. systems. solutions for large distribution systems. 5 1 R v w d Wo k Con bu 5 1 on R v w d Wo Con bu on solutions for large distribution systems. Among nume ca me hod Among OPF nume common ca me y exp hod o OPF ed a an common Among nne op nume y exp m za o ca on ed me a a go hod an hm nne OPF o op m common za on a y go exp o hm ed o a an nne op m za on a go h po2prom bo y oexhaust p u eaeo po conve b gence yanequ o p owa ema d u ethough oca conve op gence ma po o owa uqua b on d y need oca p op o ema be ma add o conve eon u ed on gence w need h owa oon be d add oca ed w ma h o with on need og be add eACS, ed po b ohe p ema u ebe gence d oca op o u need o be add eua ed w hcto constrained constrained real time time planning planning problems. problems. Furthermore, Furthermore, decision-making methods methods need also need to be to be real time planning problems. Furthermore, decision-making methods also need to be awback o MCS be ng nne op m za he h gh compu ame on me n compa d awback on o NLP MCS be ng nne op m za on he hDynam gh compu aproblems on doptimization awback MCS be ng d nne awback op m ome za MCS on be ng he h gh op m za ave on on me he n h compa gh compu on aeon M on NLP me nnco compa M NLP Artificial Artificial intelligence intelligence (AI)-based (AI)-based metaheuristics (for example (for example Artificial ACS, ABC, intelligence ABC, etc.) etc.) (AI)-based and metaheuristics example intelligence (AI)-based metaheuristics (for example ABC, etc.) and hybrid ac eArtificial ac h gh aamm p eThe h ec gh on p ec ocha on cperspective. ocha ehe cSQP cen eke cgh en compu me aon on and ac au unce aca eand aon gh n p ope ec aem ny on ope ocha aon on on caop ua on cdeal en me compu ava on and unce 5d Ake m oema Mu ob 5SQP A P m n ng oyon M Mu hod vdeal ann 2ng A M hod m n oto Mu ob vg P ann ng M hod ac aamm h gh p ec on ocha cnume emetaheuristics cowa en me on and unce ope aA ua on 5p 2The A m oon Mu ob ven P ann ng M hod ebe mo c2ec esearch en cbu en hod me The hod mp The MH mp me ekP MH hod mo me eop hod equ ech cqua en eze equ mo me eama eato hod mo eaca The compu on mp aACS, me eap on MH A me me oalso mo hod o eon numbe mo equ eon numbe ehybrid mo eog compu an on me A o eop num mo ea econstrained cey en me hod The mp MH me hod equ ecompu mo eog compu aeACS, me A o mo eop programm ng me (DynP) and programm ve ng (ES) (DynP) aMH and they exhaust prom ve se ftechniques nd (ES) aop aunce oba though opt they mum prom se to fand nd aede oba earch (ES) ae though they se to fcan nd g oba SQP and dynam cmo py og amm ng and dynam ereal he mo p eeconve og cob amm ng apromising eo ca and he me mo dynam eona ccedecision-making en og nume amm ng me ahM eePSO hod he mo ede cde en nume ca hod optimization methods can consider as optimization ways to methods deal with can complex consider MOP as promising more ways complex MOP m methods optimization consider as methods promising can be ways consider to as with complex ways MOP problems deal more complex MOP problems more Howeve p ob em mu avMe on Howeve g danne and ew p va a5p em on o can mu be ang nco po ve Howeve y g Dynam d and he ew cob va aand mu can be ve nco y po gu ao dyon ed cov on be Howeve p ob em o mu aecompu on y g d ew va aem on can ed Dynam o unc o on unc ed equ ach ed eve ode ach nea eve yaMOCRU nea h gh o y unc h gh on y ey u y equ ecompu A ed o con A ach o aTab con eve ned nea aL and uncon h and gh qua uncon ave ned eey aono u ned A con a(for ned and uncon am o unc on equ ed oon ach eve y h qua ep u A o con abeen ned and uncon ay ned aheu ccoamum me hod Me evo aheu uhod yL MH go hm hod have ke con u de y ed aohybrid go hm na u have ame been con ed ade amo ume ahm aheu c comp MH hod unnume ona aca go hm have been anea na u acommon efficiently aann future perspective. The hybrid metaheuristic show robustness (high efficiently ahe efficiently from hybrid metaheuristic future The hybrid show metaheuristic robustness (high techniques show robustness (high p og amm ng DynP and p exhau og ve ng ea DynP ch ES and age exhau hough hey ve og p om ES o DynP nd hough adecision-making g and oba hey exhau p om mum eMONTR ve oaned ea nd ch aon ES g oba anumbe hough op hey p om oem nd aaequ g oba m o ev ewed wo aob epromising hown The ch con ono bu og on ca y owith n ev Tab ewed ebe 5–8 wo add k aona eo hown ng each ch ca Tab 5–8 add eopt ng og ng DynP and exhau ve ea ch ES acompu hough hey p om o nd aproblems g oba mum mu ob mu ob ve ec ob ve em p ob em d abu eed cu d o op mu o m ze ob m Sand ec m ve Ssearch p y m ob GA au and GA aebe PSO d p PSO cu ov de p o ov op nea op nea ze ma op m ahe ma y GA and PSO p nea op The con bu on o ev ewed The con wo ko aopt eCon on hown oanne ev ch ewed ono og wo ca k aob y eea n Tab hown eneed ch 5–8 ono add emethods y n each eevo 5–8 add eadd ng each mu ob ec ve p ob em eon d cu op m ze SMOVPQ m y GA p ov nea op ma way o add ekve ng comp ex MOP p way o add Howeve ng comp he hed gh ex MOP compu p aability ob ona em equ Howeve emen and h gh compu aew ona emen way Me o add ng ex ob em he h compu au ona equ emen quality quality solutions), solutions), promises promises powerful powerful quality global optimization optimization solutions), methods promises and powerful have and have global ability to optimization address to address methods and have to add quality solutions), promises powerful global optimization methods and have ability to address o u on howeve hey have no advoca oza u o aequ howeve ge d bu on have y em no kew ed eetc.) aM ma aS o ge d bu yeand Lability ecan ego yfrom unde he MOP k MODGP ca ego MOVPQ yexp unde MOCRU he MOP and MONTR k apo MODGP pec MOVPQ y MONTR ena pec v oMOP uca on hey have no ed o avamewo d bu on yamm em kew ey ma o 5 12perspective. R vamm w dem Wo km Con bu aadvoca d bu y em ego ydeevo unde he MOP ca amewo y unde ky abu he MOP MOVPQ amewo ktechniques aweights, MODGP and MONTR eme pec y and em pec ve 5PT 1u R v pedynam w kfuture Con bu 5constrained 1on on R vcon 5constrained dhe 1p vaiming w dcon Wo bu kequ Con bu on o on ofrom acWo ge dp bu y o u on aplanning ge d on yacu em Among ca hod Among OPF nume common ca me exp hod o OPF ed aed an Among nne op nume yaecompu exp m za o ca on ed me aand ang go hod an hm OPF oadvoca op m common za on y aNLP go exp o hm ed occompu aMOCRU an nne op m za on aon go Among me hod OPF common o ed ahey an nne op m za on amu go omum po b yhdynam ogh pHoweve ema u eme po conve b gence o p d u eeve conve op ma gence o owa u on oca o op po ma add b oand u yabe ed o on w need ema h o eproblems conve egence ed w owa h dto oca op ma okew u po b y o p ebe conve gence owa dem op ma o u on need oMOP be add eneed ed w h constrained constrained real time real time planning problems. problems. Furthermore, Furthermore, decision-making methods methods also need also to need be to be real time planning real problems. planning Furthermore, problems. Furthermore, decision-making decision-making methods also methods to also be need be d awback d awback o MCS be MCS ng be ng nne m za op on m he on h gh he compu awith on me ave on n compa me compa on on M awback oema MCS be ng nne op m za on he gh compu on me n compa on M NLP Artificial Artificial intelligence (AI)-based (AI)-based metaheuristics metaheuristics Artificial (for intelligence example (for example ACS, (AI)-based ACS, ABC, ABC, metaheuristics and hybrid and (for example ABC, and hy Artificial intelligence (AI)-based (for example ABC, etc.) hybrid ac aehoweve pDynP ec on ac ocha agh eintelligence cnume h eR gh cowa p en ec me on compu ocha aglobal cMH on eke ac and cgh en ad unce ed h me gh acompu nh p ope ec aMOCRU on on ocha and ua unce on con ebe aetc.) cn en ope aand me compu ua aon on and unce amum n ope ao ua 5PT A 5Howeve 2ego m A ndynam oWo Mu n oema ob Mu ob P ann vann ng P M ann 5PT hod 2and ng A M hod m n oh Mu ob vamewo P ann ng M hod 5PT 2The A m oo Mu ob P ng M hod mo en me The mo eao mp cbu eange en MH me me hod The mp mo ek emetaheuristics compu mo me eop hod eon aoca cyqua on en equ me egh A mo o eca The mo compu eo numbe mp anne ephough MH me me A hod o equ ehm numbe ehybrid mo epromising aACS, on me A mo euanee num mo eng ed cew en me hod The mp ed MH mo eog compu aeACS, on me A o mo eon numbe considered, considered, aiming aiming at finding at finding optimal optimal weights, for FDNs. for FDNs. considered, at finding optimal weights, for FDNs. SQP and pamm og ng ek he mo eeon cop en nume ca me hod SQP dynam cpo p og amm ng amore ena he mo eooba caand en nume ca me hod SQP and og SQP and eob mo cMODGP eca og cmu amm en nume ang ewo ca he me mo hod eTab cand en nume ca me hod optimization optimization methods can can consider be as promising as promising ways to ways deal to optimization deal complex with complex methods MOP problems MOP can be problems consider more more as ways to deal with comp optimization methods can be consider as promising ways to deal with complex Howeve he p ob he em p ob o em aoca o on mu anea ve on y g ve d y and g d ew and ew abe Howeve on va can ay he can nco p ob em nco au ed o po Dynam amo on cge ve y g dyon ew va aetc.) on can be Howeve he p ob em o mu ang on ve y g ew va aem on can be nco po abe ed Dynam cNLP o unc o on unc on equ ed equ ohod ach ed oetime ach nea eve y nea h gh o y qua unc h gh on y eoptimization u y equ ehod A u ed o con A ach aTab con eve ned nea aand ned y uncon h and gh qua uncon aed ned eo aono u ned A con aem ned and uncon av o unc on equ ed oon ach eve y h qua y eva u A o con abeen ned and uncon ay ned The LF impact on the basis of interactions, aiming at MOPT problems have arranged as a big picture, Me aheu cchod MH me hod Me aheu ke evo cd u MH ona me yvamewo aconsider hod go hm Me have evo aheu u been ona cnume con y MH arobustness go de me ed hm hod ang have aon na ke u evo aov con u ona de y ed aom go aemen aDynam hm u have asystems been con de ed ade aon na am efficiently from ao future perspective. The hybrid efficiently metaheuristic from ave future techniques perspective. show The robustness hybrid (high metaheuristic techniques show robustness ( efficiently from acnume future efficiently perspective. from The amethods hybrid future metaheuristic perspective. The techniques hybrid metaheuristic show techniques (high show robustness (high p og amm ng and exhau ve ea ch p og ES amm a hough DynP hey p and om exhau o nd a p ea g og oba ch amm ES op ng a mum DynP hey and p exhau e ve ea nd ch a g ES a op hough hey p om e o con on o ev ewed wo a e hown The ch con ono bu og on ca y o n ev Tab ewed e 5–8 wo add k a e hown ng each ch og ca n Tab e 5–8 add e ng p og amm ng DynP and exhau ve ea ch ES a hough hey p om e o nd a g oba op mum mu ob mu ec ve ec p ob ve em p ob a em d a e cu o cu op mu o m ze ob m S ec m ze ve a S p y m ob GA a y and GA a PSO and d p PSO cu de p o ov op nea de m op nea ze ma S op m a ma y GA and PSO p ov nea op The con bu on o ev ewed The con wo bu e on hown o ev ch ewed ono og ca k a y e n hown e ch 5–8 ono add e n each e 5–8 add e ng each mu ob ec ve p ob em a d cu o op m ze S m y GA and PSO p ov de nea op ma way o add e ng comp ex MOP way p o ob add em e Howeve ng comp he ex h MOP gh compu p ob em a ona Howeve equ he h gh and compu ona equ emen and way o ng however comp ex MOP p ob em Howeve he h gh compu a ona equ emen and quality solutions), promises powerful global optimization methods and have ability to address quality solutions), promises powerful quality global solutions), optimization promises methods powerful and global have ability to address methods and have ability to address o u on howeve hey have o u on no howeve advoca ed hey o have a ge no d advoca bu o u on on ed howeve em a ge L kew hey d have e bu a on ma no o y advoca em L ed kew o e a a ma d o bu y L kew e a m PT ca ego y unde he MOP amewo k a MODGP PT ca ego MOVPQ y unde MOCRU he MOP and amewo MONTR k a MODGP e pec ve MOVPQ y MOCRU and MONTR e pec o u on howeve hey have no advoca ed o a ge d bu on y em L kew e a ma o 5 1 R v w d Wo k Con bu on 5 1 R v w d Wo k Con bu on o u on o u o on a ge d a ge bu bu y on em y em o a ge d bu on y em ca ego y unde he MOP PT ca amewo ego y unde k a MODGP he MOP MOVPQ MOCRU k a MODGP and MONTR MOVPQ e MOCRU pec ve and MONTR e pec ve y 5PT 1 R v w d Wo k Con bu 5 1 on R v w d Wo k Con 5 bu 1 R on v w d Wo k Con bu on a ge d bu on y em Among ca me hod Among OPF nume common ca me exp hod o OPF ed a an common Among nne op y exp m za o ca on ed me a a go hod an hm nne OPF o op m common za on y go exp hm ed o a an nne op m za on a go h Among nume ca me hod OPF common y exp o ed a an nne op m za on a go hm o uto eon they have not so advocated ut on however for arge they d have str but not on advocated systems L for kew arge se a d ma str but or on L kew se a po b y o p ema u e conve gence po owa b d y o oca p ema op ma u e conve o u on gence need owa o be d add oca op ed ma w h o u on need o be add ed po so b add y parge ema ud e conve gence owa d oca op ma o u on need o be add e ed w h constrained constrained real time real time planning planning problems. problems. Furthermore, constrained Furthermore, real decision-making time decision-making planning methods problems. methods also Furthermore, need also need to be to decision-making be methods also need vocated for str but on systems L kew se a ma or constrained real time planning problems. Furthermore, decision-making methods also need to be d awback MCS be ng op m za on d awback he hcbeen oenume compu MCS be awith ng on nne me op n compa on on M he NLP hand gh compu aon on nnco compa on M d awback be ngem nne op m on he hach gh compu on me nways compa M A ceofrom ep gence A ba me aheu o examp ACS ened cnco and d A caeoaeon gence A ed A me cLF aheu n enne ccan gence oamu examp A ba esoftware-based ACS me aheu ABC emo cu caed and o examp hyb d ecomplex ACS ABC eoqua chyb hyb d ac hmn gh ec on ac ocha aamm eaiming eaiming gh cp p en ec me on compu ocha aP cas on ecompu ac and cgh en unce h me gh an compu p ope ec aon on ocha and ua unce on cnd aza cMH n en ope ame compu ua aon on and unce amo ope ahave on ua 5Howeve 2 unc A n oequ Mu ob 5Howeve 2ba A vcaaiming Pon ann m n ng oM Mu hod ob veang ann 5Howeve 2ng A M hod m oto Mu ob vafrom ann ng M hod ac aSQP gh ec on ocha cmp eke cevo en compu on unce ahod n ope aoand ua on mo eMe cMCS en The mo eon mp eke encza cevo MH en me hod The equ mo eapromising eaMH me hod aeways equ me A mo ova ehave mo compu eprovide eNLP ccomplex numbe en aP me on me The A mp mo eaiming MH numbe me equ ego ehey compu abeen on mo each en me hod The ed MH me hod equ econ epo compu aen on me A mo eon numbe considered, considered, at finding at finding optimal optimal weights, weights, for FDNs. for FDNs. considered, considered, at optimal at finding weights, optimal for weights, FDNs. for FDNs. SQP and SQP dynam and dynam paob og amm cfinding p og ng amm eed he mo eo he ehown mo cach en egh cand en nume ca me hod me hod and dynam co p og amm ng aop eed he mo cned en nume ca me optimization optimization methods methods can be consider be consider as optimization promising to methods ways deal to deal can complex be with consider MOP as problems promising more ways more to deal with complex MOP problems m optimization methods can be consider deal with MOP problems more he pp ob mu ah he ve y ob em g d o and ew on va aas ve on y can g d be he p nco ob em aand o ed ay mu Dynam on cep be ve y po gequ d aaiming and ed Dynam ew cL ahod po ed Dyn oThe ed o oeo unc eve nea y equ h gh ed qua oand ach yem eve u nea o A y unc o h gh con on qua aem y equ eeew and ed uncon A ach o con eve aABC ned nea and y h gh uncon eke ah ned A o ama ned and ao o unc equ ed oon eve nea y h qua en A ona con aap ned and uncon ay ned The modified The methods LF methods software-based and software-based LF LF platforms provide better better performance performance The modified LF methods and LF platforms provide better performance aheu chod MH me hod ke evo u ona yamewo ahoweve go hm have de ed Me aca aon aheu na u am me hod evo u ona yme hm Me aheu ceedme MH me hod Me aheu cau MH ona y me aand go hod hm have been ona con y aono go hm aequ aeom u been con de aproblems ahm na u aaiming efficiently efficiently ah future ahe future perspective. perspective. The hybrid The hybrid metaheuristic metaheuristic efficiently techniques show aza robustness robustness perspective. (high (high The hybrid efficiently from future perspective. The hybrid metaheuristic techniques show robustness (high p og amm p og ng DynP ng and DynP exhau exhau ve ea ch ve ea aego hough ES hough hey p hey amm eand p om o ng DynP eMONTR ahod ofuture g oba nd and aon g exhau oba mum op ve ea ES ametaheuristic hough pauncon om eadd con bu on oon ev ewed wo k aed epromising ch og ca y n Tab eycan 5–8 add eoba ng each p og amm ng DynP and exhau ve ea ch ES aec hough hey pnne om eand o aMOP g op mum mu ob mu ec ob ve ec p ve em pneed ob aaiming em d amp eA cu oaMOCRU cu mu o m op ze SMOVPQ m ze ve S p y m ob GA au em y GA anne PSO d PSO cu ov de p o ov op nea de op nea ze ma S op m au ma ymethods GA and PSO p ov de nea op con bu on obe ev ewed The con wo bu The en on con hown bu ev ch ewed on ono o og wo ev ca k ewed aon y eu n Tab hown wo eon kplatforms ch ono add og eog ca ch ng each Tab og ca eowa 5–8 n Tab add eadd 5–8 ng add each eova ng each mu ob ec ve p ob em anne eA d cu op m ze m ahown y GA and PSO ov de nea op ma way oon add ng way ex MOP o add pcek em ng comp Howeve MOP he way p h o gh ob add em compu em Howeve ay ng ona comp he ex h MOP emen gh compu p ob and em acy ona Howeve emen he and acan ona equ emen and quality solutions), promises powerful quality optimization solutions), promises and powerful have ability global to optimization address to quality solutions), quality powerful solutions), global promises optimization powerful methods global and optimization have ability methods to address and have ability address o uMH hey have no advoca ed o ueA o aon ge d bu hey on have y no advoca L kew o u ed eeono aM o howeve ma o ge d bu have on no y advoca em ed kew ocon eACS aon aabe ge d ocompa bu on y e ca ego yA unde he MOP k MODGP PT ca unde MOCRU he MOP MONTR kshow ahey MODGP eza pec ve MOVPQ MOCRU and MONTR pec hey have no ed oFurthermore, aSgh ge d bu on y L kew eto ay ma o o u on u on amodified ge d aenume ge bu d bu yproblems. yweights, em o u on aeco ge d bu on em ca ego unde he MOP ca ego y unde k aon MOP MOVPQ k aom MODGP and MONTR MOVPQ ep pec eem pec ve y o u on afrom ge bu y em Among Among ca me hod ca me OPF hod common OPF common Among yob exp o5–8 nume yade ed exp ahod o ca an ed me aon hod an op nne m OPF op on m common aced go on exp o hm ed ocompu ach nne op m za on aeon go Among ca me hod OPF common y exp o atechniques nne op m za on aop go hm omum po b o p ema u eon conve gence po b y omethods p oca ema u op ma conve o gence u on need d o oca op ecgo ma ed w u h o be add ectechniques ed hav po ebe cy en o me p ema eThe conve gence owa oca op ma o od be add eadvoca w h constrained real time planning Furthermore, decision-making methods also need to be constrained time planning problems. Furthermore, real time planning decision-making problems. also need decision-making to be methods also need to be o MCS ng d nne awback m o MCS be ng he h nne gh compu op m za aES awback on on o he MCS n h gh be compu ng on ave on op NLP me m za n compa on he on h gh M compu aan on me n M N d awback be ng op m za he h gh compu aand on me n compa M NLP A ceo aMe cao aob gence eh gence ba ed ba aheu me A aheu cd co aaiming o n examp eced o gence examp eprovide ACS A ACS ba ed ABC eweights, me ced aheu and eopo caach hyb d ona examp d eneed ABC erobustness and hy A cop aaiming n eMODGP gence ba ed me cy oy examp ACS ABC een and hyb d ac aamm eme hhoweve gh pppromises ec on ac ocha ahoweve eon cnume h eMu gh cd p en ec me on ocha aglobal cweights, eke ac and cowa ach unce eway h me aand n ope ec aMOCRU on ocha and ua unce on ahe cea ope aon me on compu ua agh on on and unce aehave ope ago on ua 5PT 2comp A m neve oMCS ob vamewo P ann ng M 5d hod 2aheu A n ocompa Mu ob vamewo ann ng M hod ac aconstrained gh ec ocha cmp eme con en me compu amethods on unce ope aob on ua on 5PT 2mp A myreal n ohod Mu ob 5PT 2Me vamewo P ann m nhe ng oThe M Mu hod ob vaex ann ng M hod mo enume cy en me hod The ep MH mo me een ed hod cme en equ me mo ean The compu mp aP eem on MH me hod A mo equ eop numbe eew mo eog compu aneand me Aability mo ew nu presented in 9. The overall performance comparison the addressed MOP techniques has hoduhas eawback hod equ econsidered, mo compu aza on me A o mo eva numbe considered, aiming aiming at finding at finding optimal optimal weights, considered, for FDNs. for FDNs. finding optimal for considered, at finding optimal for FDNs. SQP and dynam paob og amm ng amu eP he mo SQP ecomp csoftware-based and en dynam nume ca p me og amm hod ng accomp enco mo ehown cand en nume me hod SQP and dynam cTable og amm aeon eMe mo eon cem en nume ca me hod op m za me hod can con de aThe ng dea w h comp ex MOP p em mo em op md za on me can be con op de m au za p on om me ng way can o be dea w de h aon p om ex MOP ng way pva ob em oaand dea mo w ePSO h ex MOP p ob em mo Howeve he pcomp ob em ounc Howeve mu ah he ve p y ob em go o and ew acon on Howeve aThe ve y can gand d be he p nco ob ew po em va o ed aetraditional mu Dynam aom can on cep be ve nco y gde d aaiming and ed va cme aeon on can be nco po aoba ed Dyn Howeve he p ob em mu on ve y g ew aon on can be po abe ed Dynam cNLP o unc on equ ed oA oewed ach unc on nea y equ gh ed qua o ach y eh eve nea A ynoevo h con gh qua a5–8 y and eebu o u uncon unc A on oetraditional con aABC ned equ aperformance ned ed uncon eve nea amum ned yu h gh qua ud ocompu con o equ ed o ach eve nea y h gh qua y eSp u A o con abeen ned and uncon ave ned The modified The modified LF methods LF methods and software-based and software-based LF platforms LF provide provide better performance performance aiming The modified LF modified methods LF and methods software-based and LF platforms LF platforms better provide better performance aiming aiming aheu aheu ccca MH cd me MH hod me ke hod evo u ona evo u y aona ona go y hm aplatforms go have hm been have de con aability de ed na aFDNs. u aDynam na u ahey aheu cobe MH me hod ke u ona age go hm have been con ed ago ahyb aca efficiently efficiently from from future ahe future perspective. perspective. hybrid efficiently hybrid metaheuristic from metaheuristic aof future techniques perspective. techniques show The show robustness hybrid robustness (high metaheuristic (high techniques show (e efficiently from future perspective. The hybrid metaheuristic show robustness (high p og DynP and p exhau og amm ve ng ea DynP ch ES and acompu exhau hough hey ve p og p om amm ES em ng o acompu DynP nd hough aon) g and oba hey exhau op p mum eMONTR ve o nd ch aAlso, g ES oba hough pHoweve om ey o nd aA op m The con bu on o ev ewed wo ko athe ehod hown ch ono og ca y The n con Tab edned bu 5–8 on add o eon ev ng ewed each wo ko aov eand ch ono ca yyneed Tab ecompu 5–8 add eneed ng mu ob ec ve pema ob em mu ang econ d ob ec cu ve p op ob m ze S eop m d auglobal cu y mu GA o op and ob m ec PSO ze ve p p m ov ob aat de em y nea GA abu ehown op d ma PSO cu p op de m nea ze S op ama y GA and PSO p ov de nea op The con bu on obe ev The con wo k bu ed on hown ev ch The ewed ono con og wo bu ca k aon y on eea o Tab hown ev eneed ewed ch ono add wo og k eawback ca atechniques ng y neach Tab ch ebetter 5–8 ono add og ca eo y ng n Tab each ew 5–8 add ng each mu ob ec ve p ob em a e d cu o op m ze S m y GA and ov de nea op ma at uncertainty at uncertainty of load of load and REG and REG generation, generation, in comparison in comparison to traditional to LF models. LF models. Also, they they at uncertainty of load and REG generation, in comparison to LF models. Also, they drawback of MCS (be ng nner opt drawback m zat on) of s MCS h (be gh ng computat nner opt on m t me zat In compar s the h son gh computat MINLP on t In compar son MIN way o add e ng comp ex MOP p em Howeve he h gh compu a ona way equ o emen add e and ng comp ex MOP p ob em he h gh a at mo on) s the h gh computat on t me In compar son MINLP way o add e ng ex way MOP o add p ob e em ng comp Howeve ex MOP he p gh ob compu em Howeve a equ he h emen gh compu a ona equ emen and quality quality solutions), solutions), promises promises powerful powerful global optimization optimization methods quality methods and solutions), have and ability have promises to address powerful to address global optimization methods and quality solutions), promises powerful global optimization methods and have ability to address o u on howeve u on howeve hey have hey no have advoca no advoca ed o ed a ge o a o d u on on y howeve on em y L em kew hey L e have kew a ma no e o a advoca ma o ed o a ge bu on y PT ca ego y unde he MOP amewo k a MODGP MOVPQ MOCRU and MONTR e pec ve y howeve hey have no advoca ed o a ge d bu y em L kew e a ma o o u o on a ge o d a ge bu on bu on em y em o a ge d bu on y em PT ca ego y unde he MOP PT ca amewo ego y PT unde k a ego MODGP y MOP unde MOVPQ amewo he MOP MOCRU k a amewo MODGP and k a MONTR MOVPQ MODGP e MOCRU MOVPQ pec ve and MOCRU and e MONTR pec y e pec ve o u on o a ge bu on y em Among nume ca me hod Among OPF nume common ca me y exp hod o ed OPF a an common Among nne op nume y m exp za o ca on ed me a a go hod an hm nne OPF o op m common za on y exp o hm ed o a an nne op m za on a go po b y o p u e po conve b gence y o p owa ema d u oca e conve op ma gence po b o owa u y on o d p oca ema o u op be e ma conve add o e gence u ed on w owa need h d o be oca add op ma ed o u h on o be add e ed w h constrained real time planning problems. Furthermore, constrained real decision-making time planning methods problems. also Furthermore, need to be decision-making methods also constrained real time planning constrained problems. real time Furthermore, planning decision-making problems. Furthermore, methods decision-making also need to be methods also need to be d awback o MCS ng nne op m za on d awback he h gh o MCS compu be a ng on nne me op n compa m za d on on M he o NLP MCS h gh compu be ng a nne on op me m n za compa on he on h M gh NLP a on me d awback o MCS be ng nne m za he h gh compu a on me n compa on M NLP A A a n c e a gence n e gence A ba A ed ba me ed aheu me A aheu c c a o c n examp e o gence examp e ACS A e ACS ABC ba ed ABC e me c aheu and e c hyb and c d hyb examp d e ACS ABC e c and hy con aThe n eaiming gence A ba ed me aheu cgo ohm examp ebeen ACS ABC een com hyb d ac aeop eA ac h gh aamm p eon hec gh on p ec ocha on cach ocha emo cmo cen eu ac me cweights, en aamm compu eway h gh compu aoen p on ec and aon on unce ocha and aeve unce n cde ope ebe aave aned cevo n on ope aand me ua on compu on ua aoon on and unce aA nnco ope aed on ua 5op 2ob A m nand oco Mu ob vfinding P ann ng M 5Howeve hod 2ned A m n oobu Mu ob vMH P ann ng M hod ac a e h gh p ec on ocha c e c en me compu a on and unce ope a on ua on 5 2mp A maiming ndynam ogh Mu 5Howeve 2 A v P ann m ng o M Mu hod ob v P ann ng M hod mo e e en me hod The mp e MH e me c en hod me equ hod The mo e mp compu e a me on hod A equ o mo e mo e numbe e compu a on me o mo e numbe mo e e on c en me hoded The e MH me hod equ e mo compu a on me A o e numbe considered, aiming at finding optimal weights, for FDNs. at finding optimal considered, weights, for FDNs. at optimal for FDNs. SQP and c p og amm SQP ng a dynam e he mo c p e og c amm en nume ng a e ca SQP he me mo and hod e dynam c c nume p og amm ca me ng hod a e he mo e c en nume ca me hod SQP and dynam c og amm ng a e he mo e c en nume ca me hod m za m on za me on hod me can hod be can con be de con a de p om op p m ng om za on ng me way dea hod o w dea can h comp be w con h comp ex MOP ex p MOP p ob em p ng ob mo way em e mo dea e w h comp ex MOP p ob em m op m za me hod can be con de a p om ng way o dea w h comp ex MOP p ob em mo e Howeve he p ob em o Howeve mu a he ve p y ob em g d o and mu ew a on va ve on y can g d be he and p nco ob ew po em va a o ed a mu on Dynam a can on c nco y po g d a and ed Dynam ew va c a on can be po a ed Dyn he p ob em o mu a on ve y g d and ew va on can be nco po a ed Dynam c o unc on equ ed o eve nea o y unc h gh on qua equ y e u ed o A ach o con nea a y h and gh qua uncon y e a u ned A o con a ned and uncon a o unc equ o considered, ach eve nea y h qua y u A con a ned and uncon a The modified The modified LF methods LF methods software-based and software-based The LF modified platforms LF platforms LF provide methods provide better and better performance software-based performance aiming LF aiming platforms provide better performance aim The modified LF methods and software-based LF platforms provide better performance aiming Me aheu c MH me hod ke evo ona Me y aheu a c MH have me hod ke de ed a u a ona na y u a a go hm have been con de a a na Me aheu c MH me hod ke evo ona y a go hm have been con de ed a a na u a e c en y om a u u e pe pec ve The hyb d me aheu c echn que how obu ne h gh e c en y om a u u e pe pec e ve c en y hyb om a d me u u aheu e pe pec c echn ve The que hyb how d me aheu ne c h gh echn que how obu ne h gh p og amm ng DynP and p exhau og ve ng ea DynP ch ES and a exhau hough hey ve p og ea p om ch ES e ng o a DynP nd hough a g and oba hey exhau op p om mum e ve o ea nd ch a ES g oba a hough op mum hey p om e o nd a g oba op m p og amm ng DynP and exhau ve ch ES hough hey p om eand o nd g oba op mum mu ec ve p ob em mu aoon ecoA d ob ec cu ve o p op ob m em ze anne eed m d aao cu yea GA oca and op m PSO ze p S m ov mu de ahey y ob GA ec op and ve ma PSO p ob p em ov akew de epec nea d cu op ma oexamp op meon ze S mem amyyza GA and PSO mu ob ec ve pema ob em anne eA d cu o op m ze SMOVPQ m ah ycomparison GA PSO p de nea op ma at uncertainty at uncertainty of load of and load REG generation, REG generation, in comparison in comparison to traditional to traditional LF models. LF models. Also, they Also, they at uncertainty at uncertainty of load and of REG load generation, and REG generation, in comparison in to traditional to traditional LF models. Also, models. they Also, they way o way add o add comp ng ex comp MOP ex p MOP ob em p ob Howeve Howeve he gh hnea gh apowerful ona aability equ ona emen equ emen and and way add eme ng comp ex MOP p5PT ob em Howeve he gh compu ona equ emen and quality quality solutions), solutions), promises promises powerful powerful global quality global optimization solutions), methods promises methods have and have global ability to optimization address to address methods and ability add quality solutions), promises powerful global optimization methods and ability address oego uca howeve hey have u on no howeve ed hey oMOVPQ have aS ge no d advoca bu o u on on ed y howeve o em an ge L kew d have eand bu aecompu on ma no on advoca em L ed ohyb eproblems. aMOVPQ age ma d oon bu yand Lmo kew eto anee PT ego unde he MOP amewo k afinding MODGP MOVPQ MOCRU ego and unde he MOP eh pec ve y kov aalimited MODGP MONTR eand pec v oygh aMu ge d bu od u on on o aetime ge d bu on y em o aza ge dhe bu on y em yeaiming unde he ca ego yoygence unde aeadvoca MODGP he MOP PT ca ego unde MOCRU ku aom MODGP he MOP and MONTR k eca acompu pec MODGP MOVPQ MONTR MOCRU ecLF ve and y pec ve o u o aem ge bu on ydynam em provide provide generation generation of various of various scenarios scenarios under under all load all load models, models, which which were were limited limited in conventional in conventional provide generation of various scenarios under all load models, which were in conventional ca me hod OPF common Among y exp oop nume aew ca an me nne hod op m OPF za on common aceom go hm y o o ed aMOCRU an nne op on aon go hm been presented Table 10. Among nume ca hod Among OPF nume common ca me y exp hod o OPF ed an common nne op yoptimization exp m on ed ahe go nne hm o op m za on aalso go hm ocompu po y ooin ppat ema u eThe conve owa d op ma o on need o be po add bmp y ed o pchave ema h u conve gence owa dw op ma uame po b yeb oob ema u eMOP conve po b gence y o owa p d u oca ebe conve ma gence owa on dem need o be add eaon o u ed on w h need o be add eto ed w h constrained real time real planning time planning problems. Furthermore, Furthermore, decision-making decision-making constrained methods real methods time also planning need to need be to be Furthermore, decision-making constrained planning problems. decision-making methods also need to be awback awback MCS o be MCS ng be ng op m za op on m za he on h gh he compu h gh d compu aw awback on ave o on naw MCS compa me be n compa ng on nne M NLP on M m za he hned gh compu aced on me d awback o MCS be ng op m za on he h compu on me compa on M A asoftware-based cng emodified aMu n eob gence A ba A ba me ed aheu me A aheu con aamewo o come examp o gence examp eaw ACS A ACS ABC ba ed ABC eaweights, me cnco aheu and epo and cexp d hyb demo ACS ABC eo hy A con n gence ba aheu cyew examp eeve ACS ABC and hyb d ac aon hdynam ec on ac ocha aamm cn h gh cu p en ec me on compu ocha cweights, on eke ac and cu aamm unce ecbeen h gh agh nMONTR pooDynP ope aon aMOCRU on on ocha and ua unce on eke anumer cde en ope aona on ua aMONTR on and unce ahave n ope ua 5d 2ob A m nedynam ok ob vop P ann ng M hod 2Furthermore, A m n oama Mu ob vamewo P ann ng M hod 5PT 2cca A n 5PT 2Among A vamewo P ann m ou M Mu hod ob vamewo P ann ng M hod mo em cp me mo emp enume ereal co MH en hod The equ ey mp mo mo ep eme MH compu eg me cen en hod aoca me on equ hod A The o ean mo compu ebeen numbe MH on hod A equ od emo mo eop numbe egh aon on me A oov eon numbe considered, aiming at optimal weights, considered, for FDNs. aiming at finding optimal for FDNs. considered, finding considered, optimal weights, at for FDNs. optimal for FDNs. SQP and pme og ng aof ep he mo SQP eona and cREG en nume cve ca og me hod ng aed eh SQP mo and eme dynam cy en nume cLF p og ca amm me ng hod amo ena he ecompu cecan en hod SQP and og amm ago eed he mo eona en nume me hod m za m on me on hod can hod can con be de con aproblems. de p m ng om za ng me way hod o dea can h comp be w con h comp ex de MOP ex aThe MOP pend ob em p ng ob way em eaheu mo o dea hoca comp ex MOP p ob m m za me hod can be con de ahm p om ng way oec dea hque ex MOP p em Howeve Howeve he peayaiming ob he em p o mu em aoca o on mu ame ve on y Howeve ve d y and gcand d he and va p ob ahm em va o ame mu on be ad can nco on be po ve amodels. y ed goob aaiming and ed cge ew va agh be nco po ed Howeve he ob em o mu ang ve y g d ew va on can be nco au ed Dynam cNLP o unc on equ ed o ach eve oaand nea unc y h gh qua equ y ed ecy u ach A o nea con y h aob gh ned qua and y uncon eop u aemum ned octo con abeen and uncon aem ned o on equ ed ach nea y hen gh qua yare econstrained A con and uncon ap ned The modified LF methods software-based LF platforms provide better performance The modified LF methods and The LF LF methods platforms provide software-based better performance LF platforms aiming provide better performance aiming SQP and dynam programm ng SQP and most dynam eff cnne ent cMOP programm numer ca ng methods are the most eff com ent ca methods e most eff ent numer methods aheu cchod MH me hod Me aheu ke evo cafinding MH y me ahough hod have evo Me u aheu ona con y ceed go de MH ed hm acan have aeop hod na u been age con evo de y aNLP aDynam go u hm aehon have con de aDyn na Me aheu ccegence MH me ke evo u yeona aze go have con ed aDynam amo na u ao eop ceex en eop yexhau cog en om ang om ap ehey u pe u epowe pec pe ve pec The ve hyb eop The con d hyb en me y aheu om me aheu ang u u echn etraditional con pe echn pec que ve obu how hyb ne obu d me h ne h gh cGA que how ne eop camm en y om ame u u ened pe pec ve The hyb d me aheu cma echn how obu ne h p amm ng and p ve ng ea DynP ch ES and ahod hey p ea p om amm ch ES ng o acompu nd hough amo g and oba hey exhau p mum ePSO ve o ea ch aAlso, ES g oba acompu hough hey pechn om oem aca gaobu oba op m p og amm ng DynP and exhau ve ea ch ES am hough hey pnne om ehave o nd ap g oba op mum mu ob ec ve p ob em anne eon d cu mu o op ob ec ve Sdea p m ob au em y GA anne ecomp and cu opo p op ov de m ze nea Sgh m op aA ma pnd de nea op muunc ob eccve p ob em o aqua eca deve cu o mDynP ze Suncertainty m au y GA PSO p ov de nea op ma at at uncertainty load of load and REG and REG generation, generation, at in uncertainty in comparison of load traditional to and traditional REG LF generation, models. in comparison Also, they they traditional LF models. Also, at uncertainty of load and generation, in comparison to LF models. Also, they way o eza comp ex p ob em way Howeve o add edecision-making he gh comp compu ex aque ona p equ em emen and he aenume ona emen enume comp MOP p ob em Howeve he h gh compu aeway equ emen and qua y o u on p om eThe ginteractions, oba op m za on hod and ab y add ehm oeMe on p om powe qua u y g o oba u on op m p om za on eba powe hod u and g oba have op za on o add me ewere and have ab y o add oaog u on howeve hey have othe u on no howeve ed hey have aexhau no advoca bu o u on on ed y howeve o em am ge L kew d have ebu ahod on ma no y o advoca em L ed kew o eh ama aM ma d o bu on yePSO L kew eneed aon m o u on howeve have no advoca ed oop am ge d bu on y em L kew eed ay o aMu ge d bu on aMu ge d bu on em o u oen ahow d bu on y em o aem ge d bu yproblems. em provide provide generation generation of various of various scenarios scenarios under under all load all models, models, which which were limited were limited in conventional generation of generation various scenarios of various under scenarios all load under models, all load which models, which limited were in conventional limited in conventional Among nume ca me hod OPF common Among y exp oope nume ed aop an hod op m OPF za on common ao go hm o nne op m on aaand go h Among ca me hod OPF nume common ca me y exp hod o OPF ad an common nne y exp za oo on ahey go an nne hm o op za on aalso go o po b po yh o b ema y o u p egence ema conve u eon gence conve gence owa d owa oca d op oca ma op o ma on o u o need be add o be add ed w ed w hcaah po byep yand o ema u eo conve gence owa d oca o u on need be add eu ed w constrained constrained real time real planning time planning problems. problems. Furthermore, constrained Furthermore, real time decision-making planning methods problems. methods also need to need be to decision-making be methods also constrained time planning Furthermore, decision-making methods also need to be d awback o MCS be ng d nne awback op oadvoca za MCS on be ng he h gh compu op m d za aog awback on me he MCS n hmo compa gh be compu ng on aaeneed M on op NLP me m za n compa he on gh compu NLP aan on me nnco compa on M A coop n emethods gence A A ba cng ed aA me n eh aheu gence caheu A o ba examp ed me A ecomparison aheu cn aload n eto o eca gence examp cmethods and A eMOP hyb ACS ba ed d ABC me aheu epo cHoweve and cexp hyb ocon examp dy eand ACS ABC eequ cpame hy A csoftware-based ay n ehod A ed aheu conume examp eproblems ABC een cFurthermore, and hyb d methods. methods. The LF The impact LF impact on the on basis basis of of interactions, aiming aiming at MOPT at MOPT problems have have arranged arranged as ah as aed methods. The LF impact on the basis of interactions, aiming at MOPT problems have arranged as ac aThe eAmong gh p ec on ocha ege co en ac me aamm compu h gh acab p on ec and on unce ocha aon nu ope em aA cin on me ua compu on aova on and unce amo ope aobu on ua 5p 2ob A m nand Mu ob voo P ann ng M hod ac eunc h phe ec on ac aamm eP caheu h gh cm p ec on me compu ocha apec cu on ey and cevo en unce me ay compu ame on and unce on acnd n ope aconventional on ua on 5way 2 eHoweve A mgh n oang 5provide 2ocha A vadd ann m 5provide non 2ome ng oen M hod m n ob Mu vthe P ann ob ng vweights, M P hod ann ng M hod mo eu cme en me hod mp ereal MH hod equ eed mo eme compu aACS on me A mo o eme mo eua ccomp numbe me hod The mp eamo MH me hod equ eza compu ane mo ey cadd en hod The mo mp eme each MH en me hod equ eo mp mo ede MH hod equ me A eABC o mo eu compu euncon numbe aACS on me o mo eop numbe considered, considered, aiming aiming at finding at finding optimal optimal weights, for FDNs. for FDNs. considered, aiming at finding optimal weights, for FDNs. considered, aiming at finding optimal weights, for FDNs. SQP and SQP dynam dynam c p og amm c p og ng amm a e ng he a mo e he e mo c en e c en nume ca SQP me and hod ca me dynam hod c p og amm ng e he mo e c en nume ca hod SQP and dynam c p og amm ng a e he mo e c en nume ca me hod op m op za m on za me on hod me can hod be can con be de con a de p om op a p m ng om za way on ng me way o dea hod o w dea can h comp be w con h comp ex de MOP ex a p MOP p ob om em p ng ob mo way em e mo dea e w h comp ex MOP ob em m op m za on can be con a p om ng way o dea w h ex MOP p ob em e p ob em o Howeve mu a on he ve p y ob g em d mu ew a on va a Howeve ve on y can g d be he and nco p ob ew po em va a ed o a mu Dynam on a can on c be ve nco y g d a and ed Dynam ew a on can be po ed Dyn o on equ ed o o unc eve nea y equ gh ed qua ach y e eve u nea unc A y o on h con gh qua equ a ned y ed and e u ach A eve o nea con a ned y h a gh ned qua and y uncon e a A ned o con a ned and uncon a ned The modified LF methods and software-based The LF modified platforms LF provide better and performance software-based aiming LF platforms provide better performance aim The modified LF The and modified LF methods LF platforms and software-based provide better LF performance platforms provide aiming better performance aiming Me aheu c MH hod ke evo ona Me y a go hm c MH have me been hod con ke de evo ed Me u a ona aheu na y a a go c MH hm me have hod been ke evo de u ed ona a y a a na go u a hm have been Me c MH me hod ke u ona y a go hm have been con de ed a a na u a e c en e y c en om a om u u a e u pe u e pec pe ve The ve hyb e The c d hyb en me aheu om me aheu a c u echn e c pe que echn pec que how ve The obu how hyb ne obu d me h ne gh aheu h gh c echn que how e c en y om a u u e pe pec ve The hyb d me aheu c echn que how obu ne h gh og amm p og ng amm DynP ng and DynP exhau and exhau ve ea ch ve p ES og ea ch a hough ES ng a DynP hough hey p and om hey exhau e p om o e ve a o g ea oba nd ch a op ES g oba mum a hough op mum hey p om e o nd a g oba op m p og ng DynP and exhau ve ea ch ES a hough hey p om e o nd a g oba mum mu ob ec ve p ob em a e d mu cu ob o ec op ve m p ze ob S em m a a y d GA cu and o PSO op p m ov ze de S nea m a op y GA ma and PSO p ov de nea op ma mu ob ec ve p ob em a e d cu o op m ze S m a y GA and PSO p ov de nea op ma at uncertainty of load and REG generation, in comparison to traditional LF models. Also, they at uncertainty of load and REG at generation, uncertainty in of comparison load and REG to traditional generation, LF in models. comparison Also, to they traditional LF models. Also, they way o add enec ng comp way ex MOP o add p ob eca em ng comp Howeve ex MOP he h p way gh ob compu em o add Howeve edec ona ng comp equ he h emen gh ex MOP compu and p aeab ob ona em equ Howeve emen he h gh compu aem ona equ way onne add enume ng comp ex MOP p ob em Howeve he h gh compu aACS ona equ emen and qua yaconsidered, qua o u y on o u p on om p ebe om powe ehave powe ung ginteractions, oba uunder qua g op oba y m o op za u m on on za me on p hod om me and edered hod powe have and u have gab oba y ab oehod op add yoca m o ehm add za on eand me hod and have ab yeaemen o aM add qua y o u on p om e powe u g oba op m za on me hod and have y o add e o u on howeve hey have u on no howeve advoca ed hey o a ge no d advoca bu o u on on ed y howeve em a ge L kew hey d have e bu a on ma no y o advoca em L ed kew e a ge a ma d o bu on y L kew e m o u on howeve hey have no advoca ed o a ge d bu on y em L kew e a ma o o a ge d bu on y em o a ge d bu on y em o u on o a ge d bu on y em provide provide generation generation of various of various scenarios scenarios under provide under all load all generation load models, models, which of various which were scenarios were limited limited in under conventional in all conventional load models, which were limited in conventi provide generation of various scenarios all load models, which were limited in conventional Among nume me hod OPF common Among exp nume ed a ca an me nne hod op m OPF za on common a go hm y exp o ed a an nne op m za on go h Among nume ca me hod Among OPF common ca me y exp hod o OPF ed a an common nne op y exp m za o on ed a go an nne hm o op m za on a go o po b y o p ema u e conve gence owa po d oca b y op o ma p ema o u u on e conve need gence o be owa add e d ed w op h ma o u on need o be add ed po b y o p ema u e conve gence owa d oca op ma o u on need o be add e ed w h con ned ea me p ann ng p ob em Fu he mo e on mak ng me hod a o need o be con a ned ea me p ann ng con p ob em a ned Fu ea he me mo p e ann dec on p mak ob em ng me Fu hod he mo a e o need dec on o be mak ng me hod a need o be d awback o MCS be ng d awback op m o za MCS on ng he nne h gh compu op m d za a awback on me o MCS n h compa gh be compu ng on nne a M on op NLP me m za n compa he on h gh M compu NLP a on me n compa on N d awback o MCS be ng nne op m za on he h gh compu a on n compa on M NLP A c a n e gence A A ba c ed a me n e aheu gence c A o ba examp ed e aheu ACS ABC c o e c examp A and c e hyb a n d ABC gence c A and ba ed hyb me d aheu c o examp ACS A c n e gence A ba ed me aheu c o examp e ACS ABC e c and hyb d methods. methods. The LF The impact LF impact on the on basis the of basis of interactions, aiming aiming at MOPT at MOPT problems problems have arranged have arranged as a as a methods. The methods. LF impact The on LF the impact basis on of the interactions, basis of interactions, aiming at MOPT aiming problems at MOPT have problems arranged have as arranged a as a ac a e h gh p ec on ocha e c en ac me a compu e h gh a p on ec and on unce ocha a n c ope e a c en on me ua compu on a on and unce a ope a on ua 5 2 A m o Mu ob v P ann ng M hod 5 2 A m n o Mu ob v P ann ng M ac a e h gh p on ac ocha a e c h e gh c p en ec on me compu ocha a c on e and c en unce me a compu n ope a on and ua unce on a ope a on ua on 5 2 A m n o Mu ob 5 2 A v P ann m n ng o M Mu hod 5 ob 2 A v P m ann n ng o Mu M hod ob v P ann ng M hod mo e e mo c en e e me c en hod The hod mp The e MH mp me e MH hod me equ hod e mo equ e compu e mo e compu a on me a on A me o mo A e o numbe mo e numbe mo e me hod The mp e MH me hod equ e mo e compu a on me A o mo e numbe considered, aiming aiming at finding at finding optimal optimal weights, considered, weights, for FDNs. for aiming FDNs. at finding optimal weights, for FDNs. Metaheur st c (MH) methods ke Metaheur evo ut onary st c a (MH) gor thms methods have been ke evo cons ut onary a as gor a natura thms have been cons dered as a na at finding optimal weights, for FDNs. evoo ut onary a gor bu thms have been cons dered aem natura SQP and dynam cHoweve pand og amm SQP ng and aas dynam ecof he mo p epe og cpe amm en nume ng a9. eoverall SQP he mo hod eona dynam cplatforms cuncon nume p og amm ca me ng hod enco he mo ed cMOP en nume ca me hod op moamm za on me hod can m con za de on me acgh p hod om can con oaob de dea op agh w p m om h za comp on ng me way ex MOP oprovide p can ob be em con haom comp de ex aLF MOP p om p ng ob em mo oMOP eahey w hemen comp ex MOP paoba em m op za on me hod can be con de aThe pca om ng way ohod dea w h comp ex MOP p em mo emum big picture, big picture, has presented has presented in Table in Table 9. The The overall performance performance comparison comparison the of addressed the addressed MOP big picture, has presented Table 9. The comparison the addressed he pyaiming ob oom mu ang on ve y Howeve g d and ew he va p ob ahm em on o can mu be nco on po ve aof yoba ed g Dynam and cge va aand on can nco po ed Dyn he p ob em Howeve mu aabe he ve p ob em g oMH and mu ew aaway on va ve on yoba can g d be and nco ew po va ed au on Dynam can be cLF po aL ed Dynam con o unc on equ ed om ach eve nea y h gh qua yon ed u A o con aom ned and o uncon unc on amo ned equ ed oaAlso, ach eve nea ydea hto gh qua ehave uhow A ocompu con am oHoweve unc on equ ed oou oem unc eve on nea y h equ qua ed o y ach ebe eve u nea A o y h con gh qua aooverall yaze and eperformance u A o aw con ned abeen ned and uncon agh ned The modified The modified LF methods LF methods and software-based and software-based LF platforms LF The provide better modified better performance performance methods aiming and aiming software-based LF platforms provide be The modified LF methods and software-based LF provide better performance aiming Me aheu Me aheu MH ccd me hod me ke hod evo ke u ona evo u y aao go yen hm aplatforms go have hm Me been have aheu con de cab con MH ed aab de me ain ed na hod u aop ke evo u ona y go hm have Me aheu MH me hod ke evo u ona go have been con de ed acomparison na u ama eat cach en econ y cog en om aon om u u aDynP ehey u pe u epowe pec ve pec The ve hyb eat The d hyb en me yned aheu d me aheu aeop cadec u echn etraditional cemen pe que echn pec que how ve The obu how hyb ne obu d me h ne aheu h gh cbu echn que ne econsidered, cREG en y om ay u u eme pec ve hyb d me aheu cdea echn que how obu ne h gh p og ng DynP and p exhau amm ve ng ea ch ES and aMOP exhau hough hey ve p og ea p om ch amm ES ey ng o aob DynP nd hough and oba hey exhau op p mum ePSO ve o ea nd ch g ES oba amo op pHoweve om eyapbe o nd anea gobu op mu ob ec p ob mu eon d ob ec cu ve o p op ob m em ze S ein m mu aba cu yu GA ob ocme ec and op ve m PSO p p S em m ov de ag aem y nea d GA op cu and ma o op p m ov ze de S nea m ahough op y GA ma PSO ov de op ma uncertainty of load and REG generation, in uncertainty comparison of to load traditional and REG generation, models. they traditional LF models. Also, at uncertainty of load at uncertainty generation, load in and comparison REG generation, to traditional in comparison LF models. to Also, they LF models. Also, they way add eve ng comp ex MOP p ob em way Howeve o add eand he ng h comp compu ex MOP agh ona p ob way equ o Howeve add eof and ng he comp h gh ex compu MOP p aew ob ona em equ and he hA gh am way o add enume ng comp ex p em Howeve he h gh compu aACS ona equ emen and qua yaop qua o u y on o u p on om p ebe om ehave powe u g oba u qua g op oba y m op za u m on on za me on p hod om me and ehod powe have and u have g y oob op add y m oeway eama add za eu me hod and ab yob obeen add qua y o u p emp powe u g op m za on me hod and have ab y o add ena o on howeve u on howeve hey have hey no advoca no advoca ed o u o on ed a howeve ge d a ge bu hey d on have bu y on em no y advoca L em kew ed e kew a ma o a d o on y em L kew e a o u on howeve have no advoca ed ge d bu on y em L kew e a o o a ge d bu y o em on o ge y em u Aon co a a ngeed gence on y em provide generation of various scenarios under all load models, which were limited in conventional provide generation of various scenarios provide under generation all load of various models, scenarios which were under limited all load in conventional models, which were limited in conventional Among nume ca me hod OPF common exp o ed a an nne op m za on go hm Among nume ca me hod Among OPF common Among ca nume y exp hod o ca OPF ed me a hod an common nne OPF op y exp common m za o on ed y a exp go an nne hm ed o a op an m nne za on op a go m za hm on o a go hm o po b y o p ema e conve po b gence y o owa p ema d u oca e conve op ma gence po o u owa on b d need o oca p o ema be ma u add e conve e o u on gence w need h owa o be d add oca ed ma w h o u on need o be add ed po b y o p ema u e conve gence owa d oca op ma o u on need o be add ed w h con ned a ea ned me ea p ann me p ng ann p ob ng em p ob Fu em con he Fu a mo ned he e mo dec ea me on mak p ann on ng mak ng me p ng hod ob me em a hod o Fu need he o need o be e dec be on mak ng me hod a o need con a ned ea me p ann ng p ob em Fu he mo e dec mak ng me hod a o need o be d awback o MCS be ng d nne awback op m o za MCS on ng he nne h gh compu op m d za a awback on me o he MCS n h compa gh be compu ng on nne a M on op NLP me m za n compa he on h gh M compu NLP a on me n compa on M N d awback o MCS be ng nne op m za on he h gh compu a on n compa on M NLP A c a n e gence A ed A aheu c a c n e o gence examp A e ba ed ABC me aheu e c and c hyb o examp d e ACS ABC e c and hy A ba ed me aheu c o examp e ACS ABC c hyb d methods. methods. The LF The impact LF impact on the on basis the basis of interactions, of methods. interactions, aiming The aiming LF at impact MOPT at MOPT on problems the problems basis have of interactions, have arranged arranged as aiming a as a at MOPT problems have arranged methods. The LF impact on the basis of interactions, aiming at MOPT problems have arranged as a Table 10. Performance comparison of algorithms, applied in multi-objective planning problems. ac a e h gh p ec on ocha c e c en ac me a compu e h a p on ec and on unce ocha a n c ope e a c en on me ua compu on a on and unce a ope a on ua ac a e h gh p ec on ac ocha a e c h e gh c p en ec on me compu ocha a c on e and c en unce me a compu n ope a on and ua unce on a ope a on ua on mo e e c en me hod The e MH me hod mo e equ e c e en mo me e compu hod The a on mp me e MH A o mo hod e numbe equ e mo e compu a on me o mo e nu mo e e c en me hod The mp e MH me hod equ e mo e compu a on me A o mo e numbe con de ed a m ng a nd ng op ma we gh o FDN con de ed a m ng a nd ng op con ma de we ed gh a m o ng FDN a nd ng op ma we gh o FDN SQP and dynam cHoweve p og amm SQP ng and ave dynam he mo cd p eed og cpe amm en nume ng ahe eo ca SQP he me mo and hod eona dynam coverall en ccto nume p og amm ca me ng hod enco he mo eof cand en nume ca me hod SQP and dynam cpresented p og amm ng acon ehey mo eand cand en nume ca me hod op m za on meem hod can op be con m za de on aThe me p hod om can ng way be ogeneration, de dea aoverall w p h om comp ng ex way MOP op o puncertainty dea m ob za em w on h mo comp me hod ex MOP can be p con ob em de mo aaeu phey eop om ng way ond dea w ha ne comp op m za on me hod can be con de a p om ng way o dea w h comp ex MOP p ob em mo e big picture, big picture, has presented has presented in Table in Table 9. The 9. overall The overall performance performance comparison comparison of the of addressed the addressed MOP MOP big picture, big has picture, presented has in presented Table 9. in The Table 9. The performance performance comparison comparison of the addressed the MOP addressed MOP he p ob em o mu a on ve y Howeve g d ew he va p ob a em on o can mu be nco on po ve a y ed g Dynam d c ew va a on can be nco po a ed Dyn Howeve he p ob o Howeve mu a on he ve p y ob em g o and mu ew a on va a ve on y can g d be nco ew po va ed a on Dynam can be c po a ed Dynam c o unc o on unc on equ ed equ ach eve o ach nea eve y h nea gh qua y h gh y qua e u y A e u o con A o a con ned and a ned uncon and uncon a ned a ned o unc on equ ed o ach eve nea y h gh qua y e u A o con a ned and uncon ned The modified The modified LF methods LF methods and software-based and software-based The LF modified platforms LF platforms LF provide methods provide better and better performance software-based performance aiming LF aiming platforms provide better performance aim The modified LF methods and software-based LF platforms provide better performance aiming Me aheu c MH me hod Me aheu ke evo c u MH ona me y a hod go hm ke have evo Me u been aheu con y a go de MH ed hm a me have a hod na u been a ke con evo u de ona ed y a a go na hm a have been con de ed a na e c en y om a u u e e pe c en pec y ve om a hyb u u e d me pec aheu ve e The c c echn hyb en que d om me how aheu u u obu e c pe ne echn pec h que ve gh The how hyb obu d me ne aheu h gh c echn que how obu e c en y om a u u e pe pec ve The hyb d me aheu c echn que how obu ne h gh techniques techniques has been has been presented in Table in Table 10. 10. techniques has been presented in Table 10. p og amm ng DynP and exhau ve ea ch p ES og amm a hough ng DynP hey p and om exhau e o nd ve a g ea oba ch op ES mum a hough p om e o a g oba op m p og amm ng DynP and p exhau og amm ve ng ea DynP ch ES and a exhau hough ve ea p om ch ES e o a hough nd a g oba hey op p om mum e o nd a g oba op mum mu ob ec ve p ob em a e d cu o op m ze S m a y GA PSO p ov mu de nea ob ec op ve ma p ob em a e d cu o m ze S m a y GA and PSO mu ob ec ve p ob em mu a e d ob cu ec o p op ob m em ze a S m e d a y cu GA and op PSO m ze p S ov de a nea y GA op and ma PSO p ov de nea op ma at uncertainty at uncertainty of load of and load REG and generation, REG in comparison in comparison at traditional to traditional LF of models. load LF models. and Also, REG they Also, generation, they in comparison to traditional at uncertainty of load and REG generation, in comparison to traditional LF models. Also, they way o way add o e add ng comp e ng ex comp MOP ex p MOP ob em p ob Howeve em Howeve he h gh way he compu h o gh add compu a ona e ng a equ comp ona emen equ ex MOP emen and p ob and em Howeve he h gh compu a way o add e ng comp ex MOP p ob em Howeve he h gh compu a ona equ emen and qua y qua o u y on o u p on om p e om powe e powe u g oba u qua g op oba y m o op za u m on on za me on p hod om me e hod powe have and u ab have g oba y ab o op add y m o e add za on me hod and have ab y o add qua yahod o u on p om end powe g oba op m za me hod and have ab yo add eop ooacomp uuecomputat on howeve hey have u on no howeve advoca ed hey ooca have aoyob ge no d advoca bu oza u y howeve ohe em acompu ge Lme kew d have eproblems bu aon on ma no oy em L ed kew eh abe age ded bu on yed em L okew eneed anee ongh op ame ge d bu on em o ame ge bu on o u on o ge den bu on y em provide generation of various scenarios under provide all load generation models, which of various were scenarios limited in under load models, which were limited in conventi provide generation of various provide scenarios generation under of all various load models, scenarios which under were all load limited models, in conventional which were limited in conventional Among nume me OPF common y exp o ed aen an Among op za ca me acomparison go hod hm OPF oadvoca common y oma aza an nne op m on aon go hm Among nume ca me hod Among OPF nume common ca y exp hod Among OPF ed nume auma an common nne ca me op hod exp m za om OPF on ahey go common nne hm oy op m owe za ed on aaddressed auDynam an go nne hm o m on acan go hm po bh y oom pMOP ema u ede gence owa po d b op o ma p ema o u eeon conve need gence oon be po owa add b d y ed oca w op h ema ma u eom oneed conve on gence need owa oas be daw add oca op w hcoza way ng ex way prob of ems However ng comp ex h gh MOP computat prob ems ona However requ rements the h and gh computat ona requ rements po b y o p u eme conve gence owa d oca ma o u on need oem add eua ed w h con acon con ned ahas ea me ea p ann me p ng ann pmp ng em p ob Fu em con he Fu anne mo he eon mo eob on dec mak p ann on ng mak ng me p ng hod ob me athe hod o Fu he mo need oas eenumbe dec be on mak ng me au o b ems However the h gh ona requ rements and con ned ea me p ann ng p ob em Fu he mo eea dec on mak ng me hod aoverall need o be d awback d awback MCS oh be MCS ng nne be ng nne m za op on m d awback on hned o compu MCS hcmo gh be compu aed ng on nne me aeba on op n compa m za n compa on NLP on M compu NLP on me nma compa on M N d awback oy MCS be ng op m za on he h gh ahave on n compa M NLP A cimpact aema n ehyb gence A ba A ceon aheu n enume gence examp A eang ed me aheu ABC eed cM and oexp examp hyb deua ACS ABC ehod and hyb d A e gence A ba me aheu cca o examp ACS ABC eTable cp and hyb methods. The LF on the basis of interactions, aiming at MOPT have arranged ave The LF impact on the basis methods. of interactions, The LF impact aiming on the at MOPT basis of problems interactions, have aiming arranged at MOPT as problems have arranged aaeA ac aamm eme h gh p ec on ocha cma eva c10. en me compu ame on and unce ao ope abe on on ac ec on ac ocha aed ecoon caheu h ened gh ac chas p en ec eme h on me gh compu pnne ocha ec on aon cmp on ey ocha and c10. unce cpicture, me ehod aed c10. n en ope me am on compu and ua unce ahod on and ope unce aconventional on ape n ope ua on on on mo eed eand cy en The mo mp eng eo cve MH en hod The equ mo eahe eem MH compu mo me edhe aed cyacompu on en equ emance hod A mo o ean mo The compu numbe mp aACS eapexp MH on me hod A oall mo equ eaagh mo eohey compu aecomp on me A o mo eop num mo econve emu en me hod The MH me hod equ eacon ep me A o mo eop numbe con de aen m ed ng ake ah m ng nd ad ng op ng op we gh con we o gh de FDN o aop FDN m ng ao nd ng op ma gh oaoohe FDN con de ed aeaddress m ng aand nd op ma we gh o FDN SQP chod p og amm SQP ng and aequ dynam ehas he mo cd p eeona og cop amm en nume ng eed ca SQP he me mo and hod egh dynam cedec ced nume og amm ca me hod eov he mo eon cmance en nume ca me hod SQP and dynam cpresented p og amm ng athe ehey mo eon cand en nume ca me op m za on me hod can be con de op aThe p m om za ng me way hod can dea be w con h comp de aThe ex p MOP p ng ob way em mo om dea h ex MOP paechn ob em op m za conaof menaddress hod can methods. be con de ahe pdynam ng way o dea w comp ex MOP ob em mo big picture, big picture, presented presented in Table in Table 9. The 9. overall The big overall performance performance has presented comparison in Table of the 9. of addressed MOP performance MOP comparison of the addressed M big picture, has presented in 9. The overall performance comparison of the addressed MOP Howeve he p ob em o mu a on ve y Howeve g d ew he va p a em on o can mu be nco on po ve a y ed g d and c ew va on be nco po ed Dyn Howeve p ob em o Howeve mu a he p y ob em g o and mu ew a a ve on y can g d be nco ew po va a on Dynam can be c nco po a Dynam c o unc on ed o ach eve nea y h gh o qua unc on y e u equ A ed o con o ach a eve ned nea and y uncon h gh qua a ned y e u o con a ned and uncon a o unc on equ ed o ach eve nea y gh qua y e u A o con a ned and uncon a ned The mod ed LF hod and o wa ba ed LF p a o p ov de be e pe o mance m ng The mod ed LF me hod and The wa mod e ba ed ed LF LF me p a hod o m and p ov o de wa be e e ba pe ed LF o p a o a m m p ng ov de be e pe o ng Me aheu c MH hod Me aheu evo c u MH me y hod go hm ke have evo Me u been aheu ona y a c go de MH ed hm a me have a hod na u been a ke con evo u de ona ed y a go na u hm a have been con de ed a a na Me c MH me hod ke evo u ona y a go hm been con de ed a a na u a e c en om a u u e e pe pec y ve om The a u u d e me pe aheu pec ve c echn hyb que d me how aheu e obu c en c ne y echn om h que gh a u how u e obu pec ne The h gh hyb d me aheu c que e c en y om a u u e pe pec ve The hyb d me aheu c echn que how obu ne h gh techniques techniques has been been presented in Table in Table 10. techniques has techniques been presented has been in presented Table in Table p og DynP exhau ve ea ch p ES og amm a hough ng DynP hey p and om exhau e o nd ve a g ea oba ch op ES mum a hough p om e o nd a g oba m p og amm ng DynP and p exhau og amm ve ng ea DynP ch ES and a exhau hough ve ea p om ch ES e o a hough nd a g oba hey op p om mum e o nd a g oba mum mu ob mu ec ve ob p ec ob ve em p ob a em a cu e d o cu op m o ze op S m m ze S y m GA and y GA PSO and p PSO de p nea ov de op nea ma op ma ob ec ve p ob em a e d cu o op m ze S m a y GA and PSO p ov de nea op ma at uncertainty at uncertainty of load of and load REG and generation, REG generation, at in uncertainty comparison in comparison of to load traditional and to traditional REG LF generation, models. LF models. Also, in comparison they Also, they to traditional LF models. Also, at uncertainty of load and REG generation, in comparison to traditional LF models. Also, they way o add e ng comp way ex MOP o add p ob e em ng comp Howeve ex MOP he p h way gh ob em compu o add Howeve a ona ng comp equ he h emen gh ex MOP compu and p a ob ona em equ Howeve emen he and h gh compu a ona equ emen qua y o u on p om qua e powe y o u u on g oba p om op e m za powe on me u qua g hod oba y and o op u m have on za ab on p om me y e hod o add powe and e u have g oba ab op y m o add za on e me hod and have ab y o add qua y o u on p om e powe u g oba op m za on me hod and have ab y o add e obegence u on howeve hey have no advoca ed o u aequ howeve ge d bu hey on have yneed em advoca Lhod kew ed earranged ma ama ge d bu on ynumbe em Lcompu eneed aon ouud howeve hey have ong u on no howeve advoca ed hey op have ahe ge no advoca bu on ed y oACS em agh ge L kew d bu ewhich aon on ma y ono em L eon aua oacompu oon ueeA ge d bu on y em o u on o aame ge d bu yegence em o on acp ge dame bu on em o aMH d bu y em provide provide generation generation of various various scenarios scenarios under under all load all models, load models, which provide which were generation were limited of in various in conventional scenarios under all load models, which were provide generation of various scenarios under all load models, were limited in conventional po b po o b p ema y oge u p eh ema conve u eon gence conve gence owa owa oca op oca ma op odec po ma u on b o u y o on p o need ema be add o eThe conve add ed w ed w owa h dao oca op okew nee po b y o p u eThe conve gence owa d oca ma o on need o be add ed w h con acon ned ahas ea me ea ann me p ng ann p ng em p ob Fu em con he Fu aom mo ned he emp mo eope on mak p ann on ng mak ng me p ng hod ob me em aayinteractions, hod o Fu need aDynam he o mo need oLF be eMH dec ocme be on mak ng auem oon con ay ned ea p p ob em Fu he mo eea dec mak ng me acompu o need o be awback oop MCS nne awback op m o za MCS on be ng h gh op m d za ahod awback on on me oew he MCS n h compa gh be compu ng nne aeba M on op NLP m za n compa on he on h gh M NLP on me nma compa on M cp aob n ehod A A ba ccimpact ed aema n eme aheu gence cma A o ba examp ed A me een con aheu n ep ABC capo gence ep cme examp A and eng hyb ed me d aheu ABC eed camance and one examp hyb d eA ACS ehod cp and hyb dem methods. The LF on the basis of interactions, methods. aiming The LF at MOPT problems the basis have of arranged as aiming ah at MOPT problems have arranged methods. impact on methods. the basis The LF interactions, on aiming the basis at MOPT interactions, problems aiming have at arranged MOPT as problems alimited have as ac aThe eon hLF gh pequ ec on ced ebeen cme en me compu ama on ac and aamm unce ed h acdec n p ec ao on ocha ua on cma eke c10. en me compu ahMOP on and unce amo ope aobu on ua ac aamm h gh ec on ac ocha aamm econ caheu h ened gh cp p en ec on me ac compu ocha ang enne h aon cma gh on eme p and c10. ec en unce on me aed ocha compu n ope cequ am eand on cA and en unce on compu n ope aconventional on on and unce on n ope ame on ua on mo eon en The mp me hod mo end eob equ cob en eTable mo ewa hod compu The aen me eo MH A mo me mo eeu hod eua cde numbe en equ mo hod ekew mp ame on eaoo me A hod mo eABC eme eov ane mo ed eocha en me hod mp MH me econ mo ep compu ame on me A o mo eop numbe con de con ed de amod m ng ake aom m ng nd aof ng op ng op we gh con we o gh de FDN o aop FDN m ng ao nd ng op we gh oaed FDN con de ed aeof m ng aand nd op we gh o FDN SQP and SQP dynam and dynam p og amm cd og amm abe ed ng he aGA mo ey SQP he eog mo cand eon dynam nume en cimpact nume ca og me amm hod ca me hod eov he mo eona cob en hod SQP and dynam cimpact pp og amm ng acon ecompu he mo eo cand en nume ca me hod op m za on me hod can op de m aThe za p on me ng hod way can be dea con w de h comp p om ex MOP ng way p em dea mo w eca h comp ex MOP ob mo op zayoca onom hod can be con de aThe om ng way dea w h comp ex MOP p ob em mo eze big picture, presented in Table 9. The overall performance comparison of the addressed MOP big picture, has presented in Table big 9. picture, The overall has presented performance in Table comparison 9. The overall of the performance addressed MOP comparison of the addressed Howeve he ped ob em o mu ang on ve y g d va on can be nco po au ed cge he em oog Howeve mu ao he Howeve ve y ob em g he oann and p mu ew aex em va o aof mu ve on y ad can on g d be ve nco ew g d va ame and ed aon on Dynam ew can va be cLF aaACS on nco can po be abe nco Dynam po anume ed Dynam cequ oway unc ed o o ach unc eve on nea y equ gh ed qua o ach y e eve u nea o A y unc o h con gh on qua ned y e u ed uncon A ach o eve a con ned nea a ned h and gh qua uncon y e a u ned o con a ned and uncon a o unc on equ ed o ach eve nea h gh qua y e u o con a ned and uncon a ned The mod The mod LF me ed LF hod me and hod o and wa o ba wa ed The ba LF mod p ed a LF ed p a LF o ov m de hod p be ov and de e be pe o e o wa pe mance e ba o m ng a p m a ng o m p ov de be e pe o mance a m The ed LF hod and o e ba ed LF o m p ov be e pe o mance a m ng Me aheu c MH me hod Me aheu evo c u MH ona me y a hod go hm ke have evo Me u been aheu ona y a c go de MH ed hm a me have a hod na u been a con evo u de ed y a go na u hm a have been con de ed a a na Me c MH me hod ke evo u ona y a go hm have been con de ed a a na u e en y a u u e pe pec ve e c hyb y d om me a aheu u u pe c echn pec que ve The how hyb obu d aheu gh c echn que how e cmen u of uma e premature pe pec ve hyb d me aheu c echn que how obu ne h gh techniques techniques has been has been presented presented in Table in Table 10. techniques 10. has been presented in Table techniques has presented in p amm ng DynP exhau ve ea ch p ES a hough ng DynP hey p and om exhau e o nd ve a g ea oba ch op ES mum a hough hey p om e o nd a g oba op p og ng DynP and p exhau og ve ng ea DynP ch ES and a exhau hough ve hey ea p om ch ES e o a hough nd g oba hey op p om mum e o nd a g oba mum mu ob ec ve p ob em a e d cu o op mu m ob S ec m ve a p y ob GA em and a PSO e d p cu de o nea op m op ze ma S m a y GA and PSO p de nea op mu ob ec ve p ob em a e d cu o op m ze S m a y and PSO p ov de nea op ma a unce a n o oad and REG gene a on n compa on o ad ona mode A o hey aHoweve unce a n y o oad REG a gene unce a a on n y n o compa oad and on REG o gene ad ona a on LF n mode compa A on o o hey ad ona LF mode A o hey poss bmeaopt ty convergence poss towards b ty of premature oca opt ma convergence so ut ons need towards to be oca addressed opt ma w so th ut ons need to be addressed o add e ng comp way ex MOP o add p ob e em ng comp Howeve MOP he p h way gh ob em compu add Howeve a e ona ng comp equ he h emen gh ex MOP compu and p a ob ona em equ Howeve emen he and h gh compu a ona equ emen owards so ut ons need to be addressed w th way o add e ng comp ex MOP p ob em Howeve he h gh compu ona equ emen and qua y o u on p om qua e powe y o u u on g oba p op om m e za powe on me u hod g oba and op have m za ab qua on me y y hod o add u on and e have p om ab e powe y o add u g e oba op m za on me hod and qua y o u on p om e powe u g oba op m za on me hod and have ab y o add e o u on howeve hey have no advoca ed o u o on a howeve ge d bu hey on have y em no advoca L kew ed e a ma a o d bu on y em L kew e a m o u on howeve hey have o u on no howeve advoca ed hey o have a ge no d advoca bu on ed y o em a ge L kew d bu e a on ma y o em L kew e a ma o Table 10. Performance 10. Performance comparison comparison of algorithms, of algorithms, applied applied in multi-objective in multi-objective planning planning problems. problems. u oen on am ge oen d ame ge bu on bu y em yma em Table Performance comparison of algorithms, applied in planning oTable u on o10. abe ge d bu on y provide provide generation generation of various of various scenarios scenarios under provide under all load generation all models, load models, which of various which were scenarios limited were limited in under conventional in all conventional load models, which were limited in conventi provide generation of various scenarios under all load models, which were limited in conventional po ydynam ppbe ema upicture, eecon po conve b gence y o p owa ema d u oca eThe conve op gence po owa b y need ooca p oMOP ema be ma u add emulti-objective conve o emu u ed on gence need h owa o d add oca eba op ed ma w h o ume on need othe be add edeM Problem/Solution conunc aoza ned ea me p con ann ng ahas ned p ob em ea me Fu p he ann mo ng eecon p dec ob em con on mak Fu aom ned ng he me mo hod eop dec abe p ann o on need mak ng p o ng ob be me em hod Fu aeproblems. he oemp mo need eMH dec ocned be on mak ng o need con amod ned ea p ann ng p ob em Fu he mo eea dec mak me hod aoverall need o d awback o be ng nne op m za on awback he hd compu MCS be aed ng on nne me op n compa m za on on M he NLP h gh compu ahave n compa M N d awback MCS ng d nne awback op o za MCS on ng he nne h gh op compu m za ame me he nh gh on ahave M on NLP me n on M A cp apresented n gence A ba ed aheu cma oed examp ACS ABC ep cme A and cng aaand n d ecompa gence A ed me cme oof examp eaon A cb ahe no eob gence A A ed cm aahas me n aheu eme gence chod A o ba examp ed ACS ABC cgo eapo ocompu cm examp and ebeen d ABC earranged ced and hyb d methods. methods. The LF The impact on the on basis the of basis interactions, of interactions, aiming aiming at MOPT methods. at MOPT problems The problems LF have impact have on the arranged as basis aas of as interactions, aaechn aiming MOPT prob methods. The impact on the basis of aiming at MOPT problems have arranged aOthers mo emu mo cba eng eMCS me cThe hod hod mp MH mp me eaequ MH hod me equ eo mo equ eap ecan mo end compu eca ahyb on cde en me aw on me o mo o numbe eaNLP numbe hod equ eme mo eat compu aACS on mo eSQP eaheu cd en me hod The mp MH hod equ eme mo epo compu aeACS on me A mo numbe con de ed de ang ed ng aDynP aom m ng ad ng nd op ng op we ma gh con we o gh de FDN ed oof FDN m ng am ng op ma we gh oana con de ed m ng amo nd ng op ma we gh o FDN SQP and ceob p og amm and dynam eLF he p eme og amm en nume ng eo ca SQP he me mo and hod egh dynam cethe en ccompu nume p og amm me ng enco he eona cob en nume ca hod op m on can op be con m za de on ach p hod om can ng be op o de dea m aThe za w p h on comp me ng hod ex way p dea ob con em w de mo comp ep om ex MOP ng way p em o dea mo w eand h comp ex p ob mo big presented in Table 9. The overall big picture, performance has presented comparison in Table of the 9. The addressed MOP performance comparison addressed bigpowe picture, has big in Table 9. presented overall in Table 9. The comparison overall addressed comparison MOP of the addressed Howeve he ob em o on ve y g d and ew va Howeve ay on can be p nco ob em auncon o ed Dynam ahyb on cA ve y go d and ew va aaheu on can be nco po aem Howeve em oog Howeve amu on he ve p y ob em g Howeve o and mu ew aex on va ap ve ob can g o be and nco ew on va ame ve aem y on Dynam g can d be cLF ew po abe ed on Dynam can be nco po aon ed Dynam cL on equ ed opicture, ach eve nea h o gh unc qua on y ehe u A ed oba o con ach ned nea and o yo h unc gh on ned y equ ede u ed A o con eve nea abe y h gh uncon qua y aehave uphod A onea con am o unc on equ ed o ach eve nea yinteractions, h gh qua y eu u A o con abeen ned and uncon ned The The mod ed LF me ed LF hod and hod o and wa o ba wa ed The eeand ba LF mod p ed aacompa LF o ed ahm LF op ov m de hod p be ov and de ehod be pe o ehave o wa pe mance eThe ba o ed m LF ng ap m ng o m p ov de be ened pe oaed mance The mod ed LF me and wa eaheu ed LF p p ov be ePSO pe o mance m ng Application in Me Me aheu cimpact ccd me MH hod me ke hod evo ke u ona evo Me y aheu aalgorithms, ona y co go MH have hm me have hod con ke con evo ed aA u de aach ed aFDN u aop go na u been con de aDyn nam Me aheu cLF MH me hod ke evo u ona y go been con amo na equa c10. y aaThe u pe pec e10. ve con en y hyb om d aeve aheu eParameter pec cGA echn ve The que hyb how d me obu aheu ne h gh que how obu ne equac en e pe pec ve The hyb dhod me aheu caob echn que how obu ne hch gh techniques has been presented in Table techniques has presented in techniques Table 10. has been presented in Table 10. p amm and exhau ve ea ch ES ahod hough hey om equa o nd ahod g oba op mum og amm ng DynP and p exhau og amm ve p ng og ea amm ES and ng aperformance DynP exhau hough and ve hey exhau ea p om ES ve ed o ea nd ch ES g oba hey ade hough op p om mum hey eh oame p nd om anva g eo oba o nd op ah mum g oba op mum mu ob ec ve pClassification/Execution em mu aex ehoweve cu ec ve pnd op ob m em ze aeway Sop eA m d ao cu y mu GA od op ob ec m PSO ze ve Sexamp p ov m ob ape y nea anne op d and ma cu o pmo op de m ze nea S m op ad ma yh GA and PSO ov de op mu ob ec ve p ob em anne ecREG d cu op m ze She m y GA and PSO p ov de nea op ma a unce a unce aop n o aDynP n oad oy and oad REG and REG ao gene on a aem n unce on compa n aperformance n y on ohm o oad and ona REG ad ona gene mode LF ade mode on A o compa A hey ou hey ogh ad ona LF mode oagh a unce aen n o oad and gene ap n compa on o ad ona LF A oay hey way o add eme ng comp way MOP o add p ob eMH em ng comp Howeve MOP he h way gh ob em compu o add Howeve ahough ona ng comp equ he h emen gh ex MOP compu and au ob ona em Howeve emen he and compu aMOP ona equ emen way o add ege ng ex p ob em Howeve he h gh compu aw ona equ emen and o u on p om egene powe u qua g oba y o op u on m za p om me eng hod powe and g oba ab op y m oaMOP za on ehm me hod and ab yed add y yo u om on a up uom ep uehoweve ged oba op za on me and ab y add eowa om u on hey have no advoca ed o o on amu howeve ge d bu hey on have y em no advoca L kew ed eao amance ooca ma ama o ge bu on yneed kew eAoeh aM oov uo on hey have o u on no howeve ed hey o have aoca ge no d advoca bu on ed y oaoca em acompa ge L d bu ead aMOPT on ma y o L kew eM aNLP oamo Table Table Performance 10. Performance comparison comparison of algorithms, of algorithms, applied applied in multi-objective in multi-objective planning planning problems. problems. aadvoca bu on yMOP em o am ge d bu on y em Table 10. Performance Table 10. Performance comparison of comparison algorithms, of in multi-objective applied in multi-objective planning problems. planning problems. ge ydynam p ov gene agence on o va ou cena o unde au oad mode wh we eme m ed conven ona p de gene aMCS on op va ou cena p ov de gene unde acomp on o va mode ou cena wh ch ointeractions, we unde eu m ed oad n conven mode ona wh ch we emode m ed n conven ona po b yabeen o pd ema u ede po conve bem y od p owa ema d u eeeve conve op ma gence po b d need o oca p o ema op be ma u add em conve o eon u ed on gence need h owa o be d add ep ed ma w h o ume on o be add ed Algorithm/Methods Compute Cost Quality po b o p ema eu conve gence owa op ma o u need o be add ed w h con aooza ned me p ann con ng p ned ob em ea Fu me he p ann mo ng eadea dec p ob on em mak Fu ng he me mo hod ehod con dec ao oo aA need ned mak o ea ng be me me hod p ann ng o need ob ocon Fu he mo eem dec on mak ng con amod ned ea p ann ng p ob em Fu he mo eu dec on mak ng me hod aov o need o be d awback MCS be ng op m za on d awback he h gh o compu MCS be aon ng on me op n compa m za on on he gh ahave on me n compa on N d awback be ng d nne awback m o za MCS on be ng he h gh compu m za he n h gh compu on ach M on n compa M NLP A coco ay A ccimpact ep ay gence nequ ehyb gence A ba ed me ba ed aheu me aheu cbeen o cap examp ep ACS ewe ABC cinteractions, and eequ ced and d hyb d cLF acimpact n enne gence A ed me aheu ckew examp eABC ACS ABC en chyb and hyb d methods. methods. The LF The LF impact on the on basis the of basis interactions, of methods. aiming The LF aiming at MOPT at on problems the basis problems have of arranged have arranged as aiming ade abe MOPT problems have arranged methods. The LF on the basis of aiming at MOPT problems have arranged as acompu mo eMe cec en me hod The mo eon mp ey en ccu MH en me hod The equ ehod mp mo eba eaMH compu mo eand eequa hod cy on en equ me A mo o epicture, mo The compu eNLP numbe mp eThe on MH me hod A o mo equ ePSO numbe eadd eat compu ade on me A ow mo nu con de ave m ng abu nd ng op ed ma aDynP m we ng gh ahave nd o FDN ng op ma con we gh de ed o FDN ng ao nd op ma gh oaed FDN con de ed ahod m ng apresented nd ng op ma we gh o FDN SQP and p og amm ng aeve emore he mo SQP ecomp cand en dynam nume cimpact ca p og me amm hod aem emode he mo eon caheu en nume ca hod SQP and dynam cea og amm SQP ng and ahas dynam eA he mo p eu og coad amm en nume ng ehey ca he me mo hod eme capplied en nume ca me hod op on me hod can be con de aThe p om way o dea w h comp ex MOP op p ob za em on mo me ep hod can be con aThe p om ng way dea haemen comp op mmu za on me hod can op be con m za de on me om hod ng can way be con o de ahm w p h om ng ex way MOP o p dea ob em w h mo comp ePSO ex MOP ob em mo eas big picture, big picture, has presented has in Table in Table 9. The 9. overall The overall performance performance big comparison comparison has of the presented of addressed the addressed in Table 9. MOP overall performance big picture, has presented in Table 9. The overall performance comparison of the addressed MOP o unc o on unc on equ ed o ach ed o ach nea eve y h nea gh qua yechn h gh y qua ehough y A eLF o u unc con on o aeng con ned equ and aproblems. ned ed uncon and o ach uncon aMOP eve ned nea aem ned y h gh qua yhow econ uoso A o con am o unc equ ed o ach nea y h gh y eu u A o con abeen ned and uncon akew ned The The mod ed me ed LF hod me and hod o and wa ohey einteractions, ba wa ed The ba LF mod p ed aacompa LF op ed p m atde p oo me ov m de hod p be ov and de eaaACS be pe o eov o wa pe mance ep ba o mance m ng ap m ame ng omulti-objective m p ov be eab pe ogcomparison mance agh The mod ed LF me hod and o wa emp ba ed LF p oein m ov de be em pe o mance aLF m ng Depend Planning Problems Features Algorithm/ Algorithm/ Classification/ Classification/ Compute Compute Problem/Solution Problem/Solution Parameter Parameter Application Application in in Others Others more eff c ent methods The s mp e more MH methods eff c ent methods requ re The computat s e MH on methods me A requ so more re more number computat on t me A more num Algorithm/ Classification/ Compute Problem/Solution Parameter Application in Others methods requ re more computat on t me A so more number aheu c MH me hod Me aheu ke evo c u MH ona y me a go ke have evo Me u aheu ona con y a c go MH ed hm a me have a hod na u a ke con evo u de ona y a a go na u hm a been de ed a na epo cm en y om a u u e e pe pec en y ve om a u u d e me pe aheu pec e ve c en c The y hyb om que a me how aheu pe obu pec c ne echn ve h que gh hyb how d me obu ne h c gh echn que obu ne h techniques been presented in Table 10. techniques has been presented in Table 10. techniques has been presented techniques in Table has 10. been presented in Table 10. p og amm ng DynP and exhau ve ea ch ES a hough p og p om amm e ng o DynP nd a g and oba exhau op mum ve ea ch ES a hough hey p om e o nd a oba op p og amm ng DynP and p exhau og amm ve ng ea ch ES and p a og exhau hough amm ng ve DynP ea p om ch ES o exhau a nd g ve oba hey ea op p ch om mum ES e a o hough nd a g hey oba p op om mum e o nd a g oba op mum ob p ob em a e d op mu m ze ob S ec m ve a p y ob GA em and a e PSO d cu ov mu de o op nea ob m ec op ze ve ma S a y GA a e and d cu p o ov op de m nea ze S op m ma a y GA and PSO mu ob ec ve p ob em a e d cu o op m ze S m a y GA and de nea op ma a unce a unce a n o a n oad y o and oad REG and gene REG a gene on a a n unce on compa n n compa y on o o oad on ad and o ona REG ad LF ona gene LF a mode on A n o compa A hey o hey on o ad ona LF mode A o a unce a n o oad and REG gene a n compa on o ad ona LF mode A o hey way o way add o e add ng comp e ng ex comp MOP ex p MOP ob em p way ob Howeve em o add Howeve e he ng h gh comp he compu h gh ex MOP compu a ona p a ob equ ona em emen equ Howeve emen and he and h gh compu a ona equ way o add e ng comp ex MOP p ob em Howeve he h gh compu ona equ emen and qua y o u on p om e powe qua u y g oba u on op m za om e me powe hod u and g oba have op ab m za y o on add e hod and have y o add e qua y o u on p om e powe u g oba op m za on me hod and have ab y o add e o u on howeve hey have no advoca ed o a ge d bu on y em L kew e ma o o u on howeve hey have o u on no howeve advoca o u on ed hey howeve have a ge no hey d advoca have bu on no ed y advoca o em a ge ed L kew d o bu a ge a on ma d y o bu em on L kew y em e a L ma o e a ma o Table Table 10. Performance 10. Performance comparison comparison of algorithms, of algorithms, Table applied applied 10. in Performance multi-objective multi-objective comparison planning planning of algorithms, problems. applied in planning problems. o u on o a ge d bu on u on y em o a ge d bu on y em o u on o a ge d bu on y em Table 10. Performance comparison of algorithms, applied in multi-objective planning problems. o a ge d bu on y em p ov de p ov gene de a gene on o a va on o ou va cena ou o cena unde o p ov unde a de oad gene a mode oad a on mode wh o va ch wh we ou ch e cena we m e ed o m n unde conven ed n a conven oad ona mode ona wh ch we e m ed n conven p ov de gene a on o va ou cena o unde a oad mode wh ch we e m ed n conven ona (Optimization) b y o ppbe ema uhod eand po conve b gence y omo p owa ema d uhe oca eThe conve op ma gence po owa b need ooca pDepend o ema op be ma u add conve o ehyb u ed on gence need h owa be d add oca eneed op ma w h ocomparison ume need o be add eneed ed po b oLF p ema ebe conve gence owa d op ma ong u on need o be add ed w h con aen ned ea me ann ng p ob con em ahod he mo ecehod me dec p ann ng mak p ng ob me em hod Fu he o mo eed ong be mak ng me hod oon con a ned ea me p ann ng p ob em Fu he mo eawback dec ng me hod aode oop need owe be d awback ounc MCS be ng nne op m za on d he h gh o compu MCS be aw on nne me op n compa m za on on M he NLP h ahave on me n compa M d awback MCS ng d nne op m oza za MCS be ng he nne hcp gh op compu m za on me he n gh compu on aDepend M on n compa M A con ay een gence A ba me aheu A cdned ch acompa o n examp gence ep ACS A ba ed eThe me cem aheu and hyb cMOP ocompu eoon ABC ecaddressed cechn and hy Aand coom aon naom gence A ba aheu examp eQuality ACS ABC eway and d me The mpac ba o n ac on acay m ng apresented MOPT p em have aoMOP anged aen ad me hod The LF mpac he me ba hod ehyb ac LF on mpac ang m ahm he MOPT ba p ob n eaoca em ac on have acco m anged ng acompu MOPT aApplication aon p ob have amo anged aexamp aA mo eaheu edynam en me hod The mo emp eand en cmak MH me hod equ mp mo ep eaMH compu mo me eov eFu on en equ me A mo o eeh The mo eNLP numbe mp aw eaProblems MH on hod A o equ ema enumbe mo em compu aACS on me A mo eaque num mo ehod eop cop me hod The mp eacon MH me hod equ eaea mo ey compu aob me o mo eamo numbe con de ed m ng aon nd con ng de ed ma apresented we m gh ng ain o nd FDN ng ma gh o FDN con de ed aaABC m ng aA nd ng op we gh FDN con de ed aThe m ng apresented nd op ma we gh o FDN SQP dynam con p og amm ng aon eed he mo SQP eawback cand en dynam nume ca p og me amm hod ng eov he mo eon cob nume ca hod SQP og amm SQP ng and ahas dynam eed he cpresented p og camm en nume ng eProblem/Solution ca he me mo hod con en nume me hod op m za op m me hod me can hod con can be de om a10. p ng om way ng o way dea o dea comp w h ex comp MOP ex p ob em p em ehey mo eop m za on me hod can con de an p om ng ocomparison dea w h comp ex p ob em mo eon big picture, big picture, has presented has Table in Table 9. 9. overall big The picture, overall performance performance comparison comparison in Table of the 9. of addressed the overall addressed MOP performance of the M big picture, has in Table 9. The overall performance of the addressed Methods Methods Execution Execution Cost Cost Quality Quality (Optimization) (Optimization) Depend Planning Planning Problems Features Features Methods Execution Cost (Optimization) Planning Problems Features o equ ed o o ach eve on nea y equ heu gh ed qua o ach ybe ed eve u nea o A y unc o con gh qua ahas ned equ y and eLF u ed uncon o A o eve con ned nea aob ned y h and gh qua uncon y eMOP adec u ned o con aem ned and uncon aN The mod ed LF me hod The mod oThe wa ed eLF ba me ed hod LF p and aThe o oem wa p The eh ba de mod ed be LF ede ed p pe aahm oca me m mance hod pach ov aeome and m de ng be o eov wa pe eMOP ba o mance ed LF ap m ame o10. p ov de be eab pe ooaed mance ana m The mod ed LF me hod and o wa eo ba ed LF p amulti-objective m ov de be ePSO pe o mance ay m ng Algorithm/ Algorithm/ Classification/ Classification/ Compute Compute Problem/Solution Parameter Parameter Application in in Others Others Algorithm/ Classification/ Algorithm/ Classification/ Compute Compute Problem/Solution Problem/Solution Parameter Parameter Application in Application Others in Others Me aheu cn MH me hod ke evo u ona Me y aheu go hm MH have me been hod ke de evo ed abu u aproblems. ona na u aNLP agh go hm been con de aemen Me cceaon MH me Me aheu ke evo c u MH ona me y a hod go ke evo have u been ona con y a go ed a have a na been u a con de ed na u equa cunc en yco ahey u u e pe pec ve The hyb d me aheu c echn que how e obu c en ne y om h gh a u u pe pec ve The hyb d me aheu e c en y a u u e pe e c pec en ve y om a u u d e me pe aheu pec ve c The echn hyb que d me how aheu obu ne c echn h gh que how obu ne h gh techniques techniques been has been presented in Table in 10. Table techniques has been presented in Table techniques has been presented in Table 10. mu ob mu ec ve ob p ec ob ve em p ob a e em a cu e o cu op m o ze op S m m ze S y m GA mu a and y ob GA PSO ec and ve p de p nea ov a de op e nea d ma cu op ma o m ze S m a y GA and PSO mu ob ec ve p ob em a e d cu o op m ze S m a GA and PSO p de nea op ma a unce a unce a n o a n oad y o and oad REG and gene REG a gene on a a unce on compa n a n compa y on o o oad on ad and o ona REG ad LF ona gene mode LF a mode on A n o compa A hey o hey on o ad ona LF mode A o a unce a n y o oad and REG gene a n compa on o ad ona LF mode A o way o add e ng comp way ex MOP o add p ob e em ng comp Howeve ex MOP he h p way gh ob compu em o add Howeve e ona ng comp equ he h emen gh ex MOP compu and p a ona em equ Howeve emen he and h gh compu a ona equ y u p om qua e powe y o u u on g oba p op om m e za powe on qua me u y hod g o oba u and op have m p za om ab e me y powe o hod add u and e g oba have op ab m za y o on add e hod and have y o add e o u on howeve hey have no advoca ed o a ge d bu o u on on y howeve em L kew hey have e a ma no o advoca ed o a ge d bu on y L kew e a m o u on howeve have o u on no howeve advoca ed hey o have u a on ge no d howeve advoca bu on hey ed y o have em a ge no L kew d advoca bu a ed on ma y o em a ge L d kew bu e on a ma y o em L kew e a ma o Table 10. Performance comparison of algorithms, applied in planning problems. o u on o ge d bu on y em u on o a ge d bu on y em o u on o a ge d on y em Table 10. Performance comparison of Table algorithms, 10. Performance applied in comparison multi-objective of algorithms, planning applied problems. in multi-objective planning o a ge d bu y em p ov de p ov gene de a gene on o a va on o ou va cena ou o cena unde o p ov unde a de oad gene a mode oad a on mode wh o va ch wh we ou ch cena we m e ed o m n unde conven ed n a conven oad ona mode ona wh ch we e m ed n conven p ov de gene a on o va ou cena o unde a oad mode wh ch we e m ed n conven ona po b po y o b p ema y o u p e ema conve u e gence conve gence owa po d owa oca b d y op o oca ma p ema op o ma u u e on conve o need u on gence o need be owa add o be e d add ed oca w e op h ed ma w h o u on need o be add ed po bng o p ema ebe conve gence owa d ma oNo u on need oo be add ed w h con ned ea me ann p con ob em aeop ned Fu ea he mo me p ann dec ng p mak ob em ng me Fu he mo eFeatures need on oMOP mak ng me hod ahave oe op need be a qua ned me p ob em Fu he mo eawback dec mak ng me hod oop need owe be d awback ounc MCS be ng nne op m za on he h gh compu aew on me naACS compa on M NLP d awback o be ng d nne op o d za MCS awback o ng he nne hp gh op compu ng m nne za aMH on m he za n h compa compu he on aed gh M on compu NLP me aABC n on compa n compa M on M NLP cgh a hod eeve gence A ba ed au me n aheu eme gence cabe A ba examp ed me eMe A aheu ACS con ABC cmo n o eMH gence ch examp and A hyb eng ba ed ABC me aheu eem canged and cne hyb examp d eed ced hy A cm gence A ba cgo ohm examp ebeen ehod and hyb me hod me The hod LF mpac LF mpac he on ba he o ba n em o me ac n on ac aconstra The m ng LF ape au m mpac acan MOPT on p ob he em ba p ob have em n ame have eced anged aProblems on anged aen aNLP au ng au abe MOPT pLF ob em amo anged me hod The LF mpac on he ba o n ac on aaon ng amo MOPT p em have aac mo eA edynam cMCS en me hod The mo eon mp eon ehe cu MH me hod The equ mp mo ep eme compu mo edov e❉ hod aoca caom on en equ me hod A oNo end The mo numbe mp eaProblems MH on hod A oFeatures equ ene em numbe eyaOpt. compu aACS on me A eaonum mo ehod eb ccy en me hod The mp eaQuality MH hod equ em ep compu aACS on me A o mo eze numbe con de ed m aCost nd ng ma con gh de ed o aop FDN m ng am ng ma we gh o FDN of funct ons sann requ red to of near funct y h ons gh qua scomp requ ty red to ts ach A eve so near y h ned gh and qua unconstra ty resu ts ned A so constra ned and unconstra con de ed a ea mty ng a pnd ng op ma we oog FDN y con h gh resu ts A so constra ned unconstra ned SQP and dynam p og amm ng he mo SQP eona cand en dynam nume ceo ca og me amm hod ed he mo eon co nume ca hod SQP pn amm SQP and ahas dynam eaob mo cepresented og cMCS amm en nume ng eed ca he me mo hod eme con en nume ca me hod op m za me hod can con de om op m ng za on me o dea hod h comp be con MOP de amode p p ob om ng mo way ehey o dea wme hcon comp ex MOP puncon ob em op m za me can be con aThe p om ng way o dea w h comp ex MOP p ob em eona High ✘ High ✘ Linear/better ❉ Simple Simple ✮ ✮ b p cand u eA ha p en ed n Tab eHigh 9ng The ove pe o mance compa on o he add eu ed MOP ε-constd. ε-constd. N/Easy No Inner Opt. Inner ✘ Linear/better ❉ Simple ✮ b gng pε-constd. coand u eThe ha ed n Tab g eadd p 9eve caThe en ove ha aeu epe en o ed mance n Tab compa 9Algorithm/ The on ove he add o MOP compa on o he add edec ed ε-constd. N/Easy No Inner Opt. Methods Execution Execution Cost (Optimization) (Optimization) Depend Depend Planning Planning N/Easy High Linear/better Simple Methods Execution Execution Cost Quality Cost Quality (Optimization) Depend Depend Planning Problems Planning o unc on equ ed oMOP o nea ypbe equ h gh ed qua ach y ed eve nea o A yaand unc ohod h gh qua agh ned equ y and eMOPT ed uncon A ach o eve con aex ned nea aproblems. ned y h and gh qua uncon y eOpt. am ned ocompu ned and a o unc equ o ach eve nea y h gh qua y A ocompu con ned and uncon ned mod ed LF me hod The and mod oem ed ep LF ba me ed hod LF p ao and o o p wa econ de ba be ed LF ede pe p ang o oo mance The pemo mod ov aop m de ng ed be LF eProblems me pe hod on mance and omo wa ng eA ba p aABC m ov de The mod ed LF hod and oLinear/better wa eaheu ba ed LF p amulti-objective p ov de be eque pe o mance ana m ng Algorithm/ Algorithm/ Classification/ Classification/ Compute Compute Problem/Solution Problem/Solution Classification/ Parameter Compute Application Application in Problem/Solution in Others Others Parameter Application in Oth Classification/ Compute Problem/Solution Parameter Application in Others Me aheu cccoen MH me hod ke evo u y aheu cov have me hod ke de evo ed abu ona na u aFeatures aop go have been con de aand na Me aheu cceon MH Me aheu ke evo MH ona me y go hm ke evo have u been ona con y go ed hm amance have am na been u aob con de amo aoad etechniques cAlgorithm/ en eMethods y en om y aPerformance om u eema ang pe u u pec eaon pe ve pec The ve hyb The hyb me aheu d me aheu cParameter echn cch que echn que obu how ne obu h h gh each en y om aon u u eohod pe pec ve The hyb d me aheu echn how obu h gh techniques been has been presented in Table in 10. Table techniques has been presented in Table 10. techniques has been in Table 10. mu ec ve p ob em mu anne eon d ec cu ve o ppresented ob em ze acresu S eeza m au cu y mu GA o and op ob PSO ze ve p S p m ob de au em y nea GA aou eREG op d and ma cu PSO o p op ov de nea Sagh m op aon ma GA and PSO pocompa ov de nea op ay unce aPerformance npach y oen oad and a REG aMethods gene y o ad on oad n and compa REG on a oon a(Optimization) on ad ona n aec n compa y LF o mode oad on A ad o hey gene LF aowa mode on n compa A hey on ogh ad ona LF mode obe a unce ade n o oad and REG gene aeQuality on n compa on o ad ona LF A oy way add eThe ng comp ex MOP p ob em way Howeve o add ein he h gh comp compu ex MOP aob p ob equ em emen Howeve and he hFeatures aeaem ona way add eyMCS comp way ex o p ob eon ng Howeve ex he p h ob gh em compu Howeve away ona equ he h emen gh compu and ona emen and qua yon o pggence om egene powe uN/Easy g oba op m on me hod and have qua yadd y ocp o add u on ehave p om eam powe u ghm oba op m za on me hod qua oA u on om qua eme powe yunce u u on g oba p op om m emp za powe on me ugene hod g oba have m za ab on y me eplanning ab y o add eaInner Table 10. Performance comparison of algorithms, Table applied 10. Performance comparison of algorithms, applied in multi-objective planning problems. obe u on u ocen on agence ge d awa ge bu d on bu yMOP em on y em o u on o aahow ge d on y em Table 10. comparison Table 10. of comparison in multi-objective of algorithms, planning problems. in multi-objective planning problems. onne on aalgorithms, ge bu y em p de p ov de aSQP gene o accLF va on o ou va cena ou o cena unde o p unde aunce oad gene mode oad ap on wh o va wh we ch ehyb cena we m eA ed o m n unde conven ed aanged conven oad ona mode ona wh ch we m ed n conven p ov de gene an on o va ou cena o unde aop mode wh we eona ed n conven ona po bdynam o ppn ema eo po b y o owa p d u oca eHigh conve op ma gence po o u owa on b d y need o oca o op be ma add u eand conve ep o u ed on gence w need h add oca eneed ed ma h o u on need o be eA ed con aob ned ea me p ann con ng aop p ned ob em ea me he p ann mo dec p con ob em a10. mak ea ng he me mo me hod eDepend p dec acMOPT ng oo on need mak em o be me Fu hod he mo o e✮ dec mak ng me hod aexamp oequ need oed d awback ou MCS ng m on he gh compu d awback on me MCS n be ng nne M op NLP m za on he h ap on on M N d awback o be ng d awback m o za ng d he awback nne h gh op compu o m MCS za ame be ng nne n h compa gh op compu m za on ach on M NLP me he h n gh compa compu on on M M con ang eSQP ba ed me aheu A cmance o n examp eMe ACS A ba ABC ed e❉ me aheu A and cov acon cde n d egh gence examp eon ACS ba ed ABC me aheu cechn and cme hyb oNLP d eaemen ACS A con aProblem/Solution n A ba ed aheu cgo o examp ABC eequ ced and hyb d me hod me The hod LF The mpac mpac on on ba he o ba o me ac n hod on ac ay The on m ng LF amo ang m mpac MOPT ane MOPT on ob he em ba p ob have o em n aoMOP have eove anged ac aA anged au amo aou amo abe aInner MOPT p ob have amo me LF mpac on ba o n egence ac on aa❉ ng amo p em have ad acompu eε-constd. eu mo eon me cme en hod me The hod The eah MH mp me eche MH hod mo me ed equ hod caaec en eo mo equ me ecompa compu eed hod ehod The compu aon on mp me aaSimple ewe MH on me o mo hod A ede o equ eme em numbe mo ecompa compu aon on me A oadd eand num mo econve each co en me hod The mp MH hod equ em eab compu aeACS on me A o mo eamo numbe con de ed ap m ng aCost nd ng con ma de we ed gh ahe m oed FDN nd ng op ma o FDN con de ed a med ngLF a me ndhod ng op ma we gh FDN ✘ Complex/better ❉ ❉ ✮/Complex ✮/Complex ✮ ✮ MCS MCS N/Difficult N/Difficult No Inner and dynam p og amm eREG mo eaeov cde en nume ca me hod Complex/better ❉ Simple ✮/Complex ✮ MCS N/Difficult No Inner SQP and og amm SQP ng and aha dynam eo he mo and pbe dynam eem og capplied amm en che nume p ng og aeeed eComplex/better amm ca me mo ng hod ao eme en he capplied en mo nume eema cann ca en me nume hod ca me hod op m za me hod can op be con za de aza me p hod om can ng way be con og dea de op aLinear/better w m h om za comp on ng me ex way MOP hod p can dea ob be em w con h comp de eSimple aThe ex p p ng ob way em oonhey dea eon w hcon comp ex MOP puncon ob em op m za hod can be con de agene p om ng way omode dea w h comp ex MOP p ob em eng High ✘ High ✘ Linear/better ❉ Simple ✮ ✮ b gov p cque b u g eA p cThe ha u eMCS p ha eequ en ed eFu n Tab ed n eHigh Tab 9ng The eon 9nea ove b The g p ove cFu pe u aoad eona o ha pe mance p o emance compa en ed compa n on Tab o eng he 9m o add he ebe add anumbe eMOP pe ed o MOP compa on on he add eanged ε-constd. ε-constd. N/Easy N/Easy No No Inner Opt. Inner Opt. High ✘ High Linear/better ✘ ❉ Linear/better ✮ Simple b g p cAlgorithm/ u ep ha eom en ed n Tab eaop 9✘ The ove pe o mance compa on o he add eed ed ε-constd. N/Easy N/Easy No No Inner Opt. Opt. Methods Methods Execution Execution Cost Quality Quality (Optimization) Methods (Optimization) Execution Depend Depend Planning Cost Planning Problems Quality Problems (Optimization) Features Features Depend Planning Problems Featu MCS N/Difficult High Complex/better No Simple /Complex Opt. Methods Execution Cost Quality (Optimization) Planning Problems Features o unc on equ ed o o unc eve on nea y equ h gh ed qua o ach y ed eve u o A yaned unc h gh on qua acompa ned equ y and u ed uncon A ach o eve amo ned nea aproblems. ned y hop and gh qua uncon y eOpt. am uw ned A ocompu a con ned and aM o unc oed ach eve nea y h gh qua y ede u A oSimple con ahow ned and uncon aMOP ned mod ed LF me hod and o wa The eacon ba mod LF ed p LF ang ome m hod p and be o wa ehave pe eom ba ed mance LF p aInner amance oaOpt. ng m ov de be em pe oobu mance abe m mod ode wa ba p ahod oop m p ov de be een pe o aaeof ng Algorithm/ Classification/ Compute Problem/Solution Parameter Application in Others Algorithm/ Classification/ Compute Classification/ Compute Parameter Problem/Solution Application in Parameter Others Application in Others Me aheu ccon MH me hod ke evo u ona y aheu hm cov MH have me been hod ke de evo ed aemen u aoan ona na y aNLP agh go hm have been de ed aona na aheu ccaap MH hod Me aheu ke evo cep ona me y aand hod go hm ke evo have u been ona con y go hm ao have abu na been u aob con de na equa cme en y om aPerformance u u egence pe pec ve The hyb em cd en me aheu om cin u u echn eu pe que pec ve The obu hyb ne d me h gh aheu cewh que how ne eprob cand en y om ave uo u eLF pe pec ve The hyb d me aheu echn que how obu h gh echn been eMH en n Tab ehe 10 echn que ha been pcomp emo en ed n echn Tab eque 10 ha been p en n Tab 10 mu ob ec poed ob em mu a e d ob ec cu ve o p op ob m em ze a S e m a cu y mu GA o and op ob m PSO ze ve p S p m ob de a em y nea GA a op d and ma cu PSO o p ov m ze nea S m op a ma y GA and PSO p ov de nea op mu ob ec ve p ob a e d cu o op m ze S m a y GA and PSO p ov de nea op ma aMe unce ap nePSO y oad and a unce REG gene n y a o on oad n compa o a ad on ona n LF mode a on unce o A a ad o n y hey ona o oad LF mode and REG A gene o n compa on o ad a unce a n y o oad and REG gene on compa on o ad ona LF mode A o hey way o add e ng comp ex MOP p ob em way Howeve o add e he ng h gh comp compu ex MOP a ona p ob equ em Howeve and he h gh a ona equ emen way add e ng way ex MOP o add ob e em ng comp Howeve ex MOP he p h ob gh em compu Howeve a equ he h emen gh compu and a ona equ emen and y qua o u y on o u p on e p om powe e u powe oba u op g oba m za op on m me za on hod me and hod have and ab have y ab o add y e o add e qua y o u on p om e powe u g oba op m za on me hod and ab y o add e Table 10. Table Performance 10. Performance comparison comparison of algorithms, algorithms, applied applied in multi-objective multi-objective Table planning 10. Performance planning problems. comparison of algorithms, applied in multi-objective pl o u on o ge d bu on u on y em a ge d bu on y em o u on o a ge d on y em Table 10. comparison of algorithms, applied in multi-objective planning problems. p ov gene on o va p ou ov de cena gene o a unde on o va a ou oad mode cena o p wh ov unde ch de we gene a e oad a m on mode ed o n va conven wh ou ch cena we ona e o m unde ed a conven oad mode ona ch we ed n conven p ov de gene a on o va ou cena o unde a oad mode wh ch we e m ed n conven ona po b y o ema u e conve gence owa po d oca b y op o ma p ema o u u e on conve need gence o be owa add e d ed oca w op h ma o u on need o be add e ed mu t -ob ect ve ems are d ff mu cu t t to -ob opt ect m ve ze prob S m ems ar are y GA d ff and cu t PSO to opt prov m ze de near S m ar opt y ma GA and PSO prov de near opt po b y o ema u e po conve b gence y o p owa ema d u oca e conve op gence ma o owa u on need oca o op be ma add o e ed on w need h o be add e ed w h con a ned ea me p ann ng p ob em Fu he mo e dec on mak ng me hod con a o a need ned o ea be me p ann ng p ob em Fu he mo e dec on mak ng to optThe m ze S m ar y GA and prov de near opt ma con a ned ea me p ann con ng p a ned ob em ea Fu me he p mo ann e ng dec p ob on em mak Fu ng he me mo hod e dec a o need on mak o be ng me hod a o need o be A ccu aA n ap gence n gence A ba A ed me ba ed aheu me aheu cba o examp ohe examp eaDepend A ACS cov accon ABC n egh eeThe ABC ced gence and e❉ canged hyb and ba d ed hyb me dcompu aheu ceon oned examp ehaeanged ACS A con aeha n gence A ba ed me aheu cgo o examp ebeen ABC eaadd cp and hyb me hod me The LF The mpac mpac on on ba he on n o n hod on eove ac aacm The on ng LF acbeen m mpac MOPT ng amo MOPT on p ob he em ba p ob o em n have eove anged aA on anged aen amo ade aau ap MOPT ob em have amo hod The LF mpac on he n eove ac aang m ng MOPT p em have aac ad High ✘ High ✘ Linear/better Linear/better ❉ ❉ Simple ✮/Complex ❉ Goal P N/Difficult N/Difficult No No IMO mo eaheu eand cec en The mo mp eon cve MH en me hod The equ eke mp mo e9nea MH compu mo emod equa hod aon en equ me me A hod o eFeatures mo The compu en numbe mp aACS on MH me A oFeatures equ eIMO numbe eop mo apem me A owuncon eInner nu High Linear/better ❉ ❉ Goal P N/Difficult No IMO con ed mpMH ng agGoal nd ng op de ed ma we m ng aN/Difficult o nd FDN ng con ma de we ed gh abe o FDN nd ng op ma we o FDN High Complex/better Complex/better ❉ ❉ Simple ✮/Complex Simple ✮/Complex ✮ ✮ MCS N/Difficult Inner Opt. dynam chod p og ng ave eha he eCompute cn en nume ca SQP and hod dynam ceed p og amm ng aeProblems e✮ mo eLF caeMOP nume ca me hod High ✘ High Complex/better ✘ Complex/better ❉ ❉ Simple ✮/Complex Simple ✮ ✮/Complex ✮ MCS N/Difficult MCS No No Inner Opt. Opt. SQP and dynam og amm SQP ng and aha dynam eob he mo cep p eem og SQP camm en and nume ng dynam a✘ eme ca he me cem mo p hod og eo amm cque en ng nume ao em ca me hod eeACS en nume ca hod op za on me hod can be con aThe p om op m ng za way on o dea hod w can h comp con ex de MOP amo op p m om ob em ng on mo way me e9have hod o✮/Complex dea can w be con hamo comp ex aIMO MOP om p ob ng way oodea eov comp op m za me hod can be con de aThe p om ng way ome dea w h comp ex MOP ob em eng High ✘ High ✘ Linear/better Linear/better ❉ ❉ High Simple Simple Linear/better ❉ Simple ✮ b p cPque b u g ehod p cThe ha uke ede p ha eequ en ed eeo n Tab ed n eMH Tab 9aheu The eo ove b g acme p chyb pe u aon eona om ha pe mance p o emance compa en ed compa on Tab oh ewa he The o add he eand add ed ame pe ed om MOP compa on ohow he add ed M ε-constd. N/Easy N/Easy ε-constd. N/Easy No No Opt. Inner Opt. No Linear/better ❉ Simple ✮ b go p cMCS eop eExecution en ed n Tab eop 9✘ The aob pe o mance compa on o he ed ε-constd. N/Easy Methods Execution Cost Quality (Optimization) Depend Planning Linear/better No Simple /Complex Goal P N/Difficult High Methods Execution Cost Methods Quality Depend Cost Quality Planning (Optimization) Problems Planning Problems o unc o on unc on equ ed equ ach ed o ach nea eve y h gh o qua y unc h gh on y qua ey u equ y A u ed o con A ach o ae✘ eve ned nea and aproblems. ned y uncon h gh qua uncon y egh am ned A ocompu con amo and am o unc o ach eve nea y h gh ede u A oen con aSimple ned and uncon aMOP ned mod ed LF me hod and o The wa eque ed LF p hod and p o de be eao ba pe ed mance p aned o aFeatures m p ng ov de pe mance m ng The mod hod and oode wa eaom ba LF pMe aamm oov m p ov de be een pe o mance aeof m ng Algorithm/ Classification/ Compute Problem/Solution Algorithm/ Classification/ Parameter Compute Application in Problem/Solution Others Parameter Application in Oth Algorithm/ Classification/ Compute Algorithm/ Classification/ Problem/Solution Parameter Problem/Solution Application in Parameter Others Application in Others aheu cceo MH me hod evo u ona y hm have de ahod aoInner na u aona Me ccame me hod Me aheu evo Me cLF u MH aheu ona me y aeve ccomparison hod go hm ke me evo have hod u been ona ke con evo y go u hm y au have am go na been u hm aob con have de been ed con de na u ed amance aeaInner na u abe eSQP cm en y aed u u eov pe econ pec en y om agh hyb u u d eon me pe pec ve eme cac The echn en d om me how aona aheu u obu u eu pe cza ne echn pec que gh ve how hyb obu d me ne aheu h gh cGA echn que obu ne eme cε-constd. en y om a(Optimization) u u eLF pe pec ve hyb d me aheu cGA echn que how obu ne h echn echn que been been emo en p ed eex en Tab ed eba 10 Tab echn 10 ha p eNo ed n Tab ehe 10 echn que ha been p eme en ed nba Tab eeo 10 mu ob ve p ob em mu a e d ec cu ve o p op ob m em ze a S e m d a cu y mu GA o and op ec m PSO ze ve p S p ov ob de a em y nea GA a op d and ma cu PSO p op ov de m ze nea S m ma y and PSO p de nea op mu ob ec p ob a e d cu o op m ze S m y and PSO p ov de nea op ma a unce a n y o oad and REG gene a a unce on a n n compa y oad on and o REG ad gene ona LF on mode n compa A o on hey o ad ona LF mode A o a unce a n y oed LF oadme and REG gene a on n compa on ad ona mode A o hey way add e ng comp ex MOP p ob em way Howeve add e he ng h gh comp compu ex MOP a ona p ob equ em emen Howeve and he h gh a ona equ emen way add e ng comp way ex MOP o add ob e em ng comp Howeve MOP he p h ob gh compu Howeve a equ he h emen gh compu and a ona equ emen and qua y o u on p om e powe u g oba qua op y m o za u on on me hod om and e powe have u ab g oba y o op add m e za on me hod and have ab y o add qua y u on p om e powe u g oba op m za on me hod and have ab y o add e Table 10. Table Performance 10. Performance comparison of algorithms, algorithms, Table applied 10. applied in Performance multi-objective in multi-objective comparison planning planning of algorithms, problems. applied in multi-objective planning problems. o u on o ge d bu on u on y em a ge d bu y em o u on o a ge d bu on y em Table 10. Performance comparison of algorithms, applied in multi-objective planning problems. o u on o a ge d bu y em p ov de gene a on o va p ou de cena gene o a unde on o a va oad ou mode cena o wh unde ch we a e oad m mode ed p ov n conven de wh gene ch ona we a on e o m va ed ou n conven cena o ona unde a oad mode wh ch we e p de gene a on o va ou cena o unde a oad mode wh ch we e m ed n conven po b y o p ema u e conve gence owa po d oca b y op o ma p ema o u e on conve need gence o be owa add e d ed oca w op h ma o u on need o be add e ed po b y o p ema u e po conve b gence y o p owa ema d u oca e conve op gence ma owa u on d need oca o op be ma add o e ed on w need h o be add e ed w h con a con ned ea a ned me ea p ann ng p ann p ob ng em p ob Fu em he Fu mo he e dec mo e on dec mak on ng mak me hod ng me a hod o need a o o need be o be con a ned ea me p ann ng p ob em Fu he mo e dec on mak ng me hod a need o be A c ng a chod ned emo gence A A ba ced ap me gence cn A o ba examp ed me A aheu ACS cway amo ABC nequ ocompa ew gence ced examp A e✘ hyb ACS ed ABC me aheu e✮ canged and cque hyb examp d eodbe ABC ecadd c✮ hy High ✘ Linear/Excellent ✮ ✮ Simple Simple ✮ ✮ ES N/Easy N/Easy No Inner Opt. Inner Opt. me hod The LF mpac me on he ba The o n mpac een ac on on aba he m ng am o me MOPT ny hod eMe ac p The on ob em LF acmo m have mpac auncon am MOPT on anged he ba p au ob ad o n have ane on anged aamo amo aep aNo MOPT p ob em have amo High ✘ Linear/Excellent ✮ Simple ✮ ES N/Easy No Inner Opt. me hod The LF mpac on n eove ac on aa❉ ng MOPT p em have aac au aam High ✘ High Linear/better ❉ Simple ✮/Complex ❉ ❉ Goal Goal P N/Difficult N/Difficult No IMO IMO eN/Easy eMCS en me hod The mp me hod mo eov equ e❉ cec en epec me ep compu hod aand on mp me eProblems MH A o mo hod ede equ em mo A ow eInner num High ✘ High Linear/better ✘ ❉ Linear/better ❉ Simple ✮/Complex Simple ❉ ✮/Complex ❉ Goal Pza N/Difficult Goal P N/Difficult No No IMO IMO mo ede e ES cMe en me The mo epe mp eES eha cop MH en me me hod hod The equ emp mo e✘ MH eba compu me hod aechn on equ me eam A o eaang mo compu eThe aeona on me oeov mo eapec numbe con de ed m ag ng ma we gh o FDN con de ed an m aA nd op ma we gh FDN con ed aaon m anequ nd con ng op de ma ed we aha m gh ap o nd FDN ng op ma we gh ogh High ✘ Complex/better Complex/better ❉ ❉ Simple High ✮/Complex ✮/Complex Complex/better ✮ ❉ Simple MCS MCS N/Difficult N/Difficult No Inner Complex/better ❉ ✮ Inner Opt. op m op on m me za hod on me can hod be con can be de con p om aLinear/better ng om ng o way o dea h comp za w on h ex comp me MOP hod ex pin MOP can ob em be p con ob mo em de em aThe ecompu om ng way o comp m on me hod can be con de agene p om ng o dea w comp ex MOP em eng ✘ Linear/better Simple ✮ b p cPque b u g ehod p cThe ha u eve p ha ep en ed eaheu en n Tab ed eHigh Tab 9aheu The eo 9op ove b g p ove ceque pe uba aFDN eona o ha pe mance o emance en ed compa n on Tab oba ehave on he 9ang The o add he ove eu add anumbe eMOP pe ed o MOP mance compa on oup he eanged ed M ε-constd. N/Easy No Inner Opt. High ✘ Linear/better ❉ Linear/better Simple ✮ ❉ b g p u eoand eyu en ed n Tab eeNo 9✘ The aob pe o mance compa on oaheu he add eed ed ε-constd. N/Easy ε-constd. N/Easy Inner No Opt. Inner Opt. Methods Execution Quality (Optimization) Methods Execution Depend Planning Quality (Optimization) Features Depend Planning Problems Featu High Linear/Excellent No Simple Methods Methods Cost Quality Execution (Optimization) Cost Depend Quality (Optimization) Planning Problems Features Planning Problems Features o ed oza o unc eve on equ h gh ed qua o ach y eve u nea A o o unc hona con on qua y and u o A ach o eve con ned aem ned h and qua uncon y eOpt. u ned A o con a✮/Complex ned and uncon a The mod LF me hod mod oem ed LF ba hod LF pTab aLinear/Excellent and o o The p wa mod de ed LF epde pe me p ohod o mance m and p ov ana o m de wa ng be eSimple eg ba pe ed o LF mance p aA o m p ng ov de eon odea mance m ng Algorithm/ Classification/ Classification/ Compute Problem/Solution Problem/Solution Parameter Algorithm/ Parameter Application Classification/ Application in in Compute Others Others Problem/Solution Parameter App Algorithm/ Classification/ Problem/Solution Parameter Application Others aheu ccompa MH me hod ke u ona y aMH go hm have been aheu con cDepend de anumbe me aPSO hod aov ke evo ona y go have been con de ed aand na MH me hod Me aheu ke evo cLF u ona y Me aed hod go aheu hm ke evo have cem MH been me con y hod go ed ke hm aed evo have ahyb u na been u aob y con anea go de hm ed ang have na been ao con de ed aaACS aPSO na aov equa cunc en y ang u u end pec The eCompute cme d en me om ade u coba u pe que how ve The eop obu ceNo en ne y d me h gh u uequ ep c✮ pe echn ve how hyb obu me ne aheu hme gh echn eop cAlgorithm/ en y om ang u eompa pe pec ve The hyb d me aheu cGA echn que how obu h gh echn echn que been p been eMH p ed eom en nhe Tab ed n ebut Tab eu echn 10 been p en ed Tab eequ 10 echn ha been p eHigh en n eo 10 mu mu ec ve ob p ec ob ve em p ob aoca eCompute em d aeou cu eA d o cu mu m o op S m m ze ve Sdea p y m ob GA au em and GA aon d and p cu PSO de oy p op nea m op ze nea ma S op aemen ma yOpt. GA and de nea op mu ec ve ob em ame eop d cu o op m ze Sha m aob y and PSO p ov de nea op ma a unce anea n y o oad and REG a10 unce aMCS aeze on n y n ou compa and on REG o gene ad ona LF n compa on o o hey ona LF mode o hey ap ov unce a n ons yao onfor oad and REG on on ona LF mode o hey so ut arge dMe str but on so ut ons for arge d str on systems way o add eN/Difficult ng comp MOP p ob em Howeve he h gh compu apowe ona emen and way add eunde ng comp way ex MOP o add p way ob eaN/Difficult ng o add comp Howeve eCost ex ng MOP comp he h ex ob gh MOP compu Howeve p away ob equ Howeve he h emen gh compu he and h aACS gh ona compu agh ona emen equ and y oExecution uaom on p om qua each powe y o u u on g oba p op ey za powe on me qua hod g oba y and op have on m za ab p on om y me eh oCost hod add and ehod uo have oba y m oMOP za on ehm me hod and have ab yoh opeaque add qua ycPque o u on p eva powe u g op m me ab yob o add eaInner Tab eob 10 Pe oha man eu on o aA go hm app n mu ob ey ve p ann ng ob em o u on o aman ge bu on u on y em awa ge d bu on y em o u on obe aned ge d bu on y em Tab eaheu Pe o eoad ompa on Tab go eo hm Pe oex man em n ompa mu ob on econ ad go p ann hm ng app ed mu ob eom ve p ann ng p ob em oecan u on aad ge d bu on em p ov de gene aevo on cena p o ov unde de gene aed oad aoad on mode oMH va ou wh ch cena we eon o m unde ed aop n conven oad mode ona wh we epe m ed conven de gene o va ou cena o an mode wh ch we ehyb m ed n conven ona po b y o p ema u egence conve gence owa po d oca b y op ma p ema o u on conve need gence o be owa add eab d ed oca w op h ma o u on need oMOP be add ed po beOPF y10 oN/Difficult p ema u eε-constd. po conve b gence y o p owa ema d eapp conve gence ma u on need oca oop be add o eneed u ed w need h be add ed w con aob ned ea me p ann ng p ob em Fu he amo ned eon dec me on mak p ann ng p ob em aProblems o Fu need he mo o be eeadd on ng hod aA con agene ned ea me pd ann ng p ob em he epe dec on mak ng me hod aeo ong be High ✘ Complex/ Excellent Excellent ✮ ✮/Complex ✮/Complex ❉ OPF OPF N/Difficult N/Difficult No No Inner Opt. High Complex/ ✮ Simple ✮/Complex OPF N/Difficult No Inner Opt. con esystems gence A A ba ced a10 me aheu gence cn o ba examp ed me A eExcellent aheu ACS cob amo ABC cmo n o eong gence cn examp and A hyb ba ed ABC me e❉ ceaanged and cne hyb omance examp ech eMOPT cn and hy High ✘ High Linear/Excellent Linear/Excellent ✮ ✮ Simple ✮ Simple ✮ ES N/Easy N/Easy Inner Opt. Inner Opt. A co aComplex/better n eme A ba ed omance examp eMOPT ABC eme cmode and hyb d hod The LF mpac on me hod ba The o LF n eom mpac ac on on ay he m ba ng aowa o MOPT n echyb ac p on em aon m have hod aehod anged The aahow p LF ob ad mpac em have he ae❉ ba anged odec n adead e(Optimization) ac aon m aA ob High ✘ High Linear/Excellent ✘ Linear/Excellent ✮ ✮ Simple Simple ✮ ES N/Easy ES N/Easy No No Inner Opt. Opt. hod LF mpac on he ba o n eove ac on m ng ame MOPT p em have aon aand ah High ✘ Linear/better Linear/better ❉ ❉ High ✘ ✮/Complex Linear/better ❉ ❉ ❉ Simple ✮/Complex Goal P P N/Difficult Goal P N/Difficult No No IMO IMO No IM mo eed me MH me hod ed equ e❉ co en ecem me ema compu hod aCompute on mp me eProblems MH A o mo hod ede numbe equ mo eyInner compu aACS on me o mo num Linear/better ❉ Simple ✮/Complex ❉ Goal mo eme cA me hod The mo eGoal mp econ eha cd MH en me me hod hod The equ emp mo e✘ eap compu hod ave on equ me eea A o eem mo compu eThe numbe aACS on me A o mo numbe de ed de m ed ah m nd ng ng ao op nd ma ng op we gh we o gh FDN ooha FDN de ed aFu m ng amo nd ng op ma we gh oza Complex/better ❉ Simple ✮/Complex ✮ High ✘ ❉ High ✘ Complex/better ✮/Complex ❉ ✮ Simple ✮/Complex ✮ MCS MCS N/Difficult No Inner No Opt. Inner Opt. op m za me hod op be con m za de on aThe hod om can ng be con oeon de dea op aSimple w m h za comp ng me ex way MOP hod o p can dea ob be em con hob mo comp de aon ex p MOP om p ng ob way em mo oawa dea w hcon comp ex p❉ ob em High ✘ Linear/better High Simple ✮ Linear/better Simple ✮ b g p cen uadd ha pme eLF en b g nES p Tab camod u eExecution 9hod ha p ove eo en aemp ed Tab o mance 9me b The compa g p ove u aPSO eona on ha pe o p o he emance en add ed eo compa n ed Tab MOP ehave on 9aehow The o✮ he ove add pe ed o MOP compa ong he eeneed ed N/Easy ε-constd. N/Easy No Inner Opt. No Inner High ✘ Linear/better ❉ High ✘ Linear/better Simple ❉ ✮ Simple ✮ b g p cque uhe eThe ha p eo ed n Tab eComplex/ 9✘ The aom pe o compa on o he add eobu ed ε-constd. N/Easy ε-constd. N/Easy No No Inner Opt. Inner Opt. Methods Methods Execution Execution Cost Cost Quality Quality (Optimization) (Optimization) Methods Depend Planning Planning Execution Cost Features Features Quality Depend Planni N/Difficult High Complex/ Simple /Complex Inner Opt. Methods Cost Quality (Optimization) Depend Planning Problems Features o unc on equ ed ach eve nea y h gh o qua unc on y eay u equ A ed o con ach ae✘ eve ned nea and y uncon h gh qua ae❉ ned y eagh u omak ame ned and uncon aM o unc on ed oMCS o ach eve on nea y equ gh ed qua o ach y eTab eve u nea A y o h con gh qua aand ned and u uncon A oen con aw ned ned and uncon ned The mod ed me hod and oem wa ba p o m p ov de be eDepend pe o mance The mod ame m ng ed hod and o eA ba ed LF p aABC o m p ov de be The mod ed LF hod The and o wa ed ePe LF ed me LF hod p aLF and m p wa ov eaheu de ba be ed eExcellent LF pe p aFDN mance m p aSimple ov m de ng be ePSO pe o mance m ng Algorithm/ Algorithm/ Classification/ Classification/ Compute Compute Problem/Solution Algorithm/ Problem/Solution Classification/ Parameter Parameter Application Application in Problem/Solution in Others Others Parameter Application in Oth Classification/ Compute Problem/Solution Parameter Application in Others econ cAlgorithm/ en econ y cen en om y ang u om u een aThe pe u u pec eon pe ve pec The ve hyb The me aheu d me aheu cand equa echn ceNo en cGA que y echn om que u obu how u eaheu ne pe pec haIMO gh ve h gh hyb d me aheu cadd echn que eme cPunc en y om aob u u eompa pe pec ve The hyb d me aheu cNo echn que obu ne h echn que echn que been ha been ege en p ed eex en n ed n eMH 10 Tab emo echn 10 que been p ed n Tab eLF 10 echn ha been p eme en ed n Tab eoma 10 mu ob ec ve p ob em mu aex egence ob cu ec ve p op ob m em ze aeway So eo m d ao cu y mu GA o and op ob m ec ze ve p So p ov m ob de au em y nea aon ew op d and ma cu o p op ov m ze nea SaMOP m op aThe ma GA and PSO phod ov de nea op ao unce aep nequ o oad and a unce REG gene n yep aba o on oad n and compa REG gene unce o aom ad aTab on n y ona n o compa LF oad mode on REG A gene ad o hey ona au on LF n mode compa A on o o hey ad ona LF mode A oemen hey way oyo ey ng comp MOP p em Howeve he h way gh compu o add e ona ng comp equ emen ex MOP and p ob em Howeve he h gh compu a ona equ way add e ng comp way ex MOP o add p ob e ng comp Howeve way o MOP add he p h e ob gh ng em compu comp Howeve a ex MOP equ he p ob h emen gh em compu Howeve and a ona he equ h gh emen compu and equ emen and qua o u on p om e powe u g oba qua op y m za u on me p hod e powe have u ab g oba y y o op o add u m on e za p om me hod e powe and u have g oba ab op y o m add za on e me hod and qua y o u on p om powe u g oba op m za on me hod ab y o add e Tab e 10 Tab Pe e 10 o man o e man e ompa on o a on go o hm a go app hm e ed 10 app n Pe mu ed o man n ob mu e e ve ompa ob p e ann ve on ng p o ann p a ob go ng em p hm ob em app ed n mu ob e ve p ann ng p ob em o u on u o on a ge o d a bu d on bu y em on y em o u on o a ge d bu on y em Tab e 10 Pe o man e ompa on o a go hm app ed n mu ob e ve p ann ng p ob em o u on o a ge d bu y em p ov de gene a on o va ou cena ov o de unde gene a a on oad o va mode ou wh cena ch o we unde e m a ed oad n conven mode ona wh ch we e m ed n conven ona p ovhod de gene aLFonmpac o va on ou cena o unde a oad mode wh ch we e m ed n conven ona High ✘ High ✘ Complex/better Complex/better ❉ ❉ Simple ✮/Complex ✘ ✘ Cone Cone P N/Difficult No Convex Convex High ✘ Complex/better ❉ Simple ✮/Complex ✘ po b y o p ema u e conve gence owa d oca op ma o on need o be add e ed w h Cone P N/Difficult No Convex po b y o ema u e po conve b y o po p owa ema b d y u o oca e p conve ema op gence ma u e conve owa u on gence d need oca owa o op be d ma add oca o e op u ed ma need h o o on be need add e o ed be w add h e ed w h con a ned ea me p ann con ng a p ned ob em ea Fu he p ann mo ng e dec p ob con on em mak Fu a ned ng he me ea mo hod me dec a p o ann on need ng mak p o ng ob be me em hod Fu a he o mo need e dec o be on mak ng me a o need con a ned ea p ann ng p ob em Fu he mo e dec on mak ng me hod a o need o be High ✘ Complex/ Complex/ Excellent Excellent ✮ ✮ Simple ✮/Complex Simple ✮/Complex ❉ ❉ OPF OPF N/Difficult N/Difficult No No Inner Opt. High ✘ Complex/ High ✘ Excellent Complex/ ✮ Excellent ✮ Simple ✮/Complex Simple ❉ ✮/Complex ❉ OPF N/Difficult OPF N/Difficult No No Inner Opt. Inner Opt. A c a n e gence A A ba c ed a me n e aheu gence c A o ba examp ed me A e aheu ACS c a ABC c n e o e gence c examp and A e hyb ba ACS ed d ABC me aheu e c and c hyb o examp d e ACS ABC e c and hy High ✘ High ✘ Linear/Excellent Linear/Excellent ✮ ✮ High Simple ✘ Simple ✮ Linear/Excellent ✮ ✮ Simple ✮ ES ES N/Easy N/Easy ES N/Easy No Inner Opt. Inner Opt. No Inner A c a n gence A ba ed me aheu c o examp e ACS ABC e c and hyb d me hod The LF mpac on he ba me o n hod e ac The on LF a m mpac ng a on MOPT he ba p ob o em n e have ac on a anged a m ng a a a MOPT p ob em have a anged Linear/Excellent ✮ Simple ✮ ES N/Easy me The he ba n e ac on a m ng a MOPT p ob em have a anged a a Linear/better ❉ ❉ Goal Punc N/Difficult No IMO mo eε-constd. eMCS me mp MH me hod mo eze equ eb❉ cec en eon mo me eMOP compu am on mp me eProblems MH mo hod em equ ea9eeO mo eNo compu acomp on me A op✮ mo eInner num High ✘ Linear/better ❉ High ✘ Simple Linear/better ✮/Complex ❉ Simple ✮/Complex Goal P N/Difficult Goal Pu N/Difficult IMO No IMO mo eop eg ce hod The mo eop mp eeve eC cec MH en me hod The equ emp mo e✘ MH em compu hod acechn on equ me eau A mo o eu mo compu eThe numbe aecomp on me A o(Optimization) mo e❉ numbe con de ed aThe m ng ah nd ng op ma we gh o de FDN ed m ng amance nd ng op ma gh o FDN con de ed aACS ng nd ng op ma we gh FDN Complex/better ❉ Simple High ✘ Complex/better ✮ ❉ Simple MCS N/Difficult High ✘ ❉ High ✘ Simple ✮/Complex ❉ ✮ Simple ✮/Complex ✮ MCS N/Difficult N/Difficult No No Inner Opt. Inner m za me hod can be con m za de on aThe me ped hod can ng be con oon de dea op w p m h om za on ng me ex way hod o punce can dea ob be em con h mo comp de eA awe ex p MOP om p ng ob way em o dea eon h ex MOP obne em op m me hod can con de abThe p om ng way o❉ dea h ex MOP p ob em mo emo High ✘ High Linear/better Linear/better ❉ Simple ✮ Simple ✮ High Linear/better ❉ Si cCen uy eme ha eoad en ed gza ned p Tab eComplex/better eme 9hod ha The p ove ente abe ed pe n Tab mance eComplex/better 9ad The compa ove aACS pe o he ohod b add go p cepe compa u ed eand MOP ha p on e(for oeProblems en he ed nnumbe Tab ed eOpt. ed MOP The ove aand pe oa✮/Complex mance compa on ε-constd. ε-constd. N/Easy N/Easy No ε-constd. N/Easy Opt. Inner No Linear/better ❉ Simple ✮ b g p u ey ha p eom en ed ned Tab eNo ove aob pe o mance compa on oov he add eobu ed N/Easy Methods Methods Execution Execution Cost Cost Quality Quality (Optimization) Methods (Optimization) Execution Depend Depend Cost Planning Quality Features Depend Planning Featu Cone P N/Difficult High Complex/better No Simple /Complex Convex Methods Execution Cost Quality (Optimization) Depend Planning Problems Features o unc on equ oom ach eve y hem gh o qua unc on y ey equ A ed con ach ast eve ned nea and yab uncon h gh qua a✘ ned y A onop con ned and uncon am o unc equ ed oMCS o ach on nea y equ gh ed qua oen ach y eeway eve nea A y h con gh qua am ned and eo uncon A oen con aw ned ahey ned and uncon aMOP ned The mod ed LF me hod LF and hod o and wa e(AI)-based o ba wa ed eque LF ba p ed aamode LF m p aABC p ov m de p be ov ey de pe be o mance pe oInner ao mance m ng m The mod ed LF me hod and o wa ecomp ba ed LF p aob ow ov de be eng pe o mance aOpt. ng A hm n mpu P9 m S u n P m App n n O h Art c a examp nte (AI)-based Art metaheur ahod st cs gence (for examp eoca metaheur etc )✮/Complex cs and hybr examp ACS ABC etc and hy A hm np C mpu hm P b C S uC n n C mpu P App u nm n n P O m App n nFeatures hm metaheur st cs f (for ABC etc )con and hybr eb cp en yyam om aon u u eb pe cen pec en y ve om aThe hyb u d me pe aheu pec ve econ cop The en hyb que d om me how aheu obu u eu cPlanning ne echn pec h que ve gh The how hyb d me ne aheu h gh cw echn que how obu echn p eTab en echn que n Tab ePe been p en ed n Tab eu echn 10 been p eNo ed n Tab eme 10 echn que ha been p eo en ed nPnea eo 10 mu ve p ob em aoca eA cu o op mu m Sapp m ve p y ob GA em and aehon PSO d p cu de o op nea op ze ma S m au y GA PSO pProblems ov nea op mu ob ec ve pema ob em mu ap eob d cu ve o p op ob m em ze ame S eop m d au cu y GA o and m PSO ze p Sha ov au de y nea GA op and PSO p ov de nea op ma a unce noha y oage gene aLF on n compa o ad ona LF a A ap o n✮/Complex o oad and REG gene ahod compa on oon ad ona unce aon n oon oad and aOPF REG unce gene aES n afmod oc on oad n compa and on gene o au on ona n compa LF on A o ad hey ona LF mode A o hey qua y qua o u on o u p on eThe p powe eon u powe g oba op g oba m za op on m me za qua on hod me y and hod o u have on and ab have y ab o eadd add powe y e✮ ong add u geng oba za on me and qua yccA o u on p om emp powe u g oba m za me have y add eme High ✘ High ✘ Complex/better ❉ Simple Simple ✮/Complex ✮/Complex ❉ ❉ eob 10 Tab Pe 10 oha man o eu man ompa e✘ ompa oon aba on go o hm ao go app hm ey app Pe mu ed o man n mu end eo ve ompa ob pma eSimple ann ve on ng p o✮/Complex ann p ap ob go p hm ob em app ed n mu ob ve p ann p) ob em o ugence am d bu o on u on y em aed d bu y u on o10 am ge d bu on y em SQP SQP N/Difficult N/Difficult No No Fast Fast High ✘ Complex/better ❉ Simple ✮/Complex Tab 10 Pe om man eo ompa o aComplex/ go ed n mu ob ve p ann ng p ob em SQP N/Difficult No p de gene abeen oCone ou de cena gene o a10 unde on aeom va ou mode cena p ov o de wh unde gene ch we ame aSed eon oad o va mode ed ou n conven wh cena ch ona o we unde eMOP m aem ed oad conven mode ona ch we eACS m ed n ona High Complex/better Complex/better ❉ ❉ Simple ✘ ✘ P✘ Cone P N/Difficult No Convex High ✘ High Complex/better ✘ Complex/better ❉ ❉ Simple Simple ✘ ✘ bon p ema u eeand conve gence owa d oca op ma po u b on need o p o ema be u add ehod conve ed gence w h owa d oca op ma o u need ong be add e em ed Cone Pov N/Difficult Cone No No Convex b y oned p u eε-constd. po b gence o p owa ema d eu conve b y gence ma o pA ema owa on d enn need conve o op gence be add owa o eMCS ed d w oca need h op o ma be add o✮/Complex umak on ed w need h oNo be add em ed hNo con aque ea me ann ng p ob Fu con he aREG mo ned eTab ea dec me on p mak ann ng ng p hod em con aoon Fu aHigh need ned he mo o ea be eob dec p ann p ob em Fu ap oh need mo ea✮/Complex oned dec be mak ng con aec ea me p ann ng p ob em Fu he mo eme dec ng me hod ache o need o be High ✘ High ✘ Complex/ Excellent Excellent ✮ Simple High ✘ ✮/Complex ✮/Complex Complex/ ❉ Excellent Simple ✮/Complex ❉ OPF N/Difficult N/Difficult OPF N/Difficult No No Inner Opt. No Inner Complex/ Excellent ✮ ❉ OPF Inner Opt. A cREG A n con eo aem gence nge ehyb gence ba A ed me ba ed aheu me A aheu cSu cuon ame o cm n examp emode gence examp eD ACS A ba ACS ed ABC me cac aheu and e❉ ceConvex hyb and cng daheu od examp eeon ABC ede chod and hy ✮ ✮ ES N/Easy Inner Opt. A ccon aueom ePp gence A ba ed me aheu oob examp ABC eom and hyb me hod The mpac on o n The eop ac LF on mpac aon m on ng aeed he ba enwe em on ang ae❉ m anged ahyb MOPT auwh ad ob em have aconven anged aed aham High Linear/Excellent ✮ High ✘ Linear/Excellent Simple ✮ ✮ ✮ ES N/Easy N/Easy No Inner No Opt. Inner Opt. me onapo ba o nme eha ac aconve m ng aned MOPT ob em have ahe anged aShyb aTab High Linear/better ❉ Simple ✮/Complex Linear/better ❉ ❉ Simple Goal P N/Difficult Goal P N/Difficult No IMO mo ehva eMCS me hod MH equ mo ema compu aC on me A mo numbe High ✘ ❉ High ✘ Linear/better Simple ✮/Complex ❉ ❉ Simple ✮/Complex ❉ Goal Pov N/Difficult Goal P N/Difficult No No IMO IMO mo epo cde en hod mo ep mp eeve eC caey MH en mo me me en ewe hod con en equ me hod mp eoad mo e✘ The MH compu mp hod acbechn ehm MH on equ me hod en A mo o eyu equ mo compu emak numbe aeACS on compu me A ao on o mo me numbe A o ey numbe con aom aThe nd ng op de ma ed m gh ng aN/Difficult o nd FDN ng op ma we gh de ed o FDN ng ao ng op ma gh oaed FDN de ed m aba ng op ma we gh o FDN Complex/better Complex/better ❉ ❉ Simple ✮/Complex ✮ High ✮ ❉ MCS N/Difficult N/Difficult ❉ ✮ op za on me hod can op be con m za de aThe p hod can ng be con o de dea op w p m h om za comp ng ex way MOP hod o p can dea ob be em w con h mo comp ehow aSimple ex p MOP p ng ob way em o eConvex w comp ex p❉ ob op m hod can be con de aThe p om ng way dea w h comp ex phave ob em mo eFast High ✘ High ✘ Linear/better Linear/better ❉ ❉ High Simple ✘ ✮ Simple Linear/better ✮ ❉ Simple ✮ b g p eLinear/better ha p epo en ed n Tab eComplex/better 9hod b The g ove chyb ep aecu ha pe p o emance mance en compa n Tab eede 9n on The ove add an pe en ed o MOP compa on oew he add eSimple M ε-constd. N/Easy N/Easy ε-constd. N/Easy No No Inner Opt. Inner Opt. No Inner High ✘ Linear/better ❉ Simple ✮ b g hod p c u The e haLFp mpac e en ed n Tab ecm 9on The ang pe o mance compa o add eLinear/Excellent ed MOP N/Easy No Inner Opt. M E n C Qu Op m n D p nd P nn nABC P b m u N/Difficult High Complex/better No Simple /Complex Fa Mhe h deSQP Eed uequ n C M hy Qu d Op Eme m uC n n D C p Qu P Op n P m b m n p u nd P nn n P b m o unc on equ ed oom ach eve nea y hon gh o qua unc on y ey equ A ed o con ae✘ eve ned nea and y uncon h gh qua a✘ ned y A ohe con uncon o unc ed oMCS o ach on nea y equ h gh ed qua o ach y eeway eve nea y o h con gh qua a✮ ned and eo uncon A o con am ned ap ned and uncon aLinear ned The mod ed LF and o wa eLinear/ ba ed The LF mod a✮ ed m LF p me ov de hod be and eve pe o o wa mance eom ba m LF ng p aLinear o m p ov de be pe oobu mance aIM m The mod ed LF me hod and o wa ehod ed LF p agene m p ov de be pe o aeMOPT ng A hm A hm C n n mpu mpu P b m A P m hm n C n P m P n m mpu App App n P n m n Se O u h n O hmo P m App nand nde O A C n C mpu P b m u n P m App n nbde O h ep en y aon u u eon econ pe cen pec en yme ve om aaThe u u d eon me pe aheu pec ve econ cop The en que d om me how aheu u obu eob pe cch ne echn pec h que ve gh The how hyb obu d me ne h gh cGA echn que how ne econ cdPunc en y ang u eompa pec ve d me aheu cGA echn que ne h echn been p eTab en echn ed que n Tab ha ePe 10 been p ehe en ed n Tab eo 10 echn que ha been p eobu en ed Tab emo 10 echn que ha been p ehod en ed ny Tab eGA 10 mu ec ve p em and epe cu o op mu m ze ob Sand ec m ve au p y ob GA em and aemo PSO d p cu ov de o op nea m op ze ma S m auComplex/better and PSO pMOP ov nea op mu ob ec ve p ob em mu aε-constd. eaob d ob ec cu ve o p op ob m em ze aou S eCann m d au cu y o and m PSO ze p S m ov au de nea op and ma PSO p ov nea op ma a unce a aza unce n o auPe nme oad yme o and oad REG and REG and gene an compa on n compa on oon on ad o ona ad LF ona mode LF mode A o hey A oegh hey ahm unce aob n y o oad and REG gene auSExcellent on n compa on o ad ona LF mode A oomance hey qua yque oaThe uove on p om qua eon powe o u u on gu oba p op om m eon za powe on me u qua hod g oba y o op have on m za ab on om me y each hod add powe and u have g oba ab op y m ong za add on edea me hod and have ab yem owe add High ✘ Linear/ Excellent ✮ Simple Simple ✮ ✮ MILP MILP N/Difficult No Linear High ✘ Linear/ Excellent ✮ Simple ✮ MILP N/Difficult No High ✘ High ✘ Complex/better ❉ Simple ✮/Complex Simple ✮/Complex ❉ ❉ e 10 Tab Pe e 10 o man o e man e ompa o a on go o hm a go Tab app hm e ed 10 app n Pe mu ed o man n ob mu e e ve ompa ob p e ann on ng p o ann p a ob go ng em p hm ob em app ed n mu ob e ve p ann ng p ob o u o a ge d bu on y em u on o a ge d bu on y em SQP SQP N/Difficult N/Difficult No No Fast Fast High ✘ High Complex/better ✘ Complex/better ❉ ❉ Simple ✮/Complex Simple ❉ ✮/Complex ❉ Tab e 10 o man e ompa on o a go hm app ed n mu ve p ann ng p ob em o u on o ge d bu o u y em o a ge d bu em SQP N/Difficult SQP N/Difficult No No Fast Fast ov de gene a on o va ou cena o unde a oad mode wh ch we e m ed p ov n conven de gene ona a on o va ou cena o unde a oad mode wh ch e p ov de gene a on o va p ou ov cena de gene o a unde on o va oad mode cena wh o unde ch we a e oad m ed mode n conven wh ona we e m ed n conven ona High ✘ Complex/better Complex/better ❉ ❉ High Simple ✘ ✮/Complex Complex/better ✘ ✘ ❉ Simple ✮/Complex ✘ Cone Cone P N/Difficult Cone P N/Difficult No Convex Convex No Con ✘ Cone P Convex con con ned ea a ned ea p ann me ng p p ob ng em p ob Fu em he Fu mo he e dec mo e con dec mak a on ng ned mak me ea hod ng me me a hod p o ann need a ng o o p need be ob em o be Fu he mo e dec on mak ng con a ned ea p ann ng p ob em Fu he mo e dec on mak ng me hod a o need o be High ✘ Complex/ Excellent ✮ Simple ✮/Complex ❉ OPF N/Difficult No High ✘ Complex/ Excellent ✮ High ✘ Simple Complex/ ✮/Complex Excellent ✮ ❉ Simple ✮/Complex ❉ OPF N/Difficult OPF N/Difficult No Inner No Opt. Inner Opt. A c a n e gence A A ed c a me aheu gence c A o ba examp ed me e A aheu ACS c a ABC c n o e gence c examp and A hyb e ba ACS d ed ABC me aheu e c and c hyb examp d e ACS ABC e c and hy Linear/Excellent ✮ High Simple ✘ ✮ Linear/Excellent ✮ Simple ✮ ES N/Easy ES N/Easy Inner Opt. No Inner me hod LF mpac on me he hod ba The o LF n e mpac ac on on a he m ba ng me a hod o MOPT n The e ac p LF ob on em mpac a m have on ng a he MOPT anged ba o a p ob a n e em ac have on a a m anged a MOPT a a p ob em have a anged a a High ✘ Linear/Excellent High ✘ Linear/Excellent Simple ✮ ✮ Simple ✮ ES N/Easy ES N/Easy No No Inner Opt. Inner Opt. High ✘ High ✘ Linear/better Linear/better ❉ ❉ Simple ✮/Complex Simple ✮/Complex ❉ High ❉ ✘ Linear/better ❉ Simple Goal P Goal P N/Difficult N/Difficult No Goal No P N/Difficult IMO IMO No mo e e c en me hod The mp e MH me hod equ mo e compu mo e e a c on en me hod A o The mo e numbe mp e MH hod equ e mo e compu a on me A o mo e num Linear/better ❉ Simple ✮/Complex ❉ Goal P N/Difficult IMO mo e e c en me hod The mo e mp e e c MH en me me hod hod mo The equ e e e mp mo c e en MH e compu me me hod hod a The on equ me mp e A mo e MH o e mo compu me e hod numbe a on equ me e mo A e o compu mo e numbe a on me A o mo e numbe con de ed m ng a nd ng op ma we gh con o de FDN ed a m ng a nd ng op ma we con gh de ed o a FDN m ng a nd ng op ma we gh o FDN opt m zat on methods can be cons opt der m as zat prom on s methods ng ways can to dea be cons w th der comp as prom ex MOP s ng prob ways ems to dea more w th comp ex MOP prob ems m con de ed a m ng a nd ng op ma we gh o FDN s prom su ng dea w th comp ex MOP prob ems more Complex/better Complex/better ❉ ❉ High Complex/better ✮ ✮ ❉ Simple MCS MCS N/Difficult Inner Opt. Inner Inner High ✘ Complex/better ❉ MCS No Inner op m za op on m me za on me can hod be con can be de con aand p de op p m ng om way on ng me o way dea hod w can h comp con h ex comp MOP aThe ex phow p MOP ob om em p ng ob mo way em eaheu mo o dea e ond w hon comp MOP p✮ obA m op m za me hod can be con aThe p ng way o dea w h comp ex MOP p✮ ob em mo eon H h ✘ n ❉ mp ✮ g p con eaC ha p eLF en ed ned Tab eg 9p caThe The u enOp ove ha au p pe em en ed mance n Tab compa eadd 9nm The on o aehyb pe add o eam mance ed MOP compa oApp he add eoobu ed MOP nach d N N nn Op H ✘ n ❉ H hde ✘ L mp ❉ ✮ bechn g p cque e ha haways pep ento Tab ecLP 9on aaed pe ohMILP mance compa on o he add ed MOP nunc dTab N Eo ny do N E N nn N Op nn Op M h d M h dve E uhod E n ued n C C Qu Qu Op M m h dza n m E n p nd n p nd C nn PA P nn b Qu nS m b Op m m ume n uOpt. D pm Pque nn ✮/Complex nex b nm u nea / Ex e✮ en No mp L nea M N/D fi un H gh h d EL u n Qu Op n D p nd P nn nde P b uOpt. o unc on equ ed oC ach eve nea y h gh qua y ey u A o con a✘ ned and uncon a✮ ned o equ oMCS o unc eve on nea o unc y equ h gh on ed qua o equ yHigh empu eve nea oon A ach y h eve con gh nea qua aS ned y h✮ gh and eD qua uncon A y oSimple eo con aw u ned ap ned onhow con and uncon aA ned and ned uncon ap ned The mod ed LF me hod The and oob wa eom ba ed LF hod pon aLinear/ oLTab m o p wa ov The eob de ba mod be ed LF ed pe p LF awh o me mance m hod apec ov and de ng be o wa emp pe ehe ba o mance LF p m o m ov de be ehow pe mance aFahm m The mod ed LF me hod and oLb wa eom ba ed LF p aoad o m p ov de be eov pe oop mance A hm A C hm n n mpu CHigh Pn b m A P S u m hm n S u n P m P no m C mpu App App P n m n S O u h n O ha Pon n O A hm Cu C mpu P b m S u n P m App nbm O h econ en yy u u eon eb pe cng pec en y ve om The anm hyb u u d eon me pe aheu pec ve equa cechn cbechn en hyb que d om me how aheu u u obu eob pe cPwe ne echn h que ve gh obu d ne h gh cGA echn ne equa cREG en y aem u eze pe pec ve hyb d me aheu cGA echn que obu ne h gh echn que ha been p eC en 10 que ha been pD edea en ed n Tab enove 10 been eed en n 10 ob ec p em aou eA cu o op mu m ze m ve aea p yed ob GA em and aebe PSO eA d cu de o nea m op ze ma S m aona yNo and PSO pPhod ov de op mu ob ec ve pom ob em mu aM d ob ec cu ve o p op ob em ame S eA m d au cu yLinear/Excellent GA oom op and m PSO ze p SC m ov au de y nea op and ma PSO p ov de nea op ma a unce aov oad and REG gene au on a n unce compa aABC n y on on o oad ad and ona REG LF gene mode aPg on n oed compa hey ad ona LF mode o aned unce aycaeoThe nN/Difficult oad and gene on n compa o ad ona LF mode A hey High ✘ Complex/better Complex/better ❉ ❉ Simple ✮/Complex ✮/Complex ❉ ❉ MINLP MINLP N/Difficult N/Difficult No No Non linear Non linear High ✘ Complex/better ❉ Simple ✮/Complex ❉ MINLP N/Difficult No Non linear qua o uove on p om qua eed powe omod un u on gu oba p op om m ema za powe me u hod g oba y and o op have on za ab p on om me ewh op hod and ehave have oba ab op y oam za add eng me hod and have ab yem oem add ✘ Excellent Linear/ Excellent ✮ ✮ Simple ✮ MILP MILP N/Difficult No Linear Linear y o u on p each powe g oba op m za on me hod and ab yHigh o add eang High ✘ High Linear/ ✘ Excellent Linear/ ✮ Excellent ✮ Simple ✮ Simple N/Difficult MILP N/Difficult No No Linear Linear High High ✘ Complex/better ❉ Simple High Simple ✘ ✮/Complex ✮/Complex Complex/better ❉ ❉ ❉ Simple ✮/Complex ❉nea Tab eLF 10 Pe omu man ee ompa Tab eom on 10 o Pe aaheu go o man hm e✘ app ompa ed n on mu o ae go ob Tab eon hm ve eec 10 p app ann Pe ng o man n mu ob ey em ompa ob epowe ve on p ou ann anhave go ng p hm ob em app ed n mu ob eon ve pm ann ng peMOPT ob o u o a ge d bu on y em o u o a ge d bu on y em SQP SQP N/Difficult N/Difficult SQP N/Difficult No No Fast Fast No Complex/better ❉ Simple ✮/Complex ❉ Tab e 10 Pe o man e ompa on o a go hm app ed n mu ve p ann ng p ob em oMC u a ge d bu o u on y em o a ge d bu y em SQP Fast p ov de p gene de a gene o a va on o va cena ou o cena unde o a unde oad a mode oad mode ch e ch m we ed e n m conven ed n conven ona p ov de gene a on o va ou cena o unde a mode wh ch we e m ed n conven ona ✘ Complex/better ❉ ✘ Cone PES Convex High ✘ Complex/better ❉ High ✘ Simple Complex/better ✮/Complex ❉ ✘ Simple ✮/Complex ✘ Cone Pon Cone P N/Difficult No Convex No Convex ned ea me p ann con a p ned ob ea Fu me he p ann mo ng e dec p ob con on em mak Fu a ned ng he me mo hod me dec a p o ann on need ng mak p o ng ob be me em hod Fu a he o mo need e dec o be on mak ng me a o need High Complex/ Excellent ✮ Simple High ✘ ✮/Complex Complex/ ❉ Excellent ✮ Simple ✮/Complex ❉ OPF N/Difficult OPF N/Difficult No No Inner High ✘ Complex/ Excellent High ✮ ✘ Complex/ Simple Excellent ✮/Complex ✮ ❉ Simple ✮/Complex ❉ OPF N/Difficult OPF N/Difficult No No Inner Opt. Inner Opt. A c a n e gence ba ed me aheu A c c o examp e gence e ACS ABC ba ed e me c aheu and hyb c d o examp e ACS ABC c and hy ✘ Linear/Excellent ✮ ✮ Simple ✮ Simple ✮ ✘ Linear/Excellent ✮ Si ES ES N/Easy N/Easy ES N/Easy Inner Opt. Inner Opt. No A a n e gence A A ba c ed a me e gence c o ba examp ed me e aheu ACS c o e c examp and e hyb ACS d ABC e c and hyb d me hod The LF mpac on he ba o n e ac on a m ng a MOPT p ob em have me hod a anged The a LF a mpac on he ba o n e ac a ng a p ob Linear/Excellent ✮ Simple ✮ N/Easy me hod The mpac on me he hod ba The o LF n e mpac ac on on a m he ba a MOPT o n e p ac ob on em a have m ng a a anged MOPT a p a ob em a anged a a High High ✘ Linear/better Linear/better ❉ ❉ Simple High ✘ ✮/Complex Simple ✮/Complex Linear/better ❉ ❉ ❉ Simple ✮/Complex ❉ Goal P Goal P N/Difficult N/Difficult Goal P N/Difficult No No IMO IMO No IM High ✘ Linear/better ❉ Simple ✮/Complex ❉ Goal P N/Difficult No IMO con de con de m ed ng a a m nd ng ng a op nd ng op we ma gh we o gh FDN o FDN con de ed a m ng a nd ng op ma we gh o FDN con de ed m ng a nd ng op ma we gh o FDN H h ✘ C mp ❉ mp ✮ C mp ✮ MC N D u N nn Op H hed ✘ ❉ H hve ✘ C mp mp ✮ mp ❉ ✮ mp ✮ C mp ✮ D u ed MC N u N nn N Op nn Op op m me hod can op be m za on aThe me hod om can ng way be con oo dea de op aque w p m h om za comp on ng ex me way MOP hod o p can dea ob be em w con h mo enhow aov ex p MOP om p ng ob way em mo oon dea eop w hcon comp ex MOP puncon obA em hn ✘ L n L n ❉ ❉ H h ✘ mp ✮ mp L n ✮ ❉ mp ✮ bMTab g pon cezaN u eon ha ped eM en b gdnREG p Tab cque uCom ede emp 9oob ha p ove eooad en aunc ed pe o Tab mance b e g 9 p c The compa u e ove ha a on p pe e o en he o ed mance add n Tab e compa ed e 9 MOP The on ove o he pe add o e mance ed MOP compa on o he add e ed MOP nach do n d N N Ep nS d N E N nn Op N nn H hH ✘ L n ❉ mp ✮ n N Eu N nn h d M h d E unea E n uD n C C Qu Qu Op M m h Op d nm En D n p nd n D p nd P C nn nm Pcomp P nn b Qu n m P ba Op m m u n u D pm nd PLF nn na Po b m u NLP N/D fi ucon H gh Comp ex/be eC No SLF mp ea /Comp ex Non nea M h d E u n C Qu Op n D p nd P nn nde P b m uOp o unc on equ ed ach eve y h gh qua y e u o A unc o con on a ned equ and ed uncon ach eve a ned nea y h gh qua y e u A o a ned and a o unc equ o o unc eve on nea y equ h gh ed qua o ach y e eve on u nea A equ y o h con gh ed qua o a ach ned y eve and e u nea uncon A y h o gh con a qua ned a y ned e and u uncon A o con a ned a ned and uncon ned The mod LF me hod and wa e ba ed The LF mod p a o ed m LF p me ov hod de be and pe o wa o mance e The ba mod ed a ng p ed LF o me m hod p ov and de be o e wa pe e o ba mance ed a p m ng m p ov de be The mod ed LF me hod and o wa e ba ed LF p a m p ov de be e pe o mance a m ng A hm A C hm C n C n mpu C mpu Pn b m A P S b u m hm n S u C P m P n m C mpu App App n P b n m n S O u h n O h P App n n O h A hm C n C mpu P b m u n P m App n n O h e c en e y c en om y a om u e a pe u u pec e pe ve pec The ve hyb e The c d en hyb me y aheu d om me aheu c u u echn e pe c que echn pec how que ve The obu how hyb ne obu d me h ne gh aheu h gh c echn que how obu ne e c en y a u u e pe pec The hyb d me aheu c echn que obu ne h gh echn ha been p e en ed echn Tab e 10 ha been p e en ed n Tab e 10 echn que ha been p e en ed n 10 High ✘ High ✘ Complex/better ❉ Simple ✮/Complex ❉ No No Real Time Real Time DynP DynP N/Difficult N/Difficult mu ob ec ve p em a e d cu op m ze S m a y GA and PSO p ov de nea op ma High ✘ Complex/better ❉ Simple ✮/Complex ❉ No Real Time DynP N/Difficult mu ob ec ve p ob em mu a e d ob ec cu mu ve o p op ob ob ec m em ze ve a p S e m ob d a em cu y a GA e o d op and cu m PSO ze o p S op m ov m a de ze y nea GA S m op and a ma y PSO GA p and de PSO nea p ov op de ma nea ma a unce a n y o oad and a unce gene n y a o on n compa and REG on gene a o a unce ad on ona a n n compa y LF o mode oad on and o A REG ad o hey gene ona LF a on mode n compa A o hey o ad ona LF mode o a unce a n o oad and REG gene a on n compa on o ad ona LF mode A o hey High ✘ Complex/better Complex/better ❉ ❉ Simple ✮/Complex Simple ✮/Complex ❉ ❉ MINLP MINLP N/Difficult No Non linear Non linear High ✘ High Complex/better ✘ Complex/better ❉ ❉ Simple ✮/Complex Simple ❉ ✮/Complex ❉ MINLP N/Difficult MINLP N/Difficult No No Non linear Non linear qua y o u on p om qua e powe y o u u on g oba p op om m e za powe on me u qua hod g oba y and o op u have on m za ab p on om me y e o hod add powe and e u have g oba ab op y m o za add on e me hod and have ab y o add High ✘ High ✘ Linear/ Linear/ Excellent Excellent ✮ ✮ High Simple ✘ Simple ✮ Linear/ ✮ Excellent ✮ Simple ✮ MILP MILP N/Difficult N/Difficult MILP N/Difficult No No Linear Linear No Lin qua y o u on p om e powe u g oba op m za on me hod and have ab y o add e Linear/ Excellent ✮ Simple ✮ MILP Linear Simple ✮/Complex ❉ Tab eD 10 Pe oshow man eA ompa on ePe 10 o aaheu Pe go oa man hm ecena app ompa ed n on mu ocon aov ob go ede hm ve p app ann ng ed nhybr ob mu Tab em ob ed 10 e✮ Pe ve p o ann man ng emp pHigh ompa ob em on oo aa a go hm app ed n mu ob eand ve p o u on o a ge d bu on y em o u on o a ge d bu on y em SQP Fast High ✘ Complex/better ❉ High ✘ Simple Complex/better ✮/Complex ❉ ❉ Simple ✮/Complex ❉ Tab eTab 10 o man ea ompa on oComplex/ ast go hm app ed nng mu ob ehyb ve p ann ng p ob em o u on o a ge d bu o u on y em o a ge d bu on y em SQP N/Difficult SQP N/Difficult No Fast No Fast p ov de gene a o va ou o unde p a de gene mode a on wh o va ch we ou e cena m ed o n unde conven ona oad mode wh ch we e m ed n p ov de a on o va ou cena o unde oad mode wh ch we e m ed n conven ona High ✘ Complex/better ❉ High ✘ Complex/better ✘ ❉ Simple ✮/Complex ✘conven Cone PES Cone Poad N/Difficult Convex No Con High ✘ Complex/better ❉ High ✘ Complex/better Simple ✮/Complex ❉ ✘ Simple ✮/Complex ✘ Cone P gene N/Difficult Cone P N/Difficult No No Convex Convex eff d c ent y frombst a future perspect eff ve c The ent y hybr from d metaheur future perspect c techn ve ques The show robustness metaheur (h st gh c techn ques show robustness ( con a ned ea me p ann con ng a p ned ob em ea Fu me p ann mo ng e dec p ob on em mak Fu a ned ng he me ea mo hod me dec a p ann o on need ng mak p o ng ob be me em hod Fu a he o mo need e dec o be on mak ng me hod a o need The hybr metaheur c techn ques robustness (h gh con a ned ea me p ann ng p ob em Fu he mo e dec on mak ng me hod a o need o be Complex/ Complex/ Excellent Excellent ✮ ✮ Simple ✮/Complex Simple ✮/Complex ❉ ❉ ✘ Complex/ Excellent ✮ Simple OPF OPF N/Difficult N/Difficult No OPF N/Difficult No Excellent ✮ ❉ OPF A c a n e gence A ba ed me aheu A c c a o n examp e gence e ACS A ABC ba ed e me c aheu and hyb c d examp e ACS ABC e c hy High ✘ High ✘ Linear/Excellent Linear/Excellent ✮ ✮ High Simple ✘ ✮ Simple Linear/Excellent ✮ ✮ Simple ✮ ES ES N/Easy N/Easy ES N/Easy No Inner Opt. Inner Opt. No Inner A c a n e gence A ba c ed a me n e gence c A o ba examp ed me e aheu ACS ABC c o e c examp and e ACS d ABC e c and hyb d me hod me The hod LF The mpac LF on mpac he ba on he o ba n e ac o n on e ac a m on ng a a m MOPT ng a MOPT p ob em p ob have em a have anged a a anged a a High ✘ Linear/Excellent ✮ Simple ✮ N/Easy No Inner Opt. me hod The LF mpac on he ba o n e ac on a m ng a MOPT p ob em have a anged a L n ❉ ❉ G P N D u MO H h ✘ L n ❉ H h ✘ mp L n ✮ C mp ❉ ❉ mp ✮ C ❉ G P N u G P N D u N N MO MO con de ed a m ng a nd con ng op de ed ma a we m gh ng a o nd FDN ng op ma con we gh ed o a FDN m a nd ng op ma we gh o FDN H h ✘ H h ✘ C mp C mp ❉ ❉ H mp h ✘ mp C mp ✮ C C ✮ mp mp ✮ ❉ mp ✮ C mp ✮ MC MC N D u MC N D N u N nn Op nn Op N nn H h ✘ C mp ❉ mp ✮ C mp ✮ MC N D u N nn Op op m za me hod be de aand p op m ng za way on oN dea hod can be con ex MOP phow p ob om em ng mo em o dea w hcompa comp ex p❉ad op za on me can op be m con za de on aeve me p hod can ng be way con o de dea aThe h om comp ng ex way MOP o p dea ob w em h mo comp eSPeh ex MOP p ob em mo eLF nComp Love n ❉ ❉ mp ✮ n ✮ ❉ mp ✮ bAechn g cque uy enp ha pem euMMINLP en ed ned Tab 9u The ove acon pe mance compa on oonhow he b add gNo p ccomp epe ed eand MOP ha pm eaLF en ed nad Tab eza 9Rea ove pe mance compa on n d nu d N N Ecan n Ecu L noad ❉ mp ✮ gMILP peam cDynP u eec ha eob en ed bem neMINLP g p co eu ey The ha p ove eLF aem pe n o Tab mance eNo 9✘ on o pe he ome mance add ew compa ed MOP on o he add eobu ed MOP n d N EE h dTab M h d9 E u n ued nmu C C Qu Op M m h Op dd n m Eme u p nd n D p nd P C nn nh P nn b Qu npe m b Op m m uop n um D pm PLF nn na b u N/D fi H gh ex/be em mp emp /Comp ex T me M h d E u n Op n D p nd P nn n P b m ugene The mod The ed mod LF me hod LF and hod o wa op ba wa ed eaob LF ba p ed a✮ LF m p aman p oNo ov m The de p be mod ov de pe ed be o mance me pe hod o ao mance m and ng end ba ed po oMOP m pn ov de be The mod ed me hod and oLQu wa eMINLP ba ed LF p agene o m p ov de be e✮ ong mance am m hm Cp n A C mpu hm C Pom ben m n S n mpu P m P b m hm S App u n n n P m C mpu O App P n m n Sm u nway O ho Pwa App n nm O hlm A hm C n C mpu Poba b m S u P m App nb O h High ✘ Complex/better Complex/better ❉ ❉ Simple Simple ✮/Complex ✮/Complex ❉ ❉ cpen om u eand pe equa pec en y om The aon hyb u u d ecompa me pe aheu pec ve eoA coba The echn cunde en hyb que y d om aheu obu u eh cch ne echn pec que gh ve The hyb d me ne aheu h gh cwh echn que how obu ne GA MH/Easy MH/Easy No/Yes No/Yes Variant/AI Variant/AI High ✘ Complex/better ❉ Simple ✮/Complex ❉ GA MH/Easy No/Yes Variant/AI ha p en echn que n Tab ha eon 10 been p eCuS✘ ed n echn Tab ewh que 10 ha been p eob en ed nu Tab eme 10 High ✘ High ✘ Complex/better ❉ Simple ✮/Complex ❉ No No Real Time Real Time DynP DynP N/Difficult N/Difficult mu ob ve paon ob aGA econ o op ze S m au yw mu GA and ec PSO ve p ov em nea aou eA op d ma cu o op ze S anThe yReal GA and PSO pPon ov de nea op High ✘ High Complex/better ✘ Complex/better ❉ ❉ Simple ✮/Complex ❉ ✮/Complex ❉ No No Real Time Time DynP N/Difficult N/Difficult mu ob ve mu ava eA d ob ec cu ve o p op ob m ze ame eA m ob d ec aed cu y ve GA p oeom ob op and em m PSO ze a✮ p SC d m ov ade yu nea o GA op op and m ma PSO Sde p ov aLePg de y GA nea and PSO ma p ov de ma unce aec yhod oad REG gene aN/Difficult ap unce n ay n on ocompa ad and ona REG LF mode a ade unce on A n ang o compa n yo hey o oad on and o REG ona amu mode on naech A opop hey ona a unce aEd n o oad and REG gene aQu on n compa on o ad ona LF mode A o hey High ✘ Complex/better Complex/better ❉ Simple High ✘ ✮/Complex ✮/Complex Complex/better ❉ ❉ ❉ Simple ✮/Complex MINLP N/Difficult Non linear Non linear No Non Non linear qua y on o p on om eed p om powe eng u powe g qua op g oba y m o za u on on me za p on hod om me and eompa hod powe have and ab have oba ab o add y e✮ o add on eng me hod and have ab yob oem add High Linear/ Excellent ✮ Simple MILP N/Difficult No Linear qua y o on p om ecp powe g op za on me hod have ab y add eang High ✘ Linear/ Excellent ✮ High Linear/ Simple Excellent ✮ Simple ✮ MILP No Linear Simple High ✘ ✮/Complex Complex/better ❉ ❉ Simple ✮/Complex ❉aconven Tab ecu 10 Pe man eC ompa on ou aG go Tab hm emo 10 app Pe ed o nn mu eLinear ob eze ve on pou ann anSimple go p hm ob em app ed ob enea ve ann ng po ob em oo u on ocng au ge d bu on yen em SQP SQP N/Difficult Fast No Fa High ✘ Complex/better High ✘ Complex/better Simple ✮/Complex ❉ ❉ Simple ✮/Complex ❉ Tab e 10 Pe o man e ompa on ode go hm app ed n mu ob eN/Difficult ve p ann ng p ob em o Eu on oaN/Difficult aLF ge dobeen bu om u on yThe o o aDynP u ge d o bu am ge on d em bu on y em SQP N/Difficult SQP N/Difficult No No Fast Fast p ov de gene ampac p ou de cena gene oang ame unde on oann a❉ va oad ou mode cena p o ov ch de we gene aenop e❉ oad am m on ed mode o n va conven wh ch cena ona we eMOP o m unde ed aop n conven oad mode ona we eme m ed p ov de gene a on o va ou cena o unde a oad mode wh we e m ed n conven ona Complex/better Complex/better ❉ ❉ ✘ High ✘ ✘ Complex/better ❉ Simple Cone Phe Cone Pem Cone P N/Difficult Convex Convex No ✘ Cone Pov Convex con a ned ea me p ann con a p ned ob em ea Fu me ann mo ng e dec p ob con on em mak Fu a ned ng he me ea mo hod me dec a p ann o on need mak p ng ob be me em hod Fu he o mo need e dec o be on mak ng hod a o need con a ned ea ng p ob em Fu he e dec on mak ng hod a o need o be High ✘ High ✘ Complex/ Complex/ Excellent Excellent ✮ Simple High ✘ ✮/Complex Simple ✮/Complex Complex/ ❉ Excellent ❉ Simple ✮/Complex ❉ OPF OPF N/Difficult N/Difficult OPF N/Difficult No No Inner Opt. Inner Opt. No Inner High ✘ Complex/ Excellent ✮ Simple ✮/Complex ❉ OPF N/Difficult No Inner Opt. A a n e gence A ba me aheu A c c a o n examp e gence e ACS ABC ba ed e me c aheu and hyb c d o examp ACS ABC e c and hy L n E n ✮ mp ✮ E N E nn Op A c a n e gence A ba c ed a me n e aheu gence o ba examp ed me e aheu ACS ABC c o e c examp and e hyb ACS d ABC e c and hyb d me hod LF mpac on he ba o n e me ac hod on a The m ng LF a mpac MOPT on p ob he em ba have o n a e anged ac on a a a m ng a MOPT p ob em have anged H h ✘ L n E n ✮ H h ✘ L n mp E ✮ n ✮ mp ✮ N E N E N nn N Op nn Op me hod The on ba o ac on a m ng MOPT p ob em have a anged a a H h ✘ H h ✘ L n L n ❉ ❉ H h ✘ L ❉ n ❉ ❉ ❉ G P G P N D u N D u P N D N u N MO MO N M L n ❉ ❉ G P N D u MO con ed a m ng a nd con ng op de ed ma we m gh ng a o nd FDN op ma con we gh de ed o FDN m ng a nd ng op ma we gh o FDN con de ed m a nd ng op ma we gh o FDN C mp C mp ❉ ❉ mp ✮ mp C mp ✮ C C ✮ mp mp ✮ ❉ mp ✮ C mp ✮ MC MC MC nn nn H h ✘ C mp ❉ mp ✮ C mp ✮ MC N nn Op op za on me hod can be con de a p om op m ng za way on me o dea hod w can h comp be con ex de MOP a p p ob om em ng mo way e o dea w h comp ex MOP p ob em m op m za on me hod can op be m con za de on a me p hod om can ng be way con o de dea a w p h om comp ng ex way MOP o p dea ob w em h mo comp e ex p ob em mo e H h ✘ H h ✘ L n L n ❉ ❉ H h ✘ mp ✮ mp L n ✮ ❉ mp ✮ b g p c b u g e p ha c u p e e ha en p ed e n en Tab ed e n 9 Tab The e ove 9 The a ove pe o a mance pe o compa mance compa on o he on add o he e add ed MOP e ed MOP n d n d N E N E n d N E N N Op Op N nn L ne ❉ ✮ b gty pN/Difficult ceComplex/better u ha pOp e❉ en ed nTab Tab eQu 9✘ The ove aand pe mance compa onon he add eVariant/AI ed n d N Eu MGA hy d om uu nMINLP M C h d Eu ued nd m Cgene n✘ D pechn nd M h Op PComplex/better m nn E nLF n Ped u boLF nA m D pem nd nn Qu bhyb Op m m n D p/A nd Pn nn n ehow bmom MH/Ea yyThe H gh Comp ex/be eApp No/Ye SuPy mp enon /Comp ex Va an M h d n C Qu Op m D p nd P nn nng P b m ugene The mod ed LF me hod and mod oQu wa LF ba me ed hod LF p and o m oExcellent p wa ov The edme de ba mod ed enman pe p aoba o ome mance hod pSimple ov a✘ and m de be o wa ePg pe eand ba o mance LF p auho m o m de be pe oobu mance au Average Average A hm CaEon n C A hm C m S n u C n mpu Pses m Ppowerfu m un n nm P m hm O C h App n n C n mpu O PMOP b Sechn ucompa PP App Average C nue C mpu Pon b m u n P App n O h High ✘ High ✘ Complex/better Complex/better ❉ ❉ Simple ✮/Complex Simple ✮/Complex ❉ ❉ ea cA en yhm om u eand pe pec ve The hyb cva d en ySPe aheu om aYes co u u echn pe que pec how ve The obu ne d me h aheu cmphod que ne GA GA MH/Easy MH/Easy No/Yes No/Yes Variant/AI ✘ Complex/better ❉ Simple ❉ ✮/Complex ❉ eMINLP en eo econ pe cGA pec en om The hyb up eman me pe aheu pec ve ceLinear/ The hyb d me how aheu obu compa ne echn h que gh how ne h gh MH/Easy GA MH/Easy No/Yes No/Yes Variant/AI Variant/AI echn que been p eCone en ed ny Tab eyPea 10 echn que ha been p eobu en ed n Tab eon 10 echn que ha been pge eapp en echn ed n que Tab 10 been p eme en ed n 10 High Complex/better High Simple ✘ ✮/Complex Complex/better ❉ ❉ ✮/Complex ❉ No Real Time Real Time No Real DynP DynP DynP N/Difficult DynP N/Difficult No Real Time unce a ampu unce n o n oad o oad REG and REG aanHigh gene an compa on n compa on on ad unce o ona ang ad n ona o mode oad LF mode A REG oed hey A o hey aOp on n on ad ona a unce ave nha y o oad and REG gene ame on n compa on o ad ona LF mode A o hey High ✘ Complex/better ❉ Simple ✮/Complex ❉ No Non linear High ✘ High ✘ Simple Complex/better ✮/Complex ❉ ❉ Simple ✮/Complex ❉ MINLP N/Difficult No Non No linear Non linear qua uton ses powerfu qua g oba so opt m prom zat on methods g have opt m ty zat to address methods and have ab ty to add g obaTab opt m zat methods and have ab ty to y ogo p om qua eG powe y o u u on gaddress oba p op om m eng za powe on qua hod g oba and o op u have m za ab p on om y me eng oCab hod add powe and emp u have oba ab op y m ogh za add eng me and have ab yochSimple owe add Linear/ ✮ High Simple ✮ Linear/ Excellent ✮ Simple ✮ MILP MILP N/Difficult Linear No LineT High ✘ Linear/ Excellent High ✮ ✘ Excellent Simple ✮ ✮ Simple ✮ MILP N/Difficult MILP N/Difficult No No Linear Linear High ✘ High ✘ Complex/better Complex/better ❉ ❉ Simple Simple ✮/Complex ❉ High ❉ ✘ Complex/better ❉ Tab eaam 10 Pe oup ehon ompa on ou Tab aG go ebSque 10 hm oapp ed ea n❉ mu ob on e✮/Complex oLF ve aob go p ann hm ng p app ob ed em n mu ob eMOPT pSimple ann ng p ob em oprom uhod on ouha aum d bu on em o u on obe aon ge d bu on y em SQP SQP N/Difficult N/Difficult No SQP No N/Difficult Fast Fast No ❉ e 10ty Pe so o man eons) ompa on ode hm ed n mu ob eN/Difficult ve p ann ng p ob em oOP ucqua on oaN/Difficult aThe ge du bu o on u on y o ge d bu oann u on yp o em ge d bu on y em SQP Fast p ov de gene on va ou cena oaaut unde ov aHigh de oad gene mode ame on wh ch ou we e❉ cena ed o unde ov n conven de agene oad ona a✮/Complex mode on o va wh ou ch cena we eann m unde ed nve aov conven oad mode ona wh Complex/better ❉ ✮/Complex ❉ PSO PSO MH/Easy MH/Easy Yes Variant/AI Variant/AI p ov de gene aN on onbons) va ou cena omp unde aeny oad mode wh ch we eC m ed n conven ona Complex/better ❉ Simple ✮/Complex PSO MH/Easy Yes Variant/AI High ✘ Complex/better ✘ ✘ ❉ Simple ✮/Complex ✘a Cone PE Cone Pem Cone P N/Difficult Convex Convex No Con High ✘ Complex/better ❉ Simple ✮/Complex ✘ P N/Difficult No Convex con a con ned ea a ned me ann ng p ann p ob ng em p ob Fu con em he Fu a mo ned he e dec ea mo e me on dec mak p ann on ng mak me p hod ng me em a hod o Fu need a he o mo o need be e dec o be on mak ng me hod a need a ned ea me ng ob em Fu he mo e dec on mak ng me hod a o need o be C mp Ee ✮ mp ✮ C mp ❉ OP N D u N nn H h ✘ CND mp E n ✮ H h ✘ C mp mp ✮ E C mp nm ✮ ❉ mp ✮ C mp ❉ N D OP N D un N nn N Op A c a n e gence A ba ed aheu c o examp e ACS ABC e c and hyb d H h ✘ L E L n n E ✮ n ✮ mp ✮ mp L n ✮ E n ✮ mp ✮ E E E N E E N E Op nn Op nn A c a n e gence A A ba c ed a me n A e aheu gence c a c n A e o gence ba examp ed me A aheu ba ACS ed me ABC c aheu o e c examp and c e hyb o ACS examp d ABC e ACS e c ABC and hyb e c d and hyb d me LF mpac on me he hod ba The o LF n e mpac ac on on a he m ng ba a me MOPT o hod n p ac The ob on em LF a have m mpac ng a a on anged MOPT he ba a p a ob o em n e have ac on a anged a m ng a a a p ob em have anged L n E n ✮ mp ✮ E nn Op me hod The LF mpac on he ba o n e ac on a m ng a MOPT p ob em have a anged a a H h ✘ L n L n ❉ ❉ H h ✘ L ❉ n ❉ ❉ ❉ G P P N u N D P N D u N MO MO N M H ✘ L n ❉ ❉ G P N D u N MO con ed a ng a nd con ng op de ed ma we m gh ng a o nd FDN op ma con we gh de ed o a FDN m ng a nd op ma we gh o FDN con de ed ng a nd ng op ma we gh o FDN H h ✘ C C mp ❉ ❉ mp ✮ mp mp ✮ C C ✮ mp mp ✮ ❉ mp ✮ C mp ✮ MC MC MC N nn Op nn C mp ❉ mp ✮ C mp ✮ MC nn op m za on me hod can be con de a p om op m ng za way on me o dea hod w can h comp be con ex de MOP a p p ob om em ng mo way e o dea w h comp ex MOP p ob em m op m za on me hod can op be m con za de on a me p hod om can ng be way con o de dea a w p h om comp ng ex way MOP o p dea ob w em h mo comp e ex MOP p ob em mo e ❉ ❉ H h ✘ L ❉ H h ✘ L n ❉ mp ✮ H h ✘ mp L n ✮ ❉ mp ✮ b g p c u e ha p e en ed Tab e 9 The ove b g a p c pe u e o ha mance p e compa en ed n on Tab o e he 9 The add ove e ed a MOP pe o mance compa on o he add e ed M n d N E n d N E N n d N E N nn Op Op N nn ❉ H h ✘ L n ❉ mp ✮ b g p c u e ha p e en ed n Tab e 9 The ove a pe o mance compa on o he add e ed MOP n d N E N Op M h d E u n M C h d Qu E u Op n m C n D p Qu nd Op P m nn n P n b m M D h p d nd u E P nn u n n P b m C Qu u Op m n D p nd P nn Comp ex/be em Ye SN/Difficult mp /Comp ex Va an /A PSO MH/Ea yy Ave age M h d E uha need Qu m n D p nd Pad nn nng Pepe b uhp The mod ed LF me hod The and mod o wa epe LF me ed hod LF p and o m oExcellent p wa ov The eme de ba mod be ed LF enman ed pe p LF awh onPu o me mance m hod p ov ae✘ and m de be o eagab pe emode ba ed mance LF aho m oNo/Yes ng m phod ovpSimple de beng ephow pe mance aFa Average Average The mod ed me hod and o10 wa ePMINLP ba ed LF p aNo/Yes o m ov de be e✮ o mance am m A hm C n C mpu A hm m So u C m C mpu App Pwa bm m n So u❉ nA O Pon mo App nab noobu O Average Average A hm C n C mpu b nm Su PMILP m nHigh n O h High ✘ ✘ Complex/better Complex/better ❉ ❉ Simple High Simple ✘ ✮/Complex ✮/Complex Complex/better ❉ ❉ ❉ Simple ✮/Complex ❉ equa cGA en y om u u ebeen pec ve The hyb equa cGA d en y aheu om aYes coad u eand pe que pec how ve The obu hyb ne d me h gh aheu cExcellent echn que ne GA MH/Easy MH/Easy MH/Easy No/Yes Variant/AI Variant/AI Varia High ✘ Complex/better ❉ Simple ✮/Complex ❉ equa caen yTab om ay u eMINLP equa pe cGA pec en ve om The aMH/Easy hyb uApp uFu d eman me pe aheu pec ve cLinear/ The echn hyb d me how aheu obu compa ne hng que gh how obu ne gh MH/Easy No/Yes Variant/AI echn que echn ha que been ha been en p ed eC✘ n en Tab ed eLinear/ n Tab 10 echn que p eeza en ed n Tab eOp 10 Real Time DynP N/Difficult High ✘ Complex/better ❉ High ✘ Simple ✮/Complex ❉ ❉ Simple ✮/Complex No Real No Real Time DynP DynP N/Difficult unce o oad and a unce REG gene n yep o on oad n and compa REG on gene a o aaComplex/better unce ad on ona aABC n n compa y LF on mode on o A REG o hey gene ona LF on n compa o hey ona mode A o High ✘ Complex/better ❉ Simple High ✘ ✮/Complex Complex/better ❉ ❉ Simple ✮/Complex ❉ N/Difficult No Non linear No Non l High ✘ Complex/better ❉ High ✘ Complex/better Simple ✮/Complex ❉ ❉ Simple ✮/Complex ❉ MINLP N/Difficult N/Difficult No No Non linear Non linear y u on p om eba powe u g oba op y m za u on on me hod om and eTime powe have u ab oba y o op add e✮ za on me and have op add Linear/ ✮ ✮ Simple Simple ✮ High ✘ Linear/ ✮ Si MILP MILP Linear Linear oacuPN/Difficult p powe yPo o u u on g10 oba p om op m ecp powe on me hod g oba and op have za ab me y op hod add and y o add eang Excellent Simple ✮ MILP Linear High High ✘ Complex/better Complex/better ❉ ❉ Simple High ✮/Complex Simple ✮/Complex Complex/better ❉ ❉ ❉ ✮/Complex eaon 10 oeom man eA ompa Tab on ea o aaheu Pe go o hm app nnu mu o Tab aG ob go ebque eE❉ 10 hm ve p o app ann ng ed ep n❉ ob mu em ob on eYes op ahave go p ann ng p app ob ed em n mu ob ve ann em SQP SQP N/Difficult N/Difficult SQP N/Difficult No No Fast Fast No High ✘ Complex/better ❉ Simple SQP N/Difficult No Fast p ov de p gene ov de aN gene aaLF va on ou va cena ou o cena oeemp unde oad a✮ mode oad mode ov de we wh gene ch m we a✮/Complex on ed eMOP o n m va conven ed ou n conven ona ona unde aad oad mode wh we ehm PSO PSO MH/Easy Yes Variant/AI Variant/AI p ov de gene aN on va ou cena omp aExcellent oad mode wh ch we eC m ed nn❉ conven ona Complex/better Simple ✮/Complex PSO MH/Easy MH/Easy Yes Variant/AI C ✘ C neMINLP u C n H h ✘ CN mp ❉ H C mp ✮ C mp ❉ ✘ mp ✮ mp ✘ C ncon PyMC D uLF C n N D u N C N n Co no ned ea me p ann con ng adba p ned em ea me he ann mo ng eed dec p ob con on em mak Fu adFDN ned ng me mo hod me aechn p opechn ann on need mak o ob be me em hod Fu he o mo need e❉ dec be ng me ac❉ och need H h ✘ mp E n E ✮ n ✮ mp ✮ mp C mp ✮ C mp C ❉ mp E ❉ n mp ✮ Chod mp ❉y OP OP N D N D u OP N D N ua nn nn Cunde mp E nPe ✮ mp ✮ C mp ❉ OP N D u nn aha ned gence A ba ed me compa o examp A emp ACS che ABC e✮ edec gence cme and A hyb ba d me aheu cMO examp ACS eMOPT and hy H ✘ L L n ✮ n ✮ H hnn ✘ mp ✮ L nhm ✮ E nMO ✮ mp E Eba E N N N Op Op N nn A aann nOp ePe gence A con ed aPu me n gence A A o ba com examp awe ed n eEo aheu gence ACS A con ba o eEon ed chave examp me and aheu edanged hyb ACS d cve ABC o emp c✮ and eVariant/AI ACS hyb d eac cmak and hyb d me hod The mpac on o n eon me ac hod on ahhmpu The ng ame mpac MOPT on p ob ba em o n ach hod eed ac on The ade LF aPmp m mpac ng aCexamp MOPT he p ba ob o n have eeon aeApp on anged abe m aLF aob aNo ob H h ✘ L n E n ✮ ✮ E N Op me hod LF mpac on he ba o n eove ac on a❉ m ng p ob em have aon ann aOp H h ✘ L non L n mp ✮ mp L ❉ n ❉ ❉ ❉ G PPE G P N D u N D u P N N MO M L n ❉ ❉ G uE con de con ed amod de m ed ap aohod m nd ng ng ano❉ op nd ma ng ma gh con we op gh de o aMOP FDN m am nd ng op ma we gh ohed FDN con de ed m aoo nd ng op ma we gh o FDN Average Average C mp H ❉ ✘ C mp mp ✮ ❉ C ✮ H h ✘ mp ✮ C C mp ❉ mp ✮ C mp ✮ N D uprob MC N nn nn Average H h ✘ mp ❉ mp ✮ mp ✮ MC N nn op m za on me hod can be con de aThe om ng way oLF dea w h comp ex MOP p ob em mo eaheu op m za me hod can op be m con za op aob me m p za om can ng me be way hod con o can de dea be aThe w con p h comp de ng aed ex way p om omp p dea ng way w em h mo oed comp dea eSN/Difficult ex w h p ob ex em MOP mo p ob eon em mo eSimple ❉ ❉ H h ✘ L ❉ H hp ✘ L n mp mp H ✘ nN ❉ b gque pnyA con uN ebcng p eeLF en ed b gn Tab cnd egene ePSO 9D ha The p ove ed en ame ed pe n Tab mance eC 9Cunde b compa The g cove u ehe abe ha pe p he o eLF mance add en eadd compa n Tab MOP eng 9C on The o ove add aanged pe eacena ed o mance MOP compa on oeng he add ed dP N E n d N E N n N Op N Op N ❉ ❉ L n ❉ mp ✮ b g p uup eThe ha erea ed n Tab eChop 9✘ The pe oea mance compa on o he add en✮ ed MOP n d N E M h E uon C Qu M h Op m E n u n D nd C P nn n P b Op m m n uABC p Pque nn nABC b so m HS MH/Ea y Ave age Comp ex/Ave age mp emp /Comp ex Va an /A hconstra d u n ndec Qu m n D p P n P b m u The mod me hod The and mod o wa ed eaheu LF ed hod LF and o m p wa ov The eaque de ba mod ed LF ed pe p ame o o m hod p ov ae10 m ng be obQu wa pe ehe ba o mance LF aem m oLDng m phod ov de oobu mance m Average Average Average The ed LF me hod and oLF wa enPom ba ed LF p aNo/Yes ov de be eaEed pe oon mance m A hm C n C A hm b C m Shave u nD n C P mpu m App m S n uExcellent n POp O h nP pe h rea tGA me p ng constra ned Furthermore tmo me p dec ann ng on-mak prob ng ems methods Furthermore aand so need dec to saaop on-mak be ng methods a✮ need Average hm C E ned CC mpu m S uu P App n n h ms AMFurthermore sechn on-mak methods ahe so need to Complex/better ❉ Simple ✮/Complex ❉ equa cGA en y om ang u u een pe pec ve hyb equa cunde d en me y aheu om ang cuuoad u u echn eob pe que pec how ve The obu hyb ne d me h gh cnd echn how ne MH/Easy Variant/AI High ✘ Complex/better High ✘ Simple Complex/better ✮/Complex ❉ ❉ Simple ✮/Complex ❉ equa cTab om u eno equa pe pec en ym ve om The ang hyb u u eand me pe aheu pec ve cp echn hyb que d how aheu obu cMOPT ne echn que gh how obu ne h gh MH/Easy GA MH/Easy No/Yes Variant/AI No/Yes echn ha been p ennn en ed n Tab eLinear/ 10 echn ha been p emance en ed n Tab eLinear/ 10 been pon en ed n Tab eaN/Difficult 10 High ✘ High ✘ Complex/better ❉ ✮/Complex ❉ No Real Time No Real T DynP N/Difficult DynP N/Difficult High ✘ Complex/better ❉ High ✘ Complex/better Simple ✮/Complex ❉ ❉ Simple ✮/Complex ❉ DynP N/Difficult DynP No No Real Time Real Time aen unce n yPe o oad and a unce REG n y aLF oad n and compa REG on gene a o aMC unce ad on ona ame n n compa y o mode on and o A REG ad o hey gene ona LF aecomp mode n compa A o on oon ad ona LF mode A oapnM a unce ade nN y o oad REG gene as on n compa on o ad ona LF mode A o hey Complex/better Complex/better ❉ ❉ Simple ✮/Complex Simple ✮/Complex ❉ High ❉ ✘ Complex/better ❉ Complex/Average Complex/Average ❉ ❉ ✮/Complex ✘ ✘ MINLP MINLP N/Difficult MINLP Non linear Non linear No HS HS MH/Easy MH/Easy -va Variant/AI Variant/AI Complex/Average ❉ Simple ✮/Complex ✘ Non MH/Easy -mp Variant/AI y o on p om eba powe u g oba op y m za u on on me p hod om and ewh powe have u ab oba y o add e✮ za me and have ab yoeOeSimple opve add High ✘ High ✘ Excellent Linear/ ✮ High Simple ✘ ✮ Simple ✮ ✮ Simple ✮ MILP MILP N/Difficult MILP N/Difficult No No Linear Linear No Lin y o uha on p eMINLP powe ycems o u u on gmpac oba p om op m ebe za on me hod g oba and op m za ab on me yob op hod and have ab y o add eang High ✘ Linear/ Excellent Simple ✮ MILP N/Difficult No Linear ❉ ea oom eque ompa on o aD go hm app ed n mu ob ede ve p ann ng ob Tab em eh Pe o man eed ompa on om aahey go hm app ed n mu QP N D H h ✘ mp ❉ H h ✘ C mp mp ✮ C mp ❉ ❉ mp ✮ C mp eTab 10 Pe oaN eman ompa Tab on eHS o10 Pe go o hm man app eo ompa ed n mu on onu ob aO go ep ve hm p ann app ng ed pD n ob mu em eou ve p ann ng p em QP N D10 uman QP N D u N N p ov de gene ame p ou de cena gene o ame unde on ahe va oad ou mode cena o wh ov ch de we gene a✮ enaMOP oad ae✮ m on mode ed o n conven ch cena ona we eMOP o m unde n conven oad mode ona wh ch we eme m ed n✘ conven Complex/better Complex/better ❉ Simple ✮/Complex ✮/Complex Complex/better ❉ Simple ✮/Complex ❉ PSO PSO MH/Easy MH/Easy PSO MH/Easy Yes Yes Variant/AI Variant/AI Yes Varia Complex/better ❉ Simple ✮/Complex ❉ PSO MH/Easy Yes Variant/AI H h ✘ H h ✘ C mp C mp ❉ ❉ H h ✘ ✮ mp C mp ✮ C C ✘ mp mp ✘ ❉ mp ✮ C mp C nva C n P u N D u C n N D N N C nnn C nbe N C C mp ✘ C n Pov u nlinear con acPng ned ea me p ann ng p ob em Fu con he aPmo ned eway dec ea me on mak p ann ng ng me p hod ob em ang oob Fu need he mo om be eVariant/AI dec mak ng hod aob need con aG ned ea p ann con aOP p ned ob em ea Fu he p ann ng epowe dec p ob em on mak Fu ng he me mo hod e✮ dec a-o o on need mak o ng be me hod a✮ o need o C mp C E mp n E ✮ C mp ❉ Ecompa ❉ nC OP N DC OP nn nn H h ✘ C mp E no ✮ mp ✮ C mp OP N D N H h ✘ L E Lu n ned EExcellent ✮ n ✮ mp ✮ mp L ✮ E n❉ ✮ mp ✮ E E E N E N E N Op Op nn me hod me The hod LF The on mpac ba on he o ba n e ac o n on e ac a m on ng a a m MOPT me ng hod a MOPT p ob The em p LF ob have mpac em a have anged on he a a ba anged a o n a e ac on a m ng a MOPT ob L n E n ✮ mp ✮ E N Ea Op me hod The LF mpac on he ba o n e ac on a m ng a MOPT p ob em have a anged a a ❉ ❉ ❉ L n ❉ H h ✘ L n ❉ ❉ H h ✘ mp ✮ C L n mp ❉ ❉ mp ✮ C mp ❉ P Dmp u G P N u G P N u N MO MO N M H h ✘ L n ❉ mp ✮ C mp ❉ G P N D u N MO con de ed a m ng a nd con ng op de ma ed we m gh ng a o nd FDN ng op ma con we gh ed o FDN m ng a nd ng op ma we gh o FDN Average Average H h ✘ C mp ❉ C mp mp ✮ ❉ C mp ✮ mp ✮ C mp H ✮ h ✘ C mp ❉ mp MC MC N D u N MC nn N Op D u nn N Average Average C mp ❉ ✮ MC nn Op op m za on me hod can be con de a p om ng way o dea op w m h za comp on me ex MOP hod p can ob be em con mo de e a p om ng way o dea w h comp ex MOP p ob em op m za on me hod can op be m con za de on a me p hod om can op ng be way m con za o on de dea me a w hod p h om comp can ng be ex con de o p dea a ob p w em om h mo comp ng e way ex o dea p ob w em h comp mo e ex MOP p ob em mo e ❉ ❉ ❉ H h ✘ L n ❉ H h ✘ mp L n ❉ mp ✮ b g p c u e ha e en ed n Tab e 9 The ove b g a p c pe u e o ha mance p e compa en n on Tab o e 9 he The b add g p ove e c u ed a e MOP pe ha o p mance e en ed n Tab e on 9 o The he ove add a e pe ed o MOP mance compa on n d N E n d N E N Op N nn ❉ HC h ✘ L n ❉ mp ✮ b g p c u e ha p e en ed n Tab e 9 The ove a pe o mance compa on o he add e ed MOP n dd N nn Op M h d E u n C M h Qu d Op E m u n n D C p nd Qu P nn n Op P m b m n D p u nd P nn n P b m u TLA MH/Ave age Low Comp ex/Ave age No Comp x a wo d Non nea M h EN Eu n Qu Op n D p nd nn n P b m The mod The ed mod LF me ed hod LF me and hod and wa e o ba wa ed The e LF ba mod p ed a LF o ed m p LF a p o me ov m de hod p be ov and e de pe be o o wa e mance pe e ba o a ed mance m LF ng p a a m o ng m p ov de be e pe o mance a m Average The mod ed LF me hod and o wa e ba ed LF p a m p ov de be e pe o mance a m ng A hm C n C A mpu hm C P b m S n u C n mpu A P m hm P b C m S App u n n n n C P mpu m O h P App b m S n u n n P O m h App n n O h Average Average High Complex/better ❉ Simple High ✘ ✮/Complex Complex/better ❉ ❉ Simple ✮/Complex ❉y Simple equa cGA en y om u u eon pe pec ve hyb d me cuoad echn que how obu ne h gh MH/Easy GA MH/Easy No/Yes Variant/AI No/Yes Varia High ✘ Complex/better ❉ High ✘ Complex/better Simple ✮/Complex ❉ ❉ Simple ✮/Complex ❉ equa cGA en yode om ayuame uo uom eo ep pe pec en yo ve om The en ao u yD eompa me pe aed aheu pec u u ve eLThe pe The echn pec hyb que ve d me how hyb aheu obu d me com ne aheu echn h que gh cme echn how que obu how ne obu h ne hmak gh MH/Easy MH/Easy No/Yes No/Yes Variant/AI echn que ha been p en echn ed n que Tab ha ep 10 been p e✘ en ed n Tab echn eowh 10 que ha been p eN en ed n Tab 10 echn been p enH en n Tab econ 10 ✘ High ✘ Complex/better ❉ DynP DynP N/Difficult N/Difficult DynP N/Difficult Real Time Real Time No Real Time DynP a unce aea n oad and a unce REG gene nme yEhyb aLF o on oad n and compa REG on gene a o aunce ad on ona aoad n n y LF o✮ mode on and o A REG ad oeme hey gene ona LF aob on mode n compa A o on hey oon ad ona LF mode AoNon o nlm a unce aob neha y o oad and REG gene ang on n compa on ad ona LF mode A o hey High High ✘ Complex/better Complex/better ❉ ❉ Simple High ✘ ✮/Complex Simple ✮/Complex Complex/better ❉ ❉ ❉ ✮/Complex ❉ Complex/Average Complex/Average ❉ ❉ ✘ ✘ MINLP MINLP MINLP N/Difficult No No Non linear Non linear No HS HS MH/Easy MH/Easy -Variant/AI Variant/AI Variant/AI High ✘ Complex/better ❉ ❉ Complex/Average ❉ ❉ Simple ✮/Complex Simple ✘ ✮/Complex ✘ MINLP N/Difficult No Non HS MH/Easy HS MH/Easy -ep Variant/AI Variant/AI y o u on om eom u g oba qua op m za u on me p hod and ewh powe have ab oba y o add m e✮ za on me hod and have ab an✘ add H h ✘ n E n ✮ mp ✮ M LP u N L n yaG u on pEopt qua eeng powe yPque u u on g oba p om op m epowe on me u hod oba and op have m za ab on me yo-ng oend hod add and have y o add engh H h ✘ LComplex/better n ✮ H hob ✘ L nn mp EThe ✮ n mp ✮ M LP N D M LP N u N Lmak N n LOp n ❉ ❉ ❉ Tab 10 Tab Pe ea oceD 10 man Pe eu o man eza ompa on oma amp go o hm aov go app hm ed app non mu ed ob no mu ve ob p ebe ann ve ng pu ann ng em p ob QP QP N D N D QP N D ❉ Tab 10 Pe od man eop ompa on aunde go hm app ed n mu ob ve pe✮ ann ng p ob em N p ov gene on p ou ov cena gene ame unde oann aou va oad ou mode cena p o ch de we gene aaheu enacompa oad aN m on mode ed o nneed va conven ou ch cena ona we eem o m unde ed aop n conven oad mode ona wh chSimple we eme m ed conven Complex/better Simple ✮/Complex PSO MH/Easy Yes Variant/AI ov de gene aN on o va cena o unde ay mode wh ch we eC m ed ned conven ona ❉ Simple Complex/better ✮/Complex ❉ ❉ Simple ✮/Complex PSO MH/Easy PSO MH/Easy Yes Variant/AI Yes h H h ✘ C mp mp ❉ ❉ H mp h ✘ ✮ mp mp ✮ C C ✘ mp mp ✘ ❉ mp ✮ C mp C nva P C n Pde u up C N C nem C N C H h ✘ C mp ❉ mp ✮ C mp ✘ C n PG u N nlinear con acGA ned ea p ann ng p ob em Fu con he aPmo ned eo dec ea me on mak p ann ng ng hod ob angab oaex Fu need he mo o be eVariant/AI dec ngob hod o anged need con ned p ann con ng aeC p ned em ea Fu he mo epowe dec p em on mak Fu ng he me mo hod eo dec a-✮ on o ng hod om need o be C C E mp n E ✮ ✮ C mp ❉ E ❉ nC ❉ OP OP N D N D OP N D nn Op nn CComplex/Average mp E no ✮ ❉ OP N D nn L n E n❉ H ✮ h ✘ LC n E n mp ✮ H h ✘ mp L ✮ E n❉ ✮ mp ✮ E N E N E E N nn Op N nn me hod The LF mpac on me he hod ba The o n e mpac ac on on a he m ba ng a me o MOPT n hod e p The ob on em LF a m have mpac a on MOPT anged he ba a p ob a o em n e have ac on a anged a m ng a a a MOPT p em have a H hng ✘ L E nac ✮ mp ✮ EQP E N Op ❉ ❉ ❉ H h ✘ L ned ❉ H h ✘ L n mp ✮ ❉ C mp ❉ H ❉ h ✘ Lng n ❉ mp P N Da u❉ P N D und N G N P MO N D u MO N L n ❉ mp ✮ C mp ❉ G N D u MO con de ed a m ng a nd ng op ma we gh con o de FDN ed m ng a ng op ma we gh o FDN ng at f nd ng ma cons we dered ghts a for m ng FDNs at f nd ng opt ma we ghts for FDNs de ed a m ng nd con op de ed ma a we m ng gh a o FDN ng we gh o FDN we ghts FDNs Average Average Average C mp ❉ H h ✘ mp C C mp mp ✮ ❉ mp ✮ C mp ✮ MC MC N D u nn N nn Average C mp mp ✮ C mp ✮ MC NNDE u a m nn ❉ H h ✘ L n ❉ H h ✘ L n mp ✮ ❉ b g p c b u g e p ha c u p e e ha en p ed e n en Tab ed e n 9 Tab The e ove 9 The a ove pe a mance pe o b compa mance g p c compa u on e ha o he p on e add o en he e add ed n Tab MOP e e ed 9 MOP The ove a pe o mance compa on n d N E n d N E N N Op nn Op ❉ ❉ HAcon h ✘ L n ❉ mp ✮ b g p c u e ha p e en ed n Tab e 9 The ove a pe mance compa on o he add e ed MOP ncons d for dered N Op M h d E u n M h d Qu E u Op n m C n M D p h Qu nd d Op P E m nn u n P n b m D C p nd Qu u P nn n Op P m b n D p u nd P nn n P b m u BB BC MH/Ave age Low Comp ex/Ave age Ye Comp Non nea The mod ed LF me hod The and mod o wa e LF ba me ed LF hod p a and o m o p wa ov The e de ba mod be ed e LF ed pe p LF a o o me mance m hod p a ov and m de ng be o wa e pe e ba o ed mance LF p a a m o m p ov de be e pe o mance a m Average Average A hm C n C mpu P b m S u n P m App n A n hm O C h n C mpu P b m S u P m App Average Average hm C n C A mpu hm C P b m S n u n C mpu P m P b m S App u n n n P m O h App n n O h ❉ ❉ High ✘ Complex/better ❉ c yo en yMH/Easy aon u u eunce pe pec ve The hyb d me aheu econ con cechn en y que om how aYes u u obu eand pe ne pec h ve gh The hyb d me aheu cwh echn que how neVaria GA MH/Easy MH/Easy No/Yes No/Yes GA MH/Easy Variant/AI Variant/AI No/Yes equa cepechn en om aom u uom eo econ pe pec en y ve om The ao u ed eompa me pe en aheu pec y ve The ahod echn u hyb u que ede d pe how pec aheu ve obu cmak hyb h que gh me aheu how chm echn ne que h how obu ne hC gh MH/Easy No/Yes Variant/AI que been pDynP en ed Tab ep 10 echn que ha been p eQP ed n Tab euuoad echn que ha been eobu ed Tab 10 echn ha p eH en ed n Tab 10 High ✘ High ✘ Complex/better Complex/better ❉ Simple High ✘ ✮/Complex Simple ✮/Complex Complex/better ❉ ❉ ✮/Complex ❉ No No Real Time Real Time No Real DynP N/Difficult N/Difficult DynP N/Difficult High ✘ Complex/better ❉ ✮/Complex ❉ No Real Time DynP N/Difficult ampac unce a aGA unce n yn o aon n oad yhyb o and oad REG and gene au gene on a an unce compa on ame n n yon on opdec o on o ona REG ad LF gene ona mode LF aob on mode A n oho compa hey A o on oon ad ona LF mode A o nnT a nqua y o oad and REG gene n compa on o ad ona LF mode A o hey C mp ❉ mp ✮ C mp ❉ Complex/Average ❉ ❉ Simple ✮/Complex ✘ ❉ ✘ M NLP N D u N nn nn HS HS MH/Easy MH/Easy HS MH/Easy -ne Variant/AI Variant/AI Varia H h ✘ C mp H hob ✘ C mp mp ✮ C mp ❉ ❉ mp ✮ C mp ❉ ✘ M NLP D ha NLP N D u❉ N N n nSimple N nen n HS MH/Easy -10 Variant/AI qua y o u on om eac u g oba op m za me hod and have ab y add e✮ H h ✘ H h ✘ Lom n E L n n ✮ n H h ✘ mp ✮ mp n ✮ Een ✮ mp ✮ M LP M LP u M LP N N L n Lnn N yaov uN on p qua eeng powe yque oaaob u gbeen oba p om op u on m epowe za powe p on om me eceLwe gFu powe oba and op u m g oba za ab op yThe m oEend za add on and ed me have hod ab and yo have add ab enngh yaOp oaYes ech H h ✘ n E n mp ✮ M LP u N L n ❉ ❉ ❉ Tab 10 Pe oD eu on oma aba go hm Tab app emo ed 10 n Pe mu o man ob eechn ve ompa p ann on ng o✮ ahave go em ed mu ob enSimple ve peme ann ng p obu ob emSimple QP QP N D N D N D ❉ Tab ede 10 oua man eCann on ou am go hm app ed nREG mu ob emp ve p ann ng ob em QP N va p ou ov de cena gene ame unde on oann aacou va oad ou mode cena p o wh ov unde ch we gene a✮ enacompa oad ao m on mode ed o nad va conven ou wh ch cena ona we eComplex/Average opL✮ m unde ed ahe conven oad mode we m ed n✘ Complex/better Simple ✮/Complex Complex/better Simple ✮/Complex PSO MH/Easy PSO MH/Easy Variant/AI p ov de gene aN on ono va cena o unde aEen oad mode wh ch we eC m ed nhave conven ona ❉ Complex/better Simple ✮/Complex ❉ Simple ✮/Complex PSO PSO MH/Easy Yes Yes Variant/AI Variant/AI h ✘ H h ✘ C mp C mp ❉ ❉ H mp h ✘ ✮ mp mp C C ✘ mp ✘ ❉ mp ✮ mp nompa P C nM P u u C n N N C napp C N CL H h ✘ mp ❉ mp ✮ C mp ✘ C n P u N na con acGA ned ea me p ann ng p ob em he ned eann dec ea me on p ann ng ng me p hod ob em ap oaeex Fu mo be eo dec mak ng o need ned ea me p ng aeC ned em ea Fu he p mo ng epe dec p em on mak Fu ng me e✮ a-o o on need mak o ng be hod o✘ need o be C mp E n ✮ mp ✮ C mp ❉ C mp E ❉ nC ❉ OP N D u❉ OP N D N OP N D nn nn CComplex/Average mp E nhave ✮ ❉ OP N D nn Op H h ✘ L n Eo n❉ ✮ L n E mp ✮ ✮ mp ✮ H ✘ Lona n Ecompa ✮ E N E N E nn Op N E Op N me hod The LF on me he hod ba The o LF n eove mpac on on aba he m ng aComplex/Average me o MOPT hod n ehe ac p The ob on em LF aon m have mpac ng ahod on MOPT anged he ba ame p ob amp o em nmance ac on a❉ anged m ng ahey a-add MOPT p ob em have aconven H h ✘ L n E n ✮ mp ✮ E Eman N me hod The on he o n eC ac on ahod m ng MOPT p ob em aneed anged ann ❉ ❉ ❉ ❉ H hmpu ✘ L n ❉ H h ✘ L nmp ❉ ❉ mp ✮ Chod mp ❉ P N D und N D ua N N L nPe ❉ ❉ GMC P ND u con de ed m ng aCy nd ng op ma gh o de FDN ed m ng ame ng op ma we gh oaed FDN con de aEgene m ng aEEp nd con op de ed ma abeen we ng gh aba FDN ng op gh o Average mp ❉ mp C C ❉ ✮ mp ✮ C mp ✮ MC MC N u nn Average Average H con ✘ Cu mp mp ✮ C mp ✮ N nn ❉ ❉ H h ✘ L ❉ H h ✘ L n ❉ mp ✮ H ✘ L mp ✮ ❉ mp ✮ gunce phned en ha eechn en ed b g np p Tab cnP u eEComplex/better eLF 9been ha The p equa en ed n Tab mance eC b The compa gG p ove caPPmo u eD on ha pe p he eLF mance add en ed eque compa n ed Tab MOP eehave 9unon The o ove an pe eno ed mance MOP on o he add eanged ed M n d N E N ncon d N En N nn Op N Op Op ❉ M u me nM Qu Op m n D nd P nMH/Easy Pn b m M h d ue E n C Qu Op m n Dpe p nd Comp ex/Ave age Ye Comp Non nea ABC MH/Ave age Ave age MhDHS hnP d n C M h dG Qu u Op n m D Qu nd Op P nn m n b m D pp nd u P nn n P b m un The mod ed LF me and o wa ep9CMO ba ed The LF mod p acompa ed m me ov de hod be and pe o o wa ehe ba ng p aMO o10 m pob ov de be eC oan❉ mance annMm Average Average Average The LF hod The mod wa ed eD LF me ed hod LF p aand m o10 wa p ov ePM de ba be ed LF pe p aNo/Yes o mance m ov am de be eza pe ong mance aLF m A A C hm C nmpac C mpu n C P b m S u bLP m n Shave u P no m P App App n nadd O hm O h Average A hm n C b m SP u PVariant/AI App n n O hon High ✘ High ✘ Complex/better Complex/better ❉ ❉ Simple High ✘ ✮/Complex Simple ✮/Complex Complex/better ❉ ❉ ❉ ✮/Complex GA GA MH/Easy MH/Easy GA No/Yes Variant/AI Variant/AI No/Yes Varia High ✘ Complex/better ❉ Simple ✮/Complex ❉ MH/Easy No/Yes Variant/AI que echn ha que ha p been eD en p ed eC n en Tab ed epL-we n Tab eOp 10 echn ha been p emp en ed n✮ Tab enona echn ha p ehod en ed nhhmpu Tab eLmu 10 R Tadd m D nP N u H h ✘ mp ❉ H hComplex/Average ✘ mp mp ✮ C mp ❉ ❉ mp ✮ C mp ❉ N R N T m R T m D uN D u ab yEPe op oad and a REG unce gene aon n y aN o on oad n compa and REG on gene ao aanC unce ad on aFDN n non y LF o mode oad on and o A REG ad o hey gene ona LF on mode n compa A o hey ad ona LF mode A H h ✘ C mp C mp ❉ ❉ mp ✮ mp C mp ✮ C C ❉ mp mp ❉ ❉ mp ✮ mp ❉ ❉ Simple ✮/Complex ✘ NLP M NLP D N u NLP N D N N nn N n N HS MH/Easy Variant/AI C mp ❉ mp ✮ C mp ❉ Complex/Average ❉ Simple Complex/Average ✮/Complex ❉ ✘ Simple ✮/Complex M NLP N D u N nprov MH/Easy MH/Easy -ch Variant/AI qua ycdPdLuN oea uued on p om eed powe u g oba op m za on me qua hod and u have on ab p om ewh om add powe oba za on me and have ab yem oPmp add H ✘ n E n n ✮ n H h ✘ mp ✮ mp L na ✮ E mp ✮ M LP M LP M N L n L nnbe N Lonnnm y omod on qua eng powe yand o u u g oba p op m emp y powe o on u me on u g oba p om and op eona m powe za ab on u me yo g op oba add op and m on ab y hod o and eng have ab o✮ H h ✘ E n mp ✮ M LP N L n H h ✘ C mp ❉ ✮ mp ❉ ❉ ❉ Tab ea 10 onom man ehm ompa Tab on eng 10 ome amp Pe go o hm man app eza ompa ed n on oL ap ob go Tab ede hm ve emo p ann app Pe ng o ed man nH ob mu em eym ompa ob ebe ve on p ou ann ahave go hm p ob app em ed mu eySimple ve padd ann p ob QP QP N DC u N D QP N D N ❉ Tab eHS 10 Pe o man eo ompa on o amp go hm app ed n mu ob emp ve pN ann p ob em QP N D p ov de p gene ov ao gene oN aN va on ou va cena ou o cena p oMOPT ov unde de oad gene a✮ mode oad ao on mode o va ch we ou eng ch cena we ed eo ong n m unde conven ed aop n conven oad ona mode wh ch we emp meng ed Complex/better Complex/better Simple ✮/Complex Simple ✮/Complex Complex/better ❉ Simple PSO PSO MH/Easy MH/Easy Yes PSO Yes MH/Easy Variant/AI Variant/AI Yes p ov de gene aThe on o va ou cena o unde aEy oad mode wh we eC m ed nnhave conven ona Complex/better Simple ✮/Complex PSO MH/Easy Yes Variant/AI mp H ❉ ✘ C mp ❉ ✘ H h ✘ mp ✮ C C mp mp ✘ ❉ mp ✮ C mp ✘ C ned u C n Pde u C n P u N C n C C H h ✘ C mp ❉ mp ✮ C mp ✘ C n PG u N CFDN n The mod fH qua ed LF methods and software-based The mod f ed LF LF p methods atforms prov and software-based de better LF p atforms a m ng de better performance a con aGA ned ea p ann ng ob em Fu he mo ea10 dec on mak ng me hod anng o need om ware-based prov better performance a m ng con aTS ned me ann con ng aque p ned ob con em ea aEC Fu me ned he p ann ea mo ng epe me dec p p ob ann em on ng mak Fu ob ng em me Fu e✮ dec he a-o mo on need emak dec onn ng on mak hod ng ame o need hod o ann o need omp be H h ✘ Co mp E np H ✮ h ✘ C mp E mp ✮ n C ✮ mp ❉ mp ✮ C mp H ❉ h ✘ C Ehod n✮ OP N D OP N D u N OP N Op D u Op N Cunde E no ✮ ❉ OP N D nn Op L E n ✮ H h ✘ L mp ✮ E ✮ mp ✮ E E E N E N nn me hod The LF mpac on me he hod ba LF n eom mpac ac on on aba he m ba ng anhod me o hod n ehe ac p The ob em LF ang m have mpac ng ahod on MOPT anged he ba ame p ob ann em n eove ac on an✘ anged abe m ng amp aN amoMOPT p ob em have aconven anged nde Eha ✮ mp ✮ NDE p nn Op me hod The LF mpac on he o n eC ac on ahod m apwh MOPT p ob em ame anged anMOP ❉ ❉ ❉ L n ❉ L n ❉ ❉ ❉ P D und G P N D u MO h ✘ L nm ❉ ✮ C ❉ GE P NLF u atforms N MO con de a m a nd ng op ma we gh con o FDN ed a m ng a nd ng op ma we gh o con de a ng a nd con op de ed ma a we m ng gh a o FDN ng op ma we gh o FDN Average Average C mp ❉ H h ✘ mp mp ✮ ❉ C mp ✮ hperformance ✘ mp C mp ✮ C mp ❉ ✮ C ✮ MC NN Dbeen uM MC MC N nn N Op nneMO Op Average Average ❉ ❉ ❉ H h ✘ L n ❉ mp ✮ H h ✘ L n ❉ b g p c u e p e en ed b g n p Tab c u e e 9 ha The p ove e en a ed n Tab mance e 9 b The compa g p c ove u e on ha pe p he e mance en add ed e compa n ed Tab MOP e 9 on The o he add a pe e ed o mance compa on o he add ed n d E N n d nn Op N E N ❉ H h ✘ L n ❉ H h ✘ L n mp ❉ ✮ mp ✮ b g p c u e ha p e en ed n Tab e 9 The ove a pe o mance compa on o he add e ed MOP n d N E n d N E N N nn Op nn Op h d M h E d u n u n C C Qu Op Qu m Op n m D p nd D p nd P nn n P P b n m P b m u u MH/Ave age Ave age L nea / Ex e en Ye S mp e /Comp ex L nea M h d E u n C Qu Op m n D p nd P nn P b m u The mod ed LF me hod and o wa e ba ed The LF mod p a o ed m LF p me ov de hod be and e pe o o wa mance e ba a ed m LF ng p a o m p ov de be e pe o mance anpnhM m Average Average Average The mod ed LF me hod The and mod o wa ed e LF ba me ed hod LF p a and m o wa p ov e de ba be ed LF e pe p a o mance m p ov a m de ng be e pe o mance a m ng A hm C n C mpu P b m A S u hm n C P m n C mpu App n n P b m S O u h n P App n n O Average A hm C n C mpu P b m S u n P m App n n O h H h ✘ C mp ❉ mp ✮ C mp ❉ GA MH E N Y V n A H h ✘ C mp ❉ H h ✘ C mp mp ✮ C mp ❉ ❉ mp ✮ C mp ❉ GA MH E GA MH E N Y V N Y n A V n A echn que ha p e en echn ed que n Tab ha e 10 been p e en ed n Tab echn e 10 que ha been p e en ed n Tab e 10 N R TL m RTon Tnnm RL T D nP D nP N N D nP N R m D nP N a unce a n y o oad and REG gene a on a n unce compa a n y on o o oad ad and ona REG LF gene mode a on A n o compa hey on ad ona LF mode A o a unce a n y o oad and a unce REG gene a n y a o on oad n and compa REG on gene o a ad on ona n compa LF mode on o A ad o hey ona LF mode A o hey H h ✘ C mp C mp ❉ ❉ H mp h ✘ ✮ mp C mp ✮ C C ❉ mp mp ❉ ❉ mp ✮ C mp ❉ Complex/Average ❉ Simple ✮/Complex Complex/Average ✘ ❉ Simple ✮/Complex ✘ M NLP M NLP DComplex/Average u D u M NLP D u Nn N nn n✮ N nmode N N HS MH/Easy HS MH/Easy -omu Variant/AI -o Varia H hng ✘ C mp ❉ mp ✮ Ceo mp ❉ ❉ Complex/Average Simple ✮/Complex ❉ ✘ Simple ✮/Complex ✘ M NLP D u N N nve n HS MH/Easy HS MH/Easy -n - ob Variant/AI Variant/AI H h ✘ L E L n n E ✮ n ✮ mp ✮ mp L n ✮ E mp ✮ M LP M LP N M LP N L n L n L n E n ✮ mp ✮ M LP n H h ✘ C mp ❉ ❉ H h ✘ mp ✮ C C mp mp ❉ ❉ mp ✮ C mp ❉ Tab e 10 Pe o man e ompa on o a go hm Tab app e 10 ed Pe n mu o man ob e ompa ve p ann on ng p a go em Tab hm e 10 app Pe ed n man mu e ob ompa e p ann o a ng go p hm ob em app ed n mu ob e ve QP N D QP N D u QP N D u N N H h ✘ C mp ❉ mp ✮ C mp ❉ Tab e 10 Pe o man e ompa on o a go hm app ed n ob e ve p ann ng p ob em QP N D u N p ov de gene a on o va p ou ov de cena gene o a unde on o a va oad ou mode cena p o wh ov unde ch de we gene a e oad a m on ed mode o n va conven ou wh ch cena ona we e o m unde ed a n conven oad ona wh ch we e m ed n conven Complex/better Complex/better ❉ Simple ✮/Complex Simple ✮/Complex Complex/ ❉ PSO PSO MH/Easy MH/Easy PSO MH/Easy Yes Yes Variant/AI Variant/AI Y V Complex/better Simple ✮/Complex PSO MH/Easy Yes Variant/AI H h ✘ C mp ❉ C mp mp ✮ ❉ C mp ✘ H ✘ h ✘ C mp ❉ mp C n P u C n P u N C P C n N D u C N ✘ C n P C n con a ned ea me p ann ng p ob em Fu he mo e dec con on mak a ned ng me ea hod me a p ann o need ng p o ob be em Fu he mo e dec on mak ng me hod a o need con a ned ea me p ann con ng a p ned ob em ea Fu me he p con ann mo a e ned dec p ob ea em on mak me Fu p ng he ann me mo ng hod e p dec ob a em o on need mak Fu o he ng be mo me hod dec a on o need mak ng o be me hod a o need o be H h ✘ C mp E n ✮ H h ✘ mp C ✮ mp C mp E ❉ n ✮ mp ✮ C mp ❉ OP N D OP N D u N nn Op N nn H h ✘ C mp E n ✮ mp ✮ C mp ❉ OP N D u N nn Op L ELF n ✮ L noob mp E ✮ n ✮ E N E N Ea Op me hod me The hod LF The mpac LF on mpac ba on o ba n eeEm me ac o hod n on ecove ac aSnmod The on ng LF ang ap m MOPT ng ann on p ob em ba p have oSpe em aadd eeove have anged on aVed ananged annOp ann m ng aA am apm MOPT p ob em have anneMO anged LN Eha ✮ mp E NE me hod The LF on he ba omp nop eove ac on aN MOPT p ob have aac anged aMOP ❉ ❉ ❉ ❉ L nE H hhe ✘ Ln n ❉ H ❉ hm ✘ mp L n ✮ Cn mp ❉ ✮ C mp ❉ Da G Dmp G P N u N MO N con de ed amod m ng aneEN nd ng op we gh o FDN con de m ng nd con op de ed ma adha we de ng gh abeen ed and m FDN ng op ama ma nd we gh ma o we FDN gh ompac FDN Average Average Average H h Cempac ❉ mp ✮ C mp ✮ H ✘ C mp ❉ mp MC MC nn N Op u N Average H hMC ✘ C mp H h❉ ✘ C mp mp ❉ C mp ✮ mp ✮ C mp ✮ MCb N D u N Op nn ❉ ❉ ❉ H h ✘ H h ✘ n L n ❉ ❉ mp mp ✮ gdGM cPque uanty eona euM en ed b g n p Tab cPMH/Easy u eEMH 9been ha The p ove euthey en ang ed pe ned Tab mance eC 9Op b The compa g p u eD on pe o he mance en add ed ede compa ned ed Tab MOP eD 9C on The o a❉ pe enaLF ed o mance compa on oeC he add ed nng n d N N N nn Op Op ❉ H L nn ❉ mp ✮ b g p cand uLP eny p eLF en ed n Tab ehson 9✘ The adbe om mance compa on he ethey MOP N E N nn h dden E um C Qu M m h n E D u p nd n P C nn n P b Qu m Op m u Pona nn n P pe b- nm Aped MH/D fi u H gh ex/Ave age No/Ye mp Va an /A M hoad ED uNuREG npnme C Qu Op m n D pwa nd P nn P b m u The mod ed LF me hod and o ba ed The p aha ed m p ov hod be and emp pe o o wa mance ehe ba ahhed m ng p aMO oD pmp ovLF de be oob mance m A The mod ed LF hod The wa ed ba me ed hod LF p agene and m oLaoPComp wa p10 ov ePM de ba ed LF eam pe p aN o mance m p ov aMOPT m de ng be e✮ ong mance m ng A n C A hm C P m S❉ n u C n A b m App C u no n n P C mpu O h App P b m nSadd unn O Pon App n O A A A hm C nn C mpu P b m SLP u✮ n P m App n nmp h H h ✘ ✘ C C mp ❉ ❉ H mp ✘ C mp ✮ C C ❉ mp ❉ ❉ mp ✮ mp GA GA MH Ee MH GA MH E Y N Y V nO A V Ynd V H h ✘ mp ❉ mp ✮ mp ❉ GA N Y A echn been p eM echn ed que n Tab 10 p e✮ en Tab echn ech 10 que been eN en n Tab eA 10 echn que ha p eo en ed nhhmpu Tab eLng N R T R N R T D nP D nP N N D nP N N Rem Ton m D nP N a unce acena n o oad and REG a n compa ape n y on o✮ oeLF oad ad ona REG LF gene mode aob on A n o compa hey ad LF mode Aso oanpnhnM a unce aChm nand oha oad and a unce REG gene a✘ y aA oha on oad and compa REG on gene o aN ad on compa LF mode on ome A ad o hey ona LF mode o hey H h ✘ C mp C mp ❉ ❉ mp ✮ mp mp ✮ C C ❉ mp mp ❉ ❉ mp ✮ C mp ❉ Complex/Average Complex/Average ❉ ❉ Simple ✮/Complex Simple ✮/Complex ✘ ✘ ❉ M NLP M NLP D u D uva NLP D u N N N n nm N HS MH/Easy -o HS -N MH/Easy Variant/AI Variant/AI C mp ❉ mp ✮ C mp ❉ Complex/Average ✮/Complex ✘ NLP D u N nm n HS MH/Easy -p Variant/AI at generat uncerta on nty n compar of oad and to REG trad generat t❉m ona on LF mode n compar sem so son to trad tNOpgo ona mode A rat on compar son trad LF mode snaNNocon so L n H ✮ ✘ n nha mp ✮ ✮ H ✘ mp L n ✮ E ✮ mp ✮ LP M N M N L ncena Lnn nN H hH ✘ n E ✮ mp ✮ M LP N N L n H h ✘ C mp ❉ mp ✮ C mp ❉ mp ✮ C mp H ❉ ✘ C mp ❉ mp Tab empu 10 Pe ePe o 10 Pe eeoa ompa man ecena ompa on oma ang go on oL hm amp go app hm ed app n mu ed ob n mu eQP Tab ve ob ehhhnn p e✮ ann Pe ve ng p o p ann man em emp p ompa ob oTnnhComplex/Average aa hm app ed n e Simple ve N D QP N D N N D u N CC mp ❉ mp ✮ C mp ❉ Tab eTab 10 o man eC ompa on aon go hm app ed mu ob eou ve p ann ng p ob em QP D p ov de gene on o va ou o unde p ov anunce de oad gene mode aed on o va we e10 m ed o n unde conven aaA oad ona mode wh we emp m eds n❉ Cn mp ❉ mp ✮ C mp ❉ PHS O MH E Y V nFDN A p de gene aC on oEng va p ou ov de gene o aThe unde on onbED oad ou mode cena o wh unde we aEhm enac oad mode n conven wh ch ona we m ed nnhave conven ona C mp ❉ C mp mp ✮ C mp ❉ ❉ mp ✮ C ❉ Pov MH Ea P O MH E Y V Y nSimple A V n A H h ✘ C mp ❉ H ✘ C mp ✘ ❉ mp ✮ C mu mp ✘aconven C n P u C nona P N D uwh C N CL mp ✘h COP nat P nuncerta u nty n C E n C mp mp ✮ E C mp n ✮ ❉ mpch ❉ OP N D N D u nn Hof hto ✘O Eu nu ✮ mp ✮ mp ❉ ND N nn Op H h ✘ L nu EE nn ✮ H ✘ L E n mp ✮ ✮ H hp ✘ L n mp E ✮ n ✮ mpC ✮ EQP ND E N Een N E N En N Op Op Op me hod The LF mpac on me he hod ba oman LF n mpac ac on on an he m ba aLOP me MOPT o n hod ecove p The ob on em LF aEnp have m mpac ng ach aand on anged MOPT he ba aeC aob o em n eove ac ape anged auApp m ng abN anYaond MOPT p ob em have anged ❉ ❉ ❉ ❉ L mp ✮ C mp ❉ H hon ✘ LOp n ❉✮ mp G PCmp N u❉ G P MO N D u N H h ✘ L n ❉ H h❉ ✘ n mp ✮ ❉ C mp ❉ mp ✮ C mp ❉ G P N D G P N D u N MO MO con de ed m a nd ng op ma we gh o FDN con de ed a m ng a nd ng op ma we gh o con de ed a m ng a nd con ng op de ed ma a we m ng gh a o con FDN ng de op ed a m we ng gh a nd o ng FDN op ma we gh o FDN Average Average A H h ✘ C mp C mp ❉ ❉ mp ✮ Cp mp mp ✮ C ✮ mp ✮ MC MC N D uThe N D und N nn nn Op Average H hN ✘ C mp ❉ mp ✮ C mp ✮ MC N u❉ N nn ❉ H h ✘ L n ❉ H h ✘ mp ✮ L n ❉ mp ✮ b g p c b u g e p ha c u p e e ha en p ed e n en Tab ed e 9 Tab The e ove 9 b The g a p pe u e o a ha mance pe e compa mance en ed compa n on Tab o e he 9 on The add o he e add ed MOP pe e ed o mance MOP compa on o he add e ed M n d n d N N Op N nn ❉ ❉ H h ✘ L n mp ✮ b g p c u e ha p e ed n Tab e 9 The ove a pe o mance compa on o he add e ed MOP n d N E N nn Op M h d E u n M C h d Qu E u Op n m C n D p Qu nd M h d Op P m nn n E P n u b n D nd C u P nn Qu n P Op b m m n u D p P nn n P b m u SA MH/D fi u H gh Comp x/ Ex n No/Ye Comp x a wo d Tun ng M h d E u n C Qu Op m D p nd P nn n P b m The mod ed LF me hod and o wa e ba ed LF p a o m p ov de be e pe o mance a m ng A A A The mod ed LF me hod The and mod o wa ed e LF ba mod me ed hod LF ed p LF a and me m o hod wa p ov and e de ba be o ed wa LF e pe p e ba a o o ed mance m LF p p ov a a m de o ng m be p e ov pe de o be mance a m o mance ng m ng A hm C n C mpu P b m A S u hm n C P m n C mpu App n P A b m hm S O u C h n P n m C mpu P m n S u n O h P m App A A hm C n C mpu P b m S u n P m App n n O h GA GA MH E MH E GA MH N E Y N Y V n A V n A N V GA MH E N Y V n A echn que ha been p e en echn ed que n Tab ha e 10 been p e en ed n Tab echn e 10 que ha been p e en ed n Tab e 10 echn que ha been p e en ed n Tab e 10 H h ✘ H h ✘ C mp C mp ❉ ❉ H mp h ✘ ✮ mp C mp ✮ C C ❉ mp mp ❉ ❉ mp ✮ C mp ❉ N N R T m R T m N R D nP D nP N D u N D u D nP N D u H h ✘ C mp ❉ mp ✮ C mp ❉ N R T m D nP N D u a unce acena yMH oN and REG gene aunde a n compa aoad nn✮ y on oMOPT oad ad and ona REG LF gene mode aEn on n ohhon compa on ad ona LF mode A a unce aThe ny o on oad and a unce gene a✘ n yMH aD on on n and compa on gene o anN ad on ona n compa LF mode on oob A ad oH hey ona LF mode A o hey C mp ❉ mp ✮ Co mp ❉ Complex/Average Complex/Average ❉ Simple ✮/Complex Simple ✮/Complex C mp ✘ A ✘ ✘conven M NLP N D u M NLP NLP N ne✮ n N nL N HS MH/Easy MH/Easy MH E -✮ -N Variant/AI Variant/AI V on T Complex/Average ❉ Simple ✮/Complex ✘A M NLP N n❉ nnn HS MH/Easy -nH Variant/AI H hn ✘ L nnoad Eu nREG ✮ LC n E❉ mp ✮ mp H ✘ L ✮ M LP M LP uou N M L ncena N D u N H h ✘ L n E ✮ mp ✮ M LP u N L n H h ✘ h ✘ mp ✮ C mp ❉ Tab eaLF 10 man ePREG ompa Tab on enon 10 ouoad aPe go o man hm ecena app ompa ed n mu ome amp ob go Tab ende hm ve 10 p app ann Pe ng ed o man p mu eem ompa ob ve on p o✮ ann ahave go ng p hm ob em app ed n mu ob enn ✮ ve pemp ann ng p ob em QP N D QP N D ua N N oMOPT ❉ ND p ov de gene aLF o va o p ov aunce oad gene mode aed on wh o va ch we ou e✘ m ed o✮ unde conven aanged ona mode we m ed n❉ C✘ C mp ❉ ❉ mp mp mp C C ❉ mp ❉ ❉ mp ✮ C mp PHS O P O E PH O MH E YE V noad A V nn Ann V ph ov de gene oEEpome va p ou ov de gene o aN unde on oED aon va oad ou mode cena omp wh unde ch we aove enmp m mode nmpac conven wh ch ona we eC m ed n❉ conven CCon mp ❉ mp ✮ C mp P O E Y V AN mp ❉ C ❉ ✘ ✘ C n n u C C anged n Op H me ✘ C mp mp ✮ C✮ mp ✘ C QP n P u N n C mp E npe H ✮ h ✘ C E mp ✮ n C ✮ mp ❉ hLP ✘ C mp mp C mp nadd ✮ ❉ mpch C ❉ OP NPe u❉ OP N D OP N D H LEe E ✮ mp ✮ H ✘ LOp EEwh ✮ E ND nn Op N Emp N hod The mpac he ba n epm ac hod on aSemod The m ng LF aYp on p ob he em ba have o nnnmp aC e✮ anged aPhey aunnona ann m ng aYOp p ob em have ann H h ✘ Lo nA H h ✘ L E n ✮ mp ✮ E N Eed EMH N N nn Op nn Op hod mpac on me he hod ba The LF mpac ac on on aned he m ba ng amance o MOPT n eSPePM ac p ob on em aha m have ng MOPT anged ap ob am em aac anMOP ❉ ❉ ❉ ❉ H hp ✘ L non L n ❉ ❉ mp ✮ C ❉ mp ❉ G G N PPdconvent D N D u N MO MO H ho L ❉ mp ✮ mp ❉ G P u N MO A H h ✘ C mp mp ✮ C mp C ✮ mp ❉ mp ✮ C mp ✮ MC MC N D N uN nn Op A A H hE ✘ C mp ❉ mp ✮ Cha ✮ MC D ued N nn Op ❉ ❉ ❉ H hh ✘ LN ❉ H h ✘ L n mp ❉ ✮ H h mp L ✮ ❉ mp ✮ b g p c e ha e en ed b g n p Tab c u e e 9 ha The p ove e en a ed n o Tab e 9 b compa The g p c u e a on pe o he e mance add en ed e compa n ed Tab MOP e 9 on The o he ove a pe e ed o mance compa on o he add e ed M nnP dnuN N ngene N E N n d❉ N E N nn Op Op N nn ❉ H h ✘ L n ❉ mp ✮ n d N E nn Op M h d E u n C Qu Op M m h d n E D u nd n P C nn n P Qu b M h Op d m u E n u D p nd C nn Qu n P b Op m m n u D p nd P nn HBMO MH/Ave age Ave age Comp ex/Ave age Ye S mp e /Comp ex Lo a M h d E u n C Qu Op m n D p nd P nn n P b m The mod LF me hod and o wa e ba ed LF a o p ov The de be e ed pe LF o me mance hod a and ng o wa e ba ed LF p a o m p ov de be e pe o mance a m A A The mod LF me hod The and mod o wa ed e LF ba me ed hod LF The p a and mod o m o ed wa p ov LF de ba me be ed hod LF pe and p a o o o mance m wa p e ov a ba m de ed ng be LF e p pe a o o m mance p ov a de m be ng e pe o mance a m ng A hm A C hm C n C mpu n C mpu P b m u b m n u P n m A P m hm App C App n n n n C n mpu O h O h P b m S u n P m App prov de generat on of var ous scenar prov os de under generat a on oad of mode var ous s wh scenar ch were os under m ted a n oad convent mode ona s wh ch were m ted n convent A A hm C n C mpu P b m S u n P m App n O h s under a oad mode s wh ch were m ted n ona H h ✘ C mp ❉ mp ✮ C mp ❉ GA GA MH E MH E GA MH N E Y N Y V n A V n A N Y V GA MH E N Y V A echn que echn ha que been ha p been e en p ed e n en Tab ed e n 10 Tab echn e 10 que been p e en ed n Tab e 10 echn que ha been p e en n Tab e 10 H h ✘ C mp H ❉ h ✘ C mp mp ✮ ❉ C mp ❉ H h ✘ mp ✮ C C mp mp ❉ ❉ mp ✮ C mp ❉ N N R T m R T m N R D N D u D nP N D u D nP N D u H h ✘ C mp ❉ mp ✮ C mp ❉ N R T m D nP N D u a unce a n y o oad and REG gene a on n compa on o ad ona LF mode A o hey a unce a y o oad and a unce REG a n a y unce a o on oad a n n and compa y o REG oad on gene and o a REG ad on gene ona n compa a LF on mode n on compa o A ad o on hey ona o LF ad mode ona LF A mode o hey A o hey C mp ❉ mp ✮ C mp H ❉ h✘ ✘ C mp ❉ mp C mp A ❉ mp ✮ C mp ✘ M NLP M NLP N D u M NLP N D u N ne✮ n nN nA N MH E V❉ noad A CoN A ❉ C mp mp ✮ A C mp ❉ ✘ mp ✮ Cng mp M NLP N nA ned H EEon H MH E nOp A V nmu A H hed ✘ LLC na E n✮ ✮ H h ✘ L mp nn ✮ Eanged n ✮ mp ✮ M LP M LP N D uman N L N MOPT nT LCN nMH nme ✮ mp ✮ MQP LP LN no ❉ ❉ Tab eDaLF 10 Pe o man ePhhe ompa Tab on emp 10 onuCMH a Pe go man hm e app ompa n on mu a ob go Tab e hm ve e 10 p app ann Pe ng o ed p n ob mu e em ompa ob ve on p o ann go hm p ob app em n ob e ve p ann ng p ob em QP N D QP N Dm H me hOP ✘hod mp ❉ mp ✮ C mp ❉ Tab e 10 Pe o man e ompa on o go hm app ed n mu ob e ve p ann ng p ob em ND u N p ov de gene a on o va ou cena o unde p ov de oad gene mode a on wh o va ch we ou e cena m ed o unde conven a ona mode wh ch we e m ed n conven C mp C mp ❉ ❉ mp mp C mp ✮ Ca C ❉ mp mp ❉ mp ✮ C mp ❉ PH O P O MH E EE o Pac O MH Y E YV V n Va n Y V LM p ov de o va p ou ov de cena gene a unde on o a va oad ou mode cena o wh unde ch we a e oad m mode ed n conven wh ch ona we e m ed n conven ona Camp mp ❉ mp ✮ C mp ❉ P O MH E Y V nN A C mp ❉ H h ✘ mp ❉ H ✘ h ✘ mp C mp ✮ C mp ❉ ✘ mp ✮ C mp ✘ CMC ngene P u C n P u n P u N C n C n C n H h ✘ mp E n ✮ mp C mp ❉ H h ✘ C mp E n ✮ mp OP N D OP nn N Op D u N H h ✘ C mp E n H ✮ h ✘ C mp E ✮ n C ✮ mp ❉ u OP D u N N nn nn Op H h ✘ L n E n n E ✮ n ✮ mp ✮ mp ✮ E E N E N N nn Op nn Op hod The LF mpac on he ba o n e me ac hod on a The ng LF a mpac MOPT on p ob he em ba have o n a e ac on a a m ng a p ob em have a anged H h ✘ L n E n ✮ mp ✮ E N E N nn Op The mpac on me hod ba The o LF n e mpac ac on on a he m ba ng a o MOPT n e p ob on em a m have ng a MOPT anged a p ob a em have a anged a ❉ ❉ ❉ H h ✘ L n ❉ H mp h ✘ ✮ C mp L ❉ n ❉ mp ✮ C mp ❉ G N D G P N D N u MO N H ✘ L ❉ mp ✮ C mp ❉ G P N D u N MO A A A C mp H ✘ C mp ✮ ❉ mp mp ✮ C C mp mp ✮ ❉ ✮ e PVed NMH N D MC N nn nn Op nnnnh A HnTab h ✘ C mp ❉ mp ✮ C mp ✮ MC N N nn ❉ ❉ h ✘ L ❉ H hApp mp ✮ n✮ ❉ H huV ✘ mp LNmp n✮ buuoad g cREG u eMC ha p eEp en ed n The ove b gOp aGA p chm pe ump eom o❉ mance compa en ed ned Tab o enD he 9C The add ove emnoAH aA MOP pe oO mance compa oC he add dnha ND d E N n d Op N E N nn Op N ❉ L n ❉ mp ✮ bhg✘ paA cBFO unnP eque eD ed bHp n phnCH Tab cM uage eLP eEhm 9nTab ha The ove a✮ ed pe o❉ Tab mance eC✘ 9nyN The compa ove aque on pe ohe mance add eand compa ed MOP o he add eed ed MOP d M h den M hgene d uEN n EP u n CS C Qu Op Qu m Op n D np p nd M D h p d nd P nh EC P P nn ubP✮ n m n Pmp b m C u Qu uC Op m ❉❉ non Dmp nd MH/Ave Ave age Comp ex/Poo S mp eCe✮ ex Comp ex M dmpu E u nE C Qu m nha D p P nn b10 u A A A hm A C ben m n un C ne mpu Pon m A P bona m Sa u C nC ne nL nP m C mpu O App P bm nS n❉ hT P App np n A H h ✘ C mp ❉ H h mp C C mp ❉ ❉ mp ✮ ❉ GA Eo GA MH N Y MH E N Y V n AnLF V n A Yo H h ✘ C mp ❉ mp ✮ mp MH E N Y nOp echn ha been pHnp eg echn ed n que ha eon 10 been p❉ e✮ en ed Tab echn ech 10 ha been p ech en non Tab H h mp ❉ mp C ❉ h ✘ R T m Red m N N D nP N D N u N R T D nP N D unce aDa10 n yPe oEen and REG gene aeuuuoad on n compa a o unce ad ona n✘ y LF mode oad and A REG oe✘ hey gene a/Comp on n compa on ad ona LF mode A on M unce aThe oCEEp oad and a unce a✘ y anD on on n auuuuou and compa unce REG ahhhh9on gene n o o aN ad oad on n compa LF REG mode gene on o A on ad o naon hey ona compa mode on ad o hey LF mode A hey H mp H h ✘ mp C C mp mp ❉ ❉ ❉ em mp C A mp A ❉ ✮ mp mp C mp ✘ mp ✘ mp ✮ C mp ✘ NLP D M NLP N D uman N n n Nm NO H MH MH E MH E V n A V n A V Ha mp ❉ mp mp ❉ C A mp ✮ C mp ✘ M u N N nunde nC Hn MH E V A L na E n✮ ✮ L mp E ✮ mp ✮ M LP LA n mp LCN ny np ✮ mp MNLP LP Lmp no H ✘ C mp ❉ H ❉ h ✘ mp C mp ✮ C mp ❉ ❉ mp C mp ❉ Tab eon man eGA ompa Tab on eov 10 o aPe go o hm ecena app ompa ed n on mu aG ob go Tab eand hm ve eEem 10 p app ann Pe ng o ed p n ob mu enP em ompa ob eMO ve on p o✮ ann ahave go ng hm p ob app em nC mu ob e ✮ ve pohave ann ng ob QP N D QP N D QP N D u N N Tab e✘ 10 Pe oC man euC ompa on o aM go hm app ed n mu ob emp ve p ann p ob em de gene aN on o va o unde oad we m ed conven ona mp C mp ❉ ❉ mp ✮ C C mp mp ❉ ❉ mp ✮ C mp ❉ PH O P O MH E MH E PH O MH Y E Y V✮ n A V nang A V p de on va ou ov de cena gene p oN aLF unde de o gene aon va aoad ou on mode o cena va oou wh cena we a❉ o❉ eEmp oad unde m mode ed amp n conven oad wh mode ch ona we eC wh m ch ed we nn❉ eanged conven m ed ona nm conven ona Cat mp ❉ mp ✮ C mp ❉ P O MH E Y V A H ✘ C mp ❉ mp ✮ Co mp ✘ H hon ✘ mp ❉ mp CD P N C n P C n N D u Np h mp ❉ H h ✘ mp ✮ C ✘ ✘ nmpact umpac C N D N N C n C nona H ✘ mp C E mp n ✮ n ✮ mp ✮ mp mp ✮ C ❉ mp ❉ OP N DC N D N N nn Op H mp nC ✮ mp ✮ D N nn Op H h ✘ LC nD E n ✮ mp ✮ L n❉ mp ✮ E E E N E Op nn MM me hod The LF mpac he ba o n eove me ac hod on aSmp The m ng LF ap mpac on p ob em ba have ong n aCadd eove a✮ annnnN ann aYN ob em have a Leanged hd ✘ L E n ✮ ✮ E N N nn Op me hod LF on me hod ba The o ePMH mpac ac on on aed ba ng o MOPT n eC ac p on amp m have ng awh MOPT anged p ob a❉ em aac anged ❉ ❉ H hh LN n ❉ H h ✘ L n ✮ ❉ mp ❉ H hhe ✘ mp Lnn C n ❉ ❉ mp ✮ ❉ N D uu G P N D uman N N uems N MO H h ✘ L n ❉ mp ✮ mp ❉ G P D N MO methods The the bas methods she nteract The LF ons mpact a❉ m on the MOPT bas smode of nteract have arranged aApp m at MOPT anmp prob nteract onsNN DDa m ngLF at MOPT prob ems have as A A C mp ❉ H h ✘ C mp ✮ C CL mp mp ✮ ems ❉ mp MC MC u MC nn N Op D u Nmp Op A C mp ❉ ✮ nn ❉ ❉ ❉ H h ✘ H h ✘ L nN L n ❉ mp ✮ mp ✮ H h ✘ n ❉p gov p cque u eOP ha p euEp en ed n Tab e❉ 9Tab b g aH p cPque pe eD o ha mance eMOPT compa en ed ned on Tab o eA he 9C The eEmp aas MOP pe om mance compa on oC he add ed ng ne N E N n N d N E nn Op nn Op N ❉ H h ✘ L n ❉ mp ✮ bChgov pAPM cG unhgene ePhm ha e op en ed b nMC p✘ Tab cof u earranged 9dN ove euD en a✮ ed pe n om Tab mance eChng 9✘ The compa ove aob on pe o o he mance add compa ed MOP on o he eed n d N N Op dC EMH u n M C h d Qu Ep u Op n m C n pm Qu nd M h d Op P m nn n E Pn u b n m D p nd C u P nn Qu n P b Op m u D pmMOPT nd P nn n PNmp b Yarranged m ACO/ACS MH/Ea yOP H gh Comp ex/Ave age Ye S mp eC /Comp ex Va an A A Cp n C A mpu hm C m S n u C n mpu P P b hm m App C uprob nD n n n Pe C mpu O Padd bng n m nSm uV O hT P n A A hm Cha n C mpu P bA m SnP uu n P m App n nmp hMOP H h ✘ mp ❉ H h ✘ C mp mp ❉ C mp ❉ mp ✮ C mp H ❉ h ✘ mp ❉ GA E be GA Y N Y Vons n MH Eo V n A H h ✘ C mp ❉ mp ✮ Cp mp ❉ GA MH E N Ved An✮ echn ha been eEEbD en ed nhe Tab eNaThe 10 echn ha been p eGA en n Tab e✮ 10 que ha been en echn que n Tab ha eNEThe 10 been p eA en n eD 10 H h ✘ C mp ❉ ❉Nconven R m N D nP N D N D uman R T m D NLP nP H ✘ C mp H h ✘ mp C mp ✮ C mp ❉ mp ✮ Cnn mp ❉ C mp C A mp A ❉ ❉ ✮ mp C ✮ C mp ✘ A ✘ ❉ mp ✮ C mp ✘ NLP u M NLP N u✮ N N nA nRO h H H MH E MH MH E V nO A V n V H pechn ✘ mp mp C mp ❉h C mp A ❉ mp ✮ C mp ✘ M u N N nunde n H MH E n A L nu E ne ✮ L nn E n✮ mp ✮ L nmp mp E ✮ n ✮ mp ✮ M LP M LP N M LP L✘ nh Lnn mp H h ✘ C mp ❉ mp ✮ C mp ❉ H hn✘ ✘ C mp ❉ mp Tab eM 10 Tab Pe ePe oD 10 man Pe eEoa ompa man ompa on amp go on o hm aov go Tab app hm eem ed 10 app n Pe mu o ed ob nY mu eQP em ve ompa ob p eMO ann ve on ng p ann ahave ob go ng em hm p ob em ed nC mu ob ea veApp pe ann ng pn ob em N u❉ N N D u N hed ✘ C mp ❉ H h ✘ mp ✮ C mp ❉ Tab e✘ 10 oC man eG ompa on o go hm app ed n mu ob emp ve p ann ng p ob em QP Ngene D u QP D N N pde ov aDD on o va ou cena o unde a oad mode p wh ch de we gene e a m on ed o n va conven ou cena ona o unde a oad mode wh ch we m ed n C mp ❉ mp ✮ C C mp mp ❉ ❉ mp ✮ C mp ❉ PEQP Ode MH E me P O MH Y P O MH E Y V n A V nn A Y V ov gene a on o va p ou ov de cena gene o a unde on o va p ov oad ou de mode cena gene a o wh on ch o we va a e ou oad m cena mode ed n o conven unde wh ch ona a we oad e m mode ed n conven wh ch we ona e m ed n conven ona Ca mp ❉ mp ✮ C mp ❉ PP O MH E Y V nN Aa H ho ✘ C mp C mp ❉ ❉ mp ✮ C ✘ mp ✘ CH n C n P u N D u N C❉ napp CL H h ✘ C mp ❉ mp ✮ C mp C n P N u N C n H h ✘ C E n ✮ mp ✮ C mp C mp ❉ E n ✮ OP N D OP N D N u nn Op H h ✘ C mp E n ✮ mp ✮ C ❉ OP N D u N nn Op L n E n ✮ H h ✘ L E n mp ✮ ✮ H h L mp n ✮ E ✮ mp ✮ N E E N E E N E N nn Op nn Op N nn hod The LF mpac on he ba o n e ac on a m ng a MOPT p ob em have a anged a a H h ✘ L n E n ✮ mp ✮ E N E N nn Op me hod The LF mpac on me he hod ba The me o LF hod n e mpac ac The on on LF a he mpac m ba ng a on o MOPT he n e ba ac p ob on o n a e m have ac ng on a MOPT anged a m ng a p a ob a MOPT em p ob a em anged have a anged a ❉ ❉ ❉ ❉ H h L n ❉ H mp h ✘ ✮ C mp L n ❉ ❉ mp ✮ C L mp n ❉ ❉ G P N u P N D N u G P N D u N MO L n ❉ ❉ G P N u MO A A A H h ✘ H h ✘ C mp C mp ❉ ❉ mp ✮ C mp mp ✮ C ✮ mp H ✮ h ✘ C mp ❉ mp MC MC N D u N D u N MC N nn Op nn Op N A H h ✘ C mp ❉ mp ✮ C mp ✮ MC nn Op ❉ HM h ✘ L nb ❉ L nh ❉ mp ✮ H h ✘ mp LCmp nbm ✮ ❉ mp ✮ g p cque u e❉ ha pE10 ep en ed n Tab e❉ 9Tab ove b g aGA p cque eD o mance compa en ned on Tab enDmp he 9CPQu The ove emwo a❉ MOP pe oOA>PSO mance compa onmp he add en TV ed ne ne N E N n d N nn Op nn ❉ g✘LP pAMcM uNLP ha eDEp enbeechn ed bAH ghn p✘ Tab cque u ehm 9dha The ove euDmp en a✮ ed pe ned oC Tab mance eN 910 The compa ove adpe on o he mance add compa ed MOP on o he add eed h ddha EN u n C h d Qu E u Op nC m C n❉h pm Qu nd M Op P m nn E no PnEp u beN n m D p nd C uo P nn n P Op b m nuN u D pm P nn noC b m u BA MH/Ave age Ave age Comp ex/Ave age Ye Comp xnb amp dVed HS h d E u ne C Qu Op D p P nn A A hm Cp n C A mpu C P m S n umpu n P P b m Spe App C unha nC n np n Pend m C mpu O App Padd n nSm u n❉ App n O hM A hm CLha n C P bA m Sm uu n P App n O hMOP H h ✘ C mp ❉ H h ✘ mp C mp ❉ ❉ mp ✮ CP ng mp ❉N GA Eo MH E N Y V nnP AN Ynd V MH N Y V nechn A ha been p ed Tab eThe been eYed en nperformance Tab e✮ 10 que been en echn ed n Tab eN been p enHuuuEhm en nmpu eD 10 R m Rem m D nP N nP N H bechn hM ❉ mp mp N R T m DGA nP N D E u C ❉ H ✘ mp mp ✮ mp H ❉ hem ✘ mp C mp C ❉ mpenof C mp C mp A ❉ ❉ mp C C mp ✘ mp A ✘ mp mp ✘ N M NLP M NLP u N N ne✮ n nA n H MH MH ED H E V A V nhT C mp A ❉ mp ✮ mp ✘ H MH E n H ✘ L nu E napp ✮ mp H hon ✘ L nN ✮nn LP Nn M LP L nh N u N pn ob H n E ✮ H ✘ L n E mp ✮ mp ✮ NN D u M LP D Nm L nmp n H h ✘ C mp ❉ mp ✮ C ❉ mp ❉ Tab emp 10 Pe man eOP ompa Tab on eovera 10 o aMH Pe go man ehhhn ompa edhC n mu ome aTab ob go ehm hm ve 10 p ann app Pe ng ed o✮ man n ob mu ethe ompa ve on p o✮ ann ahave go ng p hm ob app ed mu ob ve p✮ ann QP QP N Do N D N H ✘ C mp ❉ ✮ C mp ❉ QP N u N C mp ❉ mp ✮ C mp ❉ ❉ mp PEhod OC MH Eoof P O Een Y P O Vnn n MH A E nLC Amp Y CD mp ❉ O MH E Y V nV AnN h ✘ C mp ✮ C mp C ✘ mp ❉ ✘a C n P C nTab P❉ N C nem C n H h ✘ C mp ❉ mp ✮ C mp ✘ C hod n PM Dson uE N C nob b g pperformance cture has presented n Tab b eM g 9he p The cture has presented performance compar eThe 9✮ha son The of overa addressed MOP compar son the M The overa compar the MOP mp n H ✮ h ✘ CA mp E mp ✮ nMH C ✮ mp ❉ H h ✘ mp CC ✮ mp C mp E ❉ nnL❉ ✮ mp ✮ C mp ❉ OP D uu OP N D OP N D u N nn Op N nn H ✘ C E ne mp ✮ C mp ❉ N D N nn Op hh ✘ L n n ✮ H h ✘ mp ✮ n E non ✮ mp ✮ EEMOP ✮ N E N E EN Op N E N Op me The LF mpac on he o n ac on aba ng aN MOPT hod p The ob em LF have mpac on anged he ba aPTab ann oe✮ n e✮ aadd m aed MOPT p ob have anged L n E nn ✮ mp ✮ E N me LF mpac me hod ba The LF n eMH ac on me on hod aed he m ng The aThe LF omp MOPT n mpac eC p ob on he aE✮ m ba have ng ap o MOPT anged n eoen ac aon p ob aon em amp m ng amp aac MOPT anged p ann ob ang em an❉✮ anged aCPaddressed a❉ ❉ ❉ ❉ h ✘ L non L n ❉ mp ✮ C mp ❉ H ❉ huMO ✘ Ln n mp Gon G N P N D uha N P N D u MO N H ✘ L n ❉ mp ✮ C mp ❉ N D N MO A Ao A H h ✘ mp ❉ H h ✘ C mp mp ✮ ❉ C mp ✮ H h ✘ mp ✮ C C mp ✮ mp ✮ mp ✮ MC N D un MC N MC N uL nn Op Op N nn A ❉ ❉ L L n ❉ mp mp L/Comp n ❉ mp ✮ b g p cPPque u e❉ ha pN eCuuuEEp en ed n Tab e❉ 9m ove aac pe o mance compa o he em✮ MOP n The N Ep ne N E d N E ❉ H h ✘ L n ❉ mp ✮ p cM uNLP eddCha ha pN eD en ed b gGhned p Tab cPCdaddressed u 9dba b ha g The p cEC ove euuu en eEeen aEC ed pe n p Tab eA mance en ehN 9✘ The compa nmp Tab ove ehnbSn❉ 9aque on pe o ove o he mance add a✮ pe eG compa ed mance MOP compa o he add on o eed ed he enC nH d N nn Op M h EMH u M h d Qu u Op nC m C n pmp Qu nd M d Op P m nn E n Pn u b n m D p nd C u Qu n P Op b m n u D pmhave nd P nnmp nem b nm ✘ EA MH/Ave age Ave age Comp age S mp eC ex n ege M h E u n C Qu Op m D pm nd P nn n b uV A hm A C hm C nmpac mpu n C P b m S Pex/Ave u hm m n✮ Sem C un P nD m n P m C mpu App App nP nu Padd bm n nSm O ON h PA App n O A A hm CD n C mpu P bA m u n P App nmp On hMOP ❉ GA MH GA MH N Y N Y n Amp nmp A hu H bechn hg✘ mp mp mp ❉ GA MH E N Ye V n A ha been p enED ed n❉ Tab eLed echn ha been p eNeaNLP n que euechn en echn n Tab ha eN 10 p eH en nmpu eD 10 H ✘ mp H he ✘ mp C mp ✮ C mp ❉ ❉ mpe ✮ C ❉ N R T m R m Rem T DH N D nP D u nP N D u✮ H ✘ mp ❉ mp ✮ C mp ❉ H hV ✘ C mp A ❉ C mp A ❉ ✘ mp mp A ✘ mp C mp u❉ M N D N N n✮ nC N EA H MH MH E V n Ang V n V H ✘ mp ❉ H h ✘ C ❉ C mp ❉ ❉ C mp A ❉ mp ✮ C mp ✘ M NLP DDbeen M NLP N Dbeen N n ned N n❉ ned H MH EN H h ✘ H ✘ n Eu nnmp E ✮ n✮ mp ✮ mp ✮ LP M LP N N D uE N L nh H L n Eapp n M LP N u N L C mp ❉ mp ❉ ❉V Tab ePC 10 Pe oD man eE✮ ompa on oon a10 go hm Tab e❉ ed 10 n Pe mu o man ob ve ompa p ann on p anC10 ob go em hm app n mu ve p✮ ann p ob QP QP H h ✘que C mp ❉ mp ✮ C mp ❉ Tab e PP10 Pe ouD eM ompa Tab e✘ 10 oDC Pe aN go oC man hm eapp ompa ed nhhhhTab mu onD aD ob go eLLC hm ve p ann ng ed p✮ nN ob mu em ob emp ve p ann ng p ob em QP uman N ❉ P O MH PH Oapp MH E Y n Amp Y Vm CNN mp ❉ mp ✮ C mp ❉ P O MH E Y V n A C mp ❉ H ✘ C mp ❉ ✘ H h ✘ mp ✮ C C mp mp ✘ ❉ mp ✮ C ng mp ✘ C nnP non P N D u C n PE N D u N C CLA✮ nT N CM n H h ✘ C mp mp ✮ Co✮ mp ✘ CMC n u N COp nnV H hh C n ✮ H mp h ✘ C mp C mp ❉ E n ✮ mp C Cmp Eob ❉ n n❉ ✮ mp OP N OP N N OP nn N Op D u N nn Op N C mp E n ✮ ❉ OP nn H h ✘ L E n ✮ n ✮ mp ✮ mp ✮ H h ✘ L n E ✮ E E N E N E N E N E nn nn Op H h ✘ L n E n ✮ mp ✮ E E N Op ❉ ❉ ❉ ❉ H h ✘ L n ❉ H ✘ L n ❉ ❉ H h ✘ L n ❉ ❉ ❉ G N D u G P N D u N G P N MO MO N A A A C mp ❉ C mp mp ✮ ❉ C mp ✮ mp ✮ C C mp mp ✮ ❉ mp ✮ C mp ✮ MC MC MC N D u nn A C mp ❉ mp ✮ C mp ✮ N D u N nn Op ❉ ❉ H h ✘ L n ❉ H h ✘ L n ❉ mp ✮ H h ✘ mp L n ✮ ❉ mp ✮ b g p c u e ha p e en ed n Tab e 9 The ove a pe o mance b compa g p c u e on ha o p he e en add ed e n ed Tab MOP e 9 The ove a pe o mance compa on o he add e ed M n d N E n d N E N n d N E N nn Op Op N nn ❉ H h ✘ L n ❉ mp ✮ bechn gMnP pALPcque unPhm e ha haDDC pNMH eD en ed bTab n cM u eEhm 9ha ha The ove euDmp en am buA ed gnpe p oC Tab u mance eQu ha 9✘ The compa p eCM ove en ed adOp on pe n om❉uCTab he mance e10 add 9nnd compa ed ove aO ✮ on pe o mance he add emex ed on oaD hendAadd ed nTab Ep hgh dp M he✘ d uN nN E uAve nC C Op Qu m h n Eo n u p npThe D p nd P CMOP nn nnn P P nn bo Qu n m PbC Op b m unn MOP nuN u pA P nnmp nMOP P ❉ b nm V u DE MH/Ave age age age S mp e✮ Lo M h d E u ne Cn Qu Op m nmp D P n P m A A n C A C P bD S u n mpu PComp m A Pex/Ave bn hm m S App nC n Pend m C mpu h App Pbn/Comp n mH ncompa S O hnY P m App n O H ✘ C mp ❉ H he ✘ mp C mp ✮ C mp ❉ ❉ mpe e✮ C GA E GA MH E GA MH Eng N Y V n A N n nmpA que ha been p ed nced Tab eN 10 p eechn en echn ed n Tab echn eN que p ha eH en been ed n p eemu eN ed 10 n Tab e✮ H h C ❉ mp ✮ mp ❉ hu ✘ ❉ N R T m N D uM D nP N D u H ✘ C mp ❉ H h ✘ C mp mp ✮ ❉ C ❉ mp ✮ C mp ❉ N R Tn m R m D uD D nP N Dbeen H ✘ mp C mp ❉ ❉ C mp mp C ❉ mp ❉ C mp ❉ mp A ❉ ✘ ✘ mp ❉ M NLP NLP N u N N nY n H En H MH Een H V n MH Amp V n A H hN C mp ❉ mp mp A ✘ M NLP N u N N nA H E V nV A ques hasAAbeen presented techn edd❉MC 10 ques has been presented Tab e❉ 10 P Otechn MH h ✘ LC n EY n mp ✮ LL n ✮ mp ✮ LP M LP N L n L nh H h ✘que LMH n E ne ✮ mp uman N L H h C mp ❉ H ✘ mp mp C mp ❉ H h ✘ ✮ Ca✮ C mp ❉ ❉ mp C ng mp ❉ Tab empu 10 Pe oD on oon aen go hm Tab app e10 ed 10 nPe mu o ob eob ve ompa p ann on ng p go em app ed nC mu ve ann QP N QP N D u N QP N D u N N H h ✘ C mp ❉ mp ✮ C mp ❉ Tab Pebeen oMH eM ompa Tab on e✘ 10 oD Pe a10 go o man hm app nhhhhhTab oDC amp ob go eLC hm ve p✮ app ann ed p nN ob mu em emp ve p ann ng p ob em QP D N C mp ❉ mp Cp mp ❉ P O MH E P O MH V n Amp A CNN mp mp ✮ C mp ❉ E Y V n A H h ✘ C mp ✘ ❉ mp ✮ C Cmp mp mp ✘ n✮ ❉ CG n eP C nompa P C nY P C n u Nmp n✮ ✘ C n COp nT h ✘ H ✘ mp C E mp nPd E n✮ ✮ mp ✮ Cmp mp mp ✮ C ❉ mp H ❉ hnn❉ ✘ CC Eob ✮ OP OP N DN uman N D u OP N D nn nn N p ob H h❉ ✘ n ✮ mp ✮ CEo mp ❉ OP N DQu u N nn Op H h ✘ L nu EH n❉ ✮ nE E nE✮ mp ✮ ✮ mp n/Comp ✮ EEhm ✮ mp ✮ E ND Eu uu❉ E N Eue N E N E nn Op Op nn ❉ ❉ nE H ✘ L n ❉ ❉ H hnn ✘ Lob n ❉ ❉ ❉Vem nmp P10 N G P D uompa G N D uman N MO MO N M H h ✘ L nu ❉ ❉ G PP u N MO A A H h ✘ CnL mp ❉ C mp mp ✮ ❉ C mp ✮ mp C C mp ✮ ❉ mp MC N MC Op nn C ❉ mp ✮ C mp ✮ MC nn ❉ ❉ H h ✘ H hC ✘ LC L n ❉ ❉ ✘ mp ✮ L n ✮ ❉ mp ✮ ne d✘ n EMH N n N E N N Op Op N nn ❉ H h ✘ n ❉ mp ✮ nA E N Op M h C d EA M C h ddN E nu m C n✘ D pmp Qu nd M h dOp P m nn n En P n u b D pN nd u P nn n P b Op m u DAmp pGA nd P nn✮ n CP mp SOA MH/D fiD uLP H gh Comp ex/Ave age Ye SnN mp eeL✮ ex >PSO A A hm C C mpu P b m SM u hm n✮ Cmp P m n CH App nEQu nmp Pn✮ m Sm O nm PTm nb Ynm ✮ A hm Cmp C A mpu hm CDC PN bL m SnEOp u C nmp mpu PL bA m S App u n n P O hmpu App nbCmp hnnV H ✘nP mp ❉ mp ✮ C mp ❉ H hunn ✘ C ❉ mp GA MH Y Vh MH Amp H hed mp ❉ H hed ✘ C mp mp ❉ C ❉ mp ✮ C mp ❉ GA MH EN GA Y N Y Ve nC A Vh echn ha been p ed n eEN 10 echn que been p eGA ed Tab echn que ha pnEE AeunM en echn que n Tab ha eN been p echn en que n eLnN been p eLP en ed Tab enm 10 H ✘ mp ❉ ✮ C D nP nP N D u N Rnn TO Red H hha ✘ C mp ❉ mp ❉ D D m H h ✘ C mp ❉ mp ✮ C mp C ❉ mp ❉ mp ✮ mp ❉ C mp A ❉ mp A ✘ ❉ ✘Vem nmp NLP u NLP N N H MH H MH E n VO H hen ✘ C mp ❉ mp ✮ Cha ❉ mp C M NLP NPe Dbeen uman N nve np H MH E V n A nN Ee ned ✮ H ✘ n E n mp ✮ ✮ H mp n ✮ Ehm ✮ mpC N ✮ M LP N D uompa M N D uman N L✘ nn LAY nnT Nm ob Lnnh H h ✘ n E n ✮ mp ✮ LP u N nn hh C mp H mp ✘ C mp C mp ❉ ❉ H h ✘ mp ✮ Cmu mp ❉e ✮ ❉ Tab eC 10 Pe oD eE✮ on oon am hm Tab app eh❉ ed 10 n Pe mu o ob een ompa ann on oC p a10 ob go em nC veApp ann ng D QP N N uO QP N D u Nmp N p ob C mp mp ✮ C mp ❉ Tab e que 10 oMH eC❉ Tab e✘ oDTab Pe a10 oC man hm eapp ompa nPhhhNTab mu oDgo a10 ob go eLC hm ve p app ann ng ed pnN ob mu ob ve p ann ng p ob em QP C mp ❉ C mp mp ✮ ❉ C mp ❉ mp C mp ✮ C mp ❉ ❉ mp Cp mp ❉ PQP O P ON MH P Y MH E✮ Vh n Ang V n Amp A H h ✘ H ✘ C mp ❉ mp ✮ Cmp mp ✘ ✘ nompa non N10 P uman N D u N C nY N Pe C❉ napp CR H h❉ ✘ ✘ CMP nH PM N Dgo u Nem CLn H

MH E

HPG❉ hEn✘ ✘ C nA H h ✘ ✘C mp CLCG mp mp ✮ n❉ C✮ mp ❉❉ mp C✮ ✮ C mp E nn ❉nnn nOp ✮Op mp ✮ ✮ C mp mp ❉ NN DEE u OP N DnE u N OP N D uN NH nn MO Lmp nuD EH nh ✮ n E mp ✮ ✮ H hnn ✘ L mp ✮ Ewo ✮ mp En Pd E uN ND N N Eu Op Op N nn Mu HC h ✘✮❉ LCL n EL nEmp ✮Pnn✮ mp ✮ EhH NDu Eu N Op ❉ ❉ ❉ H ✘ L ❉ H h ✘ LMC n ❉ C mp ❉ L n ❉ ❉ ❉ GOP uM N uEC NOp P N D u N MO nb ❉ ❉ G MO And h ✘ mp ❉ mp h ✘ ✮ Cmp mp C ✮ mp ✮ ❉ mp MC N DN N nn N nn A A H ✘ mp ❉ mp ✮ C mp ✮ MC hh ❉ H h n mp ❉ ✮ LnC nbC ✮ ❉ mp ✮ dC N n d N N nn Op Op hnP d EP uE Qu M mE h d nnage En D p nd npmn P nnn P bP Qu m Op u DA pm nd P nnmp n C P ✮ b nm ✮ GSA MH/Ave age Ave age Comp ex/Ave Ye Comp xnbmp amp d❉ GA MD hnP d nn A M C dMC Qu EC Op nE m n D pm Qu nd Op P m nn b D u P nmp m A hm CGA nL C mpu P✘ b m A SM u hm nn P m CH App n nmp P mASV O u hnn n P n O A A hm Cmp A mpu hm P bn m u C nmp mpu PN m SnP App u nCC nEm P O hmpu App n hnuV H ✘ C mp ❉ mp ✮ C mp ❉ GA GA MH MH N Y Ne nO A n H✘ h✘ C mp ❉ ✮ mp MH E N YC V>PSO A H h ✘ C mp ❉ mp ✮ Cmp mp ❉ DC N D u D N Rm T RC H hQP ✘ mp ❉ mp ✮ mp ❉ NPe DEN uman N R T m C mp ❉ H C mp mp ❉ C mp ❉ mp ✮ C C mp mp ❉ ❉ mp C ❉ C mp A ❉ mp A ❉ ✘ C ✘ M NLP D NLP D u NLP N D u N N nC nn N N H E Hmp MH E✮ V nC V nmp Anh H h ✘ C mp ✮ C mp A ❉ mp ✮ C M NLP u N N nm n V n A H hh L nN E n ✮ H h ✘ mp ✮ LY n EY ✮ mp nAmp ✮ ✮App M M LP N uC M LP N Ln✘ nnp Nmp L n✮ LL n E ✮ mp ✮ M LP n h ✘ mp ❉ ✮ C mp ❉ H ❉ hnMO ✘ ❉ Tab eM 10 Pe oD man on aLoD hm app ed mu ob ve p ann p em QP QP N DCN u N D N QP N D u N Hompa h ✘ Camp mp ❉ mp ✮ CECe mp ❉ eCP10 oMH ompa Tab on e✘ 10 oD Pe aN Tab oCMH man hm eSneE 10 Pe oned man nPhhhhhoon mu eC ompa ob eLC hm on ve o p app ann a❉ go ng ed hm p ob mu app em ob ed emp mu ve ann ob ng p ve ob p❉ em ng p ob em N Dgo u N ❉ PE O E uu eG Y PV❉ O Vh n MH Ang Y mp ❉ mp ✮ C mp ❉ P OTab MH P O MH EN Y nn A Vnn H ✘ mp ❉ H ✘ mp mp ✮ C mp ✘ H h ✘ ✮ C C mp ✘ ❉ mp ✮ C mp mp ✘ C nLP N D CMC P u N C n PE N D u N C CAnMO nL N E mp n ✮ n C mp Eeapp ✮ mp nn❉ ✮ mp ❉ nL ✮nN ❉ OP OP N D OP nn H h ✘ Cgo E n ✮ mp ✮ C mp ❉ OP N D N H ✘ L nu EH n❉ ✮ H ✘ mp ✮ ✮ L mp n/Comp ✮ E Eann ✮ mp ✮ N Ngo N En N nn Op Op nn M n E nE ✮ mp ✮ Eh E Op ❉ H h ✘ nb Ln n ❉ ❉ H mp h ✘ ✮ Cmp mp Lob ❉ n ❉ ❉ mp ✮ mp P N D uompa P N D N u N H h ✘ L nb ❉ mp ✮ C mp ❉ G P N u N MO A A And hhEn ✘ C mp ❉ H ✘ C mp mp ❉ C mp ✮ C C mp mp ✮ ❉ ✮ MC N D uM N MC N nn Op nn Op ❉ ❉ H h ✘ L n ❉ h ✘ mp ✮ n ❉ mp ✮ n❉ N nSnP d❉ N E N Op N H h ✘ LD nu ❉ H h ✘ n ❉ mp mp ✮ n d N Eu nH d N E N N nn Op Op h d EP uE nN C Qu Op M m h d n E✮ D u p nd npmn P C nnn P bP Qu m Op m u D pmYnd mp P nn nn C b nm ❉ u SPEA/2 MH/D fiD uLP H gh Comp x/ Ex SnN mp e✮ ex Va an M GA h d nEn C dG Qu EC u nE C n D Qu nd Op P m nn n❉ PnnE✮ b m D u n bL uV A A A hm C nOp C mpu Ppm m A SG u hm nn C P m CH App n nu P mH SVnn O uN hnn n P App n O hT A❉ A hm C EMH C A mpu hm C P bC m SnEm u C n mpu P PL m App u nCn n n P O hmpu App n nbm hnT H h ✘ C mp mp ✮ C mp ❉ GA MH GA MH N E Y nO A N V HM hdNLP ✘ mp ❉ mp ✮ mp ❉ E N Y V n A H hed ✘ C mp ❉ H hP ✘ mp ✮ C C mp mp ❉ ❉ mp ✮ CP ng mp ❉ D nP N nP N D u D N D uman N R T m R T m N Rnn H h ✘ C mp mp ✮ C mp ❉ D nP N D u N R m H h ✘ C mp ❉ H mp h ✘ ✮ C mp C mp ❉ ❉ mp ✮ C C mp mp ❉ ❉ mp mp A ❉ C mp A mp ✮ C ❉ mp ✘ C mp A ❉ ✘ ✮ C mp ✘ M NLP D u M NLP N D N u M NLP N n n D N N n H MH E H MH H MH V A n A V n An M n n H h ✘ H h ✘ L E L n n E ✮ n ✮ mp ✮ mp ✮ h ✘ L n E n ✮ M LP M N D u u M LP N N L n L n N H h ✘ L n E ✮ mp ✮ M LP N D u N L n ❉ ❉ ❉ Tab e 10 Pe o man e ompa on o a go hm app n mu ob Tab e ve e 10 p ann Pe ng o p ob e em ompa on o a go hm app ed n mu ob e ve p ann p ob em QP N D QP N D QP Tab e 10 Pe o man e ompa Tab on e 10 o Pe a go o man hm e app ompa Tab ed e n 10 on mu o Pe a ob o go e man hm ve e p ompa app ann ng ed p on n ob mu o em a go ob e hm ve app p ann ed ng n p mu ob em ob e ve p ann ng p ob em C mp ❉ mp ✮ C mp ❉ P O P O MH E MH E Y Y V n A V n A C mp ❉ mp ✮ C mp ❉ P O MH E Y V n A H h ✘ C mp ❉ H h ✘ C mp mp ✮ ❉ C mp ✘ H h ✘ mp ✮ C C mp mp ✘ ❉ mp ✮ C mp ✘ C n P u C n P u N C n P N D u N C n C n N C H h ✘ C mp ❉ mp ✮ C mp ✘ C n P N D u N C n C nH mp nN ❉❉ C✮ mp ❉nnn nOp ❉ OP P N D N ND D OP Nmp DnE H OP nn Op nn MO Emp n ✮n✮ ❉ OP nn H h ✘ LC nCmp EnCM LG n nEx E ✮ ✮ H h ✘ mp ✮ L mp ✮ E EA nn ✮✮ mp ✮ EM N E N N Em N nn Op Op N nn HC h ✘✮❉ LCmp nOp E nE ✮C mp ✮ Eh N Eu N Op ❉ ❉ H hEC ✘ L ❉ hQu ✘ LMC n mp ❉ C mp ❉ mp Ln/Comp C nm mp ❉ ❉ ❉ G uM G PN N D uEC N P D✮ Nnn MO Mu A ❉ mp ✮ C mp C ✮ mp ❉ mp ✮ C mp ✮ MC N D u N u A A C mp ❉ mp ✮ ❉ mp ✮ mp ✮ C mp ✮ MC u MC N D u ❉ h ✘ L n ❉ H h ✘ mp ✮ L n ❉ mp ✮ n d n d E N N nn ❉ H h ✘ L n ❉ H h ✘ L ❉ mp ✮ mp ✮ n hm d N E n d N E N N Op Op h d E u n m h d n E D u p nd n P C nn n P b Qu m Op m u n D p nd P nn n P b m NSGA/ MH/D fi u H gh Comp x/ n S mp e ex Va an h d E u n C d Qu E Op n m n D p Qu nd Op P m nn n P n b D p nd u P nn n P b u A A A hm n C mpu P b m S u n P m App n n O h A AMH C n C A mpu hm C A P b hm m S n C u C n mpu n P m C P mpu b m S App u P n b m n S n P u m n O h P App m n n App n O n h O h H h ✘ ✘ CM mp ❉ H hA ✘ mp ✮ CnC C mp mp ❉ ❉ mp ✮ C mp mp ❉ GA MH EE uuM NLP GA MH EE N Yu LC GA MH E N YH V nn A Vnm N Amp Y A❉ VLn H h ✘ C mp ❉ mp ✮ C mp ❉ MH NuC YM V n A H hh ✘ C mp ❉ H mp h ✘ ✮ C mp mp ❉ ❉ mp Cmp mp D nP D nP N DC N D nP R T N m Nmp R ✮ T m DGA nP N R✘ T H h ✘ H h ✘ C mp mp ❉ ❉ mp ✮ C mp mp ✮ C ❉ mp HN ❉ hnn ❉ mp CE mp ADh ❉h mp ✮ C✮Y mp ✘ CnC ❉ M NLP N D u N D u N NLP N D u n✮ N n✮ n N H MH H V MH A E H h ✘ C mp ❉ mp ✮ C mp ❉ C mp A ❉ mp mp A ✮ C ❉ mp ✘ ✘ M NLP D u N N n MH EN H MH E V n V n A H h ✘ L n n ✮ n E n ✮ ✮ L mp ✮ E mp ✮ M LP M LP LP N L n L n H h ❉ h ✘ mp ✮ ❉ mp C ❉ QP N D QP N D QP N D N N ❉ QP N D C mp ❉ P O MH E P O MH E V n A Y V C mp ❉ mp ✮ C mp ❉ P O MH E Y V n A H ✘ C mp ❉ C mp mp ✮ ❉ C mp ✘ C mp ✘ ❉ ✘ C n P u C n P u N C n P u C n C n C n H h ✘ C mp ❉ mp ✮ C mp ✘ C n P u N C n H ✘ H C ✘ mp C E mp n E ✮ n H mp h ✘ ✮ C mp mp C ✮ mp C ❉ mp E ❉ n ✮ mp ✮ C mp ❉ OP OP N D u N D OP N D N N nn Op nn Op N nn Cmp mp EEmp n ✮nmp ❉ OP N Dmp nn Op H✘ hEu ✘ LdOp nunE m EH n ✮ n E ✮ ✮ L mp ✮ E n nn ✮ mpC mp ✮ E NunE M N NOp N Em nn ❉ ❉ ❉ H h ✘ L nC ❉ H h ✘ Ln ❉ n ❉ ❉ GMC P N DM Pnn N Dn N uN MO Nmn nu H hdh ✘ LD nu ❉ H hhm ✘ L nbALG ❉ ❉ ❉ P G P N N N MO MO A A C ❉ mp ✮ C mp C ✮ mp ❉ mp ✮ MC ❉ mp ✮ ❉ mp mp ✮ C mp ✮ MC nn nn ❉ ❉ H h ✘ L n ❉ H h ✘ mp ✮ L n ❉ mp ✮ nnP N nSEx d N E N Op N nn ❉ H h L H h ✘ n ❉ mp ✮ mp ✮ nA hm n d N E N Op Op h d EMH uE m n D p nd P nn nnn P b m u SFL fiM gh Comp x/ SOp mp /Comp ex Va an h ddGA ENDCEuN M C dmpu Qu EC E❉ n u❉ nA P DC pm Qu nd Op P m Qu n Pn✮ Op b m D p nd nbN u D P p nd n Pe✮ b✮ P mmp nn Phun u A A A hm nC C P bSnE E m S u n Pmp hm App C n✮ n nmp C O P mH SV n P n OM h A AMGMC CN n MH/D A mpu hm C P bun m u AC C nC mpu PEL m n App u C❉ mpu nCn n P m PY O m hmpu SV App nmp n nbC m O hm App O✮ h C H hh ✘ C mp H mp h ✘ C mp ❉ ✘ mp Cn mp ❉ ❉ mp MH GA MH N Y GA n MH A E Y VApp n ANmp GA MH E N V nV An✮ H h ✘ C ❉ mp ✮ C mp ❉ D nP N D u N N D nP Dmp u RC T m R mA NYn ✘ H h❉ ✘ C mp ❉ mp ✮ C mp ❉ D nP N Dhnu u N R T m H ✘ C mp ❉ H hQu ✘C C mp mp ✮ ❉ C mp ❉ C C mp ❉ ❉ mp ✮ ❉ C mp C ALC mp A ❉ ✘ ✘ M NLP D M NLP D u Nm M NLP N D N n✮ n n Nnn Hn H MH nb A V nC Amp C mp A ❉ mp ✘ H MH V A L n nh ✮ n E n✮ mp ✮ ✮ H h ✘ mp nnPN ✮ n ✮ mp ✮ M LP M LP uC Lu✘ nh❉ LA nnT N nn H h ✘ LLC n E ✮ mp ✮ M LP u N L n H ✘ C mp ❉ H h ✘ mp ✮ C mp ❉ mp ✮ C mp ❉ ❉ N DE QP N D N QP N D N ❉ N D C mp ❉ P QP O MH E uu P O MH EE Y P O❉ E Y V nN A n Y V P O MH V nN H h ✘ mp mp ❉ H mp h C mp C ✘ mp ✘ ❉ mp C C CD nLP N PND DN uE C nn P❉ N u nnn C N CLnM H ✘ mp ❉ ✮ C mp ✘ C nH P u N C n hh ✘ C mp H ✮❉ h ✘ mp E mp nMH C ✮ mp ❉ CL mp C mp E ❉ n ✮ ❉ OP N D OP N N OP N D N nn nn Op L nCompu En ✮ H h ✘ mp ✮ L EEnd nMO ✮ mp ✮ Eon E E N nn Op N L EO nM ✮ n E nn mp ✮ ✮ mp ✮ Eh NDApp Eu EQP N EuE nn Op Op ❉ ❉ H h ✘ L nCame ❉ nm ❉ ❉ GMC PP N G P N D N uY H hd ✘ LnD nu ❉ H hnd ✘ L nC ❉ ❉ ❉ P unu GMC uD N N MO A A C mp ❉ H mp h ✘ ✮ Cmp mp C ✮ mp ❉ mpnn ✮ CPNmp mp MC nn N nn A C mp ❉ mp ✮ ❉ C mp ✮ mp ✮ C mp MC nn C Pa assame ca on Compu euunHyb/D A go P hm ob em C So ass uhe on ca Pa eH App P ob ca em on So u nPmp on O Pa he ame eMO on✮ O hu ca nPM sN ❉ ❉ H h ✘ n ❉ mp ✮ nC N N Op H h ✘ L n ❉ H h ✘ Hmp hpmp ✘ ❉ mp nn ❉ mp ✮ mp ✮ neM N E n N n E dOp E N N Op N Op nn Op ca hnP ED C Qu Op m❉ D nd M he d P nn E n✮ PEm u b np m Cu Qu Op m D pex ndm P nun bYm ✮ dd N u H gh Comp x/ Ex Ye STuOp mp eCL✮ /Comp ex Comp MG dd EMH M hA ddfi Qu E n C h n❉ D✘ pEL Qu nd u n Op P m nn n CL Pn✮ b D Qu nd Op nn nnmp P bLmp D p✮ P nn n Pmp bApp A A A H h ✘ H h C mp ❉ ❉ mp C ✮ ❉ mp H ❉ hsn ✘ ❉ mp GA GA MH E MH N Y N GA Y MH V n A nC H hon ✘n❉ CL mp ❉ mp ✮ CE mp ❉ GA MH E N Y Vnn nV An H h ✘ D N D nP N N D nP N R m R T mA RCLnn C mp ❉ C mp mp ❉ C mp ❉ H hm ✘ mp ✮ C C mp mp ❉ ❉ mp ✮ C mp ❉ mp A ❉ mp ✮ Cmp mp ✘ A ✘ M NLP NLP D u M NLP N D u N n✮ n N n N N H MH E MH E Vnn nL Anu V H h ✘ C mp ❉ mp ✮ C mp ❉ CN mp A Em mp ✮ C mp ✘ M NLP D N N n❉ nC HHyb E VO n A H h ✘ L nu E nh H ✮ h CC ✘ LC nOP E n✮ ✮ ✮ mp nnC ✮ EE ✮ mp ✮ M LPP N N M LP N M LP N L n L n L n E n ✮ mp ✮ M LP n H h ✘ C mp ❉ H mp h ✘ ✮ C mp C ❉ mp ❉ ❉ mp ✮ C mp ❉ QP QP N D u N D QP N D N u N H h ✘ C mp ❉ mp ✮ C mp ❉ QP N D u N C mp ❉ ❉ mp ✮ C C mp mp ❉ ❉ mp P O MH E P O MH E Y P V n MH A E Y V n A Y P O MH E Y V n A H h ✘ C mp ❉ H h ✘ C mp mp ❉ C mp ✘ mp ✮ C mp ✘ ✘ C n N D u C n P u N C n P N C n n n mp E n ✮ mp ✮ C mp C mp ❉ n ✮ mp ✮ C mp ❉ OP N D u N D N u H h ✘ C mp E n H ✮ ✘ C mp E mp ✮ n C ✮ mp ❉ mp ✮ C mp ❉ OP D u OP N D u N N nn Op nn Op H h ✘ L n E n ✮ H h ✘ mp ✮ L E n ✮ mp ✮ E N E N E Op N nn LCMC ✮ E mp ✮ ✮ mp ✮L En P NDE E Nu E N E E ❉ ❉ ❉ nLCEL ❉ L/Comp ❉ n N✮ ❉Op mpnn ✮ C mp ❉ GMC Ph N DN Pd❉n N uN❉ MO M H GMC hEnA ✘Pd❉ Ln nu ❉ Ln n✘ G ❉ ❉ ❉✮ GMC NL MO MO A HnN❉ h ✘ mp mp ✮ Cmp ✮ A ❉ H h❉ ✘ h ✘C H hmp mp C ✮ ❉ CnP mp ✮ mp ✮ C emp mp C mp ✮FDM Dmp Dn u N❉ nn Op nn Opng mp ❉ ❉ H h ✘ LuDExecu n Ave ✮ H hnn ✘ n ❉ mp ✮ nnPd NPN N nnP ND EN N ❉ H ✘ L n n H ❉ mp hmp ✘ ✮ n Nnn mp ✮ mp ✮ d N E d❉ N N Op Op DM DM/Ave age age Comp x/ Ex S ex A A Aunn A GA MH GA MH E YCLC GA MH E N YA V nOp Amp V n N A V Execu onM Cos Me Qua hods y Op m za on Depend Cos Qua ann ng yYN P ob m ems za on Fea Depend es PN ann P ob✮ ems Fea H ✘ C mp H mp mp ✮ ❉ C mp ❉ H h ✘ mp ✮ C C mp mp ❉ ❉ C ❉ DH N D ng D nP N u N N D u R T m R Rnn Depend ann ob ems Fea es Hon h❉ ✘ D nP N N LOp R mp ❉ h C✘ mp A mp ✮ C ❉Y mp ✘ mp A ✘ ✘ NLP M NLP M NLP N n✮ n nm N MH EE uuP H MH Eu H MH E n A V nC Amp V Cmp mp ❉ mp ✮ CECCL mp ❉ mp A ❉ ✘ M NLP D u N n❉ nm H MH V n H h ✘ LLC nC n nnP❉ E ✮ ✮ H hmp ✘ mp ✮ mp n ✮ n ✮ mp ✮ M M LP N D uCE M LP N D N u n LA NY H h ✘ LN n E n ✮ mp ✮ M LP N N n H hh ✘ C mp H C mp mp ✮ ❉ C mp ❉ ✮ mp ❉ mp ✮ C mp mp ❉ QP D QP uA N QP N mp PH O P O MH MH PY O MH VC nL A V n A Y P O MH V nN AnnT H h ✘ H h ✘ C ✘ mp ❉ ✘ CD nn u CL nnn N CLnnnT H h ✘ C ❉ H h ✘ ❉ ✘ ✘ C u C uD CH nV nT mp E ✮ mp C mp mp ❉ nC ✮ mp C ❉ N D OP N D CN mp EE n❉ E mp ✮ n ✮n mp ❉ ✮ C mp ❉ OP NND DEN OP N Dmp N H h ✘ EE ✮ h ✘ mp ✮ L nC EEE✮ nMO ✮ mp ✮ EnLP ND E N Op N nn h ✘ L Eu nN❉ H ✮ h ✘ n mp ✮ ✮ mp ✮ En PPMC N EEN Op Op ❉ ❉ LnC nLC ❉A mp ✮ Cmp mp ❉ GOP PP N u N Ln nu Hmp hmp n✘ ❉ L n ❉ ❉ mp ✮ C mp mp mp ❉ GMC N u E uC GMC GE D P D✮ u MO N MO MO ✮ ❉ A nn H h ✘ C mp ❉ mp ✮ CEN mp ✮ nn H hnn ✘ C ❉ mpnn ✮ ✮ C mp mp ✮ N D N MC N D uN Op N nn A H hEnA ✘PP❉ C mp ❉ H h❉ ✘❉ CLn mp mp H ✮ ❉ hC C ✘ mp ✮ CY N mp ❉ ✮ mp ✮ C mp MC N N D uCE N nn Op N❉ nn Op Op ❉ ❉ ❉ A A A GA MH GA MH E N Y GA MH E Y V n A V n N A Y V GA MH E N V n A H h ✘ C mp H h ✘ mp mp ✮ ❉ C mp ❉ H h ✘ mp ✮ C C mp mp ❉ ❉ mp ✮ C mp ❉ D nP N D u D nP N D u N D nP N D u N R T m R T m N R H h ✘ C mp ❉ mp ✮ C mp ❉ D nP N D u N R T m H h ✘ C mp ❉ mp ✮ Cmp mp ❉ mp A ❉ h C✘C mp ✮ C mp C ✘ A ❉ mp Cmp mp AS✘mpmp ❉ mp NLP M N D uC NLP N N nn n✮ N MH E MH E Hbe V nnMH A E mp V n C A mp C mp A ❉ ✘ NLP N n✮ n LAn✮ H MH EN V hNLP ✘ L n nh ❉ ✮ H LM n✘ E mp ✮ ✮ Lmp n❉ ✮ EMe mp ✮ M N M LP DEer u N LP uN Nnn L✘ n nCnsn n H mp hmp ✘ ✮ C mp ❉ mp ✮ C mp ❉ N QP N D N N mp ✮ C ❉ mp ✮ C mp ❉ N D N D N C mp mp ✮ C mp ❉ mp ❉ PHLP OP N MH E uuM P O MH E EH PE OnPP MH E Y VOp Aon V Y Vnn H h ✘ mp ❉ C ✘ mp ❉ ✘ CeOP nM P C nn u C CLnM nT H h ✘ C ❉ H ✘ mp ❉ ✘ ✘ C n es P u C nH C ne nnN H gh ✘ ❉ H LN S nea mp ✮ ❉ e ✮✮ NNo Easy cons dnea No No Op nne SDD mp ✮ nne Op mp E ✮ H mp h ✮ C mp mp nCOp ✮ ❉ N D OP N D nn N CN mp nG ✮ mp n ✮n mp ❉ ✮ C No Exce /Easy Average/be ficu Numer ca Nmp Op m za Op annn Heur NN Den N D h ✘ LN nC En✘ ✮ mp ✮ EQP ND E H h ✘ L E nEasy H ✮Poor/D h❉ ✘ H LCmp n hgh EM n L✮ mp n ✮ ✮ E❉ n ✮ mp ✮ mp ✮ EQP EN EQP N Emp Ebe N E N Op ✮ nn ❉ ❉ ❉ H hL ✘ Lu neN mp ✮ C mp H h ✘ LC nnne ❉ mp ❉ N G N D uN MO N LnD nu ❉ LY mp H ✮ ❉ hC C ✘ mp ❉ L❉ ne mp ❉ ✮ CC✮ mp ❉E mp CAcs mp Op ❉ mp GOP PG D u GεOP PP❉ ump P❉ N N D uC MO N❉ MO MOC A A A A ❉ ❉ ❉ A A A A H ✘ C mp H hC ✘ C mp mp ✮ ❉ C mp ❉ mp ✮ C mp ❉ GA MH GA MH E GA MH E N YL V nnMH Amp V n N A Y A V GA MH N Y❉ V n ✘ mp H mp h ✮ C C ❉ mp ❉ mp mp ❉ nP D nP N D u D nP N D N u RL T m RmA N RCLnnnT H h ✘ C mp ❉ mp ✮ C mp ❉ D nP N D u N R T H hh ✘ C mp H h ✘ mp mp ✮ ❉ C mp ✮ ❉ mp C AC mp ❉ A ❉N ✘ ✘ mp ❉ ✮ mp M NLP NMH D NLP un h N M NLP NC n✘ N H H MH E MH EH HY nL A V n A C mp A ❉ ✘ H MH EN V nN An CT LC n EY✘ n ✮ H hnn ✘ mp ✮ E ✮V ✮ mpC ✮ M LP uE M LP uY n N h ✘M LLC nn E ✮ LN n E n mp ✮ ✮ mp ✮nnC M LP u EE M LP nnN n H h ✘ mp ✮ Cmp mp ❉ mpmp ❉ QP N D QP N D N H h❉ ✘❉ mp ✮ C mp ❉ mp ✮ CE L mp ❉ QP D QP N Dmp uD Nmp N C mp ❉ C mp mp ✮ ❉ C mp ❉ mp ✮ C mp ❉nn PE O N E u CD P O MH E Y PE OnP❉ MH E VOp nOp Amp V nnm AY V PP O MH E V n A C mp ❉ H h ✘ C ✘ ❉ mp ✮ C mp ✘ n u C n u n N H h ✘ ❉ C mp ❉ ✘ ✘ C n P u C n P C n H h ✘ C E ✮ mp ✮ C ❉ OP N D nn Op C mp E n H ✮ h ✘ C H mp h mp C ✮ n mp C ✮ mp E ❉ n ✮ mp ✮ C mp mp ❉ C mp ❉ OP N D OP N OP D u N D u N N nn N nn Op nn Op H h ✘ L n E n ✮ ✮ H h ✘ L E ✮ mp ✮ N N N E N nn H h ✘ E n ✮ L n E n H mp ✮ h ✘ ✮ L n E n ✮ mp ✮ mp ✮ E N E E N E E N E N nn Op ❉ ❉ ❉ A A A MH Ar fic a Ngh e✘ gence Hybr d Me hods Hyb Dec sGA on❉✮A Mak ng DM H Comp be eDHuOp ❉ H gh ✘ SCmp Comp ✮ Comp ex ex emp ✮ ❉ SNA mp exmp ✮ ❉ N DNo cu MCS N cu nne No Op nne S mp enDu ✮ Comp exA ✮ nne ❉ ❉ ❉ ❉ A A A h ✘ C mp H mp h ✘ ✮ C mp C ❉ mp ❉ mp ✮ C ❉ GA GA MH Enex MH E MH NeN E Y N YA V n❉ AnV n Y e ✮ Comp V H h❉ ✘ CNo mp ❉ mp ✮ C CL mp ❉ GA MH EN N V A T H h ✘ CMH mp H h ✘ mp mp ✮ ❉ C mp ❉ mp ✮ C mp ❉ DPHnP nP N D nP Nbe Tn m R R Lnn ✘ H mp h ✘ ✮ C C ❉ mp C mp A ❉ CC mp mp ✮ C ❉ mp ✘ A ✘ ✘ NLP N D u M NLP N D u nn n✮ N N MH E uuM MH EH H MH E V A n V h ✘ C mp ❉ H h ✘ C mp mp ✮ ❉ mp ❉ ✮ C mp ❉ M NLP M NLP N Dn uD N NY nEnnR nnn N n❉ n✘ LC E✘ n mp ✮ nmp EE mp ✮ LP n LC Eu nNnh ✮✮ LC n E n mp ✮ ✮ mp ✮ n H h ✘ C ❉ ❉ mpnn ✮ ✮ C mp mp ❉ QP N D QP N D uY N H hQP ✘PD mp ✮ C mp ❉ ❉ QP N D N mp ❉ mp ❉ mp ✮ C mp ❉ O N MH P Y PE O❉ MH E V Amp V nm AY V PP O MH V A✮ HE h ✘h C mp ❉H mp ✮ C mp ✘ CMH nLP u N nnn mp ❉ H ✘ Hmp hmp C ❉ ✘ ❉ mp ✮ C mp ✮NCL mp CM nLPPOP u E CM nLP D P D u N CL N nnV C n ❉ mp H hHO ✘ CCEC mp ✮❉ mp ✮ C mp ❉Ynn H h ✘ Cmp mp nCL ✮ mp ✮ C mp ❉ D N OP N D u N nn H h ✘ CN mp EE mp mp ✮ n hmp C ✮✘ mp ❉ C mp nnOp ✮ ❉ mp CA mp OP ND D N OP OP NnC D uCM Op N✘ Op Op ❉ ❉ A A A A ❉ ❉ N ❉ ❉n❉ A A A A H hGA ✘ mp H C mp mp ✮ ❉ C mp ❉ ✮ C C mpmp ❉ ❉ mp ✮ C mp mp ❉ GA MH E MH E N Y GA MH E N YL V n❉ A V n N Amp Y e ✮ Comp V H gh ✘ nea eMO H ✘ S✮ mp eYN nea ✮ Comp ex ❉ ex ❉ ❉ N DNo Goa PN cu No M H❉ h ✘ H h ✘ D nP nP N R RCLn eD Comp ex ❉ nPC nP N RC T m✘ RT mp mp ✮ C C ❉ mp ❉ CE mp A ❉h ✘ C mp A mp ✮ C ❉L mp ✘ mp ✮ mp C mp A ✘ ❉ ✘ M NLP D u M NLP D u nn n N N MH E u MH ED MH E Vh Amp n V ✘ mp ❉ H hP❉ ✘ CNo mp mp ✮ ❉ C mp ❉ mp ✮ C CL mp ❉ C mp A ✘ M NLP D u M NLP D uN N NC❉ N nnne n N nm nm H MH EC V n A H LC E✘ n ✮ H mp ✮ EMO ✮ mp ✮ M LP M N n nnT LC Eu nNCh ✮ LC n E n mp ✮ mp ✮n Mcu M n C mp ❉ mp ✮ Cmp mp ❉ QP N DN ube N H h ✘ hgh mp C ❉ mp ❉ mp ✮ C mp mp ✮N CL mp ❉ QP Nmp D N QP QP D u N N mp ❉ mp ✮ C mp ❉ ❉ PH OD PhHL O MH EnD MH E PH OP❉ MH E Y V nL A V nSCA AY V Ph O MH V nV A✮ H ✘ C mp mp ✮ C mp ✘be H h ✘ C ❉ mp C ✮ ✮ C mp ✘ N CD nLP N D uY C nn N H h ✘ C mp ❉ mp H ✮ h❉ C✘ mp ✘ mp ❉ ✘ mp mp ✘ mp CD nLP P Hn P SN u✮ C nLP P❉ n✘ N NnH D uC N❉ C nT n C ❉ ❉ ❉ A A A A ❉ ❉N C ❉ ❉ A A A mp ❉ H mp h ✘ ✮ C mp C ❉ mp ❉ mp ✮ C mp ❉ GA MH E GA MH N E Y V n A N Y V C mp ❉ C mp mp ✮ ❉ C mp ❉ mp ✮ C mp ❉ GA MH E GA MH E Y N Y V n A V n A H h ✘ D nP N D nP N N R T m R H h ✘ H h ✘ D nP N D nP N D u N N R T m R T m C mp ❉ H mp h ✘ ✮ C mp C ❉ mp ❉ mp ✮ C mp ❉ C mp A ❉ C mp A mp ✮ C ❉ mp ✘ mp C ✮ mp C mp A ✘ ❉ ✘ M NLP D u M NLP D u N n n N N H MH E H MH E H MH E V n A V n A V C mp ❉ C mp mp ✮ ❉ C mp ❉ mp ✮ C mp ❉ C mp A ❉ ✘ M NLP D u M NLP N n n N n n H MH E V n A H h ✘ L n E n ✮ mp ✮ M LP N N L n H h ✘ L n E n H ✮ h ✘ H L n h ✘ E n L mp n ✮ ✮ E n ✮ mp ✮ mp ✮ M LP N M LP N M D LP u N D u N N L n N L n L n HES ✘ C Exce C mp ❉ ❉ NH mp ✮ C nea mp ❉ H V hmp ✘n✮ CC mp mp ✮ C mp ❉ QP N D u C Exce N mpC ❉QP mp mp ✮❉ hC ✘ mp ❉ mp ❉ ✮ mp mp S ❉ QP P QP D N u mp DCNo uC ✘ N❉ No mp ❉ mp ✮ ❉ C mp ❉ mp mp ❉ ❉mp O MH E Ph nea O MH E Op YNgh Pmp O H MH E Y✮ AC ✮ V ✮ n CAY V nT H gh ✘u e ✮QP L en ✮ L S mp e en mp e ✮ NNo Easy N Easy nne Op nne SD nne ❉ ❉ ❉ A A A A ❉ A A A A HnN M h ✘ C mp ❉H mp ✮ C mp ❉ mpmp mp L ✮ ✮ C mp mp MH E GA MH NN E Y V nL AT NN Yn n V H C mp ❉ H hu ✘ ❉N Cn mp mp ✮ ❉ C✘ mp ❉ mp ✮ C CL mp ❉ GA MH GA MH Y N V n V nV A mp H h ✘ C ❉ DPGA nP N nP N u RC✘ N✮ R L nT h ✘❉ D nP nP N RL m RT H h mp mp ✮ mp ❉ CC C A mp AC ❉ mp ✮ mp C ✘ ❉ mp ✘ M NLP DN uE nN n✮ H H✘ MH EnDNLP MH E✮ V n A V C mp ❉ H ✘ H hmp ✘ mp ✮ ❉ mp C mp ❉ ❉ ✮ C mp ✮N mp ❉ C mp A ✘ M NLP D E M NLP D N nnT nmp N nm H MH E V A H LMH n nh ✮ H h ✘ EA mp ✮ Nu D N M LP N D uY LA N ✘ LM Eu E E ✮✘ LC E n mp h❉ ✮D LY n N E nn C ✮ mp ✮n M LP M LP LP N Nmp D u D n n C C mp ❉❉ mp ✮ C mp C ❉ mpN❉ ❉ mp ✮ C OD MH E PH O❉ MH YE V n❉ An Y V C mp ❉ CNo mp mp ✮ ❉ Cmp mp ❉ mp ✮ C mp mp ❉ Pcu O M LP N E Ph O MH E V n A nnm A nS Amp ❉ ❉ ❉A ❉ Y H gh ✘uComp Comp ex Exce ✮ H gh ✘ S✮ Comp ✮ ex Comp Exce ex en ❉ ✮ exmp ❉ ❉ N DNo OPF N EDen cu nne No OpnV nne A n SNMH mp eD ✮ exhhH❉ nne Op ❉ ❉ ❉ ❉ A A A H h ✘ Cmp ❉ mpN n✮ ✮C C mp ❉ MH EDmp GA MH NeN E Y n❉ AT NN Y e ✮ Comp V HM hA ✘ C mp ❉ mp ❉ ✮❉ C✘ mp ❉ MH GA MH N V V A✮ HANM h ✘ C mp mp ✮ Cmp mp ❉ nP N DN u RC✘N T H h❉ ✘ Hmp hmp ✘ C ❉✘ ✮ C mp ✮V mp ❉ DPGA nP D nP D nP ump DNLP u NCY RA T mA Nn❉ R H ✘ mp ✮ C mp ❉N H h ✘ C mp mp ❉ CE mp ❉ NY C A mp C ❉ ✮ C ❉ ✘ N NLP N D uY N n✮ nC N MH MH MH E Amp V n H ✘ C ❉ mp ✮ hmp C mp ❉ mp mp M NLP D E u E u DPGA N NY DC u Mmp nV nmp N❉ nm nm C mp C ❉ mpmp ✮ nC mp ❉ O MH E mp PH O ❉H MH Ymp E V n❉ An Ymp T m❉ mp V C ❉ CN mp mp ✮❉ C✮ mp ❉ mp ✮ C Cmp mp ❉A OMHNLP N MH E PhNLP O MH EC V n A V nV A n CAR ❉ ❉ ❉ A A A A ❉ ❉ ❉ ❉ A A A H MH h ✘ C mp ❉❉ mp ✮T Cm mp ❉Convex H gh ✘uComp ex ❉ H ✘ Sh❉ Comp ✮ Comp ex ex emp ✘ ❉ ✮ Comp exmp✘ ❉ GA MH E NeE Y V n❉❉ AT N DNoDPGA cu Cone PMH DA cu No No Con H GA H❉ ✘ N H hgh mp ✘ C mp ❉ ✮ C mp ✮V C V mp ❉ MH MH GA E Eh ❉ Y N V nR N A Y n A V n A S MH mp eD exComp Convex H h✘ ✘ mp mp ✮ C mp ❉V H h ✘ C mp N N nP N D uY N R T ✘ C mp ❉ mp ✮❉ Cmp ✘ mp ❉ mp ❉ mp e ❉ nP D E u✮ nP Dmp uN D nP N Nmp DC u D NCY Rbe T mA Rn CY A mp ✮ C mp ✘ ✘ MH Emp MH n A V CN A e C A ✘ mp ✮mp C Cmp mp ✘A H MH E H MH ECbe V n A nm A ✮SCmp C mp mp ✮ C Cmp ❉ mpN❉ ❉mp mpR ✮ ✮TC Cm mp ❉ PHOD E PH O ❉H MH YE V V A Y V C ❉ Cmp mp mp ✮ ❉ C mp ❉ ✮ C mp ❉ O D nP N E Ph O MH E n V n A ❉ ❉mp ❉ ❉ A ❉ A A A ❉ ❉ A A A h ✘MH mp mp ✮ CeE mp ❉V H ✘ Cmp mp mpV ✮ ✮nC C mp ❉ MH ✘ E N YA ✘ GA MH Vh Amp N mp Y e ❉ V Fa H ❉ hH ✘ mp ❉ HGA h❉ ✘ N Cmp mp mp ❉H ✮❉ Cmp ✘ mp ❉ CYComp mp mp ❉ ✮ C Cmp mp ❉ mp MH GA MH E Y Egh Vbe n A N ✘ YA V Vn n A A✮ C Amp CY mp ❉ mpn ✮n C✮ ✘ mp EP H MH n❉❉ An V mp AFas C A mp ✮ C mp ✘ mp ✮ C H MH MH ECbe n A V n H EEEgh ex eDA ❉ H Sh❉ Comp ✮ ex emp ❉ ✮ Comp ex ❉ ✘ N DNoPGA cuO GA S MH SQP N cu Fas No mp e ✮ Comp exComp ❉MHCCCMH CMH mp mp ✮ C mp ❉ PHO❉ E YE N A mp ❉❉ CNo mp mp C ✮ ❉ mp C mp ❉ C mp mp ❉ ✮V CV mp ❉ PHO OE MH E Y ❉exV A Y A S mp V n A ❉ ❉ A ❉ A ❉ A ❉A A ❉ ❉ A A A A CYMH mp A PH C✮ ✘ mpV ✮ mp ✘ H PHO MH Emp MH E An V CC A CCmp A mp ✮❉❉ C ✮ mp ✘YC mp mp ✮ C Cmp mp ✘A V V MH E MH EC mp V mpn A✮n❉ n A A ✮ C Ymp mp CS mp ❉ Cmp mpY ❉ MH E YEgh O ❉ MH E Ven Amp V mp ❉ mp mp ✮ C mp ❉ mp C mp MH E P❉ O A ❉ ❉ML ❉❉ H Egh ✘ nea Exce ✮ ✘ mpExce eV ✮ n❉ ✮ S ❉mpmp e ✮✮nCCAmp N DNoPH cuO P O MH LP N DA ❉A cu L❉ nea NoVn❉❉ Ln S mp eA✮ L en nea AHNo ❉ ❉ ❉ ❉ L nea C mp ❉ ✘ ✘✮VC V H H MH Emp n An ✘ CMH ❉E C mpA Amp C ✮mp ❉ C mpA ✘ ❉ V mpn mp C mpmp mp H MH E H E A MH A✮ C✮mp A V n A ❉ ❉ ❉gh ✘ A S mpComp Aex A ex A ❉ Acu H Egh ✘ Comp ex be e ❉ H e ✮ Comp ex be e ❉ S mp e ✮ Comp ex ❉ N DNo H cu H S MH M NLP N D No Non No nea Non mp e ✮ Comp ❉ Non nea A H C E mp ✘V A n mp ✘A ✘ MH E H MH Vmp A C Cmp V CMH mpCE mp A ❉ ❉ C mp A mp mp ✮❉ C ✮ mp ✘C mp ❉ mp ✘ mpV ✮nCAmp H MH E A n✮ V ❉ n A ✮ C mp ❉ ❉ ❉ ❉ ❉

Tab ed e 10n Per compar o Tab a gor e 10 hms Perapp ormance ed n mu compar obson ec o ve apgor ann hms ng prob appems ed n mu a gor hms app mu ormance ob ec ve p annson ng prob ems

hm P ob em A Sogo u on hods a y Op Me m za on L nea beε econs ❉d Comp ex be MCS e ❉ L nea be Goa e ❉P ES ✮ nea Exce en mp ex ExceOPF en ✮ P Comp ex beCone e ❉ Comp ex be SQP e ❉ nea Exce M enLP✮ Comp ex beMeNLP ❉ Comp ex be DynP e ❉ Comp ex be GA e ❉

Comp ex be PSO e ❉

HS ❉ mp ex Ave age

ob ec ve p ann ng prob ems

6 Requirements for Future Work and Research Directions

The paper presents four p ann ng techn ques based on the MO framework w th assoc ated c ass ficat ons methods and key nformat on from the v ewpo nt of the r usefu ness n MDP Furthermore other c ass ficat ons nc ude four extens ve ob ect ve categor zat on updated dec s on H gh ✘CompDynP beN eD ❉cu HNo gh ✘ S mpComp e ✮ Comp ex be ex e ❉ S mp e ✮ Comp ex N DNo cu Rea TNo me S mp e ✮ exComp ❉ ex Rea T me var abEasy es constra nts and modebesMH A so fifteen PTC-based probe ems nc ud ng nter nked H gh ✘Comp ex Comp e Easy ❉ H gh S mpComp e ✮ Comp ex be ex ❉ S mp e ✮ Comp ex MH GA No Yes✘ Va No an Yes A S mp e ✮ ❉ exVa No Yes an A ( nterdependent) have ustrated However the terature rev ew revea s that there are st Ave types age Ave age beMH e Easy ❉ S mpComp e ✮ Comp ex be ex e ❉ S mp e ✮ Comp ex MH Easy Yes Va an YesA S mp e ✮ Comp PSO exComp ❉ exVa Yes an A severa potent a ❉research areas from the p ann ng❉ perspect ve of MO-based frameworks that are Ave age Ave age Comp ageEasy S mp Comp e ✮ex Comp Ave ex age✘❉ S mp e ✮ Comp ex MH Easyworthy HS Va an A S mp e ✮ Comp ex ✘ ex Ave Va MH an A❉ research ❉ ❉ 6 1 D str but on Network Topo ogy The topo ogy most y cons dered n rev ewed works s rad a n nature and other configurat ons ( oop and mesh) have usua y been neg ected due to cost ssues desp te the r re ab e nature

❉ ❉

Rea T Va a

❉

Va a

✘

Va a

Energies 2017, 10, 208

32 of 44

Furthermore, futuristic distribution mechanisms are expected to be more interconnected in both nature and operation. Hence other configurations must also be evaluated for reliability and other technical perspectives to avoid missing reliable solutions on a cost basis only. The optimal weight allocation in decision making can provide a feasible solution. 6.2. Future Distribution Networks (FDN) Future/Smart distribution concepts and management models must also be considered from the MOP perspective. The key FDN concepts have been divided along the lines of utility and consumer paradigms. Major FDN concepts on the utility side include looped DNWs, meshed DNWs, micro-grid (MG), clustered or multi MG (MMG), virtual power plants (VPP) and smart cities (SCs), whereas smart homes (SHs) and buildings (SBs) represent FDN models on the consumers’ end. Despite classification, designing and planning of FDNs to meet future load demands from the smart grids’ perspective needs to be further explored. MOP can be use as a promising tool to exploit this potential research area of FDN from multiple stakeholders’ viewpoints. This issue will be addressed in future publications. 6.3. Multi-Obejctive Planning with Optimization Parameters Settings Since MOP may need large scale problem formulation, a possible way is to efficiently decompose a large problem into sub-problems on the lines of spatial (DNW’s size) and time (PP). The inner optimizations parameters (if heuristic methods have been used) can be optimally set by either the planner or adaptively improved with the support of SG technologies (SGTs). Key enabling SGTs may include advanced metering infrastructure (AMI), secure communication link, sensors and demand side management (DSM), favoring both service provider and consumer regarding their decisions and anticipated incentives. 6.4. Prioritization of Weights for Objectives in Future Distribution Networks Since most of the models and frameworks utilize heuristic methods for evaluation that is time-consuming and associated subjective weighting methods are liable to the deviation. Hence for a quick trade-off solution, planners and researchers must coordinate to find relevant weights from the perspective of requirements and other distribution planning-related issues. SG pilot projects and test beds can provide an opportunity to find accurate weights for each objective in MO problem. Moreover, efforts must be made to improve the speed of optimization processes. 6.5. Active Network Management and Smart Distribution Management System (SDMS) The active network management (ANM) enables active operation of DNW with support provided with communication, information and automation enable changing of protection settings, topology, and power-flow dynamically. Efforts are required for proposig new methods, aiming at achieving multiple objectives like high REG penetration, power quality, reliability while reducing overall cost and power losses, respectively. Since DERs are expected to be increased in ADNs and are required to be control by a smart distribution management system (SDMS), since such a system with AMI will allow real-time communication among decision makers for the transaction of useful information and prices makes it naturally a MOP problem in the ADN context. However, an increasing number of DER and loads over a planning horizon respectively adds complexity to SDMS control functionalities. There are several issues of SDMS that needs to be addressed in MOP problems. A compromise between control issues can be a potential area since central control has high computation requirements and decentralized control requires time-consuming synchronization among local agents to find a compromise solution. Also, addressing complications in load flows with small scale DGs on utility/consumer ends and avoiding delays in command signal propagations; are worthy research areas, in particular, system parameters synchronization (grid, DGs) and contingency analysis under normal and emergency scenarios respectively.

Energies 2017, 10, 208

33 of 44

6.6. Islanded Operation with Multi-Type Distributed Generation Units and Modified Topology Current utility practices discourage islanded operation and recommend indiscriminate disconnection of all DGs connected to a network. Such actions are neither suitable nor preferred in a deregulated and competitive multi-stakeholder electricity markets. Further, islanding with multiple types of DGs in DNW and other concepts (MG, MMG, VPP) must also be evaluated with a change of topology, since interconnected topologies (loop, weakly mesh) are more reliable and serves more consumers in a better way during main grid blackouts. 6.7. Advance Protection System (APS) The traditional protection systems are more specific towards cost effective solutions like relay-recloser-fuse coordination. However, when the DG penetration exceeds a certain limit, the traditional protection is not technically viable. The protection up gradation is also necessary for future distribution (interconnected) systems having complex power flows. Hence, a suitable trade-off between complexity and economy for any future protection strategy has desired for the distribution mechanism incorporating high DG/DER penetration. 6.8. Dynamic Planning with REGs and High Nonlinear Load Models The MOP and related planning issues with increasing REG penetration and nonlinear load by high percentage need to be addressed over the multi-stage planning horizon, mainly in the context of FDNs. The major concerns are ensuring system reliability, stability and power quality. Reason being, the simple load models, will not remain technically viable to access the actual benefits. 6.9. Incentive Prioritization for Owner, Investor and Consumer Based Future Market Scenarios The FDNs in a smart grid environment are expected to have competitive market scenarios and needs maximum stakeholders’ participation. Hence incentive-based approaches are required to prioritize facility (DG, devices) owners, investors, and consumers; instead of overall system-wide benefits, to ensure their participation in FDN planning processes. 6.10. Exploiting Real Options The utilization of real multiple options makes an active research area for the futuristic planning [102]. Prominently, risk-based planning under uncertainty needs to be addressed in MO framework to attain suitable investment strategy under the extreme scenario, from a futuristic point of view. 6.11. Need for Integrated Planning The literature review reveals that most of researchers have considered planning and (resource) scheduling as separate problems. However, a real-world planning problem needs to be addressed with deeper, wider and aggregated planning approach. The futuristic planning needs to be integrated from startup stage (over planning horizon of several years) aiming at achieving multiple objectives. Followed by efficient resource utilization (days to seasons in a year). Finally, ensure real-time stable operation with anticipated smart technologies (on the time scale of 15 min to one day). Such approach is expected to guarantee optimal planning solution of FDN on long term basis. 6.12. Need for Improved Load Flows The FDN of future is expected to be interconnected in nature. Hence, there is still room to proposed LF models, which addresses interconnected nature of DNW. Also, new LF models need to be developed considering various types of load models, from the viewpoint of uncertainty (generation and load) and interconnected nature of FDN. The proper formulation, new solvers and hot (improved)

Energies 2017, 10, 208

34 of 44

initialization of network load/power flows must benefit MOPT in terms of less computation cost and efficient solutions. 7. Conclusions The MDP has been motivated by various factors and features (ANM, RES, EV, ST, ADA, DR, DSM, etc.), which were limited in traditional planning. The real world planning problems are multi-objective (MO) in nature and involve a large number of stakeholders. MO planning tools can provide compromise solutions among contradictory objectives, satisfy multiple stakeholders, yet address the concerned issues. This paper presents a review of four planning techniques (PT) aim to address MOP problems with associated models and optimization methods under MDP paradigm. The primary aim of this paper is to provide a back ground for FDN planning on the basis of limitations in available works and from the perspectives of MO achievement. The reviewed works consists of 80 recent standard planning papers from various aspects and are organize on the basis of decision variables, attained objectives, abiding constraints, MOP formulations, test systems, load models and year of publications (Section 2). The promising MOPT have classified as DGP, VPQ, CRU, and NTR (Tables 1–4). The MOP planning methods have reviewed into five categories, namely; numerical, meta-heuristic, hybrid, decision-making and load flow methods respectively. Also, classification and interdependence of planning components from four MOPT perspectives, have arranged in (Tables 5–8) with associated contributions from each related work. The LF impact on the basis of interactions, aiming at MOPT problems have arranged as a big picture, has presented in Table 9. The overall performance comparison of techniques (methods) applied for MOP problems have been presented in Table 10. Furthermore, potential future directions in MDP from a MOP perspective have also highlighted. In the future, MOP has several grey research areas in accessing maximum benefits from interconnected network topology, FDN and associated concepts. However, more investigation is needed to carry out realistic (MO) planning by upgrading conventional to smart DNWs or redesign from the beginning. Also, optimal settings of optimization parameters, optimum objectives weights in FDNs, ANM techniques, smart distribution management system (SDMS) and islanded operation with multi-type DGs focusing modified topology need further research attention. In addition, advanced protection schemes (for bidirectional power flows), dynamic planning with high penetration of REGs and nonlinear load models will play an important part in futuristic planning problems. Moreover, maximum incentive-based prioritization given to stakeholders (facility owners, investors, and consumers) needs to be addressed to ensure maximum participation in FDN planning processes. The employment of real investment options concept in MO framework for suitable approach under extreme (worst case scenario) from the futuristic viewpoint is research worthy. Finally, efforts need to be made to proposed integrated planning approaches and new power flow models. Works regarding designing and planning of FDNs, from multiple aspects under MO framework, will be presented in future studies. Acknowledgments: This work has supported by Institute for Information & Communications Technology Promotion (IITP) grant funded by the Korea government (MSIP) (No. R0113-15-0002, Automotive Information & Communications Technology (ICT) based e-Call standardization, and after-market device development). Author Contributions: Syed Ali Abbas Kazmi (first author), Muhammad Khuram Shahzad (second author), and Dong Ryeol Shin (corresponding author) provide a composite review in the field of multi-objective planning focusing on four planning techniques. The authors provided a critical assessment of different classifications, approaches, key features and constraints to find potential future research directions for FDNs. Conflicts of Interest: The authors declare no conflict of interest.

Energies 2017, 10, 208

35 of 44

Abbreviations The following abbreviations have used in this paper: A B C D P Q ABC ABFO AC ACO ACS ADA AHP AI ANM ANSI AMHBMO APS AεCM BA BFO BL BLFM BSFS Cap CABC CAC CCC CCI CCT CEDGP CEL CENS CIC / CILL CL CLd CLT CLL CLLI CNU CVaR CVR/C COD Cone P CPE CPL CRU CSAIDI CSI CSM

Classification of planning techniques related to DG placement Classification of planning techniques related to VAR and power quality Classification of PT related to component reinforcements/allocations and upgradations Classification of planning techniques related to network topology change or reconfiguration Active power Reactive power Artificial bee colony Adaptive bacterial foraging optimization Annual cost Ant colony optimization Ant colony search algorithms Advanced distribution automation Analytical hierarchal process Artificial Intelligence Active network management Average network security index Adaptive modified honey bee mating Advance protection system Augmented ε constrained method Bat algorithm Bacterial foraging algorithm Balanced load Branch load flow model Backward/forward sweep load flow Capacitor Chaotic artificial bee colony Capacity adequacy cost Current carrying capacity (of branches or feeders) Current carrying index (of branches or feeders) Critical clearing time Capacity enhancement with distributed generation penetration Cost of energy losses Cost of energy not supplied Customer interruption cost/Customer interruption level Controllable/responsive/flexible load Constant load Central limit theorem Critical load level Contingency load loss index (overall) Cost of network upgrading Conditional value at risk Cost of voltage regulators and variable capacitors Cost of overall (new and old) devices Cone programming Cost of purchased energy (from grid) Cost of power losses Component reinforcements/allocations and upgradations Cost of system average interruption frequency index Customer service interruptions Capacity security margin (of transformers and feeders)

Energies 2017, 10, 208

DE DEL DER DISCO DM DG DGOI DGP DGPL DGR DGUI DNW DP DPL DR DSM DSS DRP DSM DV DynP EA EC ECE EEDG2G EI EIG EL ELC EMP ENS EOI ES ESGA FACTS FCF FDLF FDN FGP FCM FCL FCLL FCLLDG FDM FLd FLL Fr. GA GAMS GP GS GHG/ACE GHGE

Differential evolution Distributed storage system’s energy losses Distributed energy resources Distribution company Decision making Distributed generation penetration level Distributed generation unit’s owner income Distributed generation unit placement Distributed generation Distributed generation unit’s reliability Distributed generation unit unavailability index Distribution network Distribution planning DIgSILENT programming language Demand response Demand side management Distributed storage system Demand response provider Demand side management Decision variable Dynamic Programming Evolutionary algorithm Equipment cost External cost of energy (from grid) Energy export from DG (REG) to grid Economic index Energy imported from grid Energy losses Energy loss reduction of DG units and capacitors Electricity market price (risk based) Energy not supplied Expansion, operation and maintenance costs Exhaustive search Enhanced gravitational search algorithm Flexible alternating current transmission system Feeder current flow Fast decoupling load flow Future distribution network Fuzzy goal programming Fuzzy clustering method Fault current limiter Fault current level Fault current level due to distributed generation unit Fuzzy decision making Fuzzy load Fuzzy load level Feeders (of distribution network) Genetic algorithm General algebraic modeling system Goal programming Gauss-Seidel Greenhouse gases/Average (annual) greenhouse gases Greenhouse gases emissions

36 of 44

Energies 2017, 10, 208

GHGG GHGDG GSA GRC GUC HBMO HS hr. IA IBVT IC ICAll ICDGPPF IELOM ILP ILQ IMO IMOHS IMOPSO-PS InvC IntC IOC IRC ItC IVD LB Ld LdP LdM LF LLI LOADSYN LV MaVL MCDA MCDM MCS MCSPM MDP MH MILP MINLP MISOCP MiVL MLL MnC MODGP MO/MOO MOP MOPT MOshBAT MOSOA MV

Greenhouse gases emissions from grid Greenhouse gases emissions from distributed generation units Gravitational search algorithm Grid reinforcement components Grid upgradations with devices/components Honey bee mating algorithm Harmony search algorithm Hours Immune algorithm Interactive bi-objective programming with the valuable trade off I (current) line flow limit index Investment in distributed generation units and capacitors Investment in distributed generation units and passive power filters Investment in energy losses, operations and maintenance Index for active power loss (P-loss index) Index for reactive power loss (Q-loss index) Index for multi-objective performance Improved multi-objective harmony search Improved Multi-objective particle swarm optimization with preference strategy Investment cost Interruption cost Investment and operation cost Investments in reinforcement cost Installation cost of project Index for voltage deviation Load balancing Load Load profile Load model Load flow Line loading index Load model synthesis Low voltage Maximum voltage level limit Multi-criteria decision analysis Multi-criteria decision making Monte Carlo Simulations Multi-criteria stochastic planning model Modern distribution planning Meta-heuristics Mixed integer linear programming Mixed integer nonlinear programming Mixed integer second order cone programming Minimum voltage level limit Multiple load level Monetary (risk) cost Multi-objective based distributed generation unit placement Multi-objective/Multi-objective optimizations Multi-objective planning (or Multi-objective based planning) Multi-objective planning techniques Multi-objective shuffled bat algorithm Multi-objective seeker optimization algorithm Medium voltage

37 of 44

Energies 2017, 10, 208

NC NDE NL NO NPL NPV NR NS NSGA NTR OCPL OCS OF OFVC OLTC OLSSFrLN OMC OPF PBDG PBSS PBY PC PEM PF PFDE ph PLd PLF PLHD PLL P-loss PQ PP PSO PT QCP QIC Q-loss QOTLBO QPD RA RCC RCCI REG SA SAMHBMO SAIDI SAIFI SAIUI S-BB-BC SC SCI SCL

38 of 44

Normally closed switches Non-distributed energy Non-linear/Non-controllable load Normally open switches Network power losses Net present value (of components, system and project) Newton Raphson load flow Net savings Non-dominated genetic algorithm Network (distribution) reconfiguration Overall cost of power losses (during operation) Overall complete system cost (including installations, O&M, power losses and reliability) Objective function Overall fixed and variable costs Online tap changer Overload (OL) at substation (SS), feeders (Fr) and loads nodes (LN) Operation and maintenance cost Optimum power flow Power buying from DG owner Power buying from substation (grid) Payback years Penalty coefficient Point estimation method Power filter Pareto frontier differential evolution Phase Probabilistic load Probabilistic load flow P (active) power loss based harmonic distortions Probabilistic load level Real/resistive power losses Power quality Planning period (horizon) Particle swarm optimization Planning techniques Quadratic constrained programming Q (reactive) current (I) component Reactive/inductive power losses Quasi oppositional teaching learning-based optimization Q (reactive) power deviation Reclosers-automatic (or Automatic reclosers) Reserve capacity of conductor Reserve conductor capacity (RCC) index Renewable energy generation Simulated annealing Self-adaptive modified honey bee mating System average interruption duration index System average interruption frequency index System average interruption unavailability index Supervised Big Bang–Big Crunch Short circuit Short circuit index Short circuit level

Energies 2017, 10, 208

SDMS SFL SLL SOA SOF SPEA SPMC SQP SSW ST STATCOM STP SUC SVC SWC SWD TDP THD TLA TLBO TOC TS TSIAECP TSW TVLL TVV UB UL VAR VD VEPB VLd VLDG VMP VPI VPQ VR VRC VS VSI VSL VSgL VSM VS&OP VTHD VUBP WSM Yr.

Smart distribution management system Shuffled frog leaping algorithm Single load level Seeker optimization algorithm Single objective function Strength Pareto evolutionary algorithm Switch purchasing and maintenance cost Sequential quadratic programming Sectionalizing switches Storage Static synchronous compensator Short term planning System upgrade costs Static Volt-ampere reactive power (VAR) compensator Switching costs Switching devices Traditional distribution planning Total harmonic distortions Teaching learning algorithm Teaching learning-based optimization Total operation cost Tabu search algorithm Two stage immune algorithm embedding compromise programming Tie-switches Time-varying load level Total voltage variation Unbalanced combined single and three phase loads Unbalanced three phase load Volt-ampere reactive power Voltage deviation/drop Voltage error at power buses Variable load Voltage level at distributed generation unit Voltage magnitude profile Voltage profile index Volt-ampere reactive power compensation and power quality Voltage regulator Variable capacitors Voltage stability Voltage stability index Voltage stability limit Voltage sag level Voltage stability (load-ability) margin Voltage stability and optimization (tool) Voltage based total harmonic distortions Voltage unbalance profile Weighted sum method Years

39 of 44

Energies 2017, 10, 208

40 of 44

References 1. 2.

3. 4. 5. 6. 7. 8. 9. 10. 11.

12. 13. 14.

15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

Farhangi, H. The path of the smart grid. IEEE Power Energy Mag. 2010, 18–28. [CrossRef] Eajal, A.A.; El-Hawary, M.E. Optimal capacitor placement and sizing in unbalanced distribution systems with harmonics consideration using particle swarm optimization. IEEE Trans. Power Deliv. 2010, 25, 1734–1741. [CrossRef] Fang, X.; Xue, G.; Yang, D. Smart grid—The new and improved power grid: A survey. IEEE Commun. Surv. Tutor. 2012, 14, 944–980. [CrossRef] Khator, S.K.; Leung, L.C. Power distribution planning, a review of models and issues. IEEE Trans. Power Syst. 1997, 12, 1151–1159. [CrossRef] Tuballa, M.L.; Abundo, M.L. A review of the development of smart grid technologies. Renew. Sustain. Energy Rev. 2016, 59, 710–725. [CrossRef] Wang, J.J.; Jing, Y.Y.; Zhang, C.F.; Zhao, J.H. Review on multi-criteria decision analysis aid in sustainable energy decision-making. J. Renew. Sustain. Energy Rev. 2009, 13, 2263–2278. [CrossRef] Rodriguez, A.A.; Ault, G.; Galloway, S. Multi-objective planning of distributed energy resources: A review of the state-of-the-art. Renew. Sustain. Energy Rev. 2010, 14, 1353–1366. [CrossRef] Georgilakis, P.S.; Hatziargyriou, N.D. Optimal Distributed Generation Placement in Power Distribution Networks: Models, Methods, and Future Research. IEEE Trans. Power. Syst. 2013, 28, 3420–3428. [CrossRef] Kalambe, S.; Agnihotri, G. Loss minimization techniques used in distribution network: Bibliographical survey. Renew. Sustain. Energy Rev. 2014, 29, 184–200. [CrossRef] Georgilakis, P.S.; Hatziargyriou, N.D. A review of power distribution planning in the modern power systems era: Models, methods, and future research. Electr. Power Syst. Res. 2015, 121, 89–100. [CrossRef] Sultana, U.; Khairuddin, A.B.; Aman, M.M.; Mokhtar, A.S.; Zareen, N. A review of optimum DG placement based on minimization of power losses and voltage stability enhancement of distribution system. Renew. Sustain. Energy Rev. 2016, 63, 363–378. [CrossRef] Paliwal, P.; Patidar, N.P.; Nema, R.K. Planning of grid integrated distributed generators: A review of technology, objectives, and techniques. Renew. Sustain. Energy Rev. 2014, 40, 557–570. [CrossRef] Coster, E.J.; Myrzik, J.M.A.; Kruimer, B.; Kling, W.L. Integration issues of distributed generation in distribution grids. IEEE Proc. 2011, 99, 28–39. [CrossRef] Sajjadi, S.M.; Haghifam, M.R.; Salehi, J. Simultaneous placement of distributed generation and capacitors in distribution networks considering voltage stability index. J. Electr. Power Energy Syst. 2013, 46, 366–375. [CrossRef] Mahela, O.P.; Shaik, A.G. Topological aspects of power quality improvement techniques: A composite review. Renew. Sustain. Energy Rev. 2016, 58, 1129–1142. [CrossRef] Manditereza, P.T.; Bansal, R. Renewable distributed generation: The hidden challenges—A review from the protection perspective. Renew. Sustain. Energy Rev. 2016, 58, 1457–1465. [CrossRef] Lueken, C.; Carvalho, P.; Apt, J. Distribution grid reconfiguration reduces power losses and helps integrate renewables. Energy Policy 2012, 48, 260–273. [CrossRef] Arasteh, H.; Sepasian, M.S.; Vahidinasab, V. An aggregated model for coordinated planning and reconfiguration of electric distribution networks. Energy 2016, 94, 786–798. [CrossRef] Ahmadigorji, M.; Amjady, N. A new evolutionary solution method for dynamic expansion planning of DG-integrated primary distribution networks. Energy Convers. Manag. 2014, 82, 61–70. [CrossRef] Marler, R.T.; Arora, J.S. Survey of multi-objective optimization methods for engineering. Struct. Multidiscip. Optim. 2004, 26, 369–395. [CrossRef] Celli, G.; Ghiani, E.; Mocci, S.; Pilo, F. A multiobjective evolutionary algorithm for the sizing and siting of distributed generation. IEEE Trans. Power Syst. 2005, 20, 750–757. [CrossRef] Caprinelli, G.; Celli, G.; Mocci, S.; Pilo, F.; Russo, A. Optimisation of embedded generation sizing and siting by using a double trade-off method. IET Gener. Transm. Distrib. 2005, 152, 503–513. Haesen, E.; Driesen, J.; Belmans, R. Robust planning methodology for integration of stochastic generators in distribution grids. IET Renew. Power Gener. 2007, 1, 25–32. [CrossRef] Haghifam, M.R.; Falaghi, H.; Malik, O.P. Risk-based distributed generation placement. IET Gener. Transm. Distrib. 2008, 2, 252–260. [CrossRef]

Energies 2017, 10, 208

25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36.

37. 38. 39. 40. 41.

42.

43.

44. 45.

46.

47.

41 of 44

Ochoa, L.F.; Feltrin, A.P.; Harrison, G.P. Evaluating distributed time-varying generation through a multiobjective index. IEEE Trans. Power Deliv. 2008, 23, 1132–1138. [CrossRef] Kim, K.H.; Song, K.B.; Joo, S.K.; Lee, Y.J.; Kim, J.O. Multiobjective distributed generation placement using fuzzy goal programming with genetic algorithm. Eur. Trans. Electr. Power 2008, 18, 217–230. [CrossRef] Ochoa, L.F.; Feltrin, A.P.; Harrison, G.P. Time-series-based maximization of distributed wind power generation integration. IEEE Trans. Energy Convers. 2008, 23, 968–974. [CrossRef] Verma, K.S.; Singh, D. Multiobjective optimization for DG planning with load models. IEEE Trans. Power Syst. 2009, 24, 427–436. Vinothkumar, K.; Selvan, M.P. Fuzzy embedded genetic algorithm method for distributed generation planning. J. Electr. Power Compon. Syst. 2011, 39, 346–366. [CrossRef] Liu, Z.; Wen, F.; Ledwich, G. Optimal siting and sizing of distributed generators in distribution systems considering uncertainties. IEEE Trans. Power Deliv. 2011, 26, 2541–2551. [CrossRef] Moradi, M.H.; Abedini, M. A combination of genetic algorithm and particle swarm optimization for optimal DG location and sizing in distribution systems. Int. J. Electr. Power Energy Syst. 2012, 34, 66–74. [CrossRef] Vinothkumar, K.; Selvan, M.P. Distributed generation planning: A new approach based on goal programming. J. Electr. Power Compon. Syst. 2012, 40, 497–512. [CrossRef] Rotaru, F.; Chicco, G.; Cartina, G. Two-stage distributed generation optimal sizing with clustering-based node selection. Int. J. Electr. Power Energy Syst. 2012, 40, 120–129. [CrossRef] Jin, T.; Tian, Y.; Zhang, C.W.; Coit, D.W. Multicriteria planning for distributed wind generation under strategic maintenance. IEEE Trans. Power Deliv. 2013, 28, 357–367. [CrossRef] Nekooei, K.; Farsangi, M.M.; Nezamabadi-Pour, H.; Lee, K.Y. An improved multi-objective harmony search for optimal placement of DGs in distribution systems. IEEE Trans. Smart Grid 2013, 4, 557–567. [CrossRef] Nayeripour, M.; Moghaddam, E.M.; Aghaei, J.; Vahed, A.A. Multi-objective placement and sizing of DGs in distribution networks ensuring transient stability using hybrid evolutionary algorithm. Renew. Sustain. Energy Rev. 2013, 25, 759–767. [CrossRef] Moradi, M.H.; Tousi, S.M.R.; Abedini, M. Multi-objective PFDE algorithm for solving the optimal siting and sizing problem of multiple DG sources. Int. J. Electr. Power Energy Syst. 2014, 56, 117–126. [CrossRef] Shaaban, M.F.; El-Saadany, E.F. Accommodating high penetrations of PEVs and renewable DG considering uncertainties in distribution systems. IEEE Trans. Power Syst. 2014, 29, 259–270. [CrossRef] Vahidinasab, V. Optimal distributed energy resources planning in a competitive electricity market: Multiobjective optimization and probabilistic design. Renew. Energy 2014, 66, 354–363. [CrossRef] Zakariazadeh, A.R.; Jadid, S.; Siano, P. Stochastic multi-objective operational planning of smart distribution systems considering demand response programs. Electr. Power Syst. Res. 2014, 111, 156–168. [CrossRef] Mena, R.; Hennebel, M.; Li, Y.F.; Ruiz, C.; Zio, E. A risk-based simulation and multi-objective optimization framework for the integration of distributed renewable generation and storage. Renew. Sustain. Energy Rev. 2014, 37, 778–793. [CrossRef] Sultana, S.; Roy, P.K. Multi-objective quasi-oppositional teaching learning-based optimization for optimal location of distributed generator in radial distribution systems. Int. J. Electr. Power Energy Syst. 2014, 63, 534–545. [CrossRef] Ameli, A.; Bahrami, S.; Khazaeli, F.; Haghifam, M.R. A multiobjective particle swarm optimization for sizing and placement of DGs from DG owner’s and distribution company’s viewpoints. IEEE Trans. Power Deliv. 2014, 29, 1831–1840. [CrossRef] Cheng, S.; Chen, M.Y.; Fleming, P.J. Improved multi-objective particle swarm optimization with preference strategy for optimal DG integration into the distribution system. Neurocomputing 2015, 148, 23–29. [CrossRef] Nick, M.; Cherkaoui, R.; Paolone, M. Optimal Allocation of Dispersed Energy Storage Systems in Active Distribution Networks for Energy Balance and Grid Support. IEEE Trans. Power Syst. 2014, 29, 2300–2310. [CrossRef] Abdelaziz, A.Y.; Hegazy, Y.G.; El-Khattam, W.; Othman, M.M. A multi-objective optimization for sizing and placement of voltage-controlled distributed generation using supervised big bang–big crunch method. J. Electr. Power Compon. Syst. 2015, 43, 105–117. [CrossRef] Mohandas, N.; Balamurugan, R.; Lakshminarasimman, L. Optimal location and sizing of real power DG units to improve the voltage stability in the distribution system using ABC algorithm united with chaos. Int. J. Electr. Power Energy Syst. 2015, 66, 41–52. [CrossRef]

Energies 2017, 10, 208

48. 49. 50. 51.

52.

53. 54. 55. 56. 57. 58. 59.

60. 61. 62.

63.

64.

65.

66.

67.

68.

42 of 44

Kayal, P.; Chanda, C.K. A multi-objective approach to integrate solar and wind energy sources with electrical distribution network. Sol. Energy 2015, 112, 397–410. [CrossRef] Darfoun, M.A.; El-Hawary, M.E. Multi-objective optimization approach for optimal distributed generation sizing and placement. J. Electr. Power Compon. Syst. 2015, 43, 828–836. [CrossRef] Sheng, W.; Liu, K.Y.; Meng, X.; Li, Y. Optimal placement and sizing of distributed generation via an improved nondominated sorting genetic algorithm II. IEEE Trans. Power Deliv. 2015, 30, 569–578. [CrossRef] Yammani, C.; Maheswarapu, S.; Matam, S.K. A multi-objective Shuffled Bat algorithm for optimal placement and sizing of multi distributed generations with different load models. Int. J. Electr. Power Energy Syst. 2016, 79, 120–131. [CrossRef] Kayal, P.; Chanda, C.K. Strategic approach for reinforcement of intermittent renewable energy sources and capacitor bank for sustainable electric power distribution system. Int. J. Electr. Power Energy Syst. 2016, 83, 335–351. [CrossRef] Pires, D.F.; Martins, A.G.; Antunes, C.H. A multiobjective model for VAR planning in radial distribution networks based on tabu search. IEEE Trans. Power Syst. 2005, 20, 1089–1094. [CrossRef] Das, D. Optimal placement of capacitors in radial distribution system using a Fuzzy-GA method. Int. J. Electr. Power Energy Syst. 2008, 30, 361–367. [CrossRef] Huang, T.L.; Hsiao, Y.T.; Chang, C.H.; Jiang, J.A. Optimal placement of capacitors in distribution systems using an immune multi-objective algorithm. Int. J. Electr. Power Energy Syst. 2008, 30, 184–192. [CrossRef] Kuo, C.C. Capacitor placement and scheduling using interactive bi-objective programming with valuable trade off approach. Energy Conserv. Manag. 2009, 50, 991–1003. [CrossRef] Souza, B.A.; Almeida, A.M.F. Multiobjective optimization and fuzzy logic applied to planning of the volt/var problem in distributions systems. IEEE Trans. Power Syst. 2010, 25, 1271–1281. [CrossRef] Pires, D.F.; Antunes, C.H.; Martins, A.G. NSGA-II with local search for a multi-objective reactive power compensation problem. Int. J. Electr. Power Energy Syst. 2012, 43, 313–324. [CrossRef] Franco, J.F.; Rider, M.J.; Lavorato, M.; Romero, R. A mixed-integer LP model for the optimal allocation of voltage regulators and capacitors in radial distribution systems. Int. J. Electr. Power Energy Syst. 2013, 48, 123–130. [CrossRef] Fard, A.K.; Niknam, T. Considering uncertainty in the multi-objective stochastic capacitor allocation problem using a novel self-adaptive modification approach. Electr. Power Syst. Res. 2013, 103, 16–27. [CrossRef] El-Fergany, A.A.; Abdelaziz, A.Y. Efficient heuristic-based approach for multi-objective capacitor allocation in radial distribution networks. IET Gener. Transm. Distrib. 2013, 8, 70–80. [CrossRef] El-Fergany, A.A.; Abdelaziz, A.Y. Capacitor placement for net saving maximization and system stability enhancement in distribution networks using artificial bee colony-based approach. Int. J. Electr. Power Energy Syst. 2013, 54, 235–243. [CrossRef] Fard, A.K.; Samet, H. Multi-objective performance management of the capacitor allocation problem in distributed system based on adaptive modified honey bee mating optimization evolutionary algorithm. J. Electr. Power Compon. Syst. 2013, 41, 1223–1247. [CrossRef] Nasab, M.E.; Maleksaeedi, I.; Mohammadi, M.; Ghadimi, N. A New Multiobjective Allocator of Capacitor Banks and Distributed Generations using a New Investigated Differential Evolution. Complexity 2014, 19, 40–54. [CrossRef] Zeinalzadeh, A.; Mohammadi, Y.; Moradi, M.H. Optimal multi objective placement and sizing of multiple DGs and shunt capacitor banks simultaneously considering load uncertainty via MOPSO approach. Int. J. Electr. Power. Energy Syst. 2015, 67, 336–349. [CrossRef] Gampa, S.R.; Das, D. Optimum placement of shunt capacitors in a radial distribution system for substation power factor improvement using fuzzy GA method. Int. J. Electr. Power Energy Syst. 2016, 77, 314–326. [CrossRef] Su, X.; Masoum, M.A.S.; Wolfs, P.J. PSO and improved BSFS based sequential comprehensive placement and real-time multi-objective control of delta-connected switched capacitors in unbalanced radial MV distribution networks. IEEE Trans. Power Syst. 2016, 31, 427–436. [CrossRef] Mohammadi, M.; Rozbahani, A.M.; Montazeri, M. Multi criteria simultaneous planning of passive filters and distributed generation simultaneously in distribution system considering nonlinear loads with adaptive bacterial foraging optimization approach. Int. J. Electr. Power Energy Syst. 2016, 79, 253–262. [CrossRef]

Energies 2017, 10, 208

69. 70.

71. 72. 73.

74.

75.

76. 77.

78.

79. 80. 81.

82. 83. 84. 85. 86. 87. 88. 89. 90.

43 of 44

Kumar, D.; Samantaray, S.R. Design of advanced power distribution networks including DGs and D-FACT devices. INAEL Lett. 2016, 1, 15–20. [CrossRef] Kumar, D.; Samantaray, S.R. Implementation of multi-objective seeker-optimization-algorithm for optimal planning of primary distribution systems including DSTATCOM. Int. J. Electr. Power Energy Syst. 2016, 77, 439–449. [CrossRef] Mendoza, F.; Agustín, J.L.B.; Navarro, J.A.D. NSGA and SPEA Applied to Multiobjective Design of Power Distribution Systems. IEEE Trans. Power Syst. 2006, 21, 427–436. [CrossRef] Nazar, M.S.; Haghifam, M.R. Multiobjective electric distribution system expansion planning using hybrid energy hub concept. Electr. Power Syst. Res. 2009, 79, 899–911. [CrossRef] Falaghi, E.; Haghifam, M.R.; Singh, C. Ant Colony Optimization-Based Method for Placement of Sectionalizing Switches in Distribution Networks Using a Fuzzy Multiobjective Approach. IEEE Trans. Power Syst. 2009, 24, 268–276. [CrossRef] Haesen, E.; A-Rodriguez, A.D.; Driesen, J.; Belmans, R.; Ault, G. Opportunities for active DER management in deferral of distribution system reinforcements. In Proceedings of the Power Systems Conference and Exposition, PSCE’09, IEEE/PES, Seattle, WA, USA, 15–18 March 2009. Tippachon, W.; Rerkpreedapong, D. Multiobjective optimal placement of switches and protective devices in electric power distribution systems using ant colony optimization. Electr. Power Syst. Res. 2009, 79, 1171–1178. [CrossRef] Ganguly, S.; Sahoo, N.C.; Das, D. A novel multi-objective PSO for electrical distribution system planning incorporating distributed generation. Energy Syst. 2010, 1, 291–337. [CrossRef] Cossi, A.M.; Da-Silva, L.G.W.; Lázaro, R.A.R.; Mantovani, J.R.S. Primary power distribution systems planning taking into account reliability, operation and expansion costs. IET Gener. Transm. Distrib. 2011, 6, 274–284. [CrossRef] Sardou, I.G.; Banejad, M.; Hooshmand, R.; Dastfan, A. Modified shuffled frog leaping algorithm for optimal switch placement in distribution automation system using a multi-objective fuzzy approach. IET Gener. Transm. Distrib. 2011, 6, 493–502. [CrossRef] Borges, C.L.T.; Martins, V.F. Multistage expansion planning for active distribution networks under demand and Distributed Generation uncertainties. Int. J. Electr. Power Energy Syst. 2012, 36, 107–116. [CrossRef] Ganguly, S.; Sahoo, N.C.; Das, D. Multi-objective planning of electrical distribution systems incorporating sectionalizing switches and tie-lines using particle swarm optimization. Swarm Evol. Comput. 2012, 3, 15–32. Nazar, M.S.; Haghifam, M.R.; Nazar, M. A scenario driven multiobjective primary–secondary distribution system expansion planning algorithm in the presence of wholesale–retail market. Int. J. Electr. Power Energy Syst. 2012, 40, 29–45. [CrossRef] Gitizadeh, M.; Vahed, A.A.; Aghaei, J. Multistage distribution system expansion planning considering distributed generation using hybrid evolutionary algorithms. Appl. Energy 2012, 101, 655–666. [CrossRef] Ganguly, S.; Sahoo, N.C.; Das, D. Multi-objective planning of electrical distribution systems using dynamic programming. Int. J. Electr. Power Energy Syst. 2013, 46, 65–78. [CrossRef] Pereira, B.R., Jr.; Cossi, A.M.; Contreras, J.; Mantovani, J.R.S. Multiobjective multistage distribution system planning using Tabu search. IET Gener. Transm. Distrib. 2013, 8, 35–45. [CrossRef] Kumar, D.; Samantaray, S.R. Design of an advanced electric power distribution systems using seeker optimization algorithm. Int. J. Electr. Power Energy Syst. 2014, 63, 196–217. [CrossRef] Kumar, D.; Samantaray, S.R.; Kamwa, I. MOSOA-based multiobjective design of power distribution systems. IEEE Syst. J. 2015. [CrossRef] Pombo, A.V.; Pina, J.M.; Pires, V.F. A multiobjective placement of switching devices in distribution networks incorporating distributed energy resources. Electr. Power Syst. Res. 2016, 130, 34–45. [CrossRef] Elmitwally, A.; Gouda, E.; Eladawy, S. Restoring recloser-fuse coordination by optimal fault current limiters planning in DG-integrated distribution systems. Int. J. Electr. Power Energy Syst. 2016, 77, 9–18. [CrossRef] Castro, J.R.; Saad, M.; Lefebvre, S.; Asber, D.; Lenoir, L. Optimal voltage control in distribution network in the presence of DGs. Int. J. Electr. Power Energy Syst. 2016, 78, 239–247. [CrossRef] Mendoza, J.E.; Lopez, M.E.; Coello, C.A.C.; Lopez, E.A. Microgenetic multiobjective reconfiguration algorithm considering power losses and reliability indices for medium voltage distribution network. IET Gener. Transm. Distrib. 2009, 3, 825–840. [CrossRef]

Energies 2017, 10, 208

44 of 44

91.

Bernardon, D.P.; Garcia, V.J.; Ferreira, A.S.Q.; Canha, L.N. Electric distribution network reconfiguration based on a fuzzy multi-criteria decision making algorithm. Int. J. Electr. Power Syst. Res. 2009, 79, 1400–1407. [CrossRef] 92. Yin, S.A.; Lu, C.N. Distribution Feeder Scheduling Considering Variable Load Profile and Outage Costs. IEEE Trans. Power Syst. 2009, 24, 652–660. [CrossRef] 93. Amanulla, B.; Chakrabarti, S.; Singh, S.N. Reconfiguration of Power Distribution Systems Considering Reliability and Power Loss. IEEE Trans. Power Deliv. 2012, 27, 919–926. [CrossRef] 94. Chen, T.H.; Lin, E.H.; Yang, N.C.; Hsieh, T.Y. Multi-objective optimization for upgrading primary feeders with distributed generators from normally closed loop to mesh arrangement. Int. J. Electr. Power Energy Syst. 2012, 45, 413–419. [CrossRef] 95. Tomoiaga, B.; Chindris, M.; Sumper, A.; Andreu, A.S.; Robles, R.V. Pareto optimal reconfiguration of power distribution systems using a genetic algorithm based on NSGA-II. Energies 2013, 6, 1439–1455. [CrossRef] 96. Zidan, A.; Shaaban, M.F.; El-Saadany, E.F. Long-term multi-objective distribution network planning by DG allocation and feeders’ reconfiguration. Electr. Power Syst. Res. 2013, 105, 95–104. [CrossRef] 97. Vitorino, R.M.; Jorge, H.M.; Neves, L.P. Multi-objective optimization using NSGA-II for power distribution system reconfiguration. Int. J. Wiley Trans. Electr. Energy Syst. 2015, 25, 38–53. [CrossRef] 98. Fard, A.K.; Niknam, T. Multi-objective stochastic distribution feeder reconfiguration from the reliability point of view. Energy 2014, 64, 342–354. [CrossRef] 99. Tolabi, H.B.; Ali, M.H.; Rizwan, M. Simultaneous reconfiguration, optimal placement of DSTATCOM, and photovoltaic array in a distribution system based on fuzzy-aco approach. IEEE Trans. Sustain. Energy 2015, 6, 210–218. [CrossRef] 100. Moghaddam, E.M.; Narimani, M.R.; Khooban, M.H.; Azizivahed, A.; sharifi, M.J. Multi-objective distribution feeder reconfiguration to improve transient stability, and minimize power loss and operation cost using an enhanced evolutionary algorithm at the presence of distributed generations. Int. J. Electr. Power Energy Syst. 2016, 76, 35–43. [CrossRef] 101. Castillo, A.; Neill, R.P.O. Computational Performance of Solution Techniques Applied to the ACOPF. January 2013. Available online: http://www.ferc.gov/industries/electric/indus-act/market-planning/opf-papers/ acopf-5-computational-testing.pdf (accessed on 27 December 2016). 102. Schachter, J.A.; Mancarella, P. A critical review of real options thinking for valuing investment flexibility in smart grids and low carbon energy systems. Renew. Sustain. Energy Rev. 2016, 56, 261–271. [CrossRef] © 2017 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

Multi-Objective Planning Techniques in Distribution Networks - MDPI

Multi-Objective Planning Techniques in Distribution Networks - MDPI

Suggest Documents

Multiobjective Optimization of Water Distribution Networks ... - MDPI

Multiobjective Optimization of Distribution Networks Using ... - CiteSeerX

Multiobjective Optimization of Distribution Networks Using ... - CiteSeerX

Smart Distribution Networks - MDPI

Multiobjective optimization in process planning

AI Techniques for Planning Telecommunication Networks - CiteSeerX

Optimum Planning of Primary-Secondary Distribution Networks ...

MULTIOBJECTIVE OPTIMIZATION OF WATER DISTRIBUTION ...

Multiobjective flow control in telecommunication networks - CiteSeerX

Short-Circuit Calculation in Distribution Networks with ... - MDPI

an introduction to multiobjective optimization techniques

Multiobjective Control: Linear Matrix Inequality Techniques and

Multiobjective inverse planning for intensity modulated ... - CiteSeerX

Multiobjective Inverse Planning for External Beam Radiotherapy

Multiobjective Heuristic State-Space Planning - CiteSeerX

Electric Distribution Network Multiobjective Design ... - PPGEE - UFMG

Electric Distribution Network Multiobjective Design ... - PPGEE - UFMG

Video Distribution Techniques Over WiMAX Networks ... - IEEE Xplore

Cross Entropy multiobjective optimization for water distribution ...

Building Multiobjective Resilient Networks - Semantic Scholar

Multiobjective design of survivable IP networks

211. Planning and operation of active radial distribution networks for ...

on development planning of electricity distribution networks - DiVA

Planning of Distribution Networks for Medium Voltage and Low ...