Environmentally Constraint Economic Dispatch And Reactive Power Planning For Ensuring Secure Operation In Power System.
UNIVERSITI TEKNOLOGI MARA
ENVIRONMENTALLY CONSTRAINT
ECONOMIC DISPATCH AND REACTIVE
POWER PLANNING FOR ENSURING SECURE
OPERATION IN POWER SYSTEM
ELIA ERWANI BINTI HASSAN
Thesis submitted in fulfillment
of the requirements for the degree of
Doctor of Philosophy
Faculty of Electrical Engineering
(2)
CONFIRMATION BY PANEL OF EXAMINERS
I certify that a panel of examiners has met on 8th April 2015 to conduct the final examination of Elia Erwani binti Hassan on her Doctor of Philosophy thesis entitled “Environmentally Constraint Economic Dispatch and Reactive Power Planning for Ensuring Secure Operation in Power System” in accordance with Universiti Teknologi MARA Act 1976 (Akta 173). The Panel of Examniners recommends that the student be awarded the relevant degree. The panel of Examiners was as follows:
Mohd Dani Baba, PhD Professor
Faculty of Electrical Engineering Universiti Teknologi MARA (Chairman)
Muhammad Murtadha Othman, PhD Associate Professor
Faculty of Electrical Engineering Universiti Teknologi MARA (Internal Examiner)
Mohd Wazir Mustafa, PhD Professor
Faculty of Electrical Engineering Universiti Teknologi Malaysia (External Examiner)
Dipti Srinivasan, PhD Associate Professor
Faculty of Electrical Engineering
National University of Singapore, Singapore (External Examiner)
SITI HALIJJAH SHARIFF, PhD
Associate Professor Dean
Institute of Graduates Studies Universiti Teknologi MARA Date: 30 June 2015
(3)
AUTHOR’S DECLARATION
I declare that the work in this thesis was carried out in accordance with the regulations of Universiti Teknologi MARA. It is original and is results of my own work, unless otherwise indicated or acknowledged as referenced work. This thesis has not been submitted to any other academic institution or non-academic institution for any degree or qualification.
I, hereby, acknowledge that I have been supplied with the Academic Rules and Regulations for Post Graduate, Universiti Teknologi MARA, regulating the conduct of my study and research.
Name of Student : Elia Erwani Hassan Student I.D. No : 2009327671
Programme : Doctor of Philosophy (Electrical Engineering) Faculty : Electrical Engineering
Thesis Title : Environmentally Constraint Economic Dispatch and Reactive Power Planning for Ensuring Secure Operation in Power System
Signature of Student : ………. Date : July 2015
(4)
ABSTRACT
Economics and efficient energy dispatch management is necessary to address the increase in energy demand within a limited energy resources while maintaining secure power system operation. Many researches have been conducted to overcome the issues in the implementation of Economic Dispatch (ED). Conventionally, ED problems concern with minimization of total costs while satisfying several operational constraints. In this research, a new optimization technique namely the Adaptive Tumbling Bacterial Foraging Optimization (ATBFO) technique was developed to solve the ED problems. In solving for the ED problems, the impact to the environment was also taken into consideration. Hence, the ED problem is termed Secured Economic Environmental Dispatch (SEED), in which the objective of the optimization now not only minimizing the cost of generation, but also ensuring minimum emission to the environment as well as reducing the total system losses. These objective functions were first considered individually and then were combined to be one multi objective function using the weighted sum approach. The multi objective technique is called Multi objective ATBFO or MOATBFO. The application of the developed optimization technique was extended to solve the Reactive Power Planning (RPP) problems. The objective of conventional RPP problems is to minimize the total power losses in a system. However, in this study, the aspect of security was also taken into consideration in terms of voltage stability condition in solving RPP problems. Hence, the RPP problem is now termed as security constrained RPP (SCRPP). In order to ensure maximum benefit would be obtained as a result of ED and RPP implementation in terms of generation cost minimization, total power losses minimization, while ensuring secure operating condition and minimum impact to environment, the proposed ATBFO and MOATBFO were utilized to solve for the Hybrid of SEED and SCRPP problem. An additional objective function was also taken into consideration in this which is maximum loadability improvement. The performance of the proposed techniques were used in solving SEED, SCRPP and Hybrid of SEED and SCRPP (HSEEDRPP) problems for the IEEE 118 bus system and also the IEEE 57 bus system. The comprehensive analyses were also conducted between two other familiar optimization methods known as original Bacterial Foraging Optimization (BFO) algorithm and Meta heuristic Evolutionary Programming (Meta-EP). From the results it shows that the multi objective ATBFO optimization is able to give better overall improvement in the objective functions for SEED, SCRPP and Hybrid of SEED and SCRPP problems.
(5)
ACKNOWLEDGEMENT
In first, I praise Allah the Almighty for providing me this opportunity and granting me the capability to proceed successfully. My sincere appreciation goes to my research supervisor Professor Dr. Titik Khawa Abdul Rahman for her patience guidance, valuable and constructive comments during the planning and development of this research work.
My special thank is also extended to my supervisor Professor Madya Dr. Zuhaina Zakaria for her advice and assistance in keeping my progress in schedule. I am highly indebted to Universiti Teknikal Malaysia Melaka (UTeM) and Kementerian Pengajian Tinggi (KPT) in funding me for my Doctor of Philosophy.
Last but not least, I wish to thank my dearest husband Dr. Nazrulazhar Bahaman, my son Muhammad Adeeb Amsyar and my daughters Nur Aeen Insyirah and Nur Aimee Irdyinah for their great support and understanding in accomplishing my study. My deepest gratitude also for my beloved mother Maimun Yusop for her enduring prays of my successful.
(6)
TABLE OF CONTENTS
Page
CONFIRMATION BY PANEL OF EXAMINERS ii
AUTHOR’S DECLARATION iii
ABSTRACT iv
ACKNOWLEDGMENT v
TABLE OF CONTENTS vi
LIST OF TABLES xi
LIST OF FIGURES xvii
LIST OF SYMBOLS xx
LIST OF ABBREVIATIONS xxi
CHAPTER ONE: INTRODUCTION 1
1.1 Research Background 1 1.2 Problem Statement 4 1.3 Objectives of the research 6
1.4 Scope of Work 7
1.5 Significant of Research 8 1.6 Organisation of Thesis 9
CHAPTER TWO: LITERATURE REVIEW 10
2.1 Introduction 10
2.2 Secured Environmental Economic Dispatch 10 2.3 Optimal Power Flow 12 2.3.1 Reactive Power Planning 14 2.4 Secured Optimal Power Flow 16 2.5 Secured Reactive Power Planning 17 2.5.1 Load Margin Assessment 19 2.6 Hybrid secured Environmental Economic Dispatch reactive power planning 23 2.7 Deterministic techniques 23 2.8 Heuristic Techniques 24
(7)
2.9 Hybrid techniques 27
2.10Summary 28
CHAPTER THREE: RESEARCH METHODOLOGY 31
3.1 Introduction 31
3.2 Overall Research Methodology 31 3.2.1 Knowledge Acquisition and Background Study 31 3.2.2 Design of Algorithm 33 3.2.3 Execution and Construction 33 3.2.4 Experiment and Analysis 33
3.2.5 Conclusion 34
3.3 Overall optimization techniques implementation in Secured Environmental
Economic Dispatch 34
3.3.1 Objective Function for SEED 35 3.3.1.1 Total Cost Minimization 35 3.3.1.2 Emission Minimization 36 3.3.1.3 Total System Loss 36 3.3.1.4 The operational constraint 37 3.4 Overview on Optimization Techniques 40 3.4.1 Development of Meta Heuristic Evolutionary Programming 40 3.4.2 Development of Bacterial Foraging Optimization Algorithm 42 3.4.3 Development Adaptive Tumbling Bacterial Foraging Optimization
Algorithm 48
3.4.4 Development of ATBFO algorithm for single objective functions
SEED 52
3.4.4.1 Development of ATBFO algorithm for SOSEED 1 54 3.4.4.2 Development of ATBFO algorithm for SOSEED2 54 3.4.4.3 Development of ATBFO algorithm for SOSEED3 55 3.4.5 Development of BFO algorithm for single objective functions SEED 56 3.4.5.1 Development of BFO algorithm for SOSEED1 57 3.4.5.2 Development of BFO algorithm for SOSEED2 58 3.4.5.3 Development of BFO algorithm for SOSEED3 59 3.4.6 Development of Meta-EP algorithm for single objective functions
SEED 59
(8)
3.4.6.1 Development of Meta-EP algorithm for SOSEED1 60 3.4.6.2 Development of Meta-EP algorithm for SOSEED2 60 3.4.6.3 Development of Meta-EP algorithm for SOSEED3 61 3.4.7 Development of MOATBFO algorithm to optimize the multi-objective
functions for SEED problems in power system. 62 3.4.7.1 Development of MOATBFO algorithm for MOSEED 1 64 3.4.7.2 Development of MOATBFO algorithm for MOSEED 2 65 3.4.7.3 Development of MOATBFO algorithm for MOSEED3 66 3.4.7.4 Development of MOATBFO algorithm for MOSEED 4 66 3.4.8 Development of MOBFO algorithm for Multi-Objective SEED
functions 66
3.4.8.1 Development of MOBFO algorithm for MOSEED1 68 3.4.8.2 Development of MOBFO algorithm for MOSEED 2 68 3.4.8.3 Development of MOBFO algorithm MOSEED 3 68 3.4.8.4 Development of MOBFO algorithm for MOSEED4 69 3.4.9 Development of MOMeta-EP Algorithm for Multi Objective SEED
functions 70
3.4.9.1 Development of MOMeta-EP algorithm for MOSEED 1 71 3.4.9.2 Development of MOMeta-EP algorithm for MOSEED2 72 3.4.9.3 Development of MOMeta-EP algorithm for MOSEED 3 72 3.4.9.4 Development of MOMeta-EP algorithm for MOSEED 4 73 3.5 Development of Secured Reactive Power Planning 73 3.5.1 Objective functions for SCRPP 74 3.5.1.1 Maximizing MLP 74 3.5.1.2 Minimizing total system losses 77 3.5.1.3 The Important control variables 77 3.5.2 Development of ATBFO Algorithm for Single Objective Function
SCRPP 79
3.5.2.1 Development ATBFO algorithm for SOSCRPP1 during
unstressed and stressed conditions 83 3.5.2.2 Development ATBFO algorithm for SOSCRPP2 during
unstressed and stressed conditions 83 3.5.3 Development BFO algorithm for SOSCRPP1 during unstressed and
stressed conditions 83 viii
(9)
3.5.4 Development Meta-EP algorithm for SOSCRPP1 during unstressed and stressed conditions 87 3.5.5 Development of new MOATBFO algorithm to optimize the
multi-objective functions for SCRPP 89 3.5.5.1 Multi objective to maximize MLP and minimize system losses or MOSCRPP during unstressed and stressed conditions 91 3.5.6 Development of MOBFO algorithm for MOSCRPP during stressed and
unstressed conditions 91 3.5.7 Development of MOMeta-EP algorithm for MOSCRPP during stressed
and unstressed conditions 95 3.6 Optimization of Hybrid SEED and SCRPP in power system 97
3.6.1 Development of new ATBFO algorithm to optimize single objective solution for HSEEDRPP 100 3.6.1.1 Single objective function of maximizing MLP for unstressed
and stressed conditions or SOHSEEDRPP 102 3.6.2 Development of new MOATBFO algorithm to optimize the
multi-objective functions for HSEEDRPP 104 3.6.2.1 Development MOATBFO algorithm for MOHSEEDRPP1 107 3.6.2.2 Development MOATBFO algorithm for MOHSEEDRPP2 108 3.6.2.3 Development MOATBFO algorithm for MOHSEEDRPP3 109 3.6.2.4 Development MOATBFO algorithm for MOHSEEDRPP4 109 3.6.2.5 Development MOATBFO algorithm for MOHSEEDRPP5 110 3.6.2.6 Development MOATBFO algorithm for MOHSEEDRPP6 110 3.6.3 Development of MOBFO algorithm for MOHSEEDRPP6 111 3.6.4 Development of MOMeta-EP algorithm for MOHSEEDRPP6. 114 3.7 Aggregate function method 117 3.8 Chapter summary 117
CHAPTER FOUR: RESULT AND ANALYSIS 119
4.1 Introduction 119
4.2 Result for SEED optimization solution using ATBFO 119 4.3 Result for Multi Objective SEED using MOATBFO 135 4.4 Results for Secured Reactive Power Planning 146 4.5 Comparison among others optimization techniques 189
(10)
4.6 Results for Hybrid SEED and SCRPP 195 4.7 Comparison WITH others optimization techniques 214 4.8 Chapter summary 222
CHAPTER FIVE: OVERALL CONCLUSION AND
RECOMMENDATION FOR FUTURE RESEARCH 224
REFERENCES 228
APPENDICES 240
(11)
LIST OF TABLES
Tables Title Page
Table 4.1: Initial generating units with the corresponding total cost,
total emission and total loss 120 Table 4.2: Results for the minimum total cost for different number of swim
length, C 121
Table 4.3: The result for different Nc at constant swimlength , C at 0.3 123
Table 4.4 The result for different Nc at constant swimlength, C at 0.5 124
Table 4.5: The result for half Ns respected to their Nc at constant
swimlength, C at 0.3 126 Table 4.6: The results obtained for different number of Ns at constant swim
length, C at 0.3 126 Table 4.7: The results obtained for different number of Nre at constant swim
length, C at 0.3 126 Table 4.8: The results obtained for different number of Ned at constant
swim length , C at 0.3 127 Table 4. 9: Result for single objective function of minimization total cost
(fitness) and observation on total emission and total system
losses or SOSEED 1 127 Table 4.10: Optimal generating units through ATBFO for SOSEED 1 128 Table 4.11: Comparison result between three different optimization
techniques for SOSEED1 129 Table 4.12: The best performance approach between three different
optimization techniques for SOSEED1 130 Table 4.13: Result for single objective function of total emission minimization
(fitness) and observation on total cost and total system losses or
SOSEED 2 131
Table 4.14: Optimal generating units through ATBFO for SOSEED 2 131 Table 4.15: The aggregate performance approach between three different
optimization techniques for SOSEED2 132 xi
(12)
Table 4.16: The aggregate performance approach between three different
optimization techniques for SOSEED 2 133 Table 4.17: Result for single objective function minimization of total system
loss (fitness) and observation on total cost and total emission or
SOSEED 3 133
Table 4.18: The optimum generating units for SOSEED 3 134
Table 4.19: The aggregate performance approach between three different
optimization techniques for SOSEED 3 134 Table 4.20: The aggregate performance approach between three different
optimization techniques for SOSEED 3 135 Table 4.21: Result for multi-objective function total cost and total losses or
MOSEED1 136
Table 4. 22: The optimum generating units for MOSEED 1 137 Table 4.23: The aggregate performance approach between three different
optimization techniques for MOSEED 1 137 Table 4.24: Result for multi-objective function total cost and total emission
or MOSEED 2 138
Table 4.25: The optimum generating units for MOSEED 2 139 Table 4.26: The aggregate performance approach between three different
optimization techniques for MOSEED 2 140 Table 4.27: Result for multi-objective function total emission and total
losses or MOSEED3 141 Table 4.28: The optimum generating units for MOSEED 3 141
Table 4 29: The aggregate performance approach between three different
optimization techniques for MOSEED 2 142 Table 4.30: Result for multi-objective function total emission and total
losses or MOSEED4 143 Table 4.31: The optimum generating units for MOSEED 4 144 Table 4.32: The aggregate performance approach between three different
optimization techniques for MOSEED 4 144 Table 4.33: Overall results for different objective functions as well as
observations on total cost, total emission and total losses 145 xii
(13)
Table 4.34: Parameter considerations for ATBFO model 148 Table 4.35: P load increment during before the implementation of SCRPP
(Point A) for unstressed conditions 149 Table 4.36: P load increment during the implementation of SCRPP
(Point A’ and B) for unstressed conditions 150 Table 4.37: The P load increment during post-optimization (Point A’ and B)
for stressed condition 153 Table 4.38: The Q load increment during post-optimization (Point A’ and B)
for unstressed conditions 155 Table 4.39: The Q load increment during post-optimization (Point A’ and B)
for stressed condition 157 Table 4.40: The Q & P load increment during post-optimization
(Point A’ and B) for unstressed conditions 159 Table 4.41: The Q & P load increment during after the implementation
of SCRPP (Point A’ and B) for stressed conditions 161 Table 4.42: The P load increment after the implementation of SCRPP
(Point A’ and B) for stressed condition in Case 2 163 Table 4.43: The P loads increment after the implementation of SCRPP
(Point A’ and B) for stressed condition in Case 2 164 Table 4.44: The Q loads increment after the implementation of SCRPP
(Point A’ and B) for unstressed condition in Case 2 165 Table 4.45: The Q loads increment during after the implementation of SCRPP
(Point A’ and B) for stressed condition in Case 2 165 Table 4.46: The Q and P loads increment after the implementation of SCRPP
(Point A’ and B) for unstressed condition in Case 2 166 Table 4.47: The Q and P loads increment during after the implementation of
SCRPP (Point A’ and B) for stressed condition in Case 2 167 Table 4.48: The P loads increment on SOSCRPP2 after the implementation
of SCRPP (Point A’ and B) for unstressed and stressed condition
in Case 1 168
Table 4.49: The Q load increment on SOSCRPP2 at Point A’ and B for
unstressed and stressed conditions in Case 1 168 Table 4.50: The Q and P load increment on SOSCRPP2 during after the
implementation of SCRPP (Point A’ and B) for unstressed and
stressed conditions for Case 1 169 xiii
(14)
Table 4.51: The P load increment on SOSCRPP2 after the implementation of SCRPP (Point A’ and B) for unstressed and stressed conditions in
Case 2 170
Table 4.52: The Q load increment on SOSCRPP2 after the implementation of SCRPP (Point A’ and B) for unstressed and stressed conditions in
Case 2 171
Table 4 53: The Q and P load increment on SOSCRPP2 after the
implementation of SCRPP (Point A’ and B) for unstressed
and stressed conditions in Case 2 171 Table 4.54: Comparison between SOSCRPP1 and SOSCRPP2 at Point A’
(after the implementation of SCRPP) for Case 1 172 Table 4.55: Comparison between SOSCRPP1 and SOSCRPP2 at Point A’
(post optimization) for Case 2 173 Table 4.56: Maximum P load after the implementation of MOSCRPP
(Point A’ and B) for unstressed condition in Case 1. 176 Table 4.57: Maximum P load after the implementation of MOSCRPP
(Point A’ and B) for stressed condition in Case 1. 177 Table 4.58: Maximum Q load after the implementation of MOSCRPP
(Point A’ and B) for unstressed condition in Case 1. 178 Table 4.59: Maximum Q load after the implementation of MOSCRPP
(Point A’ and B) for stressed condition in Case 1 178 Table 4.60: Maximum P and Q load after the implementation of MOSCRPP
(Point A’ and B) for unstressed condition in Case 1 179 Table 4.61: Maximum P and Q load after the implementation of MOSCRPP
(Point A’ and B) for stressed condition in Case 1. 180 Table 4.62: Maximum P load after the implementation of MOSCRPP
(Point A’ and B) for unstressed condition in Case 2. 181 Table 4.63: Maximum P load after the implementation of MOSCRPP
(Point A’ and B) for stressed condition in Case 2 181 Table 4.64: Maximum Q load after the implementation of MOSCRPP
(Point A’ and B) for unstressed condition in Case 2 182 Table 4.65: Maximum Q load after the implementation of MOSCRPP
(Point A’ and B) for stressed condition in Case 2 183
(15)
Table 4.66: Maximum P and Q load after the implementation of MOSCRPP
(Point A’ and B) for unstressed condition in Case 2 183 Table 4.67: Maximum P and Q load after the implementation of MOSCRPP
(Point A’ and B) for stressed condition in Case 2 184 Table 4.68: Comparison between SOSCRPP1 and MOSCRPP at Point A’ in
Case 1 185
Table 4.69: Comparison between SOSCRPP1 and MOSCRPP at Point A’ in
Case 2 187
Table 4.70: Comparison between ATBFO and others optimization
techniques for SOSCRPP1 190 Table 4.71: Comparison between ATBFO and others optimization
techniques for SOSCRPP1 using aggregate performance 191 Table 4.72: Comparison between ATBFO and others optimization
techniques for SOSCRPP1 for overall performance 192 Table 4.73: Comparison between MOTBFO and others optimization
techniques for MOSCRPP 193 Table 4.74: Comparison between MOATBFO and others optimization
techniques for MOSCRPP using aggregate performance 194 Table 4.75: Comparison between ATBFO and others optimization
techniques for MOSCRPP for overall performance 195 Table 4.76: Maximum P load after the implementation of HSEEDRPP
(Point A’ and B) for unstressed condition in Case 1. 197 Table 4.77: The aggregate function to identify the P load increment for
unstressed condition in Case 198 Table 4.78: Maximum P load after the implementation of HSEEDRPP
(Point A’ and B) for stressed condition in Case 1. 199 Table 4.79: The aggregate function to identify the P load increment for
stressed condition in Case 1 199 Table 4.80: Maximum Q load after the implementation of HSEEDRPP
(Point A’ and B) for unstressed condition in Case 1 200 Table 4.81: The aggregate function to identify the Q load increment for
stressed condition in Case 1 201 Table 4.82: Maximum Q load after the implementation of HSEEDRPP
(Point A’ and B) for stressed condition in Case 1 201 xv
(16)
Table 4.83: The aggregate function to identify the Q load increment for
stressed condition in Case 1 202 Table 4.84: Maximum P & Q load after the implementation of HSEEDRPP
(Point A’ and B) for unstressed condition in Case 1 203 Table 4.85: The aggregate function to identify the Q and P load increment for
unstressed condition in Case 1 203 Table 4.86: Maximum P & Q load after the implementation of HSEEDRPP
(Point A’ and B) for stressed condition in Case 1 204 Table 4.87: The aggregate function to identify the Q and P load increment for
stressed condition in Case 205 Table 4.88: Maximum P load after the implementation of HSEEDRPP
(Point A’ and B) for unstressed condition in Case 2 206 Table 4.89: The aggregate function to identify the P load increment for
unstressed condition in Case 2 206 Table 4.90: Maximum P load after the implementation of HSEEDRPP (Point A’
and B) for stressed condition in Case 2 207 Table 4.91: The aggregate function to identify the P load increment for
stressed condition in Case 2 208 Table 4.92: Maximum Q load after the implementation of HSEEDRPP
(Point A’ and B) for unstressed condition in Case 2. 208 Table 4.93: The aggregate function to identify the Q load increment for
unstressed condition in Case 2 209 Table 4.94: Maximum Q load after the implementation of HSEEDRPP
(Point A’ and B) for stressed condition in Case 2. 209 Table 4. 95: The aggregate function to identify the Q load increment for
stressed condition in Case 2 210 Table 4.96: Maximum P & Q load after the implementation of HSEEDRPP
(Point A’ and B) for unstressed condition in Case 2 211 Table 4.97: The aggregate function to identify the Q and P load increment
for unstressed condition in Case 2 211 Table 4.98: Maximum P & Q load after the implementation of HSEEDRPP
(Point A’ and B) for stressed condition in Case 2 212 Table 4.99: The aggregate function to identify the Q and P load increment for
stressed condition in Case 2 213 xvi
(17)
Table 4.100: The aggregate function values for each objective function
in providing the Hybrid of SEED and SCRPP solutions 214 Table 4.101: The comparison results between ATBFO and others optimization
techniques for SOHSEEDRPP. 215 Table 4.102: The comparison aggregate values between ATBFO and others
optimization techniques for SOHSEEDRPP 217 Table 4.103: Comparison of overall aggregate values between ATBFO and
others optimization techniques. 218 Table 4.104: Results Comparison between MOATBFO and others
optimization techniques for MOHSEEDRPP6 219 Table 4.105: Aggregate Values Comparison between MOATBFO and others
optimization techniques. 221 Table 4.106: The comparison overall aggregate values between MOATBFO
and others optimization techniques. 222
(18)
LIST OF FIGURES
Figures Title Page
Figure 3.1 Overall research methodology 32 Figure 3.2 The Scope and Limitation for SEED 38 Figure 3.3 The overall methodology for SEED The comparative study in the
implementation of SEED for
obtaining the best solutions 39 Figure 3.4 Flow chart of ATBFO process for SOSEED1 55 Figure 3.5 Flowchart of BFO process for SOSEED1 58 Figure 3.6 Flowchart of Meta-EP process for SOSEED1 61 Figure 3.7 The flowchart of MOATBFO process for MOSEED1 65 Figure 3.8 The flowchart of MOBFO process for MOSEED1 69 Figure 3.9 The flowchart of MOMeta-EP process for MOSEED1 72 Figure 3.10 Scope and Limitation for SCRPP 74 Figure 3.11 Load margin assessment 745 Figure 3.12 Load margin assessment 745 Figure 3.13 Graph for comparison between pre and post SCRPP implementation
for unstressed condition 746 Figure 3.14 Graph for comparison between pre and post SCRPP implementation
for stressed condition 747 Figure 3.15 The overall methodology for SCRPP (i) 78 Figure 3.16 The overall methodology for SCRPP (ii) 82 Figure 3.17 Flowchart of ATBFO process for SOSCRPP1 for Case 1 and
Case 2 during unstressed and stressed condition 84 Figure 3.18 Flowchart of BFO process for SOSCRRP1 for Case 1 and Case 2
during unstressed and stressed condition 86 Figure 3.19 Flowchart of Meta-EP process for SOSCRPP1 for Case 1 and
Case 2 during unstressed and stressed condition 88 Figure 3.20 The flowchart of MOATBFO process for MOSCRPP for Case 1
and Case 2 during unstressed and stressed condition 92 xviii
(19)
Figure 3.21 The flowchart of MOBFO process for MOSCRPP for Case 1 and Case 2 during unstressed and stressed condition 94 Figure 3.22 The flowchart of MOMeta-EP process for MOSCRPP for Case 1
and Case 2 during unstressed and stressed conditio 97 Figure 3.23 The Scope and Limitation for HSEEDRPP 98 Figure 3.24 The overall methodology for HSEEDRPP (i) 99 Figure 3.25 The overall methodology for HSEEDRPP (ii) 103 Figure 3.26 The flowchart ATBFO process for SOHSEEDRPP for Case 1 and
Case 2 during unstressed and stressed condition 104 Figure 3.27 The flowchart MOATBFO process for MOHSEEDRPP1 for
Case 1 and Case 2 during unstressed and stressed condition 108 Figure 3.28 The flowchart MOATBFO process for MOHSEEDRPP 6 for
Case 1 and Case 2 during unstressed and stressed condition 111 Figure 3.29 The flowchart MOBFO process for MOHSEEDRPP6 for Case 1
and Case 2 during unstressed and stressed condition 114 Figure 3.30 The flowchart MOMeta-EP process for MOHSEEDRPP6 for
Case 1 and Case 2 during unstressed and stressed condition 116 Figure 4.1 Graph for results for the minimum total cost of different
number of swimlength, C 122 Figure 4.2 Graph for different Nc with constant swim length at 0.3 124 Figure 4.3 Graph for different Nc with constant swim length at 0.5. 125 Figure 4.4 Graph on outputs for three different optimization techniques for
SOSEED1 129
Figure 4.5 Graph for total emission as an objective function with an
observation on total cost and total losses 132 Figure 4.6 Graph for Total losses as an objective function observation on
total cost and total emission 135 Figure 4.7 Graph for MOSEED1 among three optimization techniques 138 Figure 4.8 Graph for MOSEED 2 among three optimization techniques 140 Figure 4.9 Graph for multi-objectives on total emission and total losses 142 Figure 4.10 Graph for MOSEED4 145 Figure 4.11 Graph to depict the Point A (before the implementation
of SCRPP) and Point B (after the implementation of SCRPP) 147 Figure 4.12 SOSCRPP 1 voltage and losses for P load increment in Case 1 151
(20)
Figure 4.13 SOSCRPP 1 voltage and losses for P load increment in Case 1
at Point A and A’ 152 Figure 4.14 SOSCRPP 1 voltage and losses for SOSCRPP 1 for P load
increment in Case1 at Point A and A’ for stressed conditions 154 Figure 4.15 SOSCRPP 1 voltage and losses for Q load increment in Case1
at Point A and A’ for unstressed conditions 156 Figure 4.16 SOSCRPP 1 voltage and losses for Q load increment in Case1
at Point A and A’ 158 Figure 4.17 SOSCRPP 1 voltage and losses for Q & P load increment in
Case1 at Point A and A’ 160 Figure 4.18 SOSCRPP 1 voltage and losses for Q & P load increment in
Case1 at Point A and A’ 162 Figure 4.19 Graph of comparison minimum voltage and losses between
SOSCRPP1 (MLP) and SOSCRPP2 (Losses) for Case1 at Point A’ 174 Figure 4.20 Graph of comparison minimum voltage and losses between
SOSCRPP1 (MLP) and SOSCRPP2 (Losses) for Case2 at Point A’ 175
(21)
LIST OF SYMBOLS
Symbols
��(���) Cost of generation for unit i
��� Power generated by unit i
��, ��, �� Cost coefficient for unit i
������ Total cost of generation
�� Number of generator units
��� Power generated by unit i
αi , βi , �i , Ɛi, λi Emission coefficient of i th generator
�� Power generated by unit i
Qi and Qj Reactive power at sending and receiving buses respectively.
��� Generated reactive power of bus i
������� Voltage magnitude at sending and receiving buses respectively
������, Total active power loss over the network �� Load bus
��� Voltage controlled bus
�� Reference (slack) bus
Pmin Minimum real power generated by unit i
Pmax Maximum real power generated by unit i
Qmin Minimum reactive power generated by unit i
Qmax Maximum reactive power generated by unit i
Vmin Minimum voltage at load buses
Vmax Maximum voltage at load buses
Xmer Transformer tap changing setting
Qinj Compensating capacitor injected
Qgs Reactive power generated k Numbers of objective function.
αi Weighting factor for ith objective function
fni Normalised value for ith objective function
(22)
LIST OF ABBREVIATIONS
Abbreviations
ED Economic Dispatch
ATBFO Adaptive Tumbling Bacterial Foraging Optimization SEED Secured Economic Environmental Dispatch
MOATBFO Multi-objective Adaptive Tumbling Bacterial Foraging Optimization
OPF Optimal Power Flow
BFO Bacterial Foraging Optimization
Meta-EP Meta heuristic Evolutionary Programming RPP Reactive Power Planning
SCOPF Secured Reactive Power Planning VSM Voltage Stability Margin
LP Linear Programming NLP Non Linear Programming
MINLP Mix Integer Non Linear Programming GA Genetic Algorithm
PSO Particle Swarm Optimization EP Evolutionary Programming SA Simulated Annealing ACO Ant Colony Optimization AIS Artificial Immune System TS Tabu Search
HSEEDRPP Hybrid Secured Environmental Economic Dispatch Reactive Power Planning
RPD Reactive Power Dispatch xxii
(23)
TTCS Transformer Tap Changer Setting CP Capacitor Placement
TTC Total Transfer Capability MLP Maximum Loading Point AI Artificial Intelligence SVC Static VAR Compensator
CEED Combined Economic Emission Dispatch EED Environmental Economic Dispatch NSGA Nondominated Sorted Genetic Algorithm NPGA Niched Pareto Genetic Algorithm
SPEA Strength Pareto Evolutionary Algorithm FVSI Fast Voltage Stability Index
SI Stability Index
VSM Voltage Stability Margin LM Load Margin
MLP Maximum Loading Point SOSEED Single Objective SEED MOSEED Multi-Objective SEED MOBFO Multi-Objective BFO MOMeta-EP Multi-Objective Meta -EP SOSCRPP Single Objective SCRPP MOSCRPP Multi- Objective SCRPP SOHSEEDRPP Single Objective HSEEDRPP MOHSEEDRPP Multi- Objective HSEEDRPP
(24)
CHAPTER ONE
INTRODUCTION
1.1 RESEARCH BACKGROUND
Power system optimization is a vital study for optimal power operation to provide smooth and sustainable load demand [1]. The rises of energy demand and insufficient of energy resources required for quality and secured dispatch. A well-coordinated and optimized power system operation helps in satisfying Economic Dispatch (ED) among users of power networks. This requires for the researches to be conducted in order to study and develop new tools so that the optimization issues in ED could be overcome.
Basically, the principal objective of load dispatch is to minimize the total fuel cost while satisfying the requirements of some important operational parameters. In today’s environment, efficient load dispatch requires not only to schedule the power generation at the least cost but also to consider the other performance factors to be optimized in power flow over the networks. The obligation of social attentions have influenced in reducing the energy conservation and pollution emission produced by power plants. For that reason, the total cost function alone is no longer suitable as the main focus in optimizing the ED problems. In order to reduce pollution as a result of electrical power generation, minimization on emission should be added to objective function of ED which is generation cost minimization [2]. However, ED problems also subjected to the operational constraints and security criteria of a power system, so that the secured and economic loads are dispatched equally. Therefore, the support to EDs is strongly related to the established Optimal Power Flow (OPF) over the power networks.
In recent development, deregulation has made a great pressure in United States (US) power industry in providing economic load dispatch [3]. Thus, they found that reactive power support is critical and vital to sustain voltage and regulate power factor in electric power systems. This is proven by the Great 2003 Blackout over northeastern US and Canada in August 2003 caused by poor planning and managing of reactive power in US power system. As a consequence, several objectives functions are suggested from researchers in this field in order adequate Reactive
(1)
Figure 3.21 The flowchart of MOBFO process for MOSCRPP for Case 1 and Case 2 during unstressed and stressed condition 94 Figure 3.22 The flowchart of MOMeta-EP process for MOSCRPP for Case 1
and Case 2 during unstressed and stressed conditio 97 Figure 3.23 The Scope and Limitation for HSEEDRPP 98 Figure 3.24 The overall methodology for HSEEDRPP (i) 99 Figure 3.25 The overall methodology for HSEEDRPP (ii) 103 Figure 3.26 The flowchart ATBFO process for SOHSEEDRPP for Case 1 and
Case 2 during unstressed and stressed condition 104 Figure 3.27 The flowchart MOATBFO process for MOHSEEDRPP1 for
Case 1 and Case 2 during unstressed and stressed condition 108 Figure 3.28 The flowchart MOATBFO process for MOHSEEDRPP 6 for
Case 1 and Case 2 during unstressed and stressed condition 111 Figure 3.29 The flowchart MOBFO process for MOHSEEDRPP6 for Case 1
and Case 2 during unstressed and stressed condition 114 Figure 3.30 The flowchart MOMeta-EP process for MOHSEEDRPP6 for
Case 1 and Case 2 during unstressed and stressed condition 116 Figure 4.1 Graph for results for the minimum total cost of different
number of swimlength, C 122
Figure 4.2 Graph for different Nc with constant swim length at 0.3 124 Figure 4.3 Graph for different Nc with constant swim length at 0.5. 125 Figure 4.4 Graph on outputs for three different optimization techniques for
SOSEED1 129
Figure 4.5 Graph for total emission as an objective function with an
observation on total cost and total losses 132 Figure 4.6 Graph for Total losses as an objective function observation on
total cost and total emission 135
Figure 4.7 Graph for MOSEED1 among three optimization techniques 138 Figure 4.8 Graph for MOSEED 2 among three optimization techniques 140 Figure 4.9 Graph for multi-objectives on total emission and total losses 142
Figure 4.10 Graph for MOSEED4 145
Figure 4.11 Graph to depict the Point A (before the implementation
of SCRPP) and Point B (after the implementation of SCRPP) 147 Figure 4.12 SOSCRPP 1 voltage and losses for P load increment in Case 1 151
(2)
Figure 4.13 SOSCRPP 1 voltage and losses for P load increment in Case 1
at Point A and A’ 152
Figure 4.14 SOSCRPP 1 voltage and losses for SOSCRPP 1 for P load
increment in Case1 at Point A and A’ for stressed conditions 154 Figure 4.15 SOSCRPP 1 voltage and losses for Q load increment in Case1
at Point A and A’ for unstressed conditions 156 Figure 4.16 SOSCRPP 1 voltage and losses for Q load increment in Case1
at Point A and A’ 158
Figure 4.17 SOSCRPP 1 voltage and losses for Q & P load increment in
Case1 at Point A and A’ 160
Figure 4.18 SOSCRPP 1 voltage and losses for Q & P load increment in
Case1 at Point A and A’ 162
Figure 4.19 Graph of comparison minimum voltage and losses between
SOSCRPP1 (MLP) and SOSCRPP2 (Losses) for Case1 at Point A’ 174 Figure 4.20 Graph of comparison minimum voltage and losses between
SOSCRPP1 (MLP) and SOSCRPP2 (Losses) for Case2 at Point A’ 175
(3)
LIST OF SYMBOLS
Symbols
��(���) Cost of generation for unit i ��� Power generated by unit i
��, ��, �� Cost coefficient for unit i ������ Total cost of generation
�� Number of generator units
��� Power generated by unit i
αi , βi , �i , Ɛi, λi Emission coefficient of i th generator �� Power generated by unit i
Qi and Qj Reactive power at sending and receiving buses respectively.
��� Generated reactive power of bus i
������� Voltage magnitude at sending and receiving buses respectively ������, Total active power loss over the network
�� Load bus
��� Voltage controlled bus
�� Reference (slack) bus
Pmin Minimum real power generated by unit i
Pmax Maximum real power generated by unit i Qmin Minimum reactive power generated by unit i Qmax Maximum reactive power generated by unit i Vmin Minimum voltage at load buses
Vmax Maximum voltage at load buses Xmer Transformer tap changing setting
Qinj Compensating capacitor injected
Qgs Reactive power generated
k Numbers of objective function.
αi Weighting factor for ith objective function fni Normalised value for ith objective function
(4)
LIST OF ABBREVIATIONS
Abbreviations
ED Economic Dispatch
ATBFO Adaptive Tumbling Bacterial Foraging Optimization SEED Secured Economic Environmental Dispatch
MOATBFO Multi-objective Adaptive Tumbling Bacterial Foraging Optimization
OPF Optimal Power Flow
BFO Bacterial Foraging Optimization
Meta-EP Meta heuristic Evolutionary Programming RPP Reactive Power Planning
SCOPF Secured Reactive Power Planning VSM Voltage Stability Margin
LP Linear Programming
NLP Non Linear Programming
MINLP Mix Integer Non Linear Programming
GA Genetic Algorithm
PSO Particle Swarm Optimization
EP Evolutionary Programming
SA Simulated Annealing ACO Ant Colony Optimization AIS Artificial Immune System
TS Tabu Search
HSEEDRPP Hybrid Secured Environmental Economic Dispatch Reactive Power Planning
RPD Reactive Power Dispatch xxii
(5)
TTCS Transformer Tap Changer Setting
CP Capacitor Placement
TTC Total Transfer Capability MLP Maximum Loading Point AI Artificial Intelligence SVC Static VAR Compensator
CEED Combined Economic Emission Dispatch EED Environmental Economic Dispatch NSGA Nondominated Sorted Genetic Algorithm NPGA Niched Pareto Genetic Algorithm
SPEA Strength Pareto Evolutionary Algorithm FVSI Fast Voltage Stability Index
SI Stability Index
VSM Voltage Stability Margin
LM Load Margin
MLP Maximum Loading Point SOSEED Single Objective SEED MOSEED Multi-Objective SEED MOBFO Multi-Objective BFO MOMeta-EP Multi-Objective Meta -EP SOSCRPP Single Objective SCRPP MOSCRPP Multi- Objective SCRPP SOHSEEDRPP Single Objective HSEEDRPP MOHSEEDRPP Multi- Objective HSEEDRPP
(6)
CHAPTER ONE
INTRODUCTION
1.1 RESEARCH BACKGROUND
Power system optimization is a vital study for optimal power operation to provide smooth and sustainable load demand [1]. The rises of energy demand and insufficient of energy resources required for quality and secured dispatch. A well-coordinated and optimized power system operation helps in satisfying Economic Dispatch (ED) among users of power networks. This requires for the researches to be conducted in order to study and develop new tools so that the optimization issues in ED could be overcome.
Basically, the principal objective of load dispatch is to minimize the total fuel cost while satisfying the requirements of some important operational parameters. In today’s environment, efficient load dispatch requires not only to schedule the power generation at the least cost but also to consider the other performance factors to be optimized in power flow over the networks. The obligation of social attentions have influenced in reducing the energy conservation and pollution emission produced by power plants. For that reason, the total cost function alone is no longer suitable as the main focus in optimizing the ED problems. In order to reduce pollution as a result of electrical power generation, minimization on emission should be added to objective function of ED which is generation cost minimization [2]. However, ED problems also subjected to the operational constraints and security criteria of a power system, so that the secured and economic loads are dispatched equally. Therefore, the support to EDs is strongly related to the established Optimal Power Flow (OPF) over the power networks.
In recent development, deregulation has made a great pressure in United States (US) power industry in providing economic load dispatch [3]. Thus, they found that reactive power support is critical and vital to sustain voltage and regulate power factor in electric power systems. This is proven by the Great 2003 Blackout over northeastern US and Canada in August 2003 caused by poor planning and managing of reactive power in US power system. As a consequence, several objectives functions are suggested from researchers in this field in order adequate Reactive