modeling, simulation and fault diagnosis in a microgrid

MODELING, SIMULATION AND FAULT DIAGNOSIS IN A MICROGRID

By

Eng. Atef Mohamed Abd El Razek Mansour A Thesis Submitted to the Faculty of Engineering at Cairo University in Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE

in ELECTRICAL POWER AND MACHINES ENGINEERING

FACULTY OF ENGINEERING, CAIRO UNIVERSITY GIZA, EGYPT 2016


By

Eng. Atef Mohamed Abd El Razek Mansour A Thesis Submitted to the Faculty of Engineering at Cairo University in Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE


Under the Supervision of Prof. Dr. Essam Aboul Zahab

Dr. Khaled Nagdy Faris

…………………………….

……………………………….

Electrical Power and Machines Department Faculty of Engineering, Cairo University

Power Electronics and Energy Conversion Department, Electronics Research Institute


MODELING, SIMULATION AND FAULT DIAGNOSIS IN A MICROGRID By

Eng. Atef Mohamed Abd El Razek Mansour

A Thesis Submitted to the Faculty of Engineering at Cairo University in Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE


Approved by the Examining Committee:

_____________________________ Prof. Dr. Essam Aboul Zahab Electrical power and machines department Faculty of Engineering-Cairo University

Thesis Main Advisor ……………………

_______________________________________

Prof. Dr. Hosam Kamal Youssef Electrical power and machines department, Faculty of Engineering-Cairo University

Internal Examiner ……………………

_______________________________________

Prof. Dr. Mohsen Taha El-hagry Power electronics and energy conversion department, Electronics Research Institute

External Examiner ……………………


Engineer: Atef Mohamed Abd El Razek Mansour Date of Birth : 91 / 6 / 9196 Nationality : Egyptian E-mail : [email protected] Phone. : 01152538330 Address : Tahway- Ashmoon- Menofia Registration Date : 1/ 10 /2010 Awarding Date : / / 2016 Degree : Master of science Department : Electrical power and machines engineering Supervisors :

Prof. Dr. Essam Mohamed Aboul zahab Dr. Khaled Nagdy Faris

Examiners :

Prof. Dr. Hosam Kamal Youssef Prof. Dr. Mohsen Taha El-hagry Prof. Dr. Essam Mohamed Aboul zahab

Title of Thesis :


Key Words: Photovoltaic and Microgrid, Fault diagnosis, Maximum Power Point Tracking, Energy Management Controller.

Summary : A modeling of the PV is introduced to study its electrical characteristics under normal and at faulty conditions such as short circuit and partial shadow faults. Also a proposed intelligent fault detection and diagnosis technique based on fuzzy logic is introduced. A microgrid system composed of PV source, energy storage systems and DC load; is modeled by using Matlab/Simulink simulation tools. A proposed modified perturb and observe algorithm is introduced to track the maximum power point. The proposed method is compared with fuzzy logic maximum power point tracking. The simulation results show that the two algorithms are nearly the same. Also a proposed energy management controller for the adopted microgrid is introduced to control the power sharing among the power source, the energy storage systems and the load. This controller has the ability to select the suitable energy storage element to be charged or discharged according to the environmental conditions and the load power demand. The proposed controller stabilizes the output voltage at the reference voltage also decreases the voltage change during the transient periods. An experimental study on a lab scale is executed to

verify some of the simulation results of the proposed maximum power point tracking algorithm.

Acknowledgement No one but Allah, the most greatest, deserves all our thanks as he gave us the faith, the patience, and the power to complete this work successfully. First of all, I would like to thank my advisor, Prof. Dr. Essam Aboul Zahab, for giving me continuous help, patience, understanding, willingness throughout the period of the research and the opportunity to research under his guidance and supervision. I would like also to express my sincere gratitude to my advisor, Dr. Khaled Nagdy for his guidance, encouragement, remarkable patience and caring support during my finishing this thesis. Also I am grateful to Pof. Dr. Mohsen Taha Elhagry for his valuable guidance and his insightful suggestions throughout this work. Also I am really indebted for the Electronic Research Institute for providing all the resources required to carry out this thesis. Also, I would like to thank all my friends in the department of power electronics and energy conversion for their always encouragement and support.

Last but not least I would like to thank all my family; my father, my mother, my brothers “Ahmed and Abdullah”, my wife , my daughter “Shahd”, and my son “Mohamed” for their support and encouragement and patience during my studying. I am indebted to all of them forever for all that they have done for me.

i

Table of contents Acknowledgement ................................................................................................................. i Table of contents .................................................................................................................. ii List of Tables ....................................................................................................................... vi List of figures ...................................................................................................................... vii List of abbreviations ........................................................................................................... xi List of symbols ................................................................................................................... xiii Abstract ............................................................................................................................. xvi Chapter 1: Introduction ...................................................................................................... 1 1.1 Preface ......................................................................................................................... 1 1.2 Thesis objectives ......................................................................................................... 2 1.3 Thesis outline............................................................................................................... 2 Chapter 2: Faults diagnosis of photovoltaic Plants in a micro grid ................................. 4 2.1 Introduction ................................................................................................................. 4 2.2 Modeling of the PV. .................................................................................................... 4 2.3 Factors that affects on the output characteristics of the PV module ......................... 10 2.4 Electrical configurations of photovoltaic modules in a micro grid ........................... 14 2.4.1 Series connection ................................................................................................ 14 2.4.2 Parallel connection of modules........................................................................... 16 2.5 Common Faults types of solar array .......................................................................... 17 2.5.1 Partial shadow fault: ........................................................................................... 18 2.5.2 Short circuit fault: ............................................................................................... 18 2.6 Proposed fuzzy logic classifier .................................................................................. 21 2.6.1 Input and output membership functions ............................................................. 22 2.6.2 Simulation results of the proposed fuzzy logic classifier ................................... 27 2.7 Solutions to decrease the faults effects on PV system .............................................. 30 Chapter 3: Energy storage systems used in microgrid ................................................... 32 3.1 Introduction ............................................................................................................... 32 ii

3.2 Mechanical energy storage systems: ......................................................................... 33 3.2.1 Pumped hydro energy storage (PHES) ............................................................... 34 3.2.2 Compressed air energy storage systems (CAESS) ............................................. 34 3.2.3 Flywheel ............................................................................................................. 35 3.3 Chemical energy storage systems .............................................................................. 36 3.3.1 Hydrogen energy storage .................................................................................... 36 3.3.2 Flow batteries ..................................................................................................... 37 3.3.3 Batteries .............................................................................................................. 38 3.3.3.1 Sodium sulfur batteries (NaS) ..................................................................... 38 3.3.3.2 Nickel cadmium battery .............................................................................. 39 3.3.3.3 Nickel metal hydride battery (NiMH) ......................................................... 39 3.3.3.4 Lithium ion battery ...................................................................................... 40 3.3.3.5 Lead acid battery ......................................................................................... 40 3.4 The electrical energy storage systems ....................................................................... 41 3.4.1 Superconducting magnetic energy storage (SMES) ....................................... 41 3.4.2 Supercapacitor (SC) ........................................................................................ 42 3.5 The adopted energy storage systems modeling ......................................................... 43 3.5.1 Modeling of lead acid battery[40],[41]........................................................... 44 3.5.2 Modeling of the supercapacitor ...................................................................... 47 Chapter 4: Proposed Energy Management controller for a Microgrid System and simulation Results .............................................................................................................. 50 4.1 Introduction ................................................................................................................ 50 4.2 System Configuration ................................................................................................ 54 4.2.1 PV module .......................................................................................................... 56 4.2.2 Boost converter[48] ............................................................................................ 56 4.2.3 Battery with bidirectional converter ................................................................... 59 4.2.4 Supercapacitor (SC) with bididrectional converter ............................................ 60 4.3 The Control strategy .................................................................................................. 60 4.3.1 The first control unit: Maximum power point tracking (MPPT) controller of the DC-DC boost converter ............................................................................................... 61 4.3.1.1 Modified P&O algorithm: ........................................................................... 61 iii

4.3.1.2. Fuzzy logic based maximum power point tracking algorithm ................... 65 4.3.1.2.1. Fuzzification ....................................................................................... 65 4.3.1.2.2 Fuzzy rules ............................................................................................ 66 4.3.1.2.3 Defuzzification ..................................................................................... 68 4.3.2 The proposed Energy management controller .................................................... 69 4.4 Simulation Results ..................................................................................................... 71 4.4.1 Case A: Shadow occurs on the PV module and the load is constant at 5A. ....... 72 4.4.1.1 Config1: Using PV module without energy storage. ................................... 72 4.4.1.2 Config2: PV with battery. ............................................................................ 76 4.4.1.3 Config3: PV with battery and SC. ............................................................... 81 4.4.2 Case B: The insolation is constant and step change occurs in the load demand. 86 4.4.2.1 Config1: PV without energy storage. .......................................................... 86 4.4.2.2 Config2: PV with battery. ............................................................................ 90 4.4.2.3 Config3: PV with battery and SC. ............................................................... 94 4.5 Conclusion ................................................................................................................. 99 Chapter 5: System Setup and Experimental results ..................................................... 100 5.1 Overview ................................................................................................................. 100 5.2 Experimental setup ................................................................................................. 100 5.3 System components ................................................................................................. 102 5.4 Experimental results and discussion ........................................................................ 107 5.4.1 The first case..................................................................................................... 108 5.4.2 The second case ................................................................................................ 108 5.4.3 The third case.................................................................................................... 109 Chapter 6 : Conclusions of the thesis and future work ................................................. 111 6.1 Conclusions ............................................................................................................. 111 6.2 Future work ............................................................................................................. 111 References ......................................................................................................................... 112 Appendices ........................................................................................................................ 118 Appendix “A” Fuzzy rules of (fault diagnosis fuzzy logic classifier) ........................... 118 Appendix “B” PV module specifications ........................................................................ 126 Appendix “C” voltage sensor Specs. ............................................................................... 127 iv

Appendix “D” Current sensor Specs. ............................................................................. 130 Appendix “E” DSPACE DS1102 .................................................................................... 133 Appendix “F” MOSFET Transistor IRFP460LC ......................................................... 135 Appendix “G” Electronic load ........................................................................................ 137 Appendix “H” battery specs ............................................................................................ 139 Published papers .............................................................................................................. 141

v

List of Tables Table 2. 1: Electrical characteristics of IS4000P 300 watt module at STC.......................... 7 Table 2. 2: simulation results at different values of short circuit ........................................ 28 Table 2. 3: simulation results at different values of Partial shadow .................................... 29 Table 3. 1: comparison between battery types .................................................................... 41

Table 4: 1 Fuzzy Rules ....................................................................................................... 67

vi

List of figures Figure 2. 1: Equivalent electrical circuit of PV cell. ............................................................. 5 Figure 2. 2: IS4000P 300W module. ..................................................................................... 7 Figure 2. 3: Simulated electrical characteristics of IS4000P 300W module. ........................ 8 Figure 2. 4: The subsystem which represent the mathematical equations of the PV module 9 Figure 2. 5: The second subsystem. ....................................................................................... 9 Figure 2. 6: MATLAB (SPS) based PV model of IS4000 PV module. ............................ 10 Figure 2. 7: I-V and P-V curves at constant temperature = 25oC, and varying insolation (G) ............................................................................................................................................. 11 Figure 2. 8: I-V and P-V curves at constant G=1000 W/m2 and varying temperature ........ 12 Figure 2. 9: I-V and P-V curves at constant G=1000 W/m2 and constant Temperature 25 oC and varying ideal factor. ...................................................................................................... 13 Figure 2. 10: String of PV modules ..................................................................................... 14 Figure 2. 11: I-V and P-V curves of one PV module and two modules connected in series ............................................................................................................................................. 15 Figure 2. 12: Group of PV modules .................................................................................... 16 Figure 2. 13: I-V and P-V characteristic curves of one module and two PV modules connected in parallel. ........................................................................................................... 17 Figure 2. 14: Two PV module connected in series with a shaded module and its corresponding I-V and P-V characteristic curves. .............................................................. 19 Figure 2. 15: I-V and P-V characteristics of PV array at different percentages of short circuit at STC ....................................................................................................................... 20 Figure 2. 16: Fault diagnosis system on MATLAB ............................................................ 21 Figure 2. 17: Input membership functions .......................................................................... 23 Figure 2. 18: Output membership functions ........................................................................ 24 Figure 2. 19: I-V and P-V curve at partial shadow 20% on one module............................. 25 Figure 2. 20: I-V and P-V curve at partial shadow 70% on one module............................. 26 Figure 2. 21: P-V curve of two modules connecting in series and in parallel without shadow and with 50% shadow on a PV module. ................................................................. 30 Figure 3. 1: Energy storage types ....................................................................................... 32 Figure 3. 2: Energy density and power density of different energy storage systems . ........ 33 Figure 3. 3: Pumped hydro energy storage. ........................................................................ 34 Figure 3. 4: Compressed air energy storage ........................................................................ 35 Figure 3. 5: Flywheel structure ............................................................................................ 36 Figure 3. 6: Fuel cell structure ............................................................................................ 37 Figure 3. 7: Flow battery architecture ................................................................................. 38 Figure 3. 8: Sodium sulfur battery structure ....................................................................... 39 vii

Figure 3. 9: Structure of SMES system ............................................................................... 42 Figure 3. 10: Supercapacitor construction ........................................................................... 43 Figure 3. 11: Model of battery ............................................................................................. 44 Figure 3. 12: Discharging curve of a battery ....................................................................... 46 Figure 3. 13: Model of the SC ............................................................................................. 47 Figure 3. 14: Current discharge profiles (a) data sheet profiles, (b) simulated profiles. .... 49 Figure 4. 1: Block diagram of the proposed micro grid and its control strategy ................. 51 Figure 4. 2 Proposed Microgrid system............................................................................... 52 Figure 4. 3: Power density versus energy density ranges for different energy storage system technologies [47] ................................................................................................................. 53 Figure 4. 4: MATLAB Simulink model of the proposed system ........................................ 55 Figure 4. 5 Topologies of boost converter (a) Boost converter circuit, (b) when SW is OFF, (c) when SW is ON.............................................................................................................. 56 Figure 4. 6 Battery with buck boost bidirectional converter .............................................. 59 Figure 4. 7 Super-capacitor with buck boost bidirectional converter.................................. 60 Figure 4. 8 Conventional perturb and observe (P&O) algorithm ........................................ 62 Figure 4. 9: Conditions of (dP/dI) by varying the operating voltage to track MPP. ........... 63 Figure 4. 10 Modified Perturb and observe flow chart........................................................ 63 Figure 4. 11 Implementation of Modified Perturb and observe in MATLAB .................... 64 Figure 4. 12 : The output duty cycle of modified P&O ....................................................... 64 Figure 4. 13 Inputs and output fuzzy sets of the fuzzy logic controller .............................. 66 Figure 4. 14: Rule editor of fuzzy logic controller block in MATLAB .............................. 67 Figure 4. 15: MATLAB Simulink structure of fuzzy MPPT algorithm .............................. 68 Figure 4. 16: The output duty cycle of fuzzy logic MPPT algorithm ................................. 69 Figure 4. 17: Proposed energy management controller block diagram ............................... 70 Figure 4. 18: the structure of the proposed energy management strategy in MATLAB. .... 71 Figure 4. 19: Case (A) config1 load power and PV power a) FLC MPPT, b) modified P&O MPPT. .................................................................................................................................. 73 Figure 4. 20: Case (A) config1 load voltage a) FLC MPPT , b) modified P&O MPPT. .... 74 Figure 4. 21 : PV module characteristics (a) I-V curve, (b) P-V curve ............................... 75 Figure 4. 22: Case (A) config2 load power, PV power and battery power (a) FLC MPPT, (b using modified P&O MPPT. ........................................................................................... 77 Figure 4. 23: Case (A) config2 load voltage. (a) using FLC MPPT, (b)using modified P&O MPPT. .................................................................................................................................. 78 Figure 4. 24 Case (A) Config2 Duty cycle of the bidirectional converter of the battery (a) using Fuzzy MPPT, (b) using modified P&O ..................................................................... 80 Figure 4. 25 : Case (A) Config2 State of charge of the battery pack .................................. 81 Figure 4. 26: Case (A) config3 load power, PV power, battery power and SC power (a) using FLC MPPT, (b) using modified P&O MPPT ............................................................ 82 viii

Figure 4. 27: Case (A) config3 load voltage. (a) using FLC MPPT, (b) using modified P&O MPPT. ........................................................................................................................ 83 Figure 4. 28: Case (A) Config3 Duty cycle of the bidirectional converter of the battery (a) using Fuzzy MPPT, (b) using modified P&O ..................................................................... 84 Figure 4. 29: Case (A) Config3 Duty cycle of the bidirectional converter of the SC (a) using Fuzzy MPPT, (b) using modified P&O ..................................................................... 85 Figure 4. 30: Load current ................................................................................................... 86 Figure 4. 31. Case (B) config1 Load power and PV power (a) using FLC MPPT, (b) using modified P&O MPPT .......................................................................................................... 87 Figure 4. 32. Case (B) config1 load voltage (a) using FLC MPPT, (b) using modified P&O MPPT. .................................................................................................................................. 88 Figure 4. 33: characteristic curve of the PV module (a) P-V curve, (b) I-V curve ............ 89 Figure 4. 34: Case (B) config2 load power, PV power, and battery power. (a) using FLC MPPT, (b) using modified P&O MPPT .............................................................................. 91 Figure 4. 35: Case (B) config2 load voltage (a) using FLC MPPT, (b) using modified P&O MPPT ................................................................................................................................... 92 Figure 4. 36: Case (B) config2 duty cycle of the bidirectional converter of the battery (a) when using FLC MPPT (b) when using modified MPPT ................................................... 93 Figure 4. 37: Case (B) config2 state of charge of the battery pack ..................................... 94 Figure 4. 38 Case (B) config3 load power, PV power, battery power and SC power (a)using FLC MPPT (b) using modified P&O MPPT. ........................................................ 95 Figure 4. 39. Case (B) config3 load voltage (a) using FLC MPPT , (b) using the modifed P&O MPPT. ........................................................................................................................ 96 Figure 4. 40 Case (B) Config3 Dty cycle of the bidirectional converter of the battery pack (a) when using FLC MPPT (b) using the Modified P&O MPPT ....................................... 97 Figure 4. 41: Case (B) Config3 the duty cycle of the bidirectional converter of the SC (a) when using FLC MPPT (b) using the Modified P&O MPPT ............................................ 98 Figure 4. 42 state of charge of the Battery and the SC ........................................................ 99 Figure 5. 1 Over view of the experimental setup .............................................................. 100 Figure 5. 2: block diagram of the system .......................................................................... 101 Figure 5. 3: Voltage sensor ............................................................................................... 102 Figure 5. 4: Current sensor ................................................................................................ 103 Figure 5. 5: Optocoupler isolation circuit .......................................................................... 103 Figure 5. 6: level shifter ..................................................................................................... 104 Figure 5. 7: Driving signal of the MOSFET ...................................................................... 104 Figure 5. 8: Boost converter .............................................................................................. 105 Figure 5. 9: DC Electronic load PLZ1004W ..................................................................... 105 Figure 5. 10: Measured I-V Curve of the IS4000P PV module (a) at normal operation, (b) at partial shadow. Y axis is the PV current measured on channel (B) with deviation (2A) ix

and x axis is the PV voltage measured in channel (D) with deviation (20 volt) and the PV power measured on channel (A). ....................................................................................... 107 Figure 5. 11: The maximum PV power measured on channel (1), the maximum PV current on channel (2), the battery current on channel (3) and the load current on channel (4) ... 108 Figure 5. 12 : The maximum PV power measured on channel (1), the maximum PV current on channel (2), the battery current on channel (3) and the load current on channel (4) ... 109 Figure 5. 13 The maximum PV power measured on channel (1), the maximum PV current on channel (2), the battery current on channel (3) and the load current on channel (4) ... 110

x

List of abbreviations CAESS

Compressed Air Energy Storage Systems

DC

Direct current

EMC

Energy Management Controller

ESS

Energy Storage Systems

IGBT

Insulated Gate Bipolar Transistor

L-ion

Lithium Ion

MPP

Maximum power Point

MPPT

Maximum Power Point Tracking

NaBr

Sodium Bromine

NaS

Sodium Sulfur

NiCd

Nickel Cadmium

NiMH

Nickel Metal Hydride

Pb-Acid

Lead Acid Battery

PHES

Pumped Hydro Energy Storage

P&O

Perturb and Observe

PV

Photovoltaic

PWM

Pulse Width Modulation

rpm

Revolution Per Minute

SC

Supercapacitor

SMES

Superconducting magnetic energy storage

xi

SPS

Simpower Systems In MATLAB

STC

Standard Test Conditions where G=1000W/m2 and T= 25oC

VBr

Vanadium Bromine

VRLA

Valve Regulated Lead Acid Battery

ZnBr

Zinc Bromine

xii

List of symbols A

Amplitude of the exponential zone (V),

Ai

Interfacial area between the electrodes and the electrolyte (m2)

B

The charge at the end of the exponential zone (Ah)-1,

c

Molar concentration (mol . m-3)

CGC

Gouy-Chapman capacitance (farads)

CH

Helmholtz capacitance (farads)

Ci

Ratio between Imax/Isc

Cv

Ratio between Vmax/Voc

D

Duty cycle

d

Molecular radius (m)

E

Open circuit voltage (V),

Eexp

Voltage of the end of exponential zone

Efull

Fully charged voltage of the battery

Egap

Band gap energy of silicon =1.12eV

ENom

Voltage at the end of the nominal zone

E0

Battery constant voltage (v),

Exp(s)

Dynamics of the exponential zone,

Ɛ

Permittivity of electrolyte material

Ɛ0

Permittivity free space (F/m)

F

Faraday constant = 96485.3383

G

Insolation W/m2

Gr

Solar insolation at STC

xiii

i

Battery current (A),

i*

Low frequency current dynamics (A),

ID

Diode current (A)

il

The current which passes through the coil of the boost converter

Imax

Maximum power current

INp

Current of number of parallel connected module respectively

I0

Saturation current or leakage current

I0r

Reverse saturation current at standard test conditions

Ip

Output current of parallel group

Iph

Photo current (A)

Ipv

PV output current (A)

Is

Equivalent output current of a PV string.

Isc

Short circuit current

iSC

Supercapacitor current (A)

Iscr

Reference short circuit current at STC (A)

it

Extracted capacity (Ah),

K

Polarization voltage (V),

k

Boltzman’s constant =1.38e-38 J/oK

K0

Short circuit current temperature coefficient

n

Ideal factor of the PV cell

Ne

Number of electrode layers

Np

Number of cells connected in parallel

Ns

Number of cells connected in series

Pmax

Maximum power

xiv

Q

Maximum capacity of the battery (Ah),

q

Electron charge =1.6e-19 C

Qc

Electric charge of one SC cell (Coulomb)

QNom

Charge at the end of the nominal zone

QT

Total electric charge (C)

R

Ideal gas constant = 8.31447

r

Internal resistance (ohm)

Rs

Series resistance

RSC

Resistance of SC module (ohm)

Rshunt

Shunt resistance

Sel(s)

Represents the mode of charging or discharging the battery,

T

The cell temperature oK

Tr

Reference temperature at STC (oK)

Vin

Input voltage

Vl

The voltage across the coil

Vmax

Maximum power voltage

Vo

Output voltage

Voc

Open circuit voltage

Vocr

Open circuit voltage at STC (V)

Vp

Output voltage of the group

Vpv

Photovoltaic operating voltage

Vs

Equivalent output voltage of PV string

Vsc

Supercapacitor output voltage (V)

xv

Abstract Renewable energy sources represent one of the most interesting energy sources which attract the researchers to study and increase its efficiency. This is because renewable energy sources have more advantages than the fossil fuel based conventional energy generation such as there are no environmental impacts, the renewable energy is not depleted and it lasts forever. Some of the disadvantages of renewable energy sources, especially, PV-system are its low efficiency and variable output power according to the environmental conditions. So that the researchers intended to increase the efficiency of the PV modules by improving the PV materials and by using different maximum power point tracking algorithms. In this thesis the PV source is adopted to be used as the main power source in a micro grid system. Also energy storage systems are used in the adopted microgrid to overcome the problems of the variable output power from the PV source. A modeling of the PV is introduced to study its electrical characteristics under normal and at faulty conditions such as short circuit and partial shadow faults. Also a proposed intelligent fault detection and diagnosis technique based on fuzzy logic is introduced. A microgrid system composed of PV source, energy storage systems and DC load; is modeled by using Matlab/Simulink simulation tools. A proposed modified perturb and observe algorithm is introduced to track the maximum power point. The proposed method is compared with fuzzy logic maximum power point tracking. The simulation results show that the two algorithms are nearly the same. Also a proposed energy management controller for the adopted microgrid is introduced to control the power sharing among the power source, the energy storage systems and the load. This controller has the ability to select the suitable energy storage element to be charged or discharged according to the environmental conditions and the load power demand. The proposed controller stabilizes the output voltage at the reference voltage also decreases the voltage change during the transient periods. An experimental study on a lab scale is executed to verify some of the simulation results of the proposed maximum power point tracking algorithm.

xvi

Chapter 1: Introduction

1.1 Preface Renewable energy systems attracts the attention of the researchers because of there is no pollution, longer life. The generation of electrical energy using conventional resources of fossil fuel has a lot of problems such as the greenhouse gas emissions and the diminishing values of the fossil fuel. So that the energy generation using small scale renewable energy sources such as photovoltaic, wind, biogas, etc at the distribution voltage level is a new trend of energy generation. Hence a microgrid is considered as an interesting trend [1]-[3]. Micro grid can be defined as a power system at a low voltage network which consists of distributed generation sources which provides the electrical power to the local loads. Micro-grids are divided into two types, such as, grid-connected to decrease the absorbed power from the utility and stand-alone systems to provide the load with its demanded power without need to the utility. The stand-alone system requires energy storage systems to supply the load with its demanded power during the periods of shortage in generation, such as, solar power [ 4,[ 5]. Microgrid offers better economics, improved service reliability, and reduce the dependency on the local utility. The power converters and control algorithms are required for the microgrid technology [ 6]. Photovoltaic represents one of the most interesting renewable energy sources because of there is no environmental concerns. The PV systems have the ability to operate as a grid connected or as a stand-alone. The PV grid connected systems inject the energy to the grid. The standalone PV systems are used in the remote area where the use of other energy sources is difficult. Typical applications of stand-alone systems such as in satellite systems or mobile applications and other applications [1]. There are many types of faults which may occur in the PV system such as partial shadow as in [ 7]-[ 9] and short circuit faults. Fault diagnosis of PV systems will increase the reliability and the efficiency of the system. There are many fault diagnosis techniques used to detect and diagnose the faults in the PV systems as in [10]-[13] Different power converters and control algorithms can be used in a microgrid to obtain the maximum power from the PV and to control the power flow among the power source, the energy storage systems and the loads [ 14]-[ 16].

1

1.2 Thesis objectives This thesis aims to build an efficient fault diagnosis technique which is able to detect and diagnose the faults which may occur in the PV system and determines its percentage. This thesis targets to model a micro grid system consists of a PV source, different energy storage systems (such as battery and supercapacitor) and proposes an energy management strategy which achieves a good performance of the adopted micro grid. This good performance means achieving the maximum power of the PV source, stabilizing the output voltage at its reference voltage, and charging or discharging of the adopted energy storage systems without stress to increase its life time.

1.3 Thesis outline The thesis is divided into six chapters as following: Chapter 1: introduces the feasibility of the renewable energy sources especially PV systems and the definition of a microgrid. Chapter 2: presents a mathematical model of the PV cell and module, builds the PV model in Matlab and shows the electrical characteristics of the PV module. Also it presents the factors that affects the output of the PV, the electrical configurations of the PV modules in a micro grid such as series connection and parallel connection. It presents also the common fault types of a PV array such as partial shadow and short circuit. This chapter also presents a proposed fuzzy logic classifier to detect and diagnose the faults which may occur in the PV array. At the end of this chapter some solutions to decrease the faults effect on the micro grid are introduced.

Chapter 3: This chapter introduces different energy storage system technologies, and comparison between different battery types. Also it presents a mathematical modeling of the adopted energy storage systems.

Chapter 4: This chapter deals with the simulation of the adopted microgrid and the proposed energy management strategy, also the simulation results are recorded in this chapter. 2

Chapter 5: Introduces an experimental implementation of a microgrid system in a lab scale. Also shows the experimental results which verify the efficiency of the modified P&O maximum power point tracking technique at different cases of load and environmental conditions. Chapter 6: Illustrates the conclusions of the thesis and the recommended future work.

3

Chapter 2: Faults diagnosis of photovoltaic Plants in a micro grid 2.1 Introduction One of the most interesting and promising renewable energy sources in a micro grid is the photovoltaic. Photovoltaic energy is an interesting energy source because it has no gas emissions, silent because there are no moving parts and aesthetically improved views of building architecture and sustainable renewable energy source. Developing the PV system is generally based on two areas: the PV cell manufacturing process and materials, and the PV power management. Improving the performance of the photovoltaic system is one of the main aims of the research. To decrease the outage period and increase the PV systems life time yield, the diagnosis and monitoring of PV system operation are very important. Faults of photovoltaic arrays are difficult to be determined and sometimes even hard to be noticed by people. So it is very important to have an automatically and accurately fault diagnosis technique to detect the fault accurately, faster and safer. Because of detecting faults depending on manually testing by the operator consumes more time, less accuracy and the operator may be in danger. Since the individual PV system may operate at hundreds of volts or ten of amperes. The early fault detection and elimination of the fault makes the system more reliable, more efficient and eliminates the possibility of damaging parts or the whole power system. There are various methods for the fault diagnosis of PV arrays such as visual methods, thermal methods, and electrical methods [17]. In this chapter a model of the PV cell and module is used, also a fault diagnosis technique based on fuzzy logic is proposed.

2.2 Modeling of the PV. The photovoltaic module composed of number of PV cells connected in series to achieve required output voltage. The PV cells are the main components of the module. The PV cell is represented by a current source connected in parallel with a single diode and shunt resistance (Rshunt) and in series with small resistance (Rs). Rs and Rshunt can be neglected as in [18], [19]. The simplest circuit model of the PV cell is represented by a current source in parallel with a diode as in Figure 2. 1 . The electrical characteristics of the PV cell are nonlinear and vary according to the solar insolation (G) and the cell temperature (T). The output current Ipv can be obtained using Kirchhoff’s current law as in equation (2.1).

4

Diode

ID

Current source Iph

Ipv Vpv

Figure 2. 1: Equivalent electrical circuit of PV cell.

(2.1)

I pv  I ph  I D

Where Iph is the photo current and ID is the diode current. The diode current is proportional to the saturation current and can be expressed by using equation (2.2) [20]

  qV I D  I 0 exp  pv  nkT 

    1  

(2.2)

Where I0 is the saturation current or leakage current , q is the electron charge =1.6e-19C, Vpv is the photovoltaic cell operating voltage, n is the ideal factor of the PV cell, k is Boltzman’s constant =1.38e-38 J/oK, T is the surface temperature of the PV cell in oK. The saturation current is proportional to the surface temperature of the PV cell as cleared in eq (2.3) [21]:

 qE gap  1 1   T  I 0  I 0 r   exp      T nk  r T r T    3

(2.3)

5

Where I0r is the reverse saturation current at standard test conditions (STC where G=1000W/m2 and T= 25oC), Tr is the reference temperature at STC, Egap is the band gap energy of silicon =1.12eV. Ior can be obtained by using equation (2.4) [21].

I or 

I scr   qv ocr exp   nkT r 

    1  

(2.4)

Where Iscr is the reference short circuit current at STC at PV output voltage equal zero, Vocr is the open circuit voltage at STC and PV output current equal zero. The photocurrent is proportional to the solar insolation and temperature as shown in equation (2.5) [21]

I ph 

G *[ I scr  K 0 T  T r Gr

]

=

Amp

(2.5)

Where G is the solar insolation on the PV cell (W/m2), Gr is the solar insolation at STC =1000w/m2, and K0 is the current temperature coefficient = 0.0006 for the selected module. From the above equation the PV cell output current can be expressed as in equation 2.6

I pv  I ph  I 0 [exp (

qV . pv ) – 1] n .k .T =

Amp

(2.6)

The output current of a PV module which is composed of series and parallel connected cells can be expressed by using equation (2.7)

Ipv  N p .I ph  N p .I 0 [exp (

qV . pv ) – 1] N s .n .k .T =

Amp

(2.7)

Where Ns is the number of cells connected in series, and Np is the number of cells connected in parallel. The IS4000P module is adopted in this study. The IS4000P has 72 series connected polycrystalline silicon cells, Figure 2. 2 show a picture of the IS4000P high power PV module. The key specifications are shown in Table 2. 1. The electrical characteristics at STC of the PV module IS4000P 300 watt are shown in Figure 2. 3. 6

Figure 2. 2: IS4000P 300W module.

Table 2. 1: Electrical characteristics of IS4000P 300 watt module at STC Parameter

Value

Maximum power (Pmax)

300 W

voltage@ maximum power point (Vmp)

37.6 V

current@ maximum power point (Imp)

7.98 A

Open circuit voltage (Voc)

45

Short circuit current (Isc)

8.54 A

Short circuit current temperature coefficient (K0) Bypass diodes

V

0.06%/oC Three bypass dides, each 24 series cells shunted with one bypass diode

7

10 X: 37.6 Y: 300

X: 37.6 Y: 7.978

5

0

0

5

10

15

20 25 PV Voltage (V)

30

35

40

PV Current (A)

PV Power (W)

200

0 45

Figure 2. 3: Simulated electrical characteristics of IS4000P 300W module.

The IS4000P 300W photovoltaic module is modeled based on MATLAB SimpowerSystems (SPS). The above mathematical equations of the PV module are comined in the subsystem as shown in Figure 2. 4 to obtain th output current from the PV module according to the available environmental conditions. The output current is fed to acontrolled current source which is connected in parallel with a bypass diode as shown in Figure 2. 5 . The previous two subsystems are comined in a general simpower PV module model as shown in Figure 2. 6. By using this model it is very easy to study the operation of the PV module under different environmental conditions, short crcuit, open circuit, partial shadow and at different configurations (sereis connection and parallel connection).

8

Figure 2. 4: The subsystem which represent the mathematical equations of the PV module

Figure 2. 5: The second subsystem.

9

Figure 2. 6: MATLAB (SPS) based PV model of IS4000 PV module.

2.3 Factors that affects on the output characteristics of the PV module The output of characteristics of the PV module afected by environmental conditions (insolation (G), and temperature T) and the ideality factor of the PV cell. The photocurrent of the PV module is proportional to the insolation on the PV module as shown in Figure 2. 7. The PV output voltage is inversely prorportional to the cell temperature as shown in Figure 2. 8. Also the ideality factor of the PV cell is inversely proportional to the PV output power with increasing the ideal factor the knee of the I-V and P-V curve becomes more flatten and the PV power decreases as shown in Figure 2. 9.

10

9

G=1000W/m2

8

G=900W/m2 G=800W/m2

7 PV Current (A)

G=700W/m2 6 G=600W/m2

5

G=500W/m2

4

G=400W/m2

3

G=300W/m2

2

G=200W/m2

1

G=100W/m2

0

0

5

10

15


300

30

35

40

45

40

45

G=1000W/m2 G=900W/m2

250

G=800W/m2 G=700W/m2

PV Power (W)

200

G=600W/m2 G=500W/m2 G=400W/m2 G=300W/m2

150

G=200W/m2 G=100W/m2

100

50

0

0

5

10

15


30

35

Figure 2. 7: I-V and P-V curves at constant temperature = 25oC, and varying insolation (G)

11

10 9 8 T=25 oC

7 PV Current (A)

T=35 oC 6

T=45 oC T=55 oC

5 4 3 2 1 0

0

10

20

30 PV voltage (V)

40

50

60

300

250 T=25 oC PV Power (W)

200

T=35 oC T=45 oC T=55 oC

150

100

50

0

0

10

20

30 PV Voltage (V)

40

50

60

Figure 2. 8: I-V and P-V curves at constant G=1000 W/m2 and varying temperature

12

9 8

n=1 A=1

7

n=1.4 A=1.47 7n=2 A=2

PV Current (A)

6 5 4 3 2 1 0

0

5

10

15

20 25 PV voltage (V)

30

35

40

45

350

n=1 300

n=1.47

A=1 A=1.47

n=2 A=2

PV Power (W)

250 200 150 100 50 0

0

5

10

15


30

35

40

45

Figure 2. 9: I-V and P-V curves at constant G=1000 W/m2 and constant Temperature 25 oC and varying ideal factor.

13

2.4 Electrical configurations of photovoltaic modules in a micro grid The performance of the PV module is affected by array configuration. The photovoltaic array is composed of number of modules connected in series and in parallel to increase the output power of the array. The electrical characteristics of series and parallel connection will be discussed as follow:

2.4.1 Series connection When a number of PV modules of the same type are connected in series and have the same operating conditions, this configuration is called a string or assembly. Figure 2. 10 shows a string of PV modules consists of Ns modules.

Figure 2. 10: String of PV modules

The output current of the string is equivalent to the output current of one module in case of there is no mismatching conditions such as partial shadow, degradation and the connected modules of the same type. Equation (2.8) shows the output current of the string with respect to the current of each module in the string. Is =I1 = I2 = ………………. = INs (in Amps)

(2.8)

In contrast the output voltage is increased and equals to the sum of module voltages as shown in equation (2.9) Vs = V1+V2+………..+ VNs

(in Volts)

(2.9)

Where: Is is the equivalent output current of a PV string; I1 is the current of the first module and so on; Vs is the equivalent output voltage of PV string; V1 is the voltage of the first module and so on till to the module Ns. Figure 2. 11 show the I-V, and P-V curves of

14

series connected modules, where the current is constant with the increase of Ns modules while the voltage increase with increase of modules connected in series. 9 8 7

PV Current (A)

6 One module

Two modules

5 4 3 2 1 0

0

10

20

30


60

70

80

90

600

500

PV Power (W)

400

300

Two modules

200 One module 100

0

0

10

20

30


60

70

80

90

Figure 2. 11: I-V and P-V curves of one PV module and two modules connected in series

15

2.4.2 Parallel connection of modules The connection of number of PV modules in parallel is called a group of PV modules. In this configuration the output current increases and equals to the sum of output current of each parallel module as shown in equation (2.10) . Figure 2. 12 show the parallel configuration of Np modules.

Figure 2. 12: Group of PV modules

Ip= I1 + I2+………………. + INP

(in Amps)

(2.10)

The output voltage of parallel connection modules of the same type and have the same operating conditions equals to the voltage of one module; so that the voltage is constant in parallel configuration as in equation (2.11). Vp= V1= V2 = …………… = VNp

(in Volts)

(2.11)

Where: Ip is the output current of parallel configuration, I1, I2 and Inp, are the current of the first, second, and Np of parallel connected modules respectively, Vp is the output voltage of the group, V1, V2 and VNp are the voltage of first, second and Np parallel connected modules. Figure 2. 13 show the I-V and P-V characteristic curves of two PV modules connected in parallel.

16

18 16 Two modules 14

PV Current (A)

12 10 8 6

One module

4 2 0

0

5

10

15


30

35

40

45

30

35

40

45

600

500

Two modules

PV Power (W)

400

300

One module

200

100

0

0

5

10

15


Figure 2. 13: I-V and P-V characteristic curves of one module and two PV modules connected in parallel.

2.5 Common Faults types of solar array There are some common faults of the PV array and its components such as in PV cell which may be degraded because of: the increase in series resistance due to rust in contacts, the decrease in shunt resistance due to the metal migration through the P-N junction or the weakness of the antireflection coating [22]. In PV module there are some faults such as open circuit or short circuit between cells and cracking of the cover or sealant materials. In 17

the PV array fault conditions may be classified as an open or short circuit between modules, wiring mistake at install and performance mismatch between modules at partial shading. These fault conditions affects badly on the PV system performance, and can lead to hot spots in the PV module or destroying the system. The open circuit and short circuit faults of PV system occur because of the ageing of PV system after long term operation. The partial shadow faults means that a part of the PV module becomes shadowed which makes mismatch operation of the PV cells in the PV modules. This leads to creating hot spots in which temperature can reach higher than 150 oC and this can damage the cell [23]. In this chapter a PV array composed of two IS4000P 300 watt modules connected in series is adopted. Short circuit and partial shadowing faults are studied at different operating conditions of the PV array. Fuzzy logic classifier is proposed to determine the percentage of the fault on the PV array.

2.5.1 Partial shadow fault: Partial shadow means that part of the PV array becomes shadowed because of the neighboring buildings, bird droppings or bird stands on the PV module, etc. The output current of the shadowed part from the PV array becomes smaller than the normal part. This shaded part will limit the output current at its low current so that hot spots will be found and the cells can be destroyed. Bypass diodes are connected in parallel with strings of cells in the PV modules to decrease the effect of shadow faults. The bypass diode on the shaded part will conduct to deliver the current of the healthy part of the array. The output current voltage I-V characteristics have multiple steps and the output power voltage P-V characteristics becomes have multiple peaks. Figure 2. 14 shows two PV modules connected in series and a shadow of different values applied on one module; the output characteristics show the effect of shadow on the PV array.

2.5.2 Short circuit fault: Short circuit in the PV array may occur because of ageing of PV modules or failure of the bypass diode. Short circuit on one module or on one of the series strings in the PV modules will decrease the output power and maximum power voltage significantly; but the output current will not be affected. The adopted PV module consisted of 72 series connected cell. Each twenty four 24 cell (called a string) shunted with a bypass diode. All the expected Short circuit condition which may occur in this PV array have been studied and simulated as shown in Figure 2. 15. The collected data from these conditions are used to build knowledge base of the fuzzy logic classifier (FLC). This FLC has the ability to determine

18

the percentage of the partial shadow or the percentage of the short circuit faults and the system healthy percentage.

PV2

PV 1

9 8

normal operation

7

PV Current (A)

6

shadow 0.8

5 4 shadow 0.5 3 2 shadow 0.3 1 0

0

10

20

30


60

70

80

90

600 Normal operation shadow 0.8 500

400 shadow 0.5 300 shadow 0.3 200

100

0

0

10

20

30

40

50

60

70

80

90

Figure 2. 14: Two PV module connected in series with a shaded module and its corresponding I-V and P-V characteristic curves.

19

9 8

short circuit four strings

short circuit two strings

Normal

PV array current (A)

7 6 5 4 3 2 1 0

0

10

20

30

40 50 60 PV array voltage (V)

70

80

90

600 Normal 500

PV Array Power (W)

short circuit Two strings 400 short circuit four strings

300

200

100

0

0

10

20

30

40 50 60 PV Array Voltage (V)

70

80

90

Figure 2. 15: I-V and P-V characteristics of PV array at different percentages of short circuit at STC

20

2.6 Proposed fuzzy logic classifier The adopted fault diagnosis technique is based on measuring the I-V and P-V curves of the PV array. Figure 2. 16 shows MATLAB Simulink model of the fault diagnosis system. Based on the on the collected data from the adopted PV array under different faults at different environmental conditions an expert knowledge is built. A fuzzy logic clasifier (FLC) is proposed to detect and diagnose these faults type and percentage which may occur in the PV array [24].

Figure 2. 16: Fault diagnosis system on MATLAB

21

2.6.1 Input and output membership functions The proposed (FLC) have five inputs [ insolation, maximum power voltage (Vmax), cell temperature (temp), maximum power (Pmax) and open circuit voltage (Voc)], and have three outputs ( short circuit percentage, healthy percentage and partial shadow percentage). Expert knowledge is used and can be expressed using linguistic variables, which are described by fuzzy sets as shown in Figure 2. 17, Figure 2. 18. Inputs fuzzy sets have different numbers of memebership functions with linguistic variables as following: insolation have 11 membership functions {high dark, dark, VVLShiny means very very low shiny, VLShiney means very low shiny, low shiny, Mshiny means medium shiny, shiny, HShiny means high shiney, VHShine means very high shiney, Ultra shiny}, Vmax fuzzy set has 7 membership functions {zero, Vsmall means very small, small, Medium, High, Vhigh, fullArray}, Temp input fuzzy set has five membership functions {Low, Normal, Med, High, Vhigh}, Pmax input fuzzy set has seven membership functions {Zero, Vsmall means very small, small, Med, high, VHigh means very high, FArray means full array}, Voc input fuzzy set has nine membership functions {Zero, VVSmall means very very small, VSmall means very small, Small, Med, High, VHigh means very high, VVHigh very very high, and full Array}. Output fuzzy sets have different numbers of membership functions as following: Partial shadow percentage output fuzzy set hase 11 membership function {No shadow, extremely low shadow, vvsma, vs shadow, small, Medium, high, Vhigh, VVhigh, Ultra, Total shadow}, Helthy percentage output fuzzy set {complete fault, Partial fault, good}, short circuit percentage {Healthy, VSmall SC means very small short circuit, small SC, Medium, High SC, VHigh, Total SC}. Then the expert knowledge for these fuzzy sets can be formulated as a rules such as: IF (insolation==UltraShiny) & (temp==Normal) & (pmax==FArray) & (voc==Full_Array) THEN (partial_shadow_percentage=Noshadow) (healthy_percentage=good). All the fuzzy rules which used in the proposed classifer are existed in Appendix (A).

22

highDark 1

dark

VVLShiny

VLShiny

Lowshiny

MShiny

shiny

Hshiny

VHSine

VVHShine

UltraShiny

0.1

0.2

0.3

0.4

0.5 insolation

0.6

0.7

0.8

0.9

1

Degree of membership

0.8

0.6

0.4

0.2

0 0


Zero 1

Vsmall

Small

Medium

High

Vhigh

FullArray

0.8 0.6 0.4 0.2 0 0

10

20

30

40

50

60

70

80

90

Vmax LOW 1

Normal

Med

High

Vhigh


0.8

0.6

0.4

0.2

0 20

22


zero 1

24

26

28

30 temp

32

34

36

38

40

VSmall

Small

MED

High

VHigh

FArray

100

200

300 pmax

400

500

600

0.8 0.6 0.4 0.2 0 0


Zero 1

VVSmall

VSmall

Small

Med

High

VHigh

VVHigh

FullA rray

0.8 0.6 0.4 0.2 0 78

80

82

84

86

88

voc

Figure 2. 17: Input membership functions 23

90

Noshadow 1

extremelylowsha

vvsma

Vs s hadow

small

medium

high

Vhigh

VVhigh

ultra

totalshadow

10

20

30

40

50 partials hadowpercentage

60

70

80

90

100


0.8

0.6

0.4

0.2

0 0

completef ault

1

PartialFault

good


0.8

0.6

0.4

0.2

0 0

10

Healthy 1

20

VSmallSC

30

40

50 healthy percentage

smallSC

60

Medium

70

HighSC

80

90

VHighSC

100

ToalSC


0.8

0.6

0.4

0.2

0 0

10

20

30

40

50 Shortcircuit percentage

60

70

80

90

100

Figure 2. 18: Output membership functions

Defuzzification by using center of gravity to show the percentage of the fault and the healthy percentage of the system. The proposed fuzzy logic classifier hase the ability to determine the percentage of the fault and its type, but sometimes at certain levels of insolation can’t discriminate between the shorcircuit faults and partial shadow faults ( i.e show a percentage of short circiut and percentage of partial shadow). So that we should determine what is the type of this fault to know the correct percentage. From the 24

simulation results and studying the characteristic curves (I-V, and P-V) of the PV module we can conclude that at the normal operation or when there is no shadow on the PV the I_V and P_V curves have one maximum power point. At This maximum point the ratio between Vmax/Voc called (Cv) nearly equal to 0.83 and the ratio between Imax/Isc called (Ci) nearly equal to 0.93. These ratios are used to increase the acuracy of the FLC to determine the correct fault type. The ratios Ci and Cv are compared with the reference ratios; So that there are three cases to determine the fault type as following: Case1: If

Ci < 0.93

(2.12)

And Cv = 0.83

(2.13)

From equations (2.12) and (2.13) there is a partial shadow and the shadow percentage is less than 50%. The maximum power point occurs at the lower current step near the open circuit voltage as shown in Figure 2. 19.

600

9

X: 76.8 Y: 502.8

8

500

X: 76.8 Y: 6.547

6

PV Array Power (W)

PV Array Current (A)

7

5 4 3

400

300

200

2

100 1 0

0

10

20

30

40 50 60 PV Array voltage (V)

70

80

0

90

0

10

20

30

40 50 60 PV Array voltage (V)

70

80

90

Figure 2. 19: I-V and P-V curve at partial shadow 20% on one module

Case 2: If the Ci ratio nearly equal to its reference value If

Ci = 0.93

(2.14)

And Cv < 0.83

(2.15)

25

From equations (2.14) and (2.15) there is partial shadow fault and its percentage is larger than 50%. The maximum power point occurs at the higher current step near the short circuit current as shown in Figure 2. 20. 600

9 8

PV Array Power (W)

PV array current (A)

500

X: 35.99 Y: 7.937

7 6 5 4 3

400 X: 35.99 Y: 285.6

300

200

2

100 1 0

0

10

20

30

40 50 60 PV array voltage (V)

70

80

90

0

0

10

20

30

40 50 60 PV Array Voltage (V)

70

80

90

Figure 2. 20: I-V and P-V curve at partial shadow 70% on one module

Case 3: IF Ci = 0.93 And Cv = 0.83

(2.16) (2.17)

From equations (2.16) and (2.17) the occurrence of partial shadow is rejected and the array working at the normal case or there is a short circuit. The prposed fuzzy logic classifier depending on data collection and training will determine the normal operation or short circuit according to the measured environmental conditions and the output power. If the insolation is high and the power is high so that this is a normal operation. If the insolation is high and the power is low so that there is a fault and the proposed classifier can determine the type of this fault and its percentage according to the previous equations (2.12 – 2.17).

26

2.6.2 Simulation results of the proposed fuzzy logic classifier Different short circuit cases which may occur in the PV Array are tested at different levels of insolation. The adopted array has six substrings, each module has 3 substrings. In the testing cases the short circuit faults are set to occure on the first substring, then the short circuit occurs on two substrings, three substrings, four substrings, and five substrings. The short circuit tests occure at different levels of insolations. To ensure that the proposed method can estimate the percentage of the short circuit and can specify the fault type correctly at different environmental conditions. Table 2.2 shows the simulation results of the proposed fault diagnosis technique when applying different short circuit percentages in the PV array at different insolation levels. In this case the output of the fuzzy logic classifier system will show the short circuit percentage and the healthy percentage of the system. The partial shadow percentage will equal to zero. This means that the proposed fault diagnosis system can determine that the fault type is short circuit with specified percentage and not a partial shadow. On the other hand different levels of partial shadow faults on a PV module of the PV array are tested at different levels of insolations. This means that at insolation 1000W/m2 on the PV array, one PV module is artificially illuminated by 900W/m2 and 800W/m2 so that this module is shaded by10% and 20% respectively. The insolation level on the PV array is varied and different shadow percentages are tested as shown in Table 2. 3. The simulation results shows that the fault diagnosis system can determine that the fault is partial shadow with estimated percentage close to the tested partial shadow percentage. The short circuit percentage at this case equals zero. The healthy percentage value around 50% this means that the fault is a partial fault. So that the system has the ability to discriminate between fault types.

27

Table 2. 2: simulation results at different values of short circuit

Short circuit conditions

One string (16.66%)

Two strings (33.33%)

Insolation level

Estimated short circuit percentage in the PV array byfuzzy logic classifier

Healthy percentage

Partial shadow percentage

1000 w/m2

16.69%

51.1%

0

800 w/m2

16.69%

50.09%

0

600 w/m2

16.67%

51.18%

0

400 w/m2

18.42%

51.15%

0

1000 w/m2

33.35%

50.89%

0

800 w/m2

27.58%

50%

0

600 w/m2

28.76%

50.93%

400 w/m2

29.83%

50.7%

50.0%

50%

0

800 w/m2

51.86%

50.73%

0

600 w/m2

47.94%

50.79%

0

400 w/m2

52.61%

50.88%

0

1000 w/m2

66.66%

50.92

0

800 w/m2

73.26%

50.79%

0

70.07%

50.84%

0

400 w/m2

63.15%

50.79%

0

1000 w/m2

83.34%

51.11%

0

800 w/m2 600 w/m2

83.9% 84.09%

52.05% 52.22%

0 0

52.38%

0

1000 w/m Three strings (50%)

Four strings (66.66%)

Five strings (83.33%)

600 w/m

2

2

400 w/m2

84.17%

28

0 0

Table 2. 3: simulation results at different values of Partial shadow

Tested partial shadow percentage

Insolation level

Estimated short Healthy circuit percentage percentage in the PV array byfuzzy logic classifier

Partial shadow percentage

1000w/m2

0

55.2%

24.47%

900 w/m2

0

54.74%

21.49%

1000w/m2

0

51.09%

45.3%

900 w/m2

0

51.05%

44.6%

800 w/m2

0

50.99%

46.47%

1000w/m2

0

50.76%

60.59%

900 w/m2

0

50.76%

61.85%

800 w/m2

0

50.7%

55%

1000w/m2

0

50.97%

76.84%

900 w/m2

0

50.76%

76.04%

800 w/m2

0

51.01%

76.09%

20%

40%

60%

80%

29

2.7 Solutions to decrease the faults effects on PV system There are some common methods to decrease faults effect in PV Array such as using bypass diodes shunted with the cells in the PV modules. As the number of bypass diodes in the PV modules increases as the mismatching faults effect decreases. Also parallell connection of the modules or cells decreases the effect of partial shadow than sereis connection [25] as seen in Figure 2. 21. Because at normal operation the parallel configuration and series configuration have only one peak point with the same value but at different values of current and voltage as shown in Figure 2. 21. But at partial shadowing the series configuration affects badly because of there is multiple peaks local and global. On the other hand in parallel cofiguration thereis one peak greater than the global peak of series configuration. In Figure 2. 21the global peak in series configuration is at 317.9 W but in parallel configuration the maximum power is at 440.6 W. From this simulation results we can note that in parallel configuration the efficiency = (440.6/600)*100=73.43% while the efficiency of series configuration=(317.9/600)*100 = 52.98%.

600 Two modules

X: 37.55 Y: 599.9

X: 37 Y: 440.6

400

PV Power (W)

Two modules in series

two modules connected in parallel partail shadow .5

500

two modules connected in series partialshadow one module= .5

One module

300

X: 77.7 Y: 317.9

X: 37.64 Y: 300

200

100

0

0

10

20

30


60

70

80

90

Figure 2. 21: P-V curve of two modules connecting in series and in parallel without shadow and with 50% shadow on a PV module.

To overcome the faults which occur in the PV array energy storage systems are used. These storage systems supply the load with its required power until the fault is cleared according to the fault diagnosis system signal. So that the energy storage systems will be very useful to overcome the variations of PV power because of partial shadowing or varying environmental conditions [26]. In chapter three different energy storage sytem

30

technologies are discussed, and the adopted energy storage systems in this thesis will be studied and modelled in details.

31

Chapter 3: Energy storage systems used in microgrid

3.1 Introduction This chapter introduces different types of energy storage systems (ESS) technologies. The energy storage systems are used to overcome the problems of variable output power nature of the renewable energy sources. So that Energy storage systems (ESS) are considered very important components in the micro grid systems especially in standalone systems. ESS keeps the balance between the generation and the load demand. There are different types of energy storage systems divided according to the type of stored energy as shown in Figure 3. 1 [27].

Energy Storage Systems types

Chemical

Electrical

Mechanical

Super conductors

Lead-Acid

Li-ion

Ni-Cd

NaS

32

Super capacitors

Pumped hydro

Compressed air

Fly wheel

Batteries

Hydrogen

Flow batteries

Figure 3. 1: Energy storage types

The different types of energy storage systems also have different power density and energy density such as shown in Figure 3. 2. In this chapter an overview of each type of ESS is introduced also the modeling and the discharging characteristics of the supercapacitor and lead acid battery are introduced.

Figure 3. 2: Energy density and power density of different energy storage systems [28].

3.2 Mechanical energy storage systems: This type of energy storage systems depends on transforming the mechanical energy into electrical energy by rotating generators to generate the electricity. The types of mechanical energy storage systems are discussed as following:

33

3.2.1 Pumped hydro energy storage (PHES) The excess in electrical energy stored in this type of energy storage by pumping the water from a lower reservoir to a higher reservoir. When there is a need for power, the pumped water in the higher reservoir is allowed to flow to the lower reservoir. While the flowing of water from the higher to the lower reservoir the turbine rotates and generates the electrical energy as shown in Figure 3. 3 . The advantages of pumped hydro are as following: The operating cost for the pumped hydro is low, more reliable, maturity of pumped hydro and it has a long life time. The efficiency of pumped hydro systems is approximately 75%. The disadvantages of the pumped hydro systems are that the needs of the specific geological formations to be constructed, and the large environmental impacts [29]

Figure 3. 3: Pumped hydro energy storage.

3.2.2 Compressed air energy storage systems (CAESS) CAESS depends on storing the electrical energy at off peak time by pumping air into underground sealed cavern to a high pressure. The air with high pressure is kept underground to be used at the peak times. At the peak times the high pressure air drives turbines and the generated power is used at the peak times. Figure 3. 4 show a schematic diagram of compressed air energy storage system [4]. CAESS is used in large scale power systems. The range of Typical capacities of a CAESS is around 50 –300 MW. The CAESS has the longest energy storage time among the other energy storage types because of the very small losses. The CAESS can be used to store the energy for more than a one year [ 31].

34

Figure 3. 4: Compressed air energy storage

3.2.3 Flywheel Flywheel is one of the electromechanical energy storage systems. The energy stored in the fly wheel is in the form of a kinetic energy of a rotor mass [32]. Figure 3. 5 show the structure of the flywheel. The flywheel charges and discharges through an electrical machine work as a motor or as a generator. If there is an excess in the electrical power (off peak times) the electrical machine works as a motor to absorb the excess in power and accelerates the rotor to the required speed (charging mode). If there is shortage in electrical power (on peak time) the electrical machine works as a generator by decelerating the rotor through consuming the stored kinetic energy in the flywheel mass (discharging mode). The energy stored in the flywheel is directly proportional to the rotor momentum and the square of angular momentum [33]. Achieving high energy storage density needs to increment the rotation speed. Because of this reason there are two types of flywheels as following: high speed flywheel its rotating speed is around 40 000 rpm, and low speed flywheel its rotating speed is around 7000 rpm. The advantages of flywheels are long life, high power density, little maintenance, and environmentally friendly. The disadvantages of flywheels are high level of self-discharging because of the bearing losses and the air resistance also fly wheels have a low current efficiency [34].

35

Figure 3. 5: Flywheel structure

3.3 Chemical energy storage systems This type depends on the electrochemical reactions to transform the chemical energy into electrical energy at peak times and vice versa. Some of the chemical energy storage systems are discussed in brief as following:

3.3.1 Hydrogen energy storage The hydrogen is produce by the electrolysis and stored in gas tanks. When the power is needed, Fuel cells are used to produce the electricity from the stored hydrogen. Pure water results from the transformation process from the fuel cells so that there are no emissions of fuel cells. Also fuel cells can store large amounts of power. The disadvantages of fuel cells are that low efficiency (25%); also high cost of hydrogen production and storage [29]. The Oxidation and reduction reaction between hydrogen and oxygen occurs through a structure (fundamental electrochemical cell) which is made up of anode and cathode electrodes; separated by an electrolyte which is a medium for transferring ions as shown in Figure 3. 6 [35].

36

Figure 3. 6: Fuel cell structure

3.3.2 Flow batteries Flow battery is a two electrolyte system. The chemical compound which is used for energy storage is in a liquid state. Flow batteries overcome the limitations of standard electrochemical accumulators such as lead acid batteries which its chemical reactions results solid compounds stored directly on the electrodes. The limited mass of the standard batteries limits its capacity. There are many types of electrolytes that are used in flow batteries using the bromine as a fundamental element such as zinc (ZnBr), vanadium (VBr), sodium (NaBr) and sodium polysulfide. A large quantity of electrical energy can be stored by using large tanks and connecting a large number of cells. The stored energy can be released by pumping the electrolytes into the tanks. The architecture of flow batter is shown in Figure 3. 7 [35].

37

Figure 3. 7: Flow battery architecture

3.3.3 Batteries Batteries are the oldest electricity energy storage form that stores the electrical energy in the form of chemical energy through chemical reactions. The battery consists of electrochemical cells. The cell consists of positive electrode (the anode) and negative electrode (the cathode) immersed in a liquid, paste or a solid electrolyte. The batteries have high energy efficiency ranges (60-95%) [30].

3.3.3.1 Sodium sulfur batteries (NaS)

The negative electrode in this battery contains liquid sodium and the positive electrode contains liquid sulfur as active materials. The two electrodes are separated by solid beta alumina ceramic electrolyte as shown in Figure 3. 8. This electrolyte allows to the positive sodium ions to move through it to combine with the sulfur to form the polysulfide. During discharging process the positive Na+ ions flow through the electrolyte while the electrons flow through the outer circuit to produce 2 V. On the other hand sodium polysulfide releases the positive sodium ions through the electrolyte to recombine as elemental sodium in the charging process. The operating cell temperature is maintained at 300 – 350 oC so that this type of batteries is good at hot atmospheric countries. Advantages of the NaS

38

battery are made from nontoxic and inexpensive materials, no self-discharging, and high efficiency 90%. The dis advantages are the high capital cost[36].

Figure 3. 8: Sodium sulfur battery structure [30]

3.3.3.2 Nickel cadmium battery

This battery has a negative electrode which is the cadmium hydroxide plate, positive electrode which is the nickel hydroxide plate, separator and alkaline electrolyte. The advantages of NiCd batteries are low maintenance, reliable and have a high energy density (50 –75) Wh/Kg. the disadvantages of these batteries are high cost (1000$/Kwh) and low life cycle (2000 – 2500). Also it has environmental concerns because of disposing Cd is toxic heavy metal [29], [36].

3.3.3.3 Nickel metal hydride battery (NiMH)

NiMH characteristics are similar to NiCd battery. It is a relatively new technology. It is a sealed alkaline battery, the energy density increases because of replacing the hydrogen instead of cadmium also becomes more environmentally because of there is no toxic cadmium. NiMH has also lower charge and discharge rate capability, less tolerant to overcharge than NiCd battery. High temperature and heavy loads reduces the life of battery 39

service. Self-discharge is higher than NiCd; lithium ion battery has a higher specific energy density than NiMH battery [37]

3.3.3.4 Lithium ion battery

Lithium has a highest potential because of the reactive behavior of it. Also it is lightest metal. L-ion battery is composed of three layers as following anode which is made of (lithium compound) , the cathode which is made of (graphite). The anode and cathode are separated by a separator which allows to lithium ions to pass through it. The three layers are immersed in an electrolyte (organic solvent) which permits the ions to pass between anode and cathode. Advantages of L-ion battery are higher cell voltage (3.7 V compared to 2 V for lead acid), high efficiency, and energy loss is low (5%per month). The dis advantages of L-ion batteries are very expensive, needs a special charging circuit, the cells are destroyed if there is complete discharge of the battery, life cycle about five years, and the lithium has a flammable nature if contacts with the atmospheric moisture [38].

3.3.3.5 Lead acid battery

Lead acid batteries are the most mature technology for more than a century because of the low cost. Lead acid battery consists of a lead dioxide plate which is the positive electrode and a lead plate which is the negative electrode and a separator to isolate the two electrodes. Dilute sulfuric acid is used as the electrolyte to provide sulfate ions for the discharging reactions. There are different types of lead acid battery such as flooded lead acid battery, sealed battery which has a gelled electrolyte and valve regulated lead acid (VRLA) has low cycle life and cannot be discharged deeply. Deep cycle battery is designed to discharge most of its stored energy without damaging also it can charge and discharge for thousands of times. Deep cycle battery produce output current less than shallow cycle batteries but it can produce this amount of current for much longer time period [36],[37]. A brief comparison of different types of battery technologies is shown in Table 3. 1.

40

Table 3. 1: comparison between battery types [37]

Pb Acid

NiCad

NiMH

Li-ion

40 - 60

30 - 80

160

60 - 75

50 -150

140 - 300

1400

70 - 92

70 - 90

66

Self-discharge rate (at 20 oC) (%)/month

3 - 20

10

30

5-10

Cycle durability (cycles)

500 - 800

2000

500 - 1000

1200

Calendar life (years)

2-8

3 -10

2-5

2-3

Nominal Cell Voltage (V)

2.0

1.2

1.2

4.0

Operating temperature(oC)

-40 to 60

-40 to 50

-20 to 50

-20 to 50

Energy/weight (Wh/kg) Energy/size Wh/liter Charge/discharge efficiency (%)

30 - 40

99.9

3.4 The electrical energy storage systems 3.4.1 Superconducting magnetic energy storage (SMES)

The energy stored in SMES in the form of magnetic field created by flowing a direct current through a superconducting coil; this coil is cooled below the superconducting critical temperature. The needed temperature of superconductivity was about 4oK in the past nowadays there are materials that can function at100oK. The fundamental component of SMES is the coil which is made from a superconducting material, and additional components like power cryogenically cooled refrigeration and power conditioning equipment such as shown in Figure 3. 9 . 41

Advantages of (SMES): Charging and discharging have no effect on its life time, very quick response time means a high power can be released in a few milliseconds, high efficiency (85% - 90%), very high output power. Disadvantages: energy density is small, and has stability problems because of the strong magnetic field. [29], [34]

Figure 3. 9: Structure of SMES system

3.4.2 Supercapacitor (SC)

Supercapacitor has a various names such as ultracapacitors and electric double layer capacitor. Supercapacitor has a higher energy and power capability than traditional capacitors. Also it has a greater capacitance than the conventional capacitors. SC is constructed from two carbon electrodes, electrolyte and separator as shown in Figure 3. 10. SC stores charges electrostatically and the charge do not transfer between the electrode and the electrolyte. SC uses the electrochemical double layer of charge for storing the energy. When the voltage is applied on a SC the charge accumulates on the electrodes surface. Because of the double layers, the increase in surface area of the SC electrodes and the decrease in the distance between the electrodes the SC achieve higher energy density than the conventional capacitor [39]. Supercapacitor has a very fast charging and discharging nature because of its lower internal resistance compared with conventional batteries. SC is durable, has a high reliability, long life time (about one million cycle), no maintenance, wide range of operating temperature, 42

high efficiency 90% and high specific power density. But SC has a low specific energy density. Due to this nature of SC it is preferred to be used in the applications that have a huge number of short charging and discharging cycles. SC is not suitable for storing energy for long periods because of the low energy density, high self-discharge and high costs [34].

Figure 3. 10: Supercapacitor construction

3.5 The adopted energy storage systems modeling In this thesis two types of energy storage technologies are adopted as a storage system in the adopted micro grid. The first energy storage system is the Lead acid battery which is used as the main energy storage system in the adopted micro grid. The lead acid battery supplies the load with its demanded power at the steady state cases and charges with the surplus power at steady state of the peak times. The second energy storage is the super capacitor which is used to store and discharge energy at transient conditions. The hybrid energy storage system of supercapacitor and 43

battery combines the advantages of high energy density and high power density energy storage systems. Because of the very fast response of the SC all the surplus power at transient conditions will be stored efficiently and vice versa without effects on the SC life time; also this property will allow the battery to charge and discharge slowly without stress which increase its life time. Hence the overall efficiency of the micro grid system increased. In this section modeling of the lead acid battery and the SC are introduced as following

3.5.1 Modeling of lead acid battery[40],[41]

A simple controlled voltage source connected in series with a constant resistance represents a generic model of the most popular type of batteries as shown in Figure 3. 11.

Figure 3. 11: Model of battery [40]

44

The lead acid battery can be modeled by the following equations of charging and discharging modes as following Charging mode where (i* 0)

f 2 (it , i *, i , Exp )  E 0  K .

 Exp (s )  Q Q .i * K .  Laplace 1  .0  Q  it Q  it  Sel (s ) 

(3.2)

Where, E is the open circuit voltage of the battery (V), E0 is the battery constant voltage (v), Q is the maximum capacity of the battery (Ah), i is the battery current (A), i* is low frequency current dynamics (A), K is polarization voltage (V), it is extracted capacity (Ah), Exp(s) dynamics of the exponential zone, A is the amplitude of the exponential zone (V), B is time constant inverse of the exponential zone (Ah)-1, Sel(s) represents the mode of charging or discharging the battery, r is the internal resistance (ohm). The parameters of the battery can be calculated by using the discharging characteristics. A typical discharging curve consists of three sections is shown in Figure 3. 12 . The first section is less wide and called the exponential voltage when the battery charges and this zone depend on the battery type this zone lies at the left of the discharging curve. The second section which represents the charge that can be absorbed from the battery and ends when the voltage drops below the nominal voltage of the battery; this section lies in the middle of the discharging curve. The third section appears at the region where the voltage 45

drops rapidly and represents total discharge of the battery; this section lies on the upper right of the discharging curve

Figure 3. 12: Discharging curve of a battery

The amplitude of the exponential zone can be calculated by using equation (3.3) A= fully charged voltage (Efull) – voltage of the end of exponential zone (Eexp)

(3.3)

The charge at the end of the exponential zone (B) can be calculated by using equation (3.4) as following

B (Ah ) 1 

3 (3.4)

Q Exp

The polarization voltage (K) can be calculated using equation (3.5) as following

K

 E Full  E Nom  A (exp(B .Q Nom )  1)  (Q Q Nom ) (3.5)

Q Nom

Where

46

E Nom is the end of the nominal zone voltage , Q Nom is the charge of the end of the nominal zone. The voltage constant E0 of the battery can be determine using the equation (3.6) as following

Eo  E full  K  r .i  A

(3.6)

The calculation of the parameters is approximately and depends on the precision of obtaining the points from the discharging curve. The parameters of the adopted 12volt (VRLA) battery are fed to the model of the battery on MATLAB to be used in the simulation study.

3.5.2 Modeling of the supercapacitor

The SC model is implemented in Simpower systems (SPS) based on the Stern model, which combines the models of Helmholtz and Gouy–Chapman [42]. The model of the SC is shown in Figure 3. 13.

RSC

VT 

+ -

+

Controlled voltage source

VT

iSC

NsQT d 2 NeN s RT QT  arsinh( ) 2 N p Ne 0 Ai F N p Ne Ai 8RT  0c

VSC QT



t

0

Figure 3. 13: Model of the SC

The capacitance (C) of the SC can be determined using equation (3.7) as following

 1 C  CH



  CGC  1

1

(3.7)

47

Where: CH is the Helmholtz capacitance (farads), CGC is the Gouy-Chapman capacitance (farads) can be calculated using equations (3.8), (3.9)

CH  CGC 

Ne 0 Ai d

(3.8)

FQc Qc sinh( 2 ) 2 Ne RT Ne Ai 8RT  0c

(3.9)

Where Ne is the number of electrode layers,  is the permittivity of electrolyte material,  0 is the permittivity of free space (F/m), Ai is the interfacial area between the electrodes and the electrolyte (m2), d is the molecular radius (m), F is Faraday constant = 96485.3383, Qc is the electric charge of one SC cell (C), R is the ideal gas constant = 8.314472, T is the operating temperature in kelvin, c is the molar concentration (mol . m-3) The SC module has a number of SC cells connected in series Ns and a number of SC cells connected in parallel Np so that the total capacitance of the supercapacitor module can be calculated from equation (3.10)

CT 

Np Ns

.C

(3.10)

The total electric charge of the Supercapacitor QT coulomb can be calculated from equation (3.11) as follow

QT  N pQc   iSC dt

(3.11)

. The output voltage of the super capacitor can be calculated using equation (3.12) as following:

VSC 

QT  iSC .RSC CT

(3.12)

48

Where Vsc supercapacitor output voltage (V), QT is the total electric charge (C), iSC is the SC current (A), RSC is the resistance of SC module (ohm). In the simulation study, a supercapacitor module Maxwell BMOD0083-P048 is used. The parameters of the SC are fed to the SC model in MATLAB as following: Rated voltage = 48 V, rated capacitance = 83 F, equivalent series resistance = 0.01 ohms, Ns= 18 capacitors, Np capacitors = 1, leakage current = 0.003A. The discharge current profiles of the supercapacitor in data sheet are verified with the discharge current profiles of the simulated discharge current profiles using the model of supercapacitor as shown in Figure 3. 14. It can be noticed that the simulated characteristics of the SC are very closed to the datasheet characteristics.

(a)

SC Voltage (V)

50

40A

40 100A

30

200A

20

1000A

10 0 -1 10

0

1

10

10

2

10

Time (s) (b)

Figure 3. 14: Current discharge profiles (a) data sheet profiles, (b) simulated profiles. 49

Chapter 4: Proposed Energy Management controller for a Microgrid System and simulation Results

4.1 Introduction

This chapter introduces the simulation results of the proposed energy management controller. In this chapter a micro grid is adopted to overcome the effects of faults which may occur in the PV source. The block diagram of the proposed energy management controller with the adopted micro grid is shown in Figure 4. 1. The adopted micro grid integrates the PV source, battery and supercapacitor (SC) as shown in Figure 4. 2. This micro grid is controlled with the proposed control strategy to be able to supply the load with its demanded power at normal operation and at fault conditions without affecting on the load demand. So that it is very important to add energy storage systems (ESS) into the grid to decrease the problems caused by the variable output of the renewable energy sources [44]. The ESS can save the surplus power from the PV source when the sun is available and the loads are light to be used at the peak loads periods; So that there is no power dissipation and all the power extracted from the PV source will be consumed by the load or stored in the ESS. Generally VRLA battery is the widely used energy storage technology in PV systems because of its low cost and wide availability [45]. The batteries have high energy density so that it can supply the load for long time. Usually, the battery life cycle is short. This is because of the chemical reactions nature of the battery. On the other hand SC has the advantages of fast charging and discharging of the energy and a longer life cycle, because of the capacitor electrostatic nature rather than chemical reactions in the batteries.

50

DC Bus

DC-DC boost converter

Super capacitor

PWM

Control strategy MPP tracking controller Energy management controller DC Bus voltage

Battery

Supercapacitor current (Isc)

Battery current(Iba)

PV current and voltage (Ipv,Vpv)

PV

DC Load PWM

PWM

Bidirectional Buck boost converter

Bidirectional Buck boost converter

Figure 4. 1: Block diagram of the proposed micro grid and its control strategy

51

PV Panel

Boost converter

DC Bus + -

SW1

Battery

SW2

SW3

Bidirectional converter

DC Load

Super Capacitor

SW4

SW5

Bidirectional converter

Figure 4. 2 Proposed Microgrid system

The SC can be charged and discharged more than one million times with its life time [46]. Therefore battery-supercapacitor hybrid energy storage system ESS will take advantages of both technologies and provide high power and energy density as in Figure 4. 3. This hybrid system will increase the efficiency of the system by decreasing the stress on the battery which represents the low power density storage system at transient conditions. So that the life time of the energy storage system will increase, and the load power will delivered rapidly.

52

To ensure the optimum operation of the PV source and increase the efficiency of the system a maximum power point tracking (MPPT) technique is developed in this micro grid. Also an energy management controller (EMC) is proposed to control the power flow between each component of the micro grid. EMC decreases the stress on the battery at sudden load variations by charging or discharging the SC until the battery reaches to its steady state and supply the load then the SC goes out. Also it decreases the voltage change at transient periods.

8

10

Power Density W/Kg

7

10

6

10

SC

5

10

4

SMES

10

Flywheel

3

10

Batteries

2

10

1

10

1

10

100

1000

Energy Density Wh/Kg Figure 4. 3: Power density versus energy density ranges for different energy storage system technologies [47]

53

4.2 System Configuration

The adopted smart micro grid is simulated using MATLAB Simpower systems (SPS) as shown in Figure 4. 4. The system consists of IS4000P PV module, connected to the DC bus through a boost converter, DC load represented by a controlled current source, two valve regulated lead acid batteries (YC50-12 VRLA) each one 50Ah, 12 volt and connected in series. This battery pack connected to the DC bus through a bidirectional buck boost converter and Maxwell Supercapacitor module 48 volt 83 Farad connected to the DC bus through buck boost bidirectional converter. Different configurations of this micro grid will be studied as following: The first configuration : the photovoltaic source is connected to the load through the boost converter without energy storage systems. The second configuration : the PV is connected with the battery and the load. The third configuration : the PV with the battery and super capacitor. Figure 4. 4 shows the structure of modeling the adopted microgrid in MATlAB Simulink. Each subsystem of the overall system is discussed in details as following.

54

55 Figure 4. 4: MATLAB Simulink model of the proposed system

4.2.1 PV module

A high power PV module IS4000P 300 watt is used in the adopted system. The PV source is connected to the DC bus through a boost converter as shown in Figure 4. 4. The boost converter is used to track the maximum power point by varying the duty cycle of the IGBT switch. The duty cycle variation occurs by using a MPPT technique as will be explained in section 4.3.1.

4.2.2 Boost converter[48] The boost converter in the system is used as the actuator of the MPPT controller which is responsible for locating the maximum power point. The boost converter optimizes matching between the photovoltaic array and battery bank with the load. The simplest circuit topology of DC-DC boost converter is shown in Figure 4. 5. The output voltage of the boost converter (Vo) is larger than the input voltage (Vin) and the output power nearly equals to the input power with some losses. So that the output current (Io) is less than the input current (Ii) and can be determined using equation (4.1). Vl

D

L

+ Vc -

SW

R C

+ Vo

io

il Vin

-

(a)

Vin

R

-

-

C

+ Vo

+ Vc -

R C

il

+ Vo

+ Vc -

Vin

D

+

io

il

io

- Vl

ic

-

ic

+ Vl

(c)

(b)

Figure 4. 5 Topologies of boost converter (a) Boost converter circuit, (b) when SW is OFF, (c) when SW is ON

Vo Iin 1   Vin Io 1  D

(4.1) 56

Figure 4. 5 (b) shows the boost converter topology when the switch SW is closed in this case the energy is saved in the coil (L). The voltage across the coil (Vl) equals to the input voltage (Vin). In case of the switch is opened as depicted in Figure 4. 5 (c), the energy saved in the coil is moved through the diode (D) to the capacitor (C) and the load (R). The voltage across the coil is reversed and added to the input voltage (Vin) so that the output voltage (Vo) is increased [48]. The adopted boost converter in the thesis is designed according to the maximum power of the PV source as following: The input voltage (Vin) to the boost converter is a variable voltage in the range of 10 to 43 volt, the output voltage from the boost converter (Vo)= 45 volt. The maximum output power Pomax =300 W. The switching frequency (fs) = 3 kHz. The maximum output current from the boost converter can be determined by using equation (4.2) I o max 

Po max 300   6.66A Vo 45

(4.2)

The minimum load resistance can be determined by using equation (4.3)

R l min 

Vo 45   6.75 I o max 6.66

(4.3)

The minimum output current from the boost converter can be determined from equation (4.4) as following:

I o min  5%Iomax  0.05*6.66  0.333 A

(4.4)

The maximum load resistance can be determined by using equation (4.5)

R L max 

Vo 45   135.13 I o min 0.333

(4.5) The minimum duty cycle of the boost converter can be determined by using equation (4.6) D min  1 

V i max 43  1  0.044 Vo 45

(4.6)

57

The maximum duty cycle of the boost converter can be determined by using equation (4.7)

D max  1 

V i min 10  1  0.77 Vo 45

(4.7)

The minimum inductance of the boost converter coil can be calculated using equation (4.8)

Lmin 

R L max * D max 135.13*0.77 (1  D max )2  (1  0.77) 2  914 H 2f s 2*3000

(4.8)

The value of the inductance must be greater than Lmin so that in our study a coil of 5 mH is used. To calculate the minimum value of the capacitor we can use equation (4.9)

C min 

D max *Vo 0.77*45   760 F f s * R L min *Vr 3000*6.75*5%*45

The capacitor which is used in our study is 2200  F

58

(4.9)

4.2.3 Battery with bidirectional converter

Two Valves regulated lead acid batteries (12 volt 50 Ah) connected in series, are used in the adopted micro grid. The battery back energy storage system is connected to the DC bus through a bidirectional buck boost converter as shown in Figure 4. 6. The bidirectional converter is used to stabilize the DC bus voltage at the reference voltage. When the DC bus voltage decreases below the reference value the bidirectional converter is controlled to discharge the battery and boosting the battery voltage from 24 V to the reference voltage 50 V. Also when the DC bus voltage increases than the reference voltage the bidirectional converter is controlled to charge the battery and step down the DC bus voltage to the battery charging voltage.

Figure 4. 6 Battery with buck boost bidirectional converter

59

4.2.4 Supercapacitor (SC) with bididrectional converter

In this simulation study a Super-capacitor module Maxwell BMOD0083-P048 is used. The SC is connected to the DC bus through a bidirectional buck boost converter as shown in Figure 4. 7. The bidirectional converter operates as a boost converter when discharging the SC to increase the voltage of the SC from 48 volt to 50 volt which is reference voltage of the DC bus. The above IGBT1 of bidirectional converter takes the inverse PWM of the PWM of IGBT2.

Figure 4. 7 Super-capacitor with buck boost bidirectional converter

4.3 The Control strategy The control strategy is divided into two main control units [58]. (1) The first control unit is responsible for MPPT and (2) the energy management controller is the second control unit. Each control unit will be explained in detail as following:

60

4.3.1 The first control unit: Maximum power point tracking (MPPT) controller of the DC-DC boost converter

The photovoltaic module has a low efficiency, so that it is very necessary to extract the maximum available power from the PV module to increase the efficiency of the photovoltaic system. Maximum power point tracking controller is used to force the operating power of the PV module to be at the maximum power point (MPP). A DC to DC converter which is interfacing between load and module to transfer the maximum power from the PV module to the load. The load impedance seen by the PV source is varied by changing the duty cycle of the DC-DC converter, to match the peak power point of the PV source. There is many techniques which are used to track MPP such as perturb and observe (P&O), incremental conductance, fuzzy, neural, short circuit current and open circuit voltage as mentioned in [49], [50], [51].

4.3.1.1 Modified P&O algorithm: P&O algorithm is widely used because of its simple implementation. In conventional P&O the PV voltage (Vpv) and PV current (Ipv) are measured. Then the PV power (Ppv) can be calculated. The PV operating voltage is disturbed by varying the duty cycle of the boost converter and observing the change in voltage dV and the change in power dP as shown in Figure 4. 8 [52], [53].

In the modified P&O algorithm the PV characteristics (Ipv) and (Vpv) are measured. From these measured values the PV power (Ppv) can be calculated. Then the operating voltage (Vpv) is increased by a small decrease (dD) in the duty cycle of the boost converter. Then the change in power (dP) and the change in current (dI) are observed. 

If (dP/dI) is negative so that the operating point lies in the left half side of the maximum power point (MPP) as shown in Figure 4. 9 . The perturbation of the

61

operating voltage will be in the same direction of increasing the operating voltage by decreasing the duty cycle of the boost converter. 

If (dP/dI) is positive the operating point lies at the right half side of the MPP. The operating voltage should be decreased to catch the MPP.



If (dP/dI = 0) the MPP is obtained and the boost converter is controlled to operate at this point. The flowchart of the modified P&O algorithm is shown in Figure 4. 10 . The implementation of the modified P&O algorithm in MATLAB is shown in figure 4.11. The output duty cycle of the modified P&O algorithm is shown in Figure 4. 12.

Measure Vpv and Ipv

Calculate Ppv = Ipv*Vpv

dp= Ppv(k) - Ppv (k-1) dV=Vpv(k)-Vpv(k-1)

Yes

NO

dP>0 Yes

NO

dV>0

dV>0

D(k)= D(k-1) - dD

Yes

NO

D(k) = D(k-1) + dD

D(k) = d(k-1) + dD

Figure 4. 8 Conventional perturb and observe (P&O) algorithm

62

D(k)= d(k-1) - dD

10

PV Current (A)

dP/dI=0 8

dP/dI0 4 2 0

0

10

20

30

40

50

60

400

PV power (W)

dP/dI=0 dP/dI0

0

0

10

20

30

40

50

60

PV Voltage (V)

Figure 4. 9: Conditions of (dP/dI) by varying the operating voltage to track MPP.

Figure 4. 10 Modified Perturb and observe flow chart

63

Figure 4. 11 Implementation of Modified Perturb and observe in MATLAB

0.3 0.25

Duty

0.2 0.15 0.1 0.05 0

0

0.05

0.1 Time (sec)

0.15

Figure 4. 12 : The output duty cycle of modified P&O

64

0.2

4.3.1.2. Fuzzy logic based maximum power point tracking algorithm

This algorithm is based on fuzzifying the Perturb and observe algorithm [54]. The fuzzy logic controller consists of three blocks: fuzzification, Fuzzy rules and inference engine, and finally Defuzzification process.

4.3.1.2.1. Fuzzification

The inputs and outputs of fuzzy must be expressed in linguistic variables to build the fuzzy rules. The input variables of FLC are dP and dI dP = Ppv(k) − Ppv(k-1)

(4.2)

dI = Ipv(k) − Ipv(k−1)

(4.3)

The output variable is dD

dD=D(k)-D(k-1)

(4.4)

Where dP is change in PV power, dI is the change in PV current and dD change in duty cycle of the boost converter. Figure 4. 13 shows the memberships function of input and output fuzzy sets. Each fuzzy set has four membership functions as following {PB (Positive Big), PS (Positive Small), NS (Negative Small) and NB (Negative Big)}.

65


1

NB

NS PS

PB

0.8 0.6 0.4 0.2 0


-5

1

-4

-3

-2

-1

NB

0 dp NS

1

2

3

4

PS

5

PB

0.8 0.6 0.4 0.2 0


-0.1

1

-0.08

-0.06 -0.04

NB

-0.02

0 di

NS

0.02

0.04

PS

0.06

0.08

0.1

PB

0.8 0.6 0.4 0.2 0 -3

-2

-1

0 dD

1

2

3 -3

x 10

Figure 4. 13 Inputs and output fuzzy sets of the fuzzy logic controller

4.3.1.2.2 Fuzzy rules

Fuzzy rules are built depending on the knowledge base from the system behavior. The fuzzy rules are 16 IF THEN rules as shown in Table 4: 1. These fuzzy rules are fed to the rule editor in the fuzzy logic controller block in MATLAB as shown in Figure 4. 14.

66

Table 4: 1 Fuzzy Rules

dI

PB

PS

NS

NB

PB

PB

PB

NB

NB

PS

PS

PS

NS

NS

NS

NS

NS

PS

PS

NB

NB

NB

PB

PB

dP

Figure 4. 14: Rule editor of fuzzy logic controller block in MATLAB 67

4.3.1.2.3 Defuzzification The process of Defuzzification calculates the crisp output of the FLC. The center of gravity method is adopted. It finds the point where a vertical line would slice the aggregate set into two equal masses. Mathematically this Centre of gravity (COG)[55] can be expressed as following by equation (4.5): 4

dD 

u  i 1 4

i

 i 1

i

(4.5) i

Where dD is the crisp output of FLC, i is the number of membership functions in the output fuzzy set,  i is value level of the output membership functions, and ui is the center of max- min composition of the output membership functions. The FLC have multiple degrees of freedom to compute duty cycles of different values to track the MPP rapidly with small oscillations. The fuzzy logic MPPT controller is built using MATLAB Simulink as depicted in Figure 4. 15. The output duty cycle of the fuzzy logic MPPT controller is shown in Figure 4. 16. It can be noticed from figures (Figure 4. 12, Figure 4. 16 ) that the duty cycle of FLC has lower oscillations than modified P&O.

Figure 4. 15: MATLAB Simulink structure of fuzzy MPPT algorithm 68

0.25

Duty

0.2 0.15 0.1 0.05 0 0

0.05

0.1 Time (sec)

0.15

0.2

Figure 4. 16: The output duty cycle of fuzzy logic MPPT algorithm

4.3.2 The proposed Energy management controller

The energy management controller is the second control unit. It is responsible for stabilizing the output voltage at the reference voltage (Vref). Also it is responsible for charging or discharging the energy storage systems. The energy management control unit consists of one outer voltage control loop and two inner current control loops as shown in Figure 4. 17.

69

Figure 4. 17: Proposed energy management controller block diagram

The first current control loop is to control the charging or discharging current of the battery. The second current control loop is to control charging or discharging current of the SC. The structure of the proposed energy management controller in MATLAB Simulink is shown in Figure 4. 18. The simulation results will clarify and prove this explanation. The operation of charging or discharging energy storage systems occurs according to the load demand and the environmental conditions; because the environmental conditions have a great effect on the available PV output power. The parameters of the PI controllers which are used to control the battery and SC bidirectional converters are estimated by using try and error method. The proposed energy management controller will enable charging or discharging the SC only at transient conditions. The battery will charge or discharge at steady state conditions. The SC charges when the voltage of the DC bus (Vact) is increased suddenly than Vref, to absorb the transient voltage. In this case the SC bidirectional converter operates in a buck mode and sw4 is conducting and charges the SC. When the voltage becomes stable the battery is charged and the bidirectional converter of the battery works in buck mode by conducting SW2.

70

PWM SC

Figure 4. 18: the structure of the proposed energy management strategy in MATLAB.

At steady state SC go out and didn’t charge or discharge. When the voltage of the dc bus suddenly drops below Vref; the SC will discharge rapidly to compensate the voltage drop and supply the load with the needed power. Then after the output voltage reaches to its steady state the controller enables the battery only to supply the load. In this case the converter of the battery operates in boost mode by conducting SW3. The SC did not supply the load until the following transient conditions. In the case of discharging; the bidirectional converter works as boost converter and sw5 is conducting. This control strategy will decrease the stress on the battery and hence will increase the life time of the battery and the efficiency of the system.

4.4 Simulation Results Different cases of the environmental conditions and load variations are studied by using the proposed control strategy. The studied cases are as following: case (A) Shadow occurs on

71

the PV source when the load is fixed at specified current value, case (B) The insolation is Constant and a step change in the load demand occurs at a certain time. These cases are studied at three different configurations of the system as following: 

Config1: PV only without energy storage,



Config2: PV with battery,



Config3: PV with battery and super capacitor.

The two cases (A) and (B) are studied using two algorithms of Maximum power point tracking (MPPT). The first MPPT algorithm is the modified P&O MPPT and the second is the fuzzy logic MPPT. The simulation results have been discussed in details as following:

4.4.1 Case A: Shadow occurs on the PV module and the load is constant at 5A. In this case the load absorbs a fixed current 5A. Initially the insolation (G) = 1000 w/m2 and the temperature (T) = 25oC then a shadow of 300w/m2 occurs suddenly on a third of the PV module at time = 4 sec. This case has been studied by using the two techniques of MPPT. Three configurations of the micro grid are studied under this case to show fault effect on each configuration.

4.4.1.1 Config1: Using PV module without energy storage.

In this configuration the grid consists of a PV source with the boost converter and the MPPT controller without energy storage. In this configuration all the extracted power from the PV source must be absorbed by the load. Initially the output power from the PV source equals to the maximum power (300 watt) at constant insolation G = 1000 W/m2, T =25 oC as shown in Figure 4. 19. The load power equals to PV power = 300 W. The load voltage oscillates around 58 V as shown in Figure 4. 20. Suddenly a shadow occurs at time = 4 sec the load voltage drops from 58 V to 36 V. Also the output power from the PV module and the load power drops from 300 W to 193 W. The load voltage is unregulated at constant output voltage, but it varied according to the environmental conditions and the load 72

demand. Because there is neither controller on the output voltage nor storage systems to substitute the shortage in power or save the surplus PV power. Load power and PV power (w) 400 Load Power PV Power

power (w)

300 200 100 0

0

1

2

3

4 time (s)

5

6

7

8

(a) Load power and PV power (w) 400 Load Power PV Power

power (w)

300 200 100 0

0

1

2

3

4 time (s)

5

6

7

8

(b) Figure 4. 19: Case (A) config1 load power and PV power a) FLC MPPT, b) modified P&O MPPT.

73

The fuzzy logic is used to track the maximum power point in Figure 4. 19 (a) while in Figure 4. 19(b) the modified P&O is used. Also Figure 4. 20(a) shows the load voltage by using FLC MPPT and Figure 4. 20(b) shows the load voltage by using modified P&O. From the two figures we can notice that the results are nearly the same but fuzzy has lower oscillations than P&O.

Load Voltage (v)

Volt (v)

60

40

20

0

0

1

2

3

4 time (s)

5

6

7

8

(a)

Load Voltage (v)

Volt (v)

60

40

20

0

0

1

2

3

4 time (s)

5

6

7

8

(b)

Figure 4. 20: Case (A) config1 load voltage a) FLC MPPT , b) modified P&O MPPT. 74

The characteristic curves of the PV module while tracking the MPP are shown in Figure 4. 21. In case of the normal operation the maximum power point oscillates around 300 W. Then when the shadow occurs the maximum power drops to the local MPP =100 W because this local maximum power point lies at the maximum operating voltage of the normal conditions; but at this instant the MPPT controller scans the I-V curve until reaches to the global MPP = 193 W and the operating voltage 24.35 V. from this result we can notice that the MPPT controller can track the global maximum and neglect the local maximum.

I-V Curve 10

Current (A)

8 6 4 2 0

0

5

10

15

20 25 Voltage (v)

30

35

40

45

30

35

40

45

(a) P-V Curve

Power (w)

300

200

100

0

0

5

10

15

20 25 Voltage (v)

(b) Figure 4. 21 : PV module characteristics (a) I-V curve, (b) P-V curve 75

4.4.1.2 Config2: PV with battery.

This configuration integrates the PV source with battery pack. The output voltage is controlled by using the proposed energy management controller. In this configuration the load power equals to 250 W and the output voltage is controlled at reference voltage (Vref = 50 V). Initially the solar insolation G = 1000 W/m2 and T= 25oC. The maximum PV output power has been tracked by using FLC algorithm as shown in Figure 4. 22(a), and by using modified P&O algorithm as shown in Figure 4. 22(b). The available maximum output power from the PV = 300 W is larger than the load power = 250 W. The battery is charging with surplus power from the PV (300-250 = 50 W). Suddenly a shadow 300 w/m2 occurs at time = 4 sec on a third of the PV module. The PV power drops from 300 watt to100 watt then oscillates around 193.4 watt, which is the global MPP (GMPP). The load power is greater than the available MPP from the PV. The battery is controlled to be discharged to substitute the shortage in the load power and the load power is maintained constant at 250 W as shown in figure 4.22. The load voltage is controlled to be constant at the reference voltage 50 volt as shown in Figure 4. 23. The proposed energy management controller controls the charging and discharging of the battery and stabilizes the output voltage by generating the suitable duty cycle to the bidirectional converter as shown in Figure 4. 24. While the first period (from time 0 to 4 sec) the bidirectional converter works as a buck converter to charge the battery pack with the surplus power from the PV.

76

Load power and PV power (w) 300

power (w)

200

PV power Load power

100

Battery power

0

-100

0

1

Load Power PV Power Battery power 2

3

4 time (s)

5

6

7

8

(a) 300

PV power

power (w)

200

Load power 100

Battery power

Load Power PV Power Battery power

0

-100

0

1

2

3

4 time (s)

5

6

7

8

(b)

Figure 4. 22: Case (A) config2 load power, PV power and battery power (a) FLC MPPT, (b using modified P&O MPPT.

77

Load Voltage (v)

Volt (v)

60

40

20

0

0

1

2

3

4 time (a) (s)

5

6

7

8

Load Voltage (v) 60

Volt (v)

50 40 30 20 10 0

0

1

2

3

4 time (s) (b)

5

6

7

8

Figure 4. 23: Case (A) config2 load voltage. (a) using FLC MPPT, (b)using modified P&O MPPT.

The duty cycles of the bidirectional converter of the battery is designed at a reference value. The reference duty cycle of the buck converter (D) which is fed to IGBT1 is shown in Figure 4. 6 and can be determined by using equation (4.6)

D

Vo V in

(4.6)

78

Where the output voltage of the buck converter is the battery charging voltage =24 volt and the input voltage is the DC bus voltage 50 volt. So that the reference duty cycle of the buck converter is calculated as following:

D

Vo 24   0.48 Vin 50

Also the reference duty cycle of the boost converter which is fed to the IGBT2 shown in Figure 4. 6 can be determined by using equation (4.1). Where the output voltage at this time is the DC bus voltage =50 volt and the input voltage of the boost converter is the battery pack voltage =24 volt. So that the reference value of the duty cycle of the boost converter is calculated as following: D  1

Vin 24  1  0.52 Vo 50

The bidirectional converter operates as buck converter when the duty cycle of the (IGBT1) shown in Figure 4. 6 is greater than its reference value (D >0.48). The bidirectional converter operates as a boost converter when the duty cycle of the (IGBT2) is greater than the reference duty cycle of the boost converter (D > 0.52).

It can be noticed from Figure 4. 24 in the first period the bidirectional converter operates as buck converter. So that the duty cycle of the buck converter is greater than its reference value 0.48. This is to generate the output charging voltage greater than the voltage of the battery pack 24 V. On the other hand the duty cycle of the IGBT2 is lower than its reference value 0.52.

At the second period (from 4 sec to 8 sec) the duty cycle of the IGBT2 becomes greater than its reference value 0.52, and the duty cycle of the IGBT1 becomes lower than its reference value so that the bidirectional converter works as a boost converter. The battery pack is controlled to be discharged. Figure 4. 25 shows the state of charge (SOC) of the 79

battery while increasing from 92.73% to 92.733% at the first period this means that the battery is charging. Then the SOC decreases to 92.7275% at the second period this means that the battery discharging. It can be noticed from figures (Figure 4. 22 to Figure 4. 24 ) that when using fuzzy MPPT the oscillations decrease in load power, load voltage and the duty cycle of the battery.

Duty of Battery converter 1 D buck Batt D boost Bat

Duty

0.8 0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

(a) Duty of Battery converter 1 D buck Batt D boost Bat

Duty

0.8 0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

Figure 4. 24 Case (A) Config2 Duty cycle of the bidirectional converter of the battery (a) using Fuzzy MPPT, (b) using modified P&O

80

Battery SOC 92.734

SOC(%)

92.732 92.73 92.728 92.726

0

1

2

3

4 time (s)

5

6

7

8

Figure 4. 25 : Case (A) Config2 State of charge of the battery pack

4.4.1.3 Config3: PV with battery and SC.

This configuration integrates the PV source with battery pack and SC. The output voltage has been controlled by using the proposed energy management controller. In this case the load demand = 250 W and the output voltage is controlled at 50 volt. The solar insolation G= 1000 w/m2 and T= 25oC in the first period (from 0 to 4 sec). The maximum PV output power is tracked by using FLC as shown in Figure 4. 26 (a), and the MPP is tracked by using the modified P&O as shown in Figure 4. 26 (b). At the first period the available maximum output power from the PV = 300 W is larger than the load power = 250 W. The battery charges with the surplus power from the PV at steady state.

But at starting time (t=0) the PV power is zero and begin to build up until reaching the MPP. The DC bus voltage equal zero at this instant. The SC will provide the demanded power to stabilize the output voltage at the reference value. When the PV power reaches to its MPP and the DC bus voltage reaches to its reference value. The battery charges with the surplus power and the SC go out. Suddenly a shadow of 300 W/m2 occurs at the second period (from 4 sec to 8 sec) on the third of the PV module. The PV power drops from 300 W to 100 watt, which is the local 81

MPP, then back to 193.4 watt which is the global MPP. At this instant the SC will discharge rapidly to substitute the shortage in load power, until the battery reaches to its steady state of discharging, and then the SC does not supply power.

The proposed energy management controller has the ability to share the power between the power elements in this configuration and decreases the stress on the battery. Also the voltage change at transient conditions is reduced greatly as shown in Figure 4. 27. Load power and PV power (w) 300 200

PV power

power (w)

load power 100 0 -100 -200

SC power

battery power

0

1

2

3

4 time (s)

5

6

7

8

6

7

8

(a) Load power and PV power (w) 300

power (w)

200

PV Power

Load power

100 0 -100 -200

battery power SC power 0

1

2

3

4 time (s)

5

(b) Figure 4. 26: Case (A) config3 load power, PV power, battery power and SC power (a) using FLC MPPT, (b) using modified P&O MPPT

82

Load Voltage (v) 60

Volt (v)

50 40 30 20 10 0

0

1

2

3

4 time (s)

5

6

7

8

(a)

Load Voltage (v) 60

Volt (v)

50 40 30 20 10 0

0

1

2

3

4 time (s)

5

6

7

8

(b) Figure 4. 27: Case (A) config3 load voltage. (a) using FLC MPPT, (b) using modified P&O MPPT.

The duty cycle of the battery converter is shown in Figure 4. 28, and the duty cycle of the supercapacitor bidirectional converter is shown in Figure 4. 29. The duty cycle of the bidirectional converter of the SC is designed at reference value when it operates as a buck converter. Where the output voltage of the buck converter at this case = 48 volt and the input voltage is the DC bus voltage = 50 V. the reference duty cycle =48/50 = 0.96.

83

The reference duty cycle of the boost converter = 1- (48/50) = 0.04. So that at the starting time (t=0) the duty cycle of the boost converter is increased rapidly to discharge power, then oscillates again at its reference value so that SC does not discharge or charge. At time = 4 sec the duty cycle of the boost converter is increased to 0.2 to discharge the SC an substitute the shortage in power as shown in Figure 4. 26, Figure 4. 29. From the simulation results it can be noticed that fuzzy and modified P&O have the same performance.


Duty

0.8 0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

(a) Duty of Battery converter 1 D buck Batt D boost Bat

Duty

0.8 0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

(b)

Figure 4. 28: Case (A) Config3 Duty cycle of the bidirectional converter of the battery (a) using Fuzzy MPPT, (b) using modified P&O 84

Duty of SC Converter 1

Duty

0.8 D buck SC D boost SC

0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

(a)

Duty of SC Converter 1 D buck SC D boost SC

Duty

0.8 0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

(b) Figure 4. 29: Case (A) Config3 Duty cycle of the bidirectional converter of the SC (a) using Fuzzy MPPT, (b) using modified P&O

85

4.4.2 Case B: The insolation is constant and step change occurs in the load demand. In this case the solar insolation is constant at 1000 w/m2 and the temperature is constant 25oC. The load current changes from 7A to 4 A at time = 4 sec. In this case we will study the performance of the system with three configurations as the previous case as following

4.4.2.1 Config1: PV without energy storage. In this configuration the grid consists of a PV source and the MPPT controller without energy storage. In this configuration all the extracted power from the PV source must be absorbed by the load. At the first period the output power of the PV source equals to the maximum power (300 W) at constant insolation G = 1000 W/m2 and T =25 oC. The load power equals to the PV power = 300 W as shown in Figure 4. 31. Suddenly the load current drops from 7 A to 4 A at the second period (from 4 to 8 sec) as shown Figure 4. 30. The output voltage increases from 42.8 V to 75 V as shown in Figure 4. 32 . The load voltage is unregulated at constant output voltage, but it varied according to the environmental conditions and the load demand; because there is neither voltage controller on the output voltage nor energy storage systems to substitute the shortage in power. The electrical characteristics of the PV module during the MPPT are shown in Figure 4. 33. From the simulation results shown in figures (4.31, 4.32) it can be noticed that FLC MPPT and the modified P&O nearly have the same results.

Load Current (A) 8

Current (A)

6 4 2 0

0

1

2

3

4 time (s)

Figure 4. 30: Load current

86

5

6

7

8


power (w)

300 250 200 Load Power PV Power

150 100 50

0

1

2

3

4 time (s)

5

6

7

8

(a)


power (w)

300 250 200

Load Power PV Power

150 100 50

0

1

2

3

4 time (s)

5

6

7

(b)

Figure 4. 31. Case (B) config1 Load power and PV power (a) using FLC MPPT, (b) using modified P&O MPPT

87

8

Load Voltage (v) 100

Volt (v)

80 60 40 20 0

0

1

2

3

4 time (s)

5

6

7

8

(a)

Load Voltage (v) 80

Volt (v)

60 40 20 0

0

1

2

3

4 time (s)

5

6

7

8

(b)

Figure 4. 32. Case (B) config1 load voltage (a) using FLC MPPT, (b) using modified P&O MPPT.

88

P-V Curve 300

Power (w)

250 200 150 100 50 0

0

5

10

15

20 25 Voltage (v)

30

35

40

45

(a)

I-V Curve 10

Current (A)

8 6 4 2 0

0

5

10

15

20 25 Voltage (v)

30

35

40

45

(b)

Figure 4. 33: characteristic curve of the PV module (a) P-V curve, (b) I-V curve

89

4.4.2.2 Config2: PV with battery.

In this configuration the battery is used as the energy storage source to store the surplus power from the PV source. The output voltage is controlled by using the proposed energy management controller at Vref = 50 V. Initially ( at the first period from 0 sec to 4 sec) the load power = 350 W and the output voltage is controlled at 50 V. The maximum PV output power = 300 W at constant G = 1000 W/m2 and T= 25oC. The maximum PV output power is tracked by using FLC algorithm as shown in Figure 4. 34 (a), and by using the modified P&O algorithm as shown in Figure 4. 34 (b). The available maximum output power from the PV = 300 W is lower than the load power = 350 W so that the battery is controlled to discharge to substitute the shortage in PV power = 50 W. Suddenly the load demand drops to 200 W at time = 4 sec. The excess in PV power = 100 W will be supplied to the battery (charging the battery). The simulation results show that the battery is forced to be charged and discharged rapidly which affects badly on its life time. Also the proposed energy management controller succeeds in maintaining the output voltage at its reference value in spite of the load variation as shown in Figure 4. 35.

The duty cycle of the bidirectional converter of the battery back is shown in Figure 4. 36. In the first period it can be noticed that the duty cycle of the boost converter is larger than its reference value so that the bidirectional converter operates as a boost converter in this period and discharging the battery pack. At the second period the duty cycle of the boost converter is decreased than its reference value and the duty cycle of the buck converter is increased than its reference value, so that the bidirectional converter operates as a buck converter and charging the battery pack. The state of charge of the battery pack is shown in Figure 4. 37. It can be noticed from that the battery state of charge decreases from 92.73 % to 92.724% because the battery is discharging at the first period; and at the second period the battery state of charge increases from 92.724 to 92.732% because the battery is charging.

90


power (w)

300 200 Load Power PV Power Bat Power

100 0 -100 -200

0

1

2

3

4 time (s)

5

6

7

8

(a)

400

power (w)

300 200 Load Power PV Power Bat Power

100 0 -100 -200

0

1

2

3

4 time (s)

5

6

7

8

(b) Figure 4. 34: Case (B) config2 load power, PV power, and battery power. (a) using FLC MPPT, (b) using modified P&O MPPT

91

Load Voltage (v) 60

Volt (v)

50 40 30 20 10 0

0

1

2

3

4 time (s) (a)

5

6

7

8

Load Voltage (v)

Volt (v)

60

40

20

0

0

1

2

3

4 time (s)

5

6

7

8

(b) Figure 4. 35: Case (B) config2 load voltage (a) using FLC MPPT, (b) using modified P&O MPPT

92


Duty

0.8 0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

(a)


Duty

0.8 0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

(b) Figure 4. 36: Case (B) config2 duty cycle of the bidirectional converter of the battery (a) when using FLC MPPT (b) when using modified MPPT

93

Battery SOC 92.732

SOC(%)

92.73

92.728

92.726

92.724

0

1

2

3

4 time (s)

5

6

7

8

Figure 4. 37: Case (B) config2 state of charge of the battery pack

4.4.2.3 Config3: PV with battery and SC. This configuration integrates the PV source with battery pack and the SC. The output voltage is controlled by using the proposed energy management controller. At the first period the load demand is 350 W and the output voltage is controlled at Vref = 50 V. The maximum PV output power is tracked at solar insolation G= 1000 W/m2 and T= 25oC using FLC as shown in Figure 4. 38 (a), and by using modified P&O as shown in Figure 4. 38 (b). The available maximum output power from the PV = 300 W is smaller than the load power = 350 W. At starting time the SC will provide power to the load to stabilize the output voltage at the reference voltage. When the PV output power reaches to its MPP and the battery reaches to its steady state, the battery pack supplies the load with the difference in power and the SC goes out as shown in Figure 4. 38. At the second period (from 4 sec to 8 sec) a step decrease in load power from 350 W to 200 W occurs. So that the SC charges immediately to absorb the surplus power from the PV source to maintain the output voltage at Vref as shown in Figure 4. 39 until the battery reachs to its steady state of charging. Then the SC goes out (did not charge) and the battery charging at the steady state. The duty cycle of the bidirectional converter of the battery is shown in Figure 4. 40. The duty cycle of the bidirectional converter of the SC is shown in Figure 4. 41. The state of charge of the battery and the SC is shown in Figure 4. 42. It can be noticed from this figure that the battery discharges and charges at transient periods slower than in config2 So that the battery operates without stress.

94

From the simulation results it can be noticed that the FLC MPPT and the modified P&O have nearly the same performance. Also the voltage change decreases than voltage change in config2 because of using the proposed energy management controller and the SC. 400

power (w)

300 Load voltage

200

PV power

100 0 Battery power

-100 -200

0

1

SC power

2

3

4

5

6

7

8

4

5

6

7

8

(a) 400

power (w)

300 Load power

200

PV power

100 0 Battery power

-100 -200

0

1

SC power 2

3

(b) Figure 4. 38 Case (B) config3 load power, PV power, battery power and SC power (a)using FLC MPPT (b) using modified P&O MPPT.

95

Load Voltage (v) 60

Volt (v)

50 40

50.6 50.4 50.2 50 49.8 49.6

30 20 10 0

0

1

2

3

1.5

4 time (s)

2 5

2.5 6

7

8

(a)

Load Voltage (v) 60

Volt (v)

50 50.6 50.4 50.2 50 49.8 49.6

40 30 20 10 0

0

1

2

3

1.5

4 time (s)

2 5

2.5 6

7

8

(b) Figure 4. 39. Case (B) config3 load voltage (a) using FLC MPPT , (b) using the modifed P&O MPPT.

96


Duty

0.8 0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

(a)


Duty

0.8 0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

(b) Figure 4. 40 Case (B) Config3 Dty cycle of the bidirectional converter of the battery pack (a) when using FLC MPPT (b) using the Modified P&O MPPT

97

Duty of SC Converter 1

Duty


0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

(a) Duty of SC Converter 1

Duty


0.6 0.4 0.2 0

0

1

2

3

4 time

5

6

7

8

Figure 4. 41: Case (B) Config3 the duty cycle of the bidirectional converter of the SC (a) when using FLC MPPT (b) using the Modified P&O MPPT

98

Battery and SC SOC 92.78

SOC(%)

92.76

Batt SOC SC SOC

92.74 92.72 92.7

0

1

2

3

4 time

5

6

7

8

Figure 4. 42 state of charge of the Battery and the SC

4.5 Conclusion The modeling and simulation of each element of the adopted micro grid has been introduced. A smart energy management control strategy has been proposed. A modified method for P&O has been used taking the (dP/dI) instead of the (dP/dV). This modified method increased the sensitivity of MPPT especially at partial shadowing faults. The method is compared with FLC based MPPT. From the simulation results it can be noticed that the two algorithms have nearly the same results.

Different configurations of the micro grid have been studied with the proposed energy management controller. The simulation results show that the configuration 3 which integrates the PV source with battery and SC is the best configuration. Because the stress on the battery is decreased which may increase the life time of the battery. Also the amplitude of the voltage change has been decreased in this configuration because of fast charging and discharging of the SC and its high power density.

99

Chapter 5: System Setup and Experimental results 5.1 Overview In this chapter some of the simulated results are verified by the introduced experimental work.Some of these experiments include implementation of the proposed modified P&O MPPT algorithm also including a case study of power sharing between the PV system and the battery bank . the experimental results have been recorded at different environmental conditions and load demands.

5.2 Experimental setup The experimental setup of the adopted microgrid is shown in Figure 5. 1. The block diagram of the system which shows the wiring of each component is shown in Figure 5. 2. The system is composed of the following components:

PC with DS1102

DS1102 card I/O interface

oscilloscope PV current clamp Current Voltage sensor sensor

PLZ1004 DC Electronic load

Boost converter IS4000P PV Module

Load current Clamp

Battery Current Clamp

Figure 5. 1 Overview of the experimental setup.

100

Battery bank

Figure 5. 2: block diagram of the system 

A high power PV module IS4000P is placed on the roof top of the electronics research institute as shown in Figure 5. 2. The specs of the PV module are shown in appendix “B”.



Boost converter.



dsPACE control card DS1102.



PC with the control desk software of the DS1102.



Optocoupler isolation circuit.



Level shifter circuit.



Current sensor.



Voltage sensor.



DC electronic load PLZ1004W.



Oscilloscope.



Digital multimeter.



Battery bank . 101

5.3 System components The output voltage of the IS4000P PV module is measured using the voltage sensor (LV 25-P) as shown in Figure 5. 3. The voltage sensor is calibrated to measure a voltage range from 0 – 45volt. The details of the voltage sensor and its calibration are discussed in appendix “C”. Analog to digital converter (ADC1) is used as an input of the dsPACE DS1102 which reads the output voltage of the voltage sensor (LV 25-P).

Figure 5. 3: Voltage sensor

The output voltage from the voltage sensor is in the range from 0 -10 volt. So that it is very important to find an equation to fit the output signal from the voltage sensor with the measured voltage. The output signal from the voltage sensor is fitted by using the fitting equation to read the real voltage value. The dsPACE also reads 10 volts applied on the ADC externally as 1 volt internally. The input voltage in the program is multiplied by a gain = 10, then fed to the fitting equation of the voltage sensor to read the real value in the program of P&O. The output current of the PV module has been measured by a Hall Effect current sensor as shown in Figure 5. 4. More details of the adopted current sensor and its calibration are shown in appendix “D”. The details of DS1102 are shown in appendix “E”. The ADC4 is used as an input to read the PV output current which is measured by the current sensor. Inside the program it must be multiply the (ADC4) by a gain = 10 to fed the program with 102

real values of the measured current. The used current sensor has no need to calibration equation because it calibrated by winding the cable as shown in Figure 5. 4, so that if the input current = (1 A) the output of the current sensor = (1 Volt).

Figure 5. 4: Current sensor

The modified P&O MPPT algorithm is built using MATLAB Simulink and loaded to the DS1102. The output control signal is obtained by using the DS1102 PWM channel at pin 54 on the I/O board of the DS1102. The output PWM is fed to the isolation circuit which is shown in Figure 5. 5 composed of the 4N33 optocoupler.

Figure 5. 5: Optocoupler isolation circuit The output signal of the optocoupler is fed to the level shifter such as shown in Figure 5. 6. The level shifter is used as the driving circuit of the MOSFET gate. The level shifter increases the input signal from lower value to higher value such as shown in Figure 5. 7. 103

Figure 5. 6: level shifter

I/P to the optocoupler

O/P of the optocoupler

O/P of the level shifter

Figure 5. 7: Driving signal of the MOSFET The output PWM from the level shifter is fed to the gate of the MOSFET transistor of the boost converter to track the MPP as shown in Figure 5. 8. The data sheet of the used MOSFET transistor is shown in appendix “F”.

104

Figure 5. 8: Boost converter

The boost converter is designed to operate at input voltage range from 10 volt to 43 volt . This range of voltage is the PV output voltage operating range. The output voltage from the boost converter is 45 volt. the output current range is 0.33 A to 6.6 A at load resistance range from 136.6 ohm to 6.8 ohm. An electronic load PLZ1004W is used as the load to absorb the PV power as shown in Figure 5. 9. The electronic load specs are shown in appendix “G”.

Figure 5. 9: DC Electronic load PLZ1004W

105

The electronic load is programmed to work in the constant current mode. The constant current mode enables to absorb fixed current at variable voltages; because the current setting can be changed easily by using the rotary knob. The electronic load is used to measure the I-V curve of the PV module as following:     

Connecting the terminals of the PV module on the terminals of the electronic load. Adjust the electronic load to work on the mode of constant current mode. Using oscilloscope to measure the terminal voltage of the PV and the PV current. Adjust the oscilloscope to work on the XY axis mode. Where the voltage is in the X axis and the current is measured on Y axis. Increase the load current from 0 A at the open circuit voltage of the PV module to the short circuit current by rotating the rotary knob of the electronic load which changes the current setting of the electronic load. The output voltage of the PV module decreases when the current increases until reaches to zero at short circuit current.

The measured I-V curve of the PV module is shown in Figure 5. 10(a,b) at normal operation and at partial shadowing. At normal operation the maximum power 188 W, and at partial shadow the maximum power was 145W.

A battery bank consists of three valve regulated lead acid (VRLA 12 volt 50 Ah) batteries connected in series is used as the energy storage unit in the system. The battery specification is shown in appendix “H” . The battery bank will supply the load if there is a shortage in the PV power and also charges if there is a surplus PV power. The battery bank will stabilize the output voltage to its terminal voltage while the operation at 36 volt.

106

(a)

(b) Figure 5. 10: Measured I-V Curve of the IS4000P PV module (a) at normal operation, (b) at partial shadow. Y axis is the PV current measured on channel (B) with deviation (2A) and x axis is the PV voltage measured in channel (D) with deviation (20 volt) and the PV power measured on channel (A).

5.4 Experimental results and discussion The MPPT algorithm is built using MATLAB Simulink package and loaded on the DSP1102 card. Different cases of load demand and environmental conditions have been 107

studied to ensure the effectiveness of the MPPT algorithm and the battery charging as following:

5.4.1 The first case In this case the available maximum power from the PV source is constant at 145W (Vpv = 30.7 V, Ipv = 4.72) as shown in Figure 5. 11. The output power = 126 W (Vo = 42V, Io=3A) the load absorbs 2.42A and the battery absorbs 0.58A. Then the load is disconnected in the second region of the figure 5.11 so that the battery absorbs all the available output current. In the third region the load is connected again and absorbs 2.42A so that the battery will absorb the excess output power. This result shows that the load absorbs its demanded power at any time and the PV output power is still at its maximum value at all times.

Figure 5. 11: The maximum PV power measured on channel (1), the maximum PV current on channel (2), the battery current on channel (3) and the load current on channel (4)

5.4.2 The second case In this case the load is connected with the battery in the first region as shown in Figure 5. 12 the available maximum power tracked at 146 W (Ipv = 4.73A, Vpv = 30.86 V). The 108

output power is 132 W; the load voltage is 42 volt, the load current 2.42 A and the battery charges by a current = 0.74 A. In the second region the battery is disconnected so that the load absorbs all the available output power at constant current 2.42 A by boosting the output voltage to 54.84 Volt. Then the battery is connected again so that the output voltage is regulated at 42 volt and the battery charges by the excess current 0.74 A.

Figure 5. 12 : The maximum PV power measured on channel (1), the maximum PV current on channel (2), the battery current on channel (3) and the load current on channel (4)

5.4.3 The third case This case is recorded at low insolation and the available maximum power obtained at 113 W the PV current is 3.63 A and the voltage is 31.13 V. the output power from the boost 109

converter 100 W because of the losses in the boost converter. The output voltage is 37 V so that the output current from the boost converter is 2.7 A but the load current demand is 3 A the battery will supply the load with 0.3 A. Then the load is disconnected so that the battery will absorb all the available power at current 2.7 A; then the load is connected again and absorbs 0.48 A at the third region so that the battery charging current decreases to 1.81 A as shown in Figure 5. 13. From these experimental results we can notice that the output power from the PV is kept at its maximum power at all conditions of load variation and environmental conditions.

Figure 5. 13 The maximum PV power measured on channel (1), the maximum PV current on channel (2), the battery current on channel (3) and the load current on channel (4)

110

Chapter 6 : Conclusions of the thesis and future work

6.1 Conclusions In this thesis a model of the PV module has been presented to show the electrical characteristics of the PV module under the normal and faulty conditions. Two different fault types in the PV such as partial shadow and short circuit have been studied. An intelligent fuzzy logic fault detection and diagnosis technique of PV systems has been introduced. Also this technique is able to determine the fault type and its percentage. A detailed model of a microgrid composed of a PV source, battery bank and a supercapacitor has been introduced and simulated in matalb simulink software package. A proposed modified P&O algorithm for MPPT has been introduced. This algorithm is compared with fuzzy logic MPPT algorithm and the simulation results showed that fuzzy logic has lower oscillations in the first configuration than the modifed P&O algorithm, and have nearly the same results in the other used configurations. A proposed energy management controller has been introduced to control the power sharing between the PV and the two different energy storage systems (battery and supercapacitor). This energy management controller has the ability to select the appropriate energy storage system to be charged or discharged according to the load demand and the environmental conditions without stress on any storage element, this may help in increasing its life time. Also decreases the voltage change especially at the transient conditions. An experimental case study has been excuted to verify some of the simulation results of the proposed modified P&O algorithm for MPPT.

6.2 Future work  



Implementation of the proposed adopted micro grid with the proposed energy management controller experimentally. Application of different optimization techniques such as genetic algoritm and particle swarm to obtain the global maximum power piont fastly with lower oscillations. Grid connection of the proposed microgrid . 111

References [1] S. J. Chiang, Hsin-Jang Shieh, and Ming-Chieh Chen “A Fuzzy Control Maximum Power Point Tracking Photovoltaic System” IEEE International Conference on Fuzzy Systems, June 27-30, 2011, Taipei, Taiwan [2] Bader N. Alajmi, Khaled H. Ahmed, Stephen J. Finney, and Barry W. Williams, “Fuzzy-Logic-Control Approach of a Modified Hill-Climbing Method for Maximum Power Point in Microgrid Standalone Photovoltaic System” IEEE Transactions On Power Electronics, Vol. 26, No. 4, April 2011 [3] Yixin Zhu, Fang Zhuo, Hongtao Shi “Power Management Strategy Research for a Photovoltaic-Hybrid Energy Storage System” 2013 IEEE ECCE Asia Downunder, pp.842-848 3-6 June 2013 , Melbourne [ 4] Chaouachi, A. ; Kamel, R.M. ; Nagasaka, K., “Microgrid Efficiency Enhancement based on Neuro-Fuzzy MPPT Control for Photovoltaic generator,” in Photovoltaic Specialists Conference (PVSC), 2010, pp. 002889 – 002894. [ 5] A. Arulampalam, N. Mithulananthan, R.C. Bansal, T.K. Saba ,“Micro-grid Control of PV Wind-Diesel Hybrid System with Islanded and Grid Connected Operations” IEEE ICSET, 6-9 Dec 2010, Kandy, Sri Lanka [ 6] Yanfei-Shen, Zhongdong-Yin, Changyin-Zhang, “PV power system energy control research in micro-grid” Asia-Pacific Power and Energy Engineering Conference (APPEEC), 25-28 March 2011 [ 7] E. D. Dorado*, A. S.Garc, C. Carrillo, and J. Cid “Influence of the shadows in photovoltaic systems with different configurations of bypass diodes” International Symposium on Power Electronics, Electrical Drives, Automation and Motion [ 8] H.Ziar, S.Mansourpour, A.Salimi, E.Afjei “Analysis of shadow effect in photovoltaic arrays using binary coding method” 2nd International Conference on Electric Power and Energy Conversion Systems (EPECS), 15-17 Nov. 2011 [ 9] R.Ramaprabha, and B.L.Mathur, “MATLAB based Modelling to Study the Influence of Shading on Series Connected SPVA” Second International Conference on Emerging Trends in Engineering and Technology, ICETET 2009. [10] Kuei-Hsiang Chaoa, Sheng-Han Hob, Meng-Hui Wang , “Modeling and fault diagnosis of a photovoltaic system” Electric Power Systems Research, vol. 78, pp. 97–105, 2008 112

[11] Il-Song Kim; "Fault detection algorithm of the photovoltaic system using wavelet transform", (IICPE), 2010 India International Conference on Power Electronics, pp. 1-6, India, 28-30 Jan. 2011,. [ 12] Xiaoli Xu, Huan Wang, Yunbo Zuo;" Method for Diagnosing Photovoltaic Array Fault in Solar Photovoltaic System", 2011 Asia-Pacific Power and Energy Engineering Conference (APPEEC), pp. 1-5 25-28 March. 2011. [13]Mario González, Bertrand Raison, Seddik Bacha, Long Bun; “Fault diagnosis in a Grid -connected photovoltaic system by Applying a signal Approach”, IECON 2011 - 37th Annual conference on IEEE industerial Electronics society, pp 1354-1359, Melbourne, Australia,7-10 November.2011. [ 14] V. Salas, E. Olias, A. Barrado, A. Lazaro, “Review of the maximum power point tracking algorithms for stand-alone photovoltaic systems,” Solar Energy Materials & Solar Cells, pp.1555–1578, 2006. [ 15] Dalila BERIBER,“MPPT Techniques for PV Systems,” 4th International Conference on Power Engineering, Energy and Electrical Drives,pp.1437-1442, 2013. [ 16] S. J. Chiang, Hsin-Jang Shieh, IEEE, and Ming-Chieh Chen,“Modeling and Control of PV Charger” System With SEPIC Converter” IEEE Transactions On Industrial Electronics, VOL. 56, NO. 11, NOVEMBER 2009.

[17] Huang Zhiqiang, “Research and Implementation of Microcomputer Online Fault Detection of Solar Array," Computer Science & Education, 2009, pp. 1052-1055. [18] Yousef Mahmoud, W. Xiao, H. H. Zeineldin,” A Simple Approach to Modeling and Simulation of Photovoltaic Modules” IEEE Transactions on sustainable energy, VOL. 3, NO. 1, JANUARY 2012 [19] Francisco M. González-Longatt,“Model of Photovoltaic Module in Matlab” 2do Congreso Iberoamericano De Estudiantes De Ingeniería Eléctrica, Electrónica Y COMPUTACIÓN . 2005. [20] Habbati Bellia, Ramdani Youcef, Moulay Fatima “A detailed modeling of photovoltaic module using MATLAB” NRIAG Journal of Astronomy and Geophysics (2014), vol. 3, pp. 53–61. [21] N. Pandiarajan and Ranganath Muthu “Mathematical Modeling of Photovoltaic Module with Simulink” International Conference on Electrical Energy Systems (ICEES 2011), 3-5 Jan 2011. pp. 314-319. 113

[22] S.R. Wenham, M.A. Green, “Applied Photovoltaics”, Printed and bound in the UK by TJ International Ltd, Padstow, Cornwall, 2007. [23] Syafaruddin, E. Karatepe, T. Hiyama “controlling of artificial neural network for fault diagnosis of photovoltaic array” 16th International Conference on Intelligent System Application to Power Systems (ISAP), Sep. 25–28, 2011, pp. 1–6. [24] Atef .M. Mansour, Khaled .N. Salama, Essam Abu Elzahab,“Fault Diagnosis of PVSystem in a Smart Grid Using Fuzzy Logic Classifier,” 3rd International Conference on Advanced Control Circuits and Systems (ACCS’013),30 Nov. – 03 Dec. 2013, Luxor, Egypt [25] Fathy.A.Syam, “Fault Detection of PV system using Advanced Simulation” PHD. Thesis, Faculty of engineering, Cairo University, Giza, Egypt, 2004. [26] M.E. Glavin, Paul K.W. Chan, S. Armstrong, and W.G Hurley, “A Stand-alone Photovoltaic Supercapacitor Battery Hybrid Energy Storage System,” IEEE 13th Power Electronics and Motion Control Conference, pp. 1688 – 1695, 1-3 Sept. 2008 [27] Abdul-mateen H. Hameed, Mohsen T. Elhagri, Khaled N. Faris, Adel A. Shaltout, “Optimal Configuration of Energy Storage System via Fuzzy Classifier”, 3rd International Conference on advanced control circuits and systems (accs’013) 30 Nov. – 03 Dec. 2013, Luxor, Egypt. [28] Jae Woong Shim, Youngho Cho, Seog-Joo Kim, Sang Won Min, and Kyeon Hur, “Synergistic Control of SMES and Battery Energy Storage for Enabling Dispatch ability of Renewable Energy Sources” , IEEE transactions on applied superconductivity, vol. 23, no. 3, june 2013. [29] E. Spahic, G. Balzer, B.Hellmich and W. Münch, "Wind Energy Storages– Possibilities", Power Tech, 2007 IEEE Lausanne, 1-5 July 2007, pp.615-620. [30] Haisheng Chen, Thang Ngoc Cong, Wei Yang, Chunqing Tan, Yongliang Li,Yulong Ding, “Progress in electrical energy storage system: A critical review”, Progress in Natural Science, Volume 19, Issue 3, 10 March 2009, Pages 291–312. [ 31] Ioannis Hadjipaschalis, Andreas Poullikkas , Venizelos Efthimiou, “Overview of current and future energy storage technologies for electric power applications”, Renewable and Sustainable Energy Reviews, vol. 13, pp.1513–1522, 2009 [32] J.D.Boyes, N.H.Clark, "Technologies for energy storage Flywheels and superconducting magnetic energy storage," in Power Engineering Society Summer Meeting, IEEE, vol. 3, pp.1548-1550, 2000.

114

[33] M.G.Molina, "Distributed energy storage systems for applications in future smart grids," in Transmission and Distribution: Latin America Conference and Exposition (T&DLA), 2012 Sixth IEEE/PES, pp.1-7, 3-5 Sept. 2012 [34] Electrical Energy Storage project team “Electrical energy storage systems”, IEC White Paper EES:2011. [35] H. Ibrahima,b, A. Ilincaa, J. Perronb, “Energy storage systems—Characteristics and comparisons” , Renewable and Sustainable Energy Reviews, vol. 12, pp. 1221–1250, 2008. [36] Faizur Rahman, Shafiqur Rehman, Mohammed Arif Abdul-Majeed, “Overview of energy storage systems for storing electricity from renewable energy sources in Saudi Arabia”, Renewable and Sustainable Energy Reviews, vol. 16, pp. 274– 283, 2012. [37] G. Fabbri, M. Paschero2, A. J. M. Cardoso, c. Boccaletti, F. M. Frattale Mascioli, “A Genetic Algorithm Based Battery Model for Stand Alone Radio Base Stations Powering” IEEE 33rd International Telecommunications Energy Conference (INTELEC), 9-13 Oct. 2011 [38] Andreas Oberhofer, Peter Meissen, “Energy Storage Technologies & Their Role in Renewable Integration”, Global Energy Network Institute (GENI), July 2012 [39] Marin S. Halper, James C. Ellenbogen, “Supercapacitors: A Brief Overview”, MITRE Nano systems Group, McLean, Virginia, USA, March 2006. [40] Mahmoud Amin, "Efficiency and Power Density Improvement of Grid-Connected Hybrid Renewable Energy Systems utilizing High Frequency-Based Power Converters", FIU Electronic Theses and Dissertations. Paper 600. (2012). [41] O. Tremblay, L.A. Dessaint, A. Dekkiche, “A Generic Battery Model for the Dynamic Simulation of Hybrid Electric Vehicles,” IEEE Vehicle Power and Propulsion Conference VPPC 2007, pp. 284-289, 9-12 Sept. 2007. [42] S. N. Motapon, L. A. Dessaint, and K. Al-Haddad,“A Comparative Study of Energy Management Schemes for a Fuel-Cell Hybrid Emergency Power System of More-Electric Aircraft,” IEEE Transactions on industerial electronics, vol. 61, no. 3, pp. 1320-1334, March 2014. [43] E. Veldman, M. Gibescu, J. Slootweg, and W. Kling, “Technical benefits of distributed storage and load management in distribution grids,” in PowerTech, 2009 IEEE Bucharest, pp.1 – 8, 28 July 2009.

115

[44] X. Vallve, A. Graillot, S. Gual, and H. Colin, “Micro storage and demand side management in distributed pv grid-connected installations,” 9th International Conference on in Electrical Power Quality and Utilisation, Oct. 2007, pp. 1-6. [45] M.E. Glavin, Paul K.W. Chan, S. Armstrong, and W.G Hurley, “A Stand-alone Photovoltaic Supercapacitor Battery Hybrid Energy Storage System,” IEEE 13th Power Electronics and Motion Control Conference, pp. 1688 – 1695, 1-3 Sept. 2008. [46] A.Z. Annuar, A.H.M. Yatim, “A Development of Fuzzy Control of Hybrid Energy System using Ultracapacitor,” 2nd IEEE International Conference on Power and Energy (PECon 08), December 1-3, 2008, Johor Baharu, Malaysia. [47] W. Li, G. Joós, “A Power Electronic Interface for a Battery Supercapacitor Hybrid Energy Storage System for Wind Applications,” IEEE annual power electronics specialists conference PESC, pp. 1762 - 1768, 2008. [48] Khalid. H. Mohamed, Taib B. Ibrahim and Nordin B. Saad, “Boost Converter Design with Stable Output Voltage for Wave Energy Conversion System,” International Journal of Information Technology and Electrical Engineering, Vol. 2, Issue. 1, February 2013. [49] Bidyadhar Subudhi, and Raseswari Pradha, “A Comparative Study on Maximum Power Point Tracking Techniques for Photovoltaic Power Systems,” IEEE Transactions on Sustainable Energy, Vol. 4, No. 1, January 2013 89 [50] Trishan Esram, and Patrick L. Chapman, “Comparison of Photovoltaic Array Maximum Power Point Tracking Techniques,” IEEE Transactions on Energy Conversion, Vol. 22, No. 2, June 2007 [51] Sumedha Sengar, “Maximum Power Point Tracking Algorithms for Photovoltaic System: A Review,” International Review of Applied Engineering Research, Vol. 4, No. 2, pp. 147-154, 2014. [52] V. Salas, E. Olias, A. Barrado, A. Lazaro, “Review of the maximum power point tracking algorithms for stand-alone photovoltaic systems,” Solar Energy Materials & Solar Cells, pp.1555–1578, 2006. [53] Dalila BERIBER, Abdelaziz TALHA,“MPPT Techniques for PV Systems,” 4th International Conference on Power Engineering, Energy and Electrical Drives, May 2013. 116

[54] B. N. Alajmi, K. H. Ahmed, S. J. Finney, and B. W. Williams, “Fuzzy-Logic-Control Approach of a Modified Hill-Climbing Method for Maximum Power Point in Microgrid Standalone Photovoltaic System,” IEEE Transactions on Power Electronics, vol. 26, no. 4, April 2011. [55] Ying Bai, Dali Wang, “Fundamentals of Fuzzy Logic Control – Fuzzy Sets, Fuzzy Rules and Defuzzifications,” Advanced Fuzzy Logic Technologies in Industrial Applications http://www.springer.com/cda/content/document/cda_downloaddocument/97818462846 87-c1.pdf [56] V.C. Kotak, Preti Tyagi, “DC To DC Converter in Maximum Power Point Tracker” International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering, Vol. 2, Issue 12, December 2013. [57] Khalid. H. Mohamed, “Boost Converter Design with Stable Output Voltage for Wave Energy Conversion System” International Journal of Information Technology and Electrical Engineering, Vol. 2, Issue 1, pp 10-15, 2013. [58] Atef.M.Mansour, Khaled.N. Fares, Essam Aboul Zahab, “Smart energy management controller for a micro grid,” International journal of engineering research, Vol. 4, Issue. 8, pp : 456-464, August 2015.

117

Appendices Appendix “A” Fuzzy rules of (fault diagnosis fuzzy logic classifier) 1. (insolation==UltraShiny) & (temp==Normal) & (pmax==FArray) & (voc==Full_Array) => (partial_shadow_percentage=Noshadow)(healthy_percentage=good) (1) 2. (insolation==UltraShiny) & (temp==Normal) & (pmax==FArray) & (voc==Full_Array) => (partial_shadow_percentage=extremelylowsha)(healthy_percentage=good) (1) 3. (insolation==UltraShiny) & (temp==Normal) & (pmax==VHigh) & (voc==VVHigh) => (partial_shadow_percentage=vvsma)(healthy_percentage=good) (1) 4. (insolation==UltraShiny) & (temp==Normal) & (pmax==VHigh) & (voc==VVHigh) => (partial_shadow_percentage=Vs_shadow)(healthy_percentage=PartialFault) (1) 5. (insolation==UltraShiny) & (temp==Normal) & (pmax==High) & (voc==VVHigh) => (partial_shadow_percentage=small)(healthy_percentage=PartialFault) (1) 6. (insolation==UltraShiny) & (temp==Normal) & (pmax==MED) & (voc==VVHigh) => (partial_shadow_percentage=medium)(healthy_percentage=PartialFault) (1) 7. 7. (insolation==UltraShiny) & (temp==Normal) & (pmax==MED) & (voc==VVHigh) => (partial_shadow_percentage=high)(healthy_percentage=PartialFault) (1) 8. (insolation==UltraShiny) & (temp==Normal) & (pmax==MED) & (voc==VHigh) => (partial_shadow_percentage=Vhigh)(healthy_percentage=PartialFault) (1) 9. (insolation==UltraShiny) & (temp==Normal) & (pmax==MED) & (voc==High) => (partial_shadow_percentage=VVhigh)(healthy_percentage=PartialFault) (1) 10. (insolation==UltraShiny) & (temp==Normal) & (pmax==MED) & (voc==Med) => (partial_shadow_percentage=ultra)(healthy_percentage=PartialFault) (1) 11. (insolation==VVHShine) & (temp==Normal) & (pmax==VHigh) & (voc==Full_Array) => (partial_shadow_percentage=extremelylowsha)(healthy_percentage=PartialFault) (1) 12. (insolation==VVHShine) & (temp==Normal) & (pmax==VHigh) & (voc==VVHigh) => (partial_shadow_percentage=vvsma)(healthy_percentage=PartialFault) (1) 118

13. (insolation==VVHShine) & (temp==Normal) & (pmax==VHigh) & (voc==VVHigh) => (partial_shadow_percentage=Vs_shadow)(healthy_percentage=PartialFault) (1) 14. (insolation==VVHShine) & (temp==Normal) & (pmax==High) & (voc==VVHigh) => (partial_shadow_percentage=small)(healthy_percentage=PartialFault) (1) 15. (insolation==VVHShine) & (temp==Normal) & (pmax==MED) & (voc==VVHigh) => (partial_shadow_percentage=medium)(healthy_percentage=PartialFault) (1) 16. (insolation==VVHShine) & (temp==Normal) & (pmax==MED) & (voc==VHigh) => (partial_shadow_percentage=high)(healthy_percentage=PartialFault) (1) 17. (insolation==VVHShine) & (temp==Normal) & (pmax==MED) & (voc==VHigh) => (partial_shadow_percentage=Vhigh)(healthy_percentage=PartialFault) (1) 18. (insolation==VVHShine) & (temp==Normal) & (pmax==MED) & (voc==High) => (partial_shadow_percentage=VVhigh)(healthy_percentage=PartialFault) (1) 19. (insolation==VVHShine) & (temp==Normal) & (pmax==MED) & (voc==Med) => (partial_shadow_percentage=ultra)(healthy_percentage=PartialFault) (1) 20. (insolation==VHSine) & (temp==Normal) & (pmax==VHigh) & (voc==VVHigh) => (partial_shadow_percentage=vvsma)(healthy_percentage=PartialFault) (1) 21. (insolation==VHSine) & (temp==Normal) & (pmax==VHigh) & (voc==VVHigh) => (partial_shadow_percentage=Vs_shadow)(healthy_percentage=PartialFault) (1) 22. (insolation==VHSine) & (temp==Normal) & (pmax==High) & (voc==VVHigh) => (partial_shadow_percentage=small)(healthy_percentage=PartialFault) (1) 23. (insolation==VHSine) & (temp==Normal) & (pmax==MED) & (voc==VHigh) => (partial_shadow_percentage=medium)(healthy_percentage=PartialFault) (1) 24. (insolation==VHSine) & (temp==Normal) & (pmax==MED) & (voc==VHigh) => (partial_shadow_percentage=high)(healthy_percentage=PartialFault) (1) 25. (insolation==VHSine) & (temp==Normal) & (pmax==Small) & (voc==High) => (partial_shadow_percentage=Vhigh)(healthy_percentage=PartialFault) (1) 26. (insolation==VHSine) & (temp==Normal) & (pmax==Small) & (voc==High) => (partial_shadow_percentage=VVhigh)(healthy_percentage=PartialFault) (1) 27. (insolation==VHSine) & (temp==Normal) & (pmax==Small) & (voc==Med) => (partial_shadow_percentage=ultra)(healthy_percentage=PartialFault) (1) 28. (insolation==Hshiny) & (temp==Normal) & (pmax==High) & (voc==VVHigh) => (partial_shadow_percentage=Vs_shadow)(healthy_percentage=PartialFault) (1) 29. (insolation==Hshiny) & (temp==Normal) & (pmax==High) & (voc==VHigh) => (partial_shadow_percentage=small)(healthy_percentage=PartialFault) (1) 30. (insolation==Hshiny) & (temp==Normal) & (pmax==MED) & (voc==VHigh) => (partial_shadow_percentage=medium)(healthy_percentage=PartialFault) (1)

119

31. (insolation==Hshiny) & (temp==Normal) & (pmax==MED) & (voc==High) => (partial_shadow_percentage=high)(healthy_percentage=PartialFault) (1) 32. (insolation==Hshiny) & (temp==Normal) & (pmax==Small) & (voc==High) => (partial_shadow_percentage=Vhigh)(healthy_percentage=PartialFault) (1) 33. (insolation==Hshiny) & (temp==Normal) & (pmax==Small) & (voc==Med) => (partial_shadow_percentage=VVhigh)(healthy_percentage=PartialFault) (1) 34. (insolation==Hshiny) & (temp==Normal) & (pmax==Small) & (voc==Small) => (partial_shadow_percentage=ultra)(healthy_percentage=PartialFault) (1) 35. (insolation==shiny) & (temp==Normal) & (pmax==MED) & (voc==VHigh) => (partial_shadow_percentage=small)(healthy_percentage=PartialFault) (1) 36. (insolation==shiny) & (temp==Normal) & (pmax==MED) & (voc==VHigh) => (partial_shadow_percentage=medium)(healthy_percentage=PartialFault) (1) 37. (insolation==shiny) & (temp==Normal) & (pmax==MED) & (voc==High) => (partial_shadow_percentage=high)(healthy_percentage=PartialFault) (1) 38. (insolation==shiny) & (temp==Normal) & (pmax==Small) & (voc==High) => (partial_shadow_percentage=Vhigh)(healthy_percentage=PartialFault) (1) 39. (insolation==shiny) & (temp==Normal) & (pmax==Small) & (voc==Med) => (partial_shadow_percentage=VVhigh)(healthy_percentage=PartialFault) (1) 40. (insolation==shiny) & (temp==Normal) & (pmax==Small) & (voc==Med) => (partial_shadow_percentage=ultra)(healthy_percentage=PartialFault) (1) 41. (insolation==MShiny) & (temp==Normal) & (pmax==MED) & (voc==VHigh) => (partial_shadow_percentage=medium)(healthy_percentage=PartialFault) (1) 42. (insolation==MShiny) & (temp==Normal) & (pmax==MED) & (voc==High) => (partial_shadow_percentage=high)(healthy_percentage=PartialFault) (1) 43. (insolation==MShiny) & (temp==Normal) & (pmax==Small) & (voc==High) => (partial_shadow_percentage=Vhigh)(healthy_percentage=PartialFault) (1) 44. (insolation==MShiny) & (temp==Normal) & (pmax==VSmall) & (voc==Med) => (partial_shadow_percentage=VVhigh)(healthy_percentage=PartialFault) (1) 45. (insolation==MShiny) & (temp==Normal) & (pmax==VSmall) & (voc==Small) => (partial_shadow_percentage=ultra)(healthy_percentage=PartialFault) (1) 46. (insolation==Lowshiny) & (temp==Normal) & (pmax==Small) & (voc==High) => (partial_shadow_percentage=high)(healthy_percentage=PartialFault) (1) 47. (insolation==Lowshiny) & (temp==Normal) & (pmax==Small) & (voc==Med) => (partial_shadow_percentage=Vhigh)(healthy_percentage=PartialFault) (1) 48. (insolation==Lowshiny) & (temp==Normal) & (pmax==VSmall) & (voc==Med) => (partial_shadow_percentage=VVhigh)(healthy_percentage=PartialFault) (1) 49. (insolation==VLShiny) & (temp==Normal) & (pmax==Small) & (voc==Med) => (partial_shadow_percentage=Vhigh)(healthy_percentage=PartialFault) (1) 50. (insolation==VLShiny) & (temp==Normal) & (pmax==VSmall) & (voc==Small) => (partial_shadow_percentage=VVhigh)(healthy_percentage=PartialFault) (1) 120

51. (insolation==VLShiny) & (temp==Normal) & (pmax==VSmall) & (voc==VSmall) => (partial_shadow_percentage=ultra)(healthy_percentage=PartialFault) (1) 52. (insolation==VVLShiny) & (temp==Normal) & (pmax==VSmall) & (voc==Small) => (partial_shadow_percentage=VVhigh)(healthy_percentage=PartialFault) (1) 53. (insolation==VVLShiny) & (temp==Normal) & (pmax==VSmall) & (voc==VVSmall) => (partial_shadow_percentage=ultra)(healthy_percentage=PartialFault) (1) 54. 54. (insolation==dark) & (temp==Normal) & (pmax==VSmall) & (voc==VVSmall) => (partial_shadow_percentage=ultra)(healthy_percentage=PartialFault) (1) 55. (insolation==UltraShiny) & (Vmax==High) & (temp==Normal) & (pmax==VHigh) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VSmallSC) (1) 56. (insolation==UltraShiny) & (Vmax==Medium) & (temp==Normal) & (pmax==High) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=smallSC) (1) 57. (insolation==UltraShiny) & (Vmax==Medium) & (temp==Normal) & (pmax==MED) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=Medium) (1) 58. (insolation==UltraShiny) & (Vmax==Small) & (temp==Normal) & (pmax==Small) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=HighSC) (1) 59. 59. (insolation==UltraShiny) & (Vmax==Vsmall) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VHighSC) (1) 60. (insolation==UltraShiny) & (Vmax==Zero) & (temp==Normal) & (pmax==zero) => (healthy_percentage=complete_fault)(Shortcircuit_percentage=ToalSC) (1) 61. (insolation==VVHShine) & (Vmax==High) & (temp==Normal) & (pmax==High) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VSmallSC) (1) 62. (insolation==VVHShine) & (Vmax==Medium) & (temp==Normal) & (pmax==High) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=smallSC) (1) 63. (insolation==VVHShine) & (Vmax==Medium) & (temp==Normal) & (pmax==MED) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=Medium) (1) 64. (insolation==VVHShine) & (Vmax==Small) & (temp==Normal) & (pmax==Small) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=HighSC) (1) 121

65. (insolation==VVHShine) & (Vmax==Vsmall) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VHighSC) (1) 66. (insolation==VVHShine) & (Vmax==Zero) & (temp==Normal) & (pmax==zero) => (healthy_percentage=good)(Shortcircuit_percentage=ToalSC) (1) 67. (insolation==VHSine) & (Vmax==High) & (temp==Normal) & (pmax==High) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VSmallSC) (1) 68. (insolation==VHSine) & (Vmax==Medium) & (temp==Normal) & (pmax==MED) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=smallSC) (1) 69. (insolation==VHSine) & (Vmax==Medium) & (temp==Normal) & (pmax==Small) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=Medium) (1) 70. (insolation==VHSine) & (Vmax==Small) & (temp==Normal) & (pmax==Small) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=HighSC) (1) 71. (insolation==VHSine) & (Vmax==Vsmall) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VHighSC) (1) 72. (insolation==VHSine) & (Vmax==Zero) & (temp==Normal) & (pmax==zero) => (healthy_percentage=good)(Shortcircuit_percentage=ToalSC) (1) 73. (insolation==Hshiny) & (Vmax==High) & (temp==Normal) & (pmax==MED) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VSmallSC) (1) 74. (insolation==Hshiny) & (Vmax==Medium) & (temp==Normal) & (pmax==MED) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=smallSC) (1) 75. (insolation==Hshiny) & (Vmax==Small) & (temp==Normal) & (pmax==Small) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=Medium) (1) 76. (insolation==Hshiny) & (Vmax==Small) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=HighSC) (1) 77. (insolation==Hshiny) & (Vmax==Vsmall) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VHighSC) (1) 78. (insolation==Hshiny) & (Vmax==Zero) & (temp==Normal) & (pmax==zero) => (healthy_percentage=good)(Shortcircuit_percentage=ToalSC) (1) 79. (insolation==shiny) & (Vmax==High) & (temp==Normal) & (pmax==MED) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VSmallSC) (1) 80. (insolation==shiny) & (Vmax==Medium) & (temp==Normal) & (pmax==Small) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=smallSC) (1) 81. (insolation==shiny) & (Vmax==Small) & (temp==Normal) & (pmax==Small) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=Medium) (1)

122

82. (insolation==shiny) & (Vmax==Small) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=HighSC) (1) 83. 83. (insolation==shiny) & (Vmax==Vsmall) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VHighSC) (1) 84. (insolation==shiny) & (Vmax==Zero) & (temp==Normal) & (pmax==zero) => (healthy_percentage=good)(Shortcircuit_percentage=ToalSC) (1) 85. (insolation==MShiny) & (Vmax==High) & (temp==Normal) & (pmax==MED) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VSmallSC) (1) 86. (insolation==MShiny) & (Vmax==Medium) & (temp==Normal) & (pmax==Small) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=smallSC) (1) 87. (insolation==MShiny) & (Vmax==Medium) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=Medium) (1) 88. (insolation==MShiny) & (Vmax==Small) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=HighSC) (1) 89. (insolation==MShiny) & (Vmax==Vsmall) & (temp==Normal) & (pmax==zero) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VHighSC) (1) 90. (insolation==MShiny) & (Vmax==Zero) & (temp==Normal) & (pmax==zero) => (healthy_percentage=good)(Shortcircuit_percentage=ToalSC) (1) 91. (insolation==Lowshiny) & (Vmax==High) & (temp==Normal) & (pmax==Small) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VSmallSC) (1) 92. (insolation==Lowshiny) & (Vmax==Medium) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=smallSC) (1) 93. (insolation==Lowshiny) & (Vmax==Small) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=Medium) (1) 94. (insolation==Lowshiny) & (Vmax==Small) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=HighSC) (1) 95. (insolation==Lowshiny) & (Vmax==Vsmall) & (temp==Normal) & (pmax==zero) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VHighSC) (1) 96. 96. (insolation==Lowshiny) & (Vmax==Zero) & (temp==Normal) & (pmax==zero) => (healthy_percentage=good)(Shortcircuit_percentage=ToalSC) (1) 97. (insolation==VLShiny) & (Vmax==High) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VSmallSC) (1)

123

98. (insolation==VLShiny) & (Vmax==Medium) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=smallSC) (1) 99. (insolation==VLShiny) & (Vmax==Small) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=Medium) (1) 100. (insolation==VLShiny) & (Vmax==Small) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=HighSC) (1) 101. (insolation==VLShiny) & (Vmax==Vsmall) & (temp==Normal) & (pmax==zero) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VHighSC) (1) 102. 102. (insolation==VLShiny) & (Vmax==Zero) & (temp==Normal) & (pmax==zero) => (healthy_percentage=good)(Shortcircuit_percentage=ToalSC) (1) 103. (insolation==VVLShiny) & (Vmax==High) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VSmallSC) (1) 104. (insolation==VVLShiny) & (Vmax==Medium) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=smallSC) (1) 105. (insolation==VVLShiny) & (Vmax==Small) & (temp==Normal) & (pmax==VSmall) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=Medium) (1) 106. (insolation==VVLShiny) & (Vmax==Small) & (temp==Normal) & (pmax==zero) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=HighSC) (1) 107. (insolation==VVLShiny) & (Vmax==Vsmall) & (temp==Normal) & (pmax==zero) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VHighSC) (1) 108. (insolation==VVLShiny) & (Vmax==Zero) & (temp==Normal) & (pmax==zero) => (healthy_percentage=good)(Shortcircuit_percentage=ToalSC) (1) 109. (insolation==dark) & (Vmax==High) & (temp==Normal) & (pmax==zero) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VSmallSC) (1) 110. (insolation==dark) & (Vmax==Medium) & (temp==Normal) & (pmax==zero) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=smallSC) (1) 111. (insolation==dark) & (Vmax==Small) & (temp==Normal) & (pmax==zero) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=Medium) (1) 112. (insolation==dark) & (Vmax==Small) & (temp==Normal) & (pmax==zero) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=HighSC) (1) 124

113. (insolation==dark) & (Vmax==Vsmall) & (temp==Normal) & (pmax==zero) => (healthy_percentage=PartialFault)(Shortcircuit_percentage=VHighSC) (1) 114. (insolation==dark) & (Vmax==Zero) & (temp==Normal) & (pmax==zero) => (healthy_percentage=complete_fault)(Shortcircuit_percentage=ToalSC) (1) (insolation==Lowshiny) & (temp==Normal) & (pmax==VSmall) & (voc==Small) => (partial_shadow_percentage=ultra)(healthy_percentage=PartialFault) (1)

125

Appendix “B” PV module specifications

126

Appendix “C” voltage sensor Specs. Voltage Transducer LV 25-P For the electronic measurement of voltages : DC, AC, pulsed with a galvanic isolation between the primary circuit (high voltage) and the secondary circuit (electronic circuit).

Calibration equation The voltage sensor is calibrated to measure output voltage from the PV in the range of 0 45 volt. and the output voltage from the voltage sensor in the range of 0 – 10 volt. So that the output voltage from the voltage sensor is fed to the equation which calculates the real value of the measured voltage in the program as following If the measured voltage is 45 volt so the output voltage from the voltage sensor is 10 volt, in the program of P&O we need to read the real value 45 not 10 volt. So the following equation is used in the program: Y= 0.0197 + 0.22105x X= (y-0.0197)/0.22105 Where y is the output voltage from the voltage sensor x is the measured voltage from the PV

Connection

A. 1 Connection of voltage transducer LV 25-p

127

Secondary terminals Terminal + : supply voltage + 12 .. 15 V Terminal M : measure Terminal - : supply voltage - 12 .. 15 V

Instructions for use of the voltage transducer model LV 25-P Primary resistor R 1 : the transducer’s optimum accuracy is obtained at the nominal primary current. As far as possible, R 1 should be calculated so that the nominal voltage to be measured corresponds to a primary current of 10 mA. Example: Voltage to be measured VPN = 250 V a) R1 = 25 kW / 2.5 W, IP = 10 mA Accuracy = ± 0.8 % of VPN (@ TA = + 25°C) b) R1 = 50 kW / 1.25 W, IP = 5 mA Accuracy = ± 1.6 % of VPN (@ TA = + 25°C) Operating range (recommended) : taking into account the resistance of the primary windings (which must remain low compared to R1, in order to keep thermal deviation as low as possible) and the isolation, this transducer is suitable for measuring nominal voltages from 10 to 500 V.

128

129

Appendix “D” Current sensor Specs. Current transducer RS 257-436 This current transducer measures the load current in the range of 0-200 (A) this meets output signal 0-10 volt. In our system this current range which measured 0-10 (A) so that it will meets output signal in the range of 0- 0.5 Volt. The current sensor is calibrated by turning the cable 20 wined as shown in Figure C. 1 to increase the range of output voltage from 0-10 Volt.

Figure C. 1 current sensor

130

131

132

Appendix “E” DSPACE DS1102 Product Name: DS1102 Platforms Supported: PC, Sun, HP Devices Supported: TMS320C31

Features and Benefits • Single-unit controller. • TI ’C31 running at 40 MHz. • Four on-board A/D converters, four on-board D/A converters. • Two incremental encoder interfaces. • ’P14 working as an intelligent I/O subsystem.

Product Description

As the entry-level board for the dSPACE development system, the DS1102 combines the ’C31’s computing performance of up to 40 MFLOPS with aversatile set of on-board I/O: four analog input channels (16 bit, 10 s, and 12 bit, 3 s), four analog output channels (12 bit), two incremental encoder channels, and a complete subsystem for digital I/O, based on a TMS320P14 DSP. Using provided software, this subsystem can be accessed like conventional I/O channels, and/or programmed as a slave DSP. The high performance design of the DS1102 supports the maximum usage of the ’C31’s 50-ns cycle time. 128 k 32-bit zero-wait-state memory are on-board. This on-board memory can be accessed by the host for program download as well as data transfer while the DSP is running. All boards are standard ISA cards. For Sun and HP hosts, a network interface to an expansion box is required.

133

D. 1 DSPACE DS1102

D. 2 The block diagram of the DS1102

134

Appendix “F” MOSFET Transistor IRFP460LC

Ultra Low Gate Charge Reduced Gate Drive Requirement Enhanced 30V Vgs Rating Reduced Ciss, Coss, Crss Isolated Central Mounting Hole Dynamic dv/dt Rated Repetitive Avalanche Rated Absolute Maximum Ratings Parameter Max. Units ID @ TC = 25°C Continuous Drain Current, V GS @ 10V 20 ID @ TC = 100°C Continuous Drain Current, V GS @ 10V 12 A IDM Pulsed Drain Current 80 PD @TC = 25°C Power Dissipation 280 W Linear Derating Factor 2.2 W/°C VGS Gate-to-Source Voltage ±30 V EAS Single Pulse Avalanche Energy 960 mJ IAR Avalanche Current 20 A EAR Repetitive Avalanche Energy 28 mJ dv/dt Peak Diode Recovery dv/dt 3.5 V/ns TJ Operating Junction and -55 to + 150 TSTG Storage Temperature Range °C Soldering Temperature, for 10 seconds 300 (1.6mm from case) Mounting torque, 6-32 or M3 screw. 10 lbf•in (1.1N•m) Parameter Min. Typ. Max. Units -to-Case –––– –––– 0.45 -to-Sink, Flat, Greased Surface –––– 0.24 –––– °C/W -to-Ambient –––– –––– 40 Thermal Resistance Description This new series of Low Charge HEXFET Power MOSFETs achieve significantly lower gate charge over conventional MOSFETs. Utilizing advanced Hexfet technology the device improvements allow for reduced gate drive requirements, faster switching speeds and increased total system savings. These device improvements combined with the proven ruggedness and reliability of HEXFETs offer the designer a new standard in power transistors for switching applications. The TO-247 package is preferred for commercial-industrial applications where higher power levels preclude the use of TO-220 devices. The TO-247 is similar but superior to the earlier TO-218 package because of its isolated mounting hole. 135

136

Appendix “G” Electronic load The electronic load refers to a device that uses semiconductors (transistors) in place of variable resistor to act as a load. Because voltage and current can be controlled freely on a semiconductor, the load can be controlled arbitrarily by embedding a control circuit. A PLZ1004W electronic load has been used in this thesis. The maximum operating current of this load is 200 A, the operating voltage range from 1.5 to 150 (V), and the wattage is 1000 (W). The Basic operation modes of the electronic load: 1- Constant current mode (CC mode): when the PLZ1004W is used in CC mode, the PLZ1004W operates as a constant current load as shown in fig. the PLZ1004W sinks the specified current (I) independent of the output voltage of the constant voltage power supply (V1)

Fig: equivalent circuit of the constant current load and operation 23456-

Constant resistance mode (CR mode) Constant power mode (CP mode) Constant voltage mode (CV mode) Constant current and constant voltage (CC + CV mode) Constant resistance and constant voltage (CR + CV mode)

137

138

Appendix “H” battery specs

139

140

Published papers

1- Atef .M. Mansour, Khaled .N. Salama, Essam Abu Elzahab,“Fault Diagnosis of PV-System in a Smart Grid Using Fuzzy Logic Classifier,” 3rd International Conference on Advanced Control Circuits and Systems (ACCS’013),30 Nov. – 03 Dec. 2013, Luxor, Egypt. 2- Atef.M.Mansour, Khaled .N. Fares, Essam Aboul Zahab, “Smart energy management controller for a micro grid,” International journal of engineering research, Vol. 4, Issue. 8, pp : 456-464, August 2015.

141

‫ تم عمل نموذج معممى لتحقيق بعض النتائج النظرية لأللسمو المقترح لتتبع نقطة القدرة العظمى‪.‬‬‫وتتكون الرلسالة من لستة فصول باآلضافة إلى المراجع والمالحق‪.‬‬ ‫الفصل األول‪ :‬يحتوى ىذا الفصل عمى مقدمة عن أىمية الطاقة المتجددة ومميزاتيا وعيوبيا وايضا تعريف لمشبكة‬ ‫المصغرة وكذا مقدمة ألنواع األعطال المختمفة التى قد تحدث فى الخاليا الشملسية وايضا مقدمة لمحتوى الرلسالة ‪.‬‬ ‫الفصل الثانى‪ :‬يقدم ىذا الفصل نموذجا لمخاليا الشملسية‪ .‬وكذا درالسة لمعوامل المؤثرة عمى خصائصيا الكيربية‪ .‬أيضا‬ ‫تم درالسة ألنواع مختمفة من األعطال فى الخاليا الشملسية‪ .‬يقدم ىذا الفصل ايضا نموذجا مقترحا ذكيا يقوم عمى‬ ‫المنطق الضبابى إلكتشاف وتشخيص األعطال فى الخاليا الشملسية‪ .‬يعرض ىذا البا‬

‫أيضا بعض النتائج النظرية‬

‫ليذا النموذج‪.‬‬ ‫الفصل الثالث‪ :‬فى ىذا الفصل تم درالسة أنواع مختمفة لنظم تخزين الطاقة ‪ .‬وقد تم تبنى نوعين مختمفين متكاممين من‬

‫نظم تخزين الطاقة ‪ -‬وىما البطاريات والمكثف الفائق‪ -‬وعمل نموذج مفصل ليما‪.‬‬

‫الفصل الرابع‪ :‬قدم ىذا الفصل عمل محاكاة لشبكة مصغرة تتكون من مصدر طاقة لمخاليا الشملسية ومصادر تخزين‬ ‫الطاقة (البطاريات والمكثف الفائق) بالستخدام ‪ .Matlab/Simulink simulation Tools‬تقديم السمو‬ ‫أللسمو‬

‫‪ P&O‬ومقارنتو بالسمو‬

‫مقترح معدل‬

‫ذكى لتتبع نقطة القدرة العظمى بالستخدام المنطق الضبابى‪ .‬و يحتوى ىذا الفصل‬

‫ايضا عمى مقترح لحاكم ادارة الطاقة فى الشبكة المصغرة‪ .‬ودرالسة نظم مختمفة لمشبكة المصغرة ومقارنة لمنتائج النظرية‪.‬‬

‫الفصل الخامس‪ :‬قدم ىذا الفصل عرضا لجميع النتائج التى تم إجرائيا معمميا بالستخدام االلسمو‬

‫المعدل لتتبع نقطة‬

‫اقصى قدرة ‪.‬‬ ‫الفصل اللسادس‪ :‬يمخص ىذا الفصل ما تم إنجازه فى الرلسالة و كذلك التوصيات لمدرالسات الملستقبمية فى ىذا المجال‪.‬‬

‫ملخص الرسالة‬ ‫تعد مصادر الطاقة المتجددة من أىم مصادر انتاج الطاقة اليامة التى تجذ‬ ‫االلستف ادة منيا ‪ .‬ذلك ألنيا ال تنض‬

‫الكثير من الباحثين لدرالستيا وتعظيم‬

‫عمى عكس مصادر الطاقة التى تعتمد عمى الوقود الحفرى بشكل السالسى فى‬

‫توليد الطاقة‪ .‬أيضا ليس ليا تاثير ضار عمى البيئو من انبعاثات تؤدى الى ظاىرة االحتباس الحرارى مثل الوقود‬ ‫الحفرى‪.‬‬ ‫ولكن من عيو‬

‫بعض مصادر الطاقة المتجددة مثل الخاليا الشملسية ان القدرة الناتجة منيا تكون متغيرة عمى حلس‬

‫الظروف المناخية وقمة كفائتيا‪ .‬لذلك يمجأ الباحثون الى تحلسين كفاءة الخاليا الشملسية عن طريق تطوير المواد التى‬ ‫تصنع منيا الخاليا الشملسية وأيضا عن طريق تتبع نقطة القدرة العظمى بالستخدام الطرق المختمفة‪.‬‬

‫فى ىذه الرلسالة تم اختيار الخاليا الشملسية كمصدر ألسالسى لمطاقة فى شبكة مصغرة والستخدام مصادر لتخزين الطاقة‬ ‫لمتغم‬

‫عمى تغير القدرة الناتجة من الخاليا الشملسية و أيضا تزويد الحمل بالقدرة الالزمة طوال فترة التشغيل‪.‬‬

‫وتعرضت الرلسالة لمنقاط االتية‪:‬‬ ‫ تم عمل نموذج لمخاليا الشملسية يوضح الخصائص الكيربية ليا فى ظروف التشغيل الطبيعية وأيضا فى حالة‬‫حدوث أعطال‪ .‬ولقد تم درالسة نوعين مختمفين من األعطال التى تتعرض ليا الخاليا الشملسية وىما الظل الجزئى‬ ‫و دوائر القصر لذا تم تقديم ألسمو‬

‫ذكى بالستخدام المنطق الضبابى )‪ (Fuzzy Logic‬إلكتشاف وتشخيص‬

‫أعطال أنظمة الخاليا الشملسية وقد تم اقتراح طرق لمتغم‬

‫والتقميل من تاثير ىذه األعطال‪ .‬ايضا تم درالسة أنواع‬

‫مختمفة من طرق تخزين الطاقة الكيربية وقد تم عمل نموذج مفصل لمنوعين الملستخدمين فى الرلسالة‪.‬‬ ‫ أيضا تم عمل نموذج مفصل لشبكة مصغرة (‪ (microgrid‬تتكون من خاليا شملسية كمصدر ألسالسى لمطاقة و‬‫أنظمة تخزين الطاقة الكيربية مثل البطاريات (‪ )lead acid battery‬والمكثف الفائق)‪(supercapacitor‬‬ ‫بالستخدام ‪.Matlab/Simulink simulation Tools‬‬ ‫ تم تقديم ألسمو مقترح تم بناؤه بتعديل طريقة االثارة والمالحظة (‪ )P&O perturb and observe‬لتتبع نقطة‬‫اقصى قدرة (‪ )MPPT‬ومقارنة ىذا األلسمو بالمنطق الضبابى لتتبع نقطة أقصى قدرة‪ .‬فأوضحت نتائج المحاكاة‬ ‫التقار الممحوظ بين األلسموبين فى نتائجيما‪.‬‬ ‫ تم إقتراح حاكم إلدارة الطاقة فى الشبكة المصغرة بين الخاليا الشملسية وعناصر تخزين الطاقة المختمفة وكذا‬‫الحمل‪ .‬ويعد ىذا الحاكم لو القدرة عمى اختيار عنصر تخزين الطاقة المنالس‬

‫ليتم شحنو او تفريغو طبقا لحالة‬

‫الحمل والظروف المناخية بدون تعرضو لالجياد وايضا يتم تقميل مقدار التغير فى الجيد وخاصة فى الفترات‬ ‫العابرة (‪.)transient periods‬‬

‫مهنــــــــــــــــــدس‪:‬‬

‫تاريــخ الميــــالد‪:‬‬

‫عاطف محمد عبدالرازق منصور‬ ‫‪91‬‬

‫‪91 66 / 6 /‬‬

‫الجنسيـــــــــــــــة‪:‬‬

‫مصرى‬

‫تاريخ التسجيل‪:‬‬

‫‪10 /1‬‬

‫تــــاريخ المنــــح‪:‬‬ ‫القســـــــــــــــــــــم‪:‬‬ ‫الدرجــــــــــــــــــــة‪:‬‬ ‫المشرفون ‪:‬‬

‫‪/‬‬

‫‪2010/‬‬

‫‪2016/‬‬

‫ىندلسة القوى واآلالت الكيربية‬ ‫ماجلستير العموم‬ ‫أ‪.‬د عصام محمد أبو الذى ‪( .‬كمية اليندلسة –جامعة القاىرة)‬

‫الممتحنـــــــون ‪:‬‬

‫د‪ .‬خالد نجدى فارس‬

‫(معيد بحوث االلكترونيات )‬

‫أ‪.‬د حلسام كمال يولسف‬

‫(كمية اليندلسة –جامعة القاىرة)‬

‫أ‪.‬د محلسن طو الحاجرى (معيد بحوث االلكترونيات)‬

‫أ‪.‬د عصام محمد أبو الذى ‪( .‬كمية اليندلسة –جامعة القاىرة)‬ ‫عنـــــوان الرسالــة ‪:‬‬

‫نمذجة ومحاكاة وتشخيص األعطال فى شبكة مصغرة‬ ‫الكلمات الدالة ‪- :‬‬ ‫شبكة مصغرة و الخاليا الشمسية ‪ ،‬تشخيص األعطال ‪ ،‬تتبع نقطة القدرة العظمى ‪ ،‬حاكم ادارة الطاقة‬ ‫ملخـــــص البحــــــث ‪:‬‬

‫ تم عمل نموذج لمخاليا الشملسية يوضح الخصائص الكيربية ليا فى ظروف التشغيل الطبيعية وأيضا فى حالة‬‫حدوث أعطال‪ .‬ولقد تم درالسة نوعين مختمفين من األعطال التى تتعرض ليا الخاليا الشملسية وىما الظل الجزئى‬

‫و دوائر القصر لذا تم تقديم ألسمو‬

‫ذكى بالستخدام المنطق الضبابى )‪ (Fuzzy Logic‬إلكتشاف وتشخيص‬

‫أعطال أنظمة الخاليا الشملسية وقد تم اقتراح طرق لمتغم‬

‫والتقميل من تاثير ىذه األعطال‪ .‬ايضا تم درالسة أنواع‬

‫مختمفة من طرق تخزين الطاقة الكيربية وقد تم عمل نموذج مفصل لمنوعين الملستخدمين فى الرلسالة‪.‬‬ ‫ أيضا تم عمل نموذج مفصل لشبكة مصغرة (‪ (microgrid‬تتكون من خاليا شملسية كمصدر ألسالسى لمطاقة و‬‫أنظمة تخزين الطاقة الكيربية مثل البطاريات (‪ )lead acid battery‬والمكثف الفائق)‪(supercapacitor‬‬

‫بالستخدام ‪.Matlab/Simulink simulation Tools‬‬

‫ تم تقديم ألسمو مقترح تم بناؤه بتعديل طريقة االثارة والمالحظة (‪ )P&O perturb and observe‬لتتبع نقطة‬‫اقصى قدرة (‪ )MPPT‬ومقارنة ىذا األلسمو بالمنطق الضبابى لتتبع نقطة أقصى قدرة‪ .‬فأوضحت نتائج المحاكاة‬ ‫التقار‬

‫الممحوظ بين األلسموبين فى نتائجيما‪ .‬تم إقتراح حاكم إلدارة الطاقة فى الشبكة المصغرة بين الخاليا‬

‫الشملسية وعناصر تخزين الطاقة المختمفة وكذا الحمل‪ .‬ويعد ىذا الحاكم لو القدرة عمى اختيار عنصر تخزين‬

‫الطاقة المنالس‬

‫ليتم شحنو او تفريغو طبقا لحالة الحمل والظروف المناخية بدون تعرضو لالجياد وايضا يتم تقميل‬

‫مقدار التغير فى الجيد وخاصة فى الفترات العابرة (‪.)transient periods‬‬ ‫تم عمل نموذج معممى لتحقيق بعض النتائج النظرية لأللسمو المقترح لتتبع نقطة القدرة العظمى‪.‬‬

‫نًزجت ويحاكاة وحشخُص األعطال فً شبكت يصغشة‬ ‫إعــــــذاد‬

‫انًهنـذط ‪ /‬عاطف يحًذ عبذانشاصق ينصىس‬ ‫رســانــت يقــديــت إنً كهيت انهندست ‪ -‬جـايعــت انقـاهــرة‬ ‫كجزء ين يتطهباث انحصىل عهً درجت ياجستير انعهىو‬ ‫فً هندست انقىي واآلالث انكهربيت‬

‫َعخًذ ين نجنت انًًخحنُن‪:‬‬ ‫انًششف انشئُسً‪..............‬‬

‫األسخار انذكخىس‪ :‬عصاو انذَن ابى انزهب‬ ‫األسخار بقسى انقىي و االالث انكهشبُت كلية انهنذست ‪-‬جايعت انقاهشة‪.‬‬

‫انًًخحن انذاخهً‪................‬‬

‫األسخار انذكخىس ‪ :‬حساو كًال َىسف‬ ‫األسخار بقسى انقىي و االالث انكهشبُت كلية انهنذست ‪-‬جايعت انقاهشة‬

‫األس خار ان ذكخىس‪ :‬يحس ن ط ج انح اجشي‬

‫انًً خحن انخ اسجٍ‪..............‬‬

‫أسخار بقسى انكخشونُاث انقىٌ و ححىَم انطاقت بًعهذ بحىد االنكخشونُاث‬

‫كهُت انهنذست ‪ -‬جـايعــت انقـاهــشة‬ ‫انجُضة ‪ -‬جًهىسَت يصش انعشبُت‬ ‫‪1026‬‬

‫نًزجت ويحاكاة وحشخُص األعطال فً شبكت يصغشة‬

‫إعــــــذاد‬

‫انًهنـذط‪ /‬عاطف يحًذ عبذانشاصق ينصىس‬

‫رســانــت يقــديــت إنً كهيت انهندست ‪ -‬جـايعــت انقـاهــرة‬ ‫كجزء ين يتطهباث انحصىل عهً درجت ياجستير انعهىو‬ ‫فً هندست انقىي واآلالث انكهربيت‬

‫ححج إششاف‬ ‫أ‪.‬د‪ /‬عصاو أبى انزهب‬

‫د‪ /‬خانذ نجذي فاسط‬

‫أستــاذ انقىي واالالث انكهربيت‬

‫قسى انكترونياث انطاقت انعانيت وتحىيم انطاقت‬

‫كهيت انهندست ‪,‬جـايعــت انقـاهــرة‬

‫يعهد بحىث االنكترونياث‬


‫نًزجت ويحاكاة وحشخُص األعطال فً شبكت يصغشة‬

‫إعــــــذاد‬

‫انًهنـذط‪ /‬عاطف يحًذ عبذانشاصق ينصىس‬

‫رســانــت يقــديــت إنً كهيت انهندست ‪ -‬جـايعــت انقـاهــرة‬ ‫كجزء ين يتطهباث انحصىل عهً درجت ياجستير انعهىو‬ ‫فً هندست انقىي واآلالث انكهربيت‬


modeling, simulation and fault diagnosis in a microgrid

modeling, simulation and fault diagnosis in a microgrid

Suggest Documents

MICROGRID MODELLING AND SIMULATION

Linear Microcircuit Fault Modeling and Simulation

microgrid modelling and simulation

Fault Detection and Fault Diagnosis Techniques

Aalborg Universitet Fault Diagnosis and Fault

Fault Diagnosis and Fault-tolerant Control - eolss

Open Circuit Fault Diagnosis and Fault Tolerance

Fault Modeling for Monitoring and Diagnosis of Sensor ... - CiteSeerX

Modeling and Fault Diagnosis of Flat Inland Navigation Canals - UPC

Modeling and Fault Diagnosis of Automotive Lead-Acid Batteries

Unknown input modeling and robust fault diagnosis using black box ...

Rectifier Fault Diagnosis and Fault Tolerance of a Doubly Fed ...

Modeling a Hybrid Microgrid Using Probabilistic

Workshop on “Fault Diagnosis and Fault Tolerant Control in large ...

Fault Detection and Diagnosis Methods in Permanent

Decentralised Fault Detection and Diagnosis in ...

Resistive Bridge Fault Modeling, Simulation and Test Generation1

Modeling and Simulation of SI Engines for Fault ...

Modeling and Simulation of SI Engines for Fault Detection - IntechOpen

Fault Detection and Diagnosis

Fault Detection and Diagnosis

INDUSTRIAL FAULT DETECTION AND DIAGNOSIS

Microgrid Modeling, Planning and Operation ...

Behavioral Fault Modeling in a VHDL Synthesis