Applications of Modern Heuristic Optimization Methods in Power and Energy Systems
Description
Written by a team of international experts, this book describes the use of metaheuristic applications in the analysis and design of electric power systems. This includes a discussion of optimum energy and commitment of generation (nonrenewable & renewable) and load resources during day-to-day operations and control activities in regulated and competitive market structures, along with transmission and distribution systems.
Applications of Modern Heuristic Optimization Methods in Power and Energy Systems begins with an introduction and overview of applications in power and energy systems before moving on to planning and operation, control, and distribution. Further chapters cover the integration of renewable energy and the smart grid and electricity markets. The book finishes with final conclusions drawn by the editors.
Applications of Modern Heuristic Optimization Methods in Power and Energy Systems:
* Explains the application of differential evolution in electric power systems' active power multi-objective optimal dispatch
* Includes studies of optimization and stability in load frequency control in modern power systems
* Describes optimal compliance of reactive power requirements in near-shore wind power plants
* Features contributions from noted experts in the field
Ideal for power and energy systems designers, planners, operators, and consultants, Applications of Modern Heuristic Optimization Methods in Power and Energy Systems will also benefit engineers, software developers, researchers, academics, and students.
More details
Other editions
Additional editions
Persons
KWANG Y. LEE, PhD, is a Professor and Chair of Electrical and Computer Engineering at Baylor University. He is active in the Intelligent Systems Subcommittee and Station Control Subcommittee of the IEEE Power and Energy Society. He served as Editor of IEEE Transactions on Energy Conversion and Associate Editor of IEEE Transactions on Neural Networks and IFAC Journal on Control Engineering Practice.
ZITA A. VALE, PhD, is a Full Professor in the Electrical Engineering Department at the School of Engineering of the Polytechnic of Porto and Director of GECADResearch Group on Intelligent Engineering and Computing for Advanced Innovation and Development. She has published over 800 works, including more than 100 papers in international scientific journals.
Content
Preface xv
Contributors xvii
List of Figures xxi
List of Tables xxxiii
Chapter 1 Introduction 1
1.1 Background 1
1.2 Evolutionary Computation: A Successful Branch of CI 3
1.2.1 Genetic Algorithm 6
1.2.2 Non-dominated Sorting Genetic Algorithm II 8
1.2.3 Evolution Strategies and Evolutionary Programming 8
1.2.4 Simulated Annealing 9
1.2.5 Particle Swarm Optimization 10
1.2.6 Quantum Particle Swarm Optimization 10
1.2.7 Multi-objective Particle Swarm Optimization 11
1.2.8 Particle Swarm Optimization Variants 12
1.2.9 Artificial Bee Colony 13
1.2.10 Tabu Search 14
References 15
Chapter 2 Overview Of Applications In Power And Energy Systems 21
2.1 Applications to Power Systems 21
2.1.1 Unit Commitment 23
2.1.2 Economic Dispatch 24
2.1.3 Forecasting in Power Systems 25
2.1.4 Other Applications in Power Systems 27
2.2 Smart Grid Application Competition Series 28
2.2.1 Problem Description 29
2.2.2 Best Algorithms and Ranks 30
2.2.3 Further Information and How to Download 32
References 32
Chapter 3 Power System Planning And Operation 39
3.1 Introduction 39
3.2 Unit Commitment 40
3.2.1 Introduction 40
3.2.2 Problem Formulation 40
3.2.3 Advancement in UCP Formulations and Models 42
3.2.4 Solution Methodologies, State-of-the-Art, History, and Evolution 46
3.2.5 Conclusions 56
3.3 Economic Dispatch Based on Genetic Algorithms and Particle Swarm Optimization 56
3.3.1 Introduction 56
3.3.2 Fundamentals of Genetic Algorithms and Particle Swarm Optimization 58
3.3.3 Economic Dispatch Problem 60
3.3.4 GA Implementation to ED 63
3.3.5 PSO Implementation to ED 71
3.3.6 Numerical Example 79
3.3.7 Conclusions 87
3.4 Differential Evolution in Active Power Multi-Objective Optimal Dispatch 87
3.4.1 Introduction 87
3.4.2 Differential Evolution for Multi-Objective Optimization 88
3.4.3 Multi-Objective Model of Active Power Optimization for Wind Power Integrated Systems 97
3.4.4 Case Studies 100
3.4.5 Analyses of Dispatch Plan 105
3.4.6 Conclusions 106
3.5 Hydrothermal Coordination 106
3.5.1 Introduction 106
3.5.2 Hydrothermal Coordination Formulation 107
3.5.3 Problem Decomposition 110
3.5.4 Case Studies 111
3.5.5 Conclusions 114
3.6 Meta-Heuristic Method for Gms Based on Genetic Algorithm 115
3.6.1 History 115
3.6.2 Meta-heuristic Search Method 116
3.6.3 Flexible GMS 119
3.6.4 User-Friendly GMS System 131
3.6.5 Conclusion 141
3.7 Load Flow 143
3.7.1 Introduction 143
3.7.2 Load Flow Analysis in Electrical Power Systems 144
3.7.3 Particle Swarm Optimization and Mutation Operation 148
3.7.4 Load Flow Computation via Particle Swarm Optimization with Mutation Operation 150
3.7.5 Numerical Results 153
3.7.6 Conclusions 160
3.8 Artificial Bee Colony Algorithm for Solving Optimal Power Flow 161
3.8.1 Optimization in Power System Operation 162
3.8.2 The Optimal Power Flow Problem 162
3.8.3 Artificial Bee Colony 166
3.8.4 ABC for the OPF Problem 168
3.8.5 Case Studies 170
3.8.6 Conclusions 176
3.9 OPF Test Bed and Performance Evaluation of Modern Heuristic Optimization 176
3.9.1 Introduction 176
3.9.2 Problem Definition 177
3.9.3 OPF Test Systems 178
3.9.4 Differential Evolutionary Particle Swarm Optimization: DEEPSO 183
3.9.5 Enhanced Version of Mean-Variance Mapping Optimization Algorithm: MVMO-PHM 187
3.9.6 Evaluation Results 193
3.9.7 Conclusions 196
3.10 Transmission System Expansion Planning 197
3.10.1 Introduction 197
3.10.2 Transmission System Expansion Planning Models 198
3.10.3 Mathematical Modeling 199
3.10.4 Challenges 201
3.10.5 Application of Meta-heuristics to TEP 202
3.10.6 Conclusions 210
3.11 Conclusion 210
References 210
Chapter 4 Power System And Power Plant Control 227
4.1 Introduction 227
4.2 Load Frequency Control - Optimization and Stability 228
4.2.1 Introduction 228
4.2.2 Load Frequency Control 229
4.2.3 Components of Active Power Control System 230
4.2.4 Designing LFC Structure for an Interconnected Power System 232
4.2.5 Parameter Optimization and System Performance 237
4.2.6 System Stability in the Presence of Communication Delay 242
4.2.7 Conclusions 244
4.3 Control of Facts Devices 244
4.3.1 Introduction 244
4.3.2 Role of FACTS 246
4.3.3 Static Modeling of FACTS devices 247
4.3.4 Power Flow Control using FACTS 255
4.3.5 Optimal Power Flow Using Suitability FACTS devices 259
4.3.6 Use of Particle Swarm Optimization 281
4.3.7 Conclusions 283
4.4 Hybrid of Analytical and Heuristic Techniques for facts Devices 284
4.4.1 Introduction 284
4.4.2 Heuristic Algorithms 285
4.4.3 SVC and Voltage Instability Improvement 288
4.4.4 FACTS Devices and Angle Stability Improvement 293
4.4.5 Selection of Supplementary Input Signals for Damping Inter-area Oscillations 295
4.4.6 TCSC and Improvement of Total Transfer Capability 302
4.4.7 Conclusions 305
4.5 Power System Automation 305
4.5.1 Introduction 305
4.5.2 Application of PSO on Power System's Corrective Control 307
4.5.3 Genetic Algorithm-aided DTs for Load Shedding 322
4.5.4 Power System-Controlled Islanding 324
4.5.5 Application of the method on the IEEE - 30 buses test system 326
4.5.6 Application of the method on the IEEE - 118 buses test system 327
4.5.7 Conclusions 327
4.5.8 Appendix 328
4.6 Power Plant Control 334
4.6.1 Introduction 334
4.6.2 Coal Mill Modeling 335
4.6.3 Nonlinear Model Predictive Control of Reheater Steam Temperature 340
4.6.4 Multi-objective Optimization of Boiler Combustion System 345
4.6.5 Conclusions 355
4.7 Predictive Control in Large-Scale Power Plant 355
4.7.1 Introduction 355
4.7.2 Particle Swarm Optimization Algorithm 356
4.7.3 Performance Prediction Model Development Based on NARMA Model 357
4.7.4 Design of Intelligent MPOC Scheme 361
4.7.5 Control Simulation Tests 364
4.7.6 Conclusions 367
4.8 Conclusion 368
References 369
Chapter 5 Distribution System 381
5.1 Introduction 381
5.2 Active Distribution Network Planning 382
5.2.1 Introduction 382
5.2.2 Problem Formulation 382
5.2.3 Overview of the Solution Techniques for Distribution Network Planning 385
5.2.4 Genetic Algorithm Solution to Active Distribution Network Planning Problem 385
5.2.5 Numerical Results 388
5.2.6 Conclusions 392
5.3 Optimal Selection of Distribution System Architecture 392
5.3.1 Introduction 392
5.3.2 Deterministic Optimization Techniques 393
5.3.3 Stochastic Optimization Techniques 394
5.3.4 Multi-Objective Optimization 400
5.3.5 Mathematical Modeling for Power System Components 401
5.3.6 AC/DC Power Flow in Hybrid Networks 405
5.3.7 Pareto-Based Multi-Objective Optimization Problem 409
5.4 Conservation Voltage Reduction Planning 418
5.4.1 Introduction 418
5.4.2 Conservation Voltage Reduction 418
5.4.3 CVR Based on PSO 420
5.4.4 CVR Based on AHP 423
5.4.5 Case Studies for CVR in Korean Power System 424
5.4.6 Conclusion 427
5.5 Dynamic Distribution Network Expansion Planning with Demand Side Management 427
5.5.1 Introduction 427
5.5.2 Expansion Options 431
5.5.3 Problem Formulation 436
5.5.4 Optimization Algorithm 442
5.5.5 Case Studies 450
5.5.6 Conclusions 460
5.6 GA-Guided Trust-Tech Methodology for Capacitor Placement in Distribution Systems 467
5.6.1 Introduction 467
5.6.2 Overview of the Trust-Tech Method 469
5.6.3 Computing Tier-One Local Optimal Solutions 472
5.6.4 The GA-Guided Trust-Tech Method 474
5.6.5 Applications to Capacitor Placement Problems 478
5.6.6 Numerical Study 481
5.6.7 Conclusions 488
5.7 Network Reconfiguration 489
5.7.1 Introduction 489
5.7.2 Modern Distribution Systems: A Concept 490
5.7.3 Distribution System Reconfiguration 493
5.7.4 Distribution System Service Restoration 496
5.7.5 Multi-Agent System for Distribution System Reconfiguration 501
5.7.6 Conclusions 510
5.8 Distribution System Restoration 510
5.8.1 Introduction 510
5.8.2 Power System Restoration Process 511
5.9 Group-based PSO for System Restoration 531
5.9.1 Introduction 531
5.9.2 Group-Based PSO Method 533
5.9.3 Overview of the Service Restoration Problem 539
5.9.4 Application to the Service Restoration Problem 542
5.9.5 Numerical Results 545
5.9.6 Conclusions 552
5.10 MVMO for Parameter Identification of Dynamic Equivalents for Active Distribution Networks 553
5.10.1 Introduction 553
5.10.2 Active Distribution System 553
5.10.3 Need for Aggregation and the Concept of Dynamic Equivalents 554
5.10.4 Proposed Approach with MVMO 556
5.10.5 Adaptation of MVMO for Identification Problem 558
5.10.6 Case Study 562
5.10.7 Application to Test Case 568
5.10.8 Analysis 569
5.10.9 Reflections 572
5.10.10 Conclusions 572
5.11 Parameter Estimation of Circuit Model for Distribution Transformers 573
5.11.1 Introduction 573
5.11.2 Transformer Winding Equivalent Circuit 574
5.11.3 Signal Comparison Indicators 576
5.11.4 Coefficients Estimation Using Heuristic Optimization 578
5.11.5 Coefficients Estimation Results and Conclusion 582
5.11.6 Conclusions 586
References 590
Chapter 6 Integration Of Renewable Energy In Smart Grid 613
6.1 Introduction 613
6.2 Renewable Energy Sources 613
6.2.1 Renewable Energy Sources Management Overview 613
6.2.2 Energy Resource Scheduling - Problem Formulation 615
6.2.3 Energy Resources Scheduling - Particle Swarm Optimization 617
6.2.4 Energy Resources Scheduling - Simulated Annealing 618
6.2.5 Practical Case Study 621
6.2.6 Appendix 632
6.2.7 Conclusions 634
6.3 Operation and Control of Smart Grid 635
6.3.1 Introduction 635
6.3.2 Problems for Systems Configuration or Systems Design 636
6.3.3 Systems Operation and Systems Control 638
6.3.4 System's Management 640
6.3.5 Conclusion 645
6.4 Compliance of Reactive Power Requirements in Wind Power Plants 645
6.4.1 Introduction 645
6.4.2 Problem Definition 646
6.4.3 NN-Based Wind Speed Forecasting Method 648
6.4.4 Mean Variance Mapping Optimization Algorithm 650
6.4.5 Case Studies 654
6.4.6 Conclusions 665
6.5 Photovoltaic Controller Design 667
6.5.1 Introduction 667
6.5.2 Maximum Power Point Tracking in PV System 668
6.5.3 Particle Swarm Optimization 674
6.5.4 Application of Particle Swarm Optimization in MPPT 674
6.5.5 Illustration of PSO Technique for MPPT During Different Irradiance Conditions 676
6.5.6 Conclusion 678
6.6 Demand Side Management and Demand Response 680
6.6.1 Introduction 680
6.6.2 Methodology for Consumption Shifting and Generation Scheduling 683
6.6.3 Quantum PSO 685
6.6.4 Numeric Example 687
6.6.5 Conclusions 691
6.7 EPSO-Based Solar Power Forecasting 691
6.7.1 Introduction 691
6.7.2 General Radial Basis Function Network 693
6.7.3 k-Means 695
6.7.4 Deterministic Annealing Clustering 695
6.7.5 Evolutionary Particle Swarm Optimization 697
6.7.6 Hybrid Intelligent Method 698
6.7.7 Case Studies 699
6.7.8 Conclusion 704
6.8 Load Demand and Solar Generation Forecast for PV Integrated Smart Buildings 704
6.8.1 Introduction 704
6.8.2 Literature Review of Forecasting Techniques 714
6.8.3 Ensemble Forecast Methodology for Load Demand and PV Output Power 717
6.8.4 Numerical Results and Discussion 722
6.8.5 Conclusions 728
6.9 Multi-Objective Planning of Public Electric Vehicle Charging Stations 729
6.9.1 Introduction 729
6.9.2 Multi-Objective Electric Vehicle Charging Station Layout Planning Model 730
6.9.3 An Improved SPEA2 for Solving EVCSLP Problem 733
6.9.4 Case Study 737
6.9.5 Conclusion 740
6.10 Dispatch Modeling Incorporating Maneuver Components, Wind Power, and Electric Vehicles 741
6.10.1 Introduction 741
6.10.2 Proposed Economic Dispatch Formulation 743
6.10.3 Population-Based Optimization Algorithms 751
6.10.4 Test System and Results Analysis 753
6.10.5 Conclusion 756
6.11 Conclusions 757
References 757
Chapter 7 Electricity Markets 775
7.1 Introduction 775
7.2 Bidding Strategies 777
7.2.1 Introduction 777
7.2.2 Context Analysis 779
7.2.3 Strategic Bidding 780
7.3 Market Analysis and Clearing 781
7.3.1 Introduction 781
7.3.2 Electricity Market Simulators 782
7.3.3 Didactic Example 785
7.4 Electricity Market Forecasting 793
7.4.1 Introduction 793
7.4.2 Artificial Neural Networks for Electricity Market Price Forecasting 794
7.4.3 Support Vector Machines for Electricity Market Price Forecasting 795
7.4.4 Illustrative Results 796
7.5 Simultaneous Bidding of V2G In Ancillary Service Markets Using Fuzzy Optimization 798
7.5.1 Introduction 798
7.5.2 Fuzzy Optimization 799
7.5.3 FO-based Simultaneous Bidding of Ancillary Services Using V2G 801
7.5.4 Case Study 806
7.5.5 Results and Discussions 807
7.5.6 Conclusion 811
7.6 Conclusions 812
References 812
Index 819
LIST OF FIGURES
Figure 1.1.1 Publications of CI in power systems and smart grid (2002-2018).
Figure 1.2.1 Number of publications by EC methods in power systems and smart grid.
Figure 1.2.2 Genetic algorithm process flowchart.
Figure 1.2.3 Lagged particles and waiting phenomena in (a) PSO and (b) QPSO.
Figure 1.2.4 Multi-objective particle swarm optimization flowchart.
Figure 2.2.1 Overview of the aggregator energy management problem.
Figure 3.2.1 UCP solution paths.
Figure 3.2.2 UCP solution using heuristic methods.
Figure 3.2.3 Initial generation of population.
Figure 3.2.4 ISO activities.
Figure 3.2.5 Co-ordination between ISO and GENCO.
Figure 3.3.1 Structure of simple genetic algorithm.
Figure 3.3.2 Search mechanism of PSO.
Figure 3.3.3 Piecewise quadratic cost function of a generator.
Figure 3.3.4 Cost function with five valves.
Figure 3.3.5 A unit with prohibited operating zone.
Figure 3.3.6 Encoding schemes. (a) Series encoding and (b) embedded encoding.
Figure 3.3.7 Pseudocode for the new crossover technique.
Figure 3.3.8 Pseudocode for the deterministic crowding GA.
Figure 3.3.9 Artificial immune genetic algorithm flowchart.
Figure 3.3.10 Flowchart of the genetic algorithm based on the Lagrange method.
Figure 3.3.11 Adjustment strategy for an individual's position within boundary.
Figure 3.3.12 Comparison of inertia weights for IWA and CIWA.
Figure 3.3.13 Convergence characteristics of the CSPSO.
Figure 3.3.14 PSO with penalty for Gworst search mechanism.
Figure 3.3.15 Schematic of dynamic space reduction strategy.
Figure 3.4.1 Solution distribution by different selection strategies. (a) Crowding distance strategy. (b) Dynamic crowding distance strategy.
Figure 3.4.2 The calculation flowchart of IDEMO algorithm.
Figure 3.4.3 Pareto fronts of different algorithms.
Figure 3.4.4 Convergence curves of outer solutions for cost.
Figure 3.5.1 Simulation-based optimization for solving hydrothermal coordination.
Figure 3.5.2 Hydro generation (MW) - Solution Baseline (gray) versus Grouping I (black).
Figure 3.5.3 Monthly total hydro generation for the planning horizon (MW).
Figure 3.6.1 Methods for analyzing the generator maintenance scheduling problem.
Figure 3.6.2 Configuration of the solutions searched in vertex by conventional DP.
Figure 3.6.3 Configuration of the solutions searched by time axis shift method.
Figure 3.6.4 Possible time range of generator maintenance of the example problem.
Figure 3.6.5 Initial time range of generator maintenance of the example problem.
Figure 3.6.6 Concept of flexibility.
Figure 3.6.7 Year load curve (weekly load peaks).
Figure 3.6.8 Convergence of the objective function (µ).
Figure 3.6.9 Standard deviation of supply reserve rate and LOLP.
Figure 3.6.10 Standard deviation of EDNS according to iteration.
Figure 3.6.11 Maintenance powers at each week.
Figure 3.6.12 A search method using genetic algorithm.
Figure 3.6.13 Flowchart of genetic algorithm.
Figure 3.6.14 Generator maintenance scheduling system flowchart.
Figure 3.6.15 Starting screen of visualization of the GMS.
Figure 3.6.16 Preferences of the GMS.
Figure 3.6.17 Running process of the GMS by GA.
Figure 3.6.18 User-friendly visualization results of the GMS.
Figure 3.6.19 User-friendly visualization (calendar style) results of the GMS.
Figure 3.6.20 Total system result.
Figure 3.6.21 The share of power production average of all cases.
Figure 3.7.1 Three bus power system.
Figure 3.7.2 Flowchart of the proposed hybrid PSO algorithm.
Figure 3.7.3 Six-bus power system [28].
Figure 3.7.4 Decrease of global score for the 6-bus power system.
Figure 3.7.5 Voltage profile of the system buses of 6-bus power system.
Figure 3.7.6 Fourteen-bus power system.
Figure 3.7.7 Decrease of global score for the 14-bus power system.
Figure 3.7.8 Voltage profile of the system buses of 14-bus power system.
Figure 3.8.1 IEEE 30-bus system.
Figure 3.8.2 Convergence characteristics of ABC method in case 1.
Figure 3.8.3 Effect of valve-point loading on a quadratic cost function.
Figure 3.8.4 Convergence characteristics of ABC method in case 2.
Figure 3.8.5 Convergence characteristics of ABC method in case 3.
Figure 3.8.6 Convergence characteristics of ABC method in case 4.
Figure 3.8.7 Voltage profiles for case 1 and case 4.
Figure 3.9.1 Offshore WPP with optimization-based management of reactive power sources.
Figure 3.9.2 Simple 2D illustration of the Movement Rule in DEEPSO.
Figure 3.9.3 Algorithmic procedure of MVMO-PHM. The fitness evaluation and candidate solution counters are denoted by i and k, whereas Np, ?FE, and rand stand for number of candidate solutions, number of fitness evaluations, and uniform random numbers between [0, 1], respectively.
Figure 3.9.4 Layout of the set of solution archive.
Figure 3.9.5 Procedure for parent selection in MVMO-PHM.
Figure 3.10.1 Simple divide-to-conquer strategy (branching).
Figure 4.2.1 Turbine mechanical power variation following a step change in gate position.
Figure 4.2.2 Speed power response of governor.
Figure 4.2.3 Two-area system.
Figure 4.2.4 Two-area thermal power system.
Figure 4.2.5 System performance with different cost functions.
Figure 4.2.6 System performance with different parameter optimization.
Figure 4.2.7 Single-area hydro system.
Figure 4.2.8 Impact of communication delay on frequency.
Figure 4.2.9 Impact of secondary controller gain in the presence of communication delay.
Figure 4.3.1 Representation of different controllers.
Figure 4.3.2 Model of transmission line.
Figure 4.3.3 Model of TCSC.
Figure 4.3.4 Injection model of TCSC.
Figure 4.3.5 Equivalent circuit of TCPAR.
Figure 4.3.6 Injection model of TCPAR.
Figure 4.3.7 Equivalent circuit of UPFC.
Figure 4.3.8 Vector diagram of UPFC.
Figure 4.3.9 Equivalent circuit injection model of UPFC.
Figure 4.3.10 Equivalent circuit of GUPFC.
Figure 4.3.11 Injection model of GUPFC.
Figure 4.3.12 5-Bus system.
Figure 4.4.1 One-line diagram of a 5-area study system.
Figure 4.4.2 The critical eigenvector and the corresponding bus number.
Figure 4.4.3 Bus voltage magnitude profile when system is heavily stressed.
Figure 4.4.4 Voltage profile of the system after placing 145 Mvar at bus 40.
Figure 4.4.5 Convergence of GA and PSO on the level of compensation.
Figure 4.4.6 Convergence of GA and PSO on the average best-so-far.
Figure 4.4.7 The critical eigenvector and the corresponding bus number in system.
Figure 4.4.8 Bus voltage magnitude profile when system is heavily stressed in system.
Figure 4.4.9 The RGA-number of candidate sets 3, 6, and 9 (pre-fault).
Figure 4.4.10 The RGA-number of candidate sets 34 and 37 (pre-fault).
Figure 4.4.11 The RGA-number of candidate sets 31 and 37 (pre-fault).
Figure 4.4.12 The RGA-number of candidate sets 3, 6, and 9 (post-fault).
Figure 4.4.13 The RGA-number of candidate sets 34, 37, and 40 (post-fault).
Figure 4.4.14 The RGA-number of candidate sets 22 and 37 (post-fault).
Figure 4.4.15 The convergence characteristic of fitness function with PSO to find solution.
Figure 4.4.16 The convergence characteristic of fitness function with GCPSO to find solution.
Figure 4.4.17 The convergence characteristic of fitness function with GA to find solution.
Figure 4.5.1 Framework of the proposed...
