Design and Optimization of Hybrid Micro-Grid System - ScienceDirect

Available online at www.sciencedirect.com

ScienceDirect

Availableonline onlineatatwww.sciencedirect.com www.sciencedirect.com Available Energy Procedia 00 (2017) 000–000

ScienceDirect ScienceDirect

www.elsevier.com/locate/procedia

Energy (2017)000–000 95–103 Energy Procedia 117 00 (2017) www.elsevier.com/locate/procedia

1st International Conference on Power Engineering, Computing and CONtrol, PECCON-2017, 24 March 2017, VIT University, Chennai Campus

Design Optimization of Hybrid Micro-Grid System Theand 15th International Symposium on District Heating and Cooling Jayachandran. Ma,*, Ravi. Gb demand-outdoor Assessing the feasibility of using the heat Department of function Electrical and Electronics Pondicherrydistrict engineering college, Puducherry, 605014, India temperature for aEngineering, long-term heat demand forecast a,b

Abstract

I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc

a

IN+microgrid Center forsystems Innovation, Technology and Policy Researchparallel - Instituto Superior Técnico, Av. Rovisco Paiswith 1, 1049-001 Lisbon, Portugal Hybrid (HMGS) comprise of several connected distributed resources electronically controlled b Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France strategies, which are capable to operate in both islanded and grid connected mode. HMGS based on renewable energy sources c Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France (RES) is the cost-effective option for solving the power supply problem in remote areas, which are located far from grids. In this paper, the wind and solar meteorological data for Sundarban (India) station are used to design Islanded HMGS for providing necessary electricity. Cost effectiveness and system reliability are major factors considered for designing HMGS to achieve better power management scheme. Particle Swarm Optimization (PSO) scheme is applied to identify the sizing of wind turbines (WT), Abstract photovoltaic (PV) module, battery energy storage system (BESS) and diesel generator, and find the optimal configuration of HMGS designare andcommonly optimal operation systemashas developed and validated MATLAB Districtsystem. heating The networks addressed ofin HMGS the literature onebeen of the most effective solutionsthrough for decreasing the software. greenhouse gas emissions from the building sector. These systems require high investments which are returned through the heat ©sales. 2017 The Authors. Published by Elsevier Ltd. to the changed climate conditions renovation policies, heat demand in the future could decrease, © 2017 Due The Authors. Published by Elsevier Ltd. and buildingthe 1st International Conference on Power Engineering, Peer-review under responsibility of the scientific committee of prolonging the investment return period. Peer-review under responsibility of the scientific committee of the 1st International Conference on Power Engineering, Computing and CONtrol. The main scope of this paper is to assess the feasibility of using the heat demand – outdoor temperature function for heat demand Computing and CONtrol. forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 Keywords: HMGS; PSO algorithm; Power Management Scheme. buildings that vary in both construction period and typology. Three weather scenarios (low, medium, high) and three district renovation scenarios were developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were compared with results from a dynamic heat demand model, previously developed and validated by the authors. 1.The Introduction results showed that when only weather change is considered, the margin of error could be acceptable for some applications (the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation In India,theover people(depending still don’tonhave electricity. There are scenarios twenty-eight thousand villages scenarios, errorthree valuehundred increasedmillion up to 59.5% the weather and renovation combination considered). not electricity [1]. Sufficient systems regionally Wind and PV energy Theaccessing value of slope coefficient increased stand-alone on average within the for range of 3.8% accessible up to 8% per decade, thatrenewable corresponds to the decreasehave in thebecome number recommended of heating hoursoption of 22-139h during areas the heating season and (depending on the combination weather and sources for remote [2]. BESS diesel generators are usedof for backup renovation scenarios considered). On the other intercept increased for energy 7.8-12.7% decade (depending on the system to overwhelmed the stochasticity of hand, wind function and irregular nature of solar [3].perON/OFF and continuous coupledstrategies scenarios).are The values suggested could be to modify the function for the scenarios considered, control involving the operation of used the diesel generator withinparameters HMGS [4,5]. PSO algorithm is used and to improve the accuracy of heat demand estimations.

© 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and * Corresponding author. Tel.: +919345932131 Cooling. E-mail address: [email protected]

Keywords: Heat demand; Forecast; Climate change 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the scientific committee of the 1st International Conference on Power Engineering, Computing and CONtrol.

1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling. 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the scientific committee of the 1st International Conference on Power Engineering, Computing and CONtrol. 10.1016/j.egypro.2017.05.111

Jayachandran M et al. / Energy Procedia 117 (2017) 95–103 M.Jayachandran / Energy Procedia 00 (2017) 000–000

96 2

minimize the annual cost of the HMGS and it is directly depending on its reliability [6]. PSO will be recognized as a simple concept and shorter computation time when compared to GA [7,8]. PSO is the most promising and powerful strategies to find the best configuration in HMGS [9]. The main goal of this paper, is to choose a cost effective and reliable HMGS by using PSO method, owing to find high Renewable Factor (RF), the lowest Price of Electricity (POE) and Loss of Power Supply Probability (LPSP). Meteorological data for Sundarban station is used to design HMGS which is situated in North east of India and lies on geographical coordinates of latitude 21.95° N and longitude 89.1833° E. The paper is structured as follows. Section 1 provides an introduction of Islanded HMGS. Section 2 describes the mathematical model of hybrid microgrid system. Section 3 and 4 briefly introduces power management scheme and particle swarm optimization algorithm respectively. Design considerations of islanded HMGS explain in Section 5. Section 6 and 7 presents economic analysis and simulation results. 2. Hybrid microgrid system HMGS is designed as low voltage distribution network to supply 220V, 50 Hz, 1Φ AC system and detailed model depicted in Fig.1 (a). Load profile determination is the primary step for designing HMGS. In India, most of the loads are lights, fans, Television, Mixer, Laptop, Mobile phone and others [10]. The average power requirement to meet the load demand is approximately 6kWh per day and peak load is approximately 1.5kW per house as tabulated in Table. 1. The load profile of 15 houses in Sundarban site as shown in Fig.1 (b) with a peak load of 8.75 kW and the hourly load profile as depicted in Fig. 2 (a). 2.1. Mathematical model for HMGS Fig. 2(b) shows the hourly solar incident of the horizontal for Sundarban site. It is clear from the figure that the average horizontal solar radiation is 239 W/m2 and the peak solar irradiance exceeds 1000 W/m2 [11]. The input data of PV generator could be the hourly solar radiation on the horizontal surface. The solar irradiation of PV can be calculated by [12], ��

�

��

(1)

�1 � �� − ��

(2)

�� = �� ∗ η��

(3)

�� = �� 0.0�� × �)

Where, � and �� are solar radiation (W/m2) and solar irradiation at reference conditions (�� = 1000 �� )) respectively. �� is the temperature coefficient of maximum power ( �� = −3.7 × 10�� (1/oC)) for mono and polycrystalline silicon. �� , is the PV cell temperature at reference condition (�� = 25oC). �� , is cell temperature and �� is the ambient temperature as shown in Fig. 2(c). �� is the rated power of the PV panel at reference conditions and �� is hourly output power of PV cell as shown in Fig. 2(d). The bi-directional inverter offers a path from the DC bus to the AC load and acts like a rectifier, which converts AC supply to DC voltage to charge the battery banks. The inverter has been selected 20% more than the rated power of AC loads. a

b

Average daily load power

10 8 6 4 2 0

0

5

10

Hour

15

Fig. 1. (a) Hybrid Microgrid system; (b) Load profile consumption per day for 15 houses

20

25


97 3

Table 1. Peak Load and Power Requirement per Day for Rural Homes Appliances

Power

Quantity

in Watts Fluorescent Light Lamps (Bulb) Lamps (CFL) Ceiling Fan Table Fan Laptop/PC Television Mixer Grinder Mobile phone Mosquito Repellent Total

40 60 15 75 75 60 100 500 300 7 9

5 1 2 2 1 2 1 1 1 2 1

No of hours

Electric load

Power required in

used per day

in Watts

Watts hour/ day

6 3 8 9 5 3 11 1 1 3 12

200 60 30 150 75 120 100 500 300 14 9 1558

1200 180 240 1350 375 360 1100 500 300 42 108 5755

a

b Load Profile Solar Radiation (W/m2)

10 8 6 4 2 0

1000

2000

c

3000

4000

5000

Hour of the year

6000

7000

8000

d

Temperature

Temperature (0 C)

40

Power (Watts)

35 30 25 20

0

1000

2000

3000

e

5000

6000

7000

8000

f

Wind Speed

10

Wind Speed (m/s)

4000

hour of the year

8 6 4 2 0

0

1000

2000

3000

4000

5000

Hour of the year

6000

7000

8000

Wind Generator Power (KW)

0

Fig. 2. (a) Hourly power consumption by load; (b) Hourly solar radiation; (c) Hourly ambient temperature; (d) Hourly PV power; (e) Hourly wind speed; (f) Hourly WT power

RES such as PV generators may not be able to compensate the power required by load during night times. Batteries are connected between the DC bus and the load of HMGS. The storage capacity �� (Watt-h) can be obtained from the following relation [16], �� =

��

��

(4)

Where, �� is the average load in a day for the whole village (�� = peak residential load*number of houses in a village). How long system to run without recharging is known as autonomy days (AD is typically 3-5 days). How


98 4

deeply you are willing to discharge the battery pack is called depth of discharge (DOD is 50-80% common for deep cycle batteries). The inverter and battery efficiencies are �� = 92% and �� = 85% respectively. Energy is available in the wind for each hour is calculated by using wind speed [11]. The wind data are available from the National Solar Radiation Database for Sundarban site. The average wind speed of at a height of 40m is 4.14 m/s and peak wind speed exceeds 8.37 m/s as shown in Fig. 2(e). In this study, 5 kW rated capacity of wind turbine is considered. The wind speed at hub height is considered to compute output power of WT, [13], �

��

= �

��

�

(5)

�

�

�� = ��

��

� ��

�� = ��

�

� � ��

��

� ��

� � ��

�

��

(6)

��

(7)

��

Where, � is the wind speed at desired height (ℎ�� = 4�� in current time step and �� is the wind speed (m/s) at reference height (ℎ�� = 4��); � is the ground surface friction coefficient (� = ��) and � = ��25 for heavily forested landscape. �� is rated power of wind turbine (�� = 5��). �� is cut-in wind speed (�� = 2�5 ��), �� is rated wind speed (�� = 9�5 ��), �� is cut-out wind speed (�� = 4� ��). Even when the wind speed is very low, the wind turbines can operate efficiently. The blade diameter is 6.4m of wind turbine. The swept area is the product of pi and square of radius of the blade and the efficiency of the wind turbine (η wind) is 95% [14-15]. Hourly output power of wind turbine as depicted in Fig. 2(f). During peak demand, diesel generator acts as a secondary energy storage. The diesel generator fuel consumption FDG is obtained by [18-19], �� = ��

(8)

Where, �� (kW) is the nominal power. The average load power demand for every hour is 3.54 kW, hence, the diesel generator rated power is P�� = 4kW is considered. �� (kW) is the diesel generator output power. �� = ��24� ��ℎ and �� = ��8�45 ��ℎ. �� and �� are diesel consumption coefficients (load/kWh). 3. Power management scheme

In this methodology, the wind, PV, BESS and a diesel generator are utilized to keep a continuous power to meet the power required by load in HMGS. Case1: Renewable energy source has the highest priority to supply the load demand. The excess energy is utilized to charge the BESS. Case 2: PV & wind energy is not sufficient to meet the power required by load, hence BESS as well as renewable energy sources are turned on. Case 3: BESS, PV and wind energy are not enough to meet the power required by load, then the diesel generator has turned on to supply power to the load and charge the BESS. Case 4: PV and/or wind energy is more than the power required by load and BESS; the surplus energy is distributed to dump loads [12]. 4. Particle Swarm Optimization HMGS optimization is considered as a multi-objective problem. Using linear scalarization method, multi-objective function is transformed in to a single-objective function and objectives could be either linear function or constraints [17]. It can be defined by objective (fitness) function and constraints defined as,

Jayachandran M et al.Procedia / Energy00 Procedia (2017) 95–103 M.Jayachandran / Energy (2017) 117 000–000

�� = �� ∑��

min �� (x) ≥ 0

�� (�)

��

599

� �� ≥ 0 and ∑�� = 1

(9) (10)

Where, k is the number of objectives, f is the fitness function, �� is the upper bound of the ith cost function and x is the vector decision variables. In this paper, the tness function is dened as high RF and lowest POE and LPSP. The procedure for PSO as follows, 1) Define the objective function and set the constraints which are renewable factor [> 0.01], PV [5,50], autonomy days, [0,5], wind turbines, [0,5] and number of houses, [1,15]. 2) Initialize position, velocity of a particle, Population Array, Global Best and Population members. 3) Randomly select velocity and position of particles, generate the initial population and nd the optimum tness value from whole swarm. 4) Set the personal best. The smallest POE and LPSP have selected as a global best and update iteration. 5) Update personal best position and global best position. Apply stopping condition. a

Table 2. (a) PSO Results for sizing of HMGS; (b) Inputs for the simulation program b Price of Number Autonomy PV Number of Item Electricity of Power Wind Turbines Days Wind turbine cost (Rs/kW) (Rs/kWh) PV cost (Rs/kW) Houses (kW) (Pr wind = 5kW) PV installation cost (Rs/kW) 15 1 4 4 74.74 Bi-directional inverter cost (Rs) 15 2 5 2 62.37 PV regulator cost (Rs) 15 3 6 1 62.27 PV Installation cost (RS) 15 3 10 0 62.34 Wind regulator cost (Rs) 15 4 7 0 63.36 Wind Generator Installation cost (RS) 15 5 7 0 73.05 Battery cost (Rs/kWh) Diesel generator cost (Rs/kVA) Operational and Management cost (Rs)

Value 80,000 65,000 10,000 7,600 4,000 6,000 4,000 6,000 12,500 32,200 1,400

The PSO scheme is applied to attain the best configuration and sizing the components of HMGS. The results show that PSO provides the different configuration of optimum solar, wind and battery ratings and the best solution for 15 houses is displayed in Table 2(a). Swarms of motion in 20 iterations have considered. Optimum value for LPSP and POE as shown in Fig. 3 (a). The fitness value of each particle has evaluated. Then the convergence curve for 20 iterations depicted in Fig. 3 (b). It is evident from the convergence curve that the POE has decreased and maintained constant at 6th iteration with the lowest cost of 62.27 Rs/kWh for 6kW PV power, 5kW wind turbine and 3autonomy days. 5. Islanded HMGS design A 48V DC bus is considered where the BESS is connected. The output of charge controller and bidirectional inverter should be at this level of voltage. Full sun hours (kWh/m2/day) are termed as insolation. It varies from 2-6 depending on the time of year and part of the country. It is thought that full sun hours occur from 10am to 3pm (5 hours) a day. Polycrystalline PV panels of 24V, 250W peak power with efficiency (�� ) of 80% are selected. From the simulation results, the optimized value of PV power is 6kW. The number of PV panels is attained as NPV = 6kW/250W = 24 PV modules. In this case, 12 parallel strings, each string has two modules has coupled in series to reach 48V. The orientation of PV panels should be positioned in a north-south direction and north side of the panel is elevated at 21.950 which are equal to the latitude of Sundarban region. The rated power of PV charge controller 6kW has chosen. The rated power of wind turbine has selected depending on wind speed in Sundarban region. Hence, 1 phase 220 V, 50Hz output voltage with the rated power of 5kW has chosen. It has 3 blades with 6.4 m diameters. The rated power of rectifier and wind charge controller of 5 kW can be supported to have 48V DC rated voltage as its output. For sizing of Bidirectional inverter, peak load demand which is 8.75 kW should be reckoned. The rated power of the bi-directional inverter should be 20% more than the peak load, hence, 48V, 10kW power rating of a bidirectional


100 6

inverter with efficiency (ηinv) of 92% has been chosen to design the hybrid microgrid system. Four 12V lead acid batteries are connected in series to attain 48V at the DC bus. Capacity of battery in Watt-hour is obtained approximately 43kWh by using equation 4. Ampere-hour capacity is required for sizing of batteries for the period of 3 autonomy days is �� = 43kWh/12V = 3583Ah. Lead acid batteries with 100Ah have selected number of batteries b

a

Price of Electricity vs LPSP

Best Cost

0.3 0.2 0.1 0 40

60

80

100

Price of Electricity (Rs/KWh)

120

140

Fig. 3. (a) Optimum value for LPSP and Price of Electricity; (b) Convergence curve of Price of Electricity for different iterations

required to cover the load demand is �� = 3583/100 = 36 batteries. In this case, 9 parallel strings are connected and each string 4 batteries are connected in series. Average load demand for Sundarban site is 3.8kW is reckoned for sizing of the diesel generator. Hence, the 4kW diesel generator is required for this place to satisfy the load requirements. 6. Economic analysis The hybrid microgrid system has designed for 15 houses in the rural area. The constant price per unit is calculated by well-known parameter called price of electricity, which is calculated by, ��

��

×

�(��)�

(��)� ��

(11)

Where, n is the system life period and�� is the interest rate [16]. POE production in Rs/kWh in HMGS includes initial capital costs of PV panels with inverters, wind turbines with regulators, BESS and Diesel generator, operation and maintenance (O&M) cost, and replacement cost is tabulated in Table. 2(b). The probability of power supply failure is defined by [4], ��

��∑�� (��)� ��(��) ]�� ] ∑ ��

(12)

Reliability is evaluated in the worst condition P(t)Load > P(t)generator

��

∑�� (��) �� ] ∑ ��

(13)

A boundary between diesel energy to renewable energy is defined as Renewable factor (RF) which is described by [15], ��(%) � �� ∑

∑ ��

�� ∑ ��

� × ��

(14)


101 7

Table 3. (a) Results for the simulation program; (b) Price of Electricity, Renewable Factor and LPSP for different values of Ads b a Parameter AD=1 AD=2 AD=3 AD=3 AD POE Renewable PV power (kW) 4 5 10 6 (Rs/kWh) Factor (%) WT power (kW) 5 5 5 5 Number of WTs 4 2 0 1 0 42.55 43.06 Diesel generator power(kW) 4 4 4 4 0.5 43.71 63.71 PV contribution (%) 27 38 61 46.4 1 44.98 65.97 WT contribution (%) 44 25 0 12.2 1.5 50.33 69.40 BESS contribution (%) 18 26 28 30.4 2 55.67 73.81 Diesel generator contribution (%) 11 11 11 11 2.5 56.93 76.84 Dump energy (kW) 22.2 11.72 20.59 77.19 3 62.27 81.33 Reliability 2.51 4.73 6.75 6.62 3.5 67.60 86.39 LPSP 0.09 0.08 0.05 0.05 4 72.95 89.50 RF (%) 84.8 80.59 81.31 81.33 4.5 78.31 90.40 POE (Rs/kWh) 74.74 62.37 62.34 62.27 5 83.67 90.64 Number of Iterations & Particles 20 20 20 20

LPSP 0.13 0.16 0.15 0.13 0.12 0.07 0.05 0.05 0.03 0.03 0.02

7. Simulation Results Simulation results for different configuration and device sizing of HMGS are obtained for 15 houses for different ADs are tabulated in Table. 3(a). Results for POE, RF and LPSP for different values of ADs are presented in Table 3(b). The results provide evidence that the POE and RF are increased as the number of autonomy days and the higher POE as well as lower LPSP. Thus, the lowest POE for Sundarban site is 42.55 Rs/kWh at 27% PV contribution. It is important to note from Table 3(a) that the lowest POE is Rs.62.27/kW with a high renewable factor of 81.33% is achieved in Sundarban region. In Fig. 4, the simulation through the year of the case of 46.4% PV, 12.2% Wind and 30.4% battery contribution for 3-ADs can be observed. The energy produced by the PV is more than the power required by load during June-September. Hence, diesel generator has not contributed as illustrated in Fig. 4(a). This excess energy is utilized to charge the battery and distributed to the dump load as shown in Fig. 4(b). It is observed that high quantities of dump energy are stored in the battery is 77.19kW. It can be utilized for street lighting, heating, water pumping and refrigeration. As per our assumption, the depth of discharge is 80%. It can be observed from Fig. 4(c), the state of charge (SOC) of battery is not less than 20%. It is important to note from Table 3(a) that the lowest price of electricity is Rs.62.27/kW with the high renewable factor.

3 2 1 0

0

1000

2000

3000

4000

5000

Hour of the year

6000

7000

8000

8 6 4 2 0 0

d

c Battery Energy Battery SOC (KWh)

40 30 20 10 0

0

1000

2000

3000

4000

5000

Hour of the year

e

6000

Dump Energy

b 10 Dump Energy (KWh)

a

Diesel Generator Energy

4

7000

8000

1000

2000

3000

4000

5000

Hour of the year

6000

7000

8000


102 8

Fig. 4. (a) Diesel Generator Output Energy; (b) Battery Dump Energy during the year; (c) Battery Energy during the year; (d) Sensitivity Analysis of PV, WT and Diesel Energy; (e) Energy Contribution for PV, WT, BESS and diesel generator in HMGS System.

Sensitivity analysis of HMGS system can be observed in Fig. 4(d) that the POE is more affected by the power of wind turbines. However, the POE will be decreased by increasing the number of PV panels. The percentage of PV, WT, BESS and diesel energy over a year and the contribution of Solar PV energy is higher than the wind, battery and diesel energy is depicted pie chart in Fig. 4(e). 8. Conclusion In this paper, Hybrid microgrid system (HMGS) has been designed and investigated in islanded mode. Comprehensive analysis on cost optimization, energy flow management, and device sizing of HMGS has been reviewed in Sundarban region. By applying PSO technique, sizing of the system components and the best configuration of the hybrid system have been obtained. Reliability has been evaluated in the worst condition and sensitivity analysis has been conducted to validate the results. It is observed from the results that HMGS basically work on wind and solar energy due to a high potential of renewable energy source. Hence, using renewable energy resource in Sundarban region can be viewed as the best solution to increase energy access. References [1] Jay Giri, “Rural Electrification, Microgrids and Renewables,” May-2015 IEEE newsletters. http://smartgrid.ieee.org/newsletters/may2015/rural-electrification-microgrids-and-renewables. [2] A. Chauhan and R. P. Saini, “Renewable energy based power generation for stand-alone applications: A review,” 2013 Int. Conf. Energy Effic. Technol. Sustain., pp. 424–428, 2013. [3] M. Fadaee and M. a. M. Radzi, “Multi-objective optimization of a stand-alone hybrid renewable energy system by using evolutionary algorithms: A review,” Renew. Sustain. Energy Rev., vol. 16, no. 5, pp. 3364–3369, 2012. [4] A. Kashefi Kaviani, G. H. Riahy, and S. M. Kouhsari, “Optimal design of a reliable hydrogen-based stand-alone wind/PV generating system, considering component outages,” Renew. Energy, vol. 34, no. 11, pp. 2380–2390, 2009. [5] K. Kusakana, “Optimal scheduled power flow for distributed photovoltaic/wind/diesel generators with battery storage system,” IET Renew. Power Gener., vol. 9, no. 8, pp. 916–924, 2015. [6] R. Luna-Rubio, M. Trejo-Perea, D. Vargas-Vázquez, and G. J. Ríos-Moreno, “Optimal sizing of renewable hybrids energy systems: A review of methodologies,” Sol. Energy, vol. 86, no. 4, pp. 1077–1088, 2012. [7] M. Mohammadi, S. H. Hosseinian, and G. B. Gharehpetian, “GA-based optimal sizing of microgrid and DG units under pool and hybrid electricity markets,” Int. J. Electr. Power Energy Syst., vol. 35, no. 1, pp. 83–92, 2012. [8] E. S. Sreeraj, K. Chatterjee, and S. Bandyopadhyay, “Design of isolated renewable hybrid power systems,” Sol. Energy, vol. 84, no. 7, pp. 1124–1136, 2010. [9] C. Wang and M. H. Nehrir, “Power management of a stand-alone wind/photovoltaic/fuel cell energy system,” IEEE Trans. Energy Convers., vol. 23, no. 3, pp. 957–967, 2008. [10] http://www.tangedco.gov.in/load%20calculator.htm [11] http://rredc.nrel.gov/solar/old_data/nsrdb/ [12] A. K. Daud and M. S. Ismail, “Design of isolated hybrid systems minimizing costs and pollutant emissions,” Renew. Energy, vol. 44, pp. 215–224, 2012. [13] H. Kord and A. Rohani, “An integrated hybrid power supply for off-grid applications fed by wind/photovoltaic/fuel cell energy systems,” 24th International Power System Conference (PSC2009), pp. 1-11, 2009. [14] L. Wang and C. Singh, “PSO-based multi-criteria optimum design of a grid-connected hybrid power system with multiple renewable sources of energy,” Proc. 2007 IEEE Swarm Intell. Symp. SIS 2007, no. Sis, pp. 250–257, 2007.


103 9

[15] H. Borhanazad, S. Mekhilef, V. Gounder Ganapathy, M. Modiri-Delshad, and A. Mirtaheri, “Optimization of micro-grid system using MOPSO,” Renew. Energy, vol. 71, pp. 295–306, 2014. [16] M. M. Mahmoud and I. H. Ibrik, “Techno-economic feasibility of energy supply of remote villages in Palestine by PV-systems, diesel generators and electric grid,” Renew. Sustain. Energy Rev., vol. 10, no. 2, pp. 128–138, 2006. [17] S. Diaf, D. Diaf, M. Belhamel, M. Haddadi, and A. Louche, “A methodology for optimal sizing of autonomous hybrid PV/wind system,” Energy Policy, vol. 35, no. 11, pp. 5708–5718, 2007. [18] C. Darras et al., “Sizing of photovoltaic system coupled with hydrogen/oxygen storage based on the ORIENTE model,” Int. J. Hydrogen Energy, vol. 35, no. 8, pp. 3322–3332, 2010. [19] R. Dufo-López and J. L. Bernal-Agustín, “Multi-objective design of PV-wind-diesel-hydrogen-battery systems,” Renew. Energy, vol. 33, no. 12, pp. 2559–2572, 2008.