Summary

Cutting-edge research indicates that evolutionary programming is set to emerge as the dominant optimisation technique in the fast-changing power industry. Combining theory and practice, Intelligent System Applications in Power Engineering capitalises on the potential of neural networks and evolutionary computation to resolve real-world power engineering problems such as load forecasting, power system operation and planning optimisation. Unlike existing optimisation methods, these novel computational intelligence techniques provide power utilities with innovative solutions for improved performance. Features include:
* Introduction to evolutionary programming and neural networks serving as a foundation for later discussion of the benefits of hybrid systems
* Practical application of evolutionary programming to reactive power planning and dispatch for speedy, cost-effective increases in transmission capacity plus generator parameter estimation
* Examination of economic dispatch, power flow control in FACTS and co-generation scheduling and fault diagnosis for HVDC systems and transformers
* Consideration of power frequency and harmonic evaluation to maximise supply quality
* Employment of distance protection, faulty section estimation and calculation of fault clearing time for transient stability assessment
Graduate students in electric power engineering will value Lai s broad coverage of the applications of evolutionary programming and neural networks in the field. This unique reference will be a boon to engineers, computer application specialists, consultants and utility managers wishing to understand the benefits intelligent systems can bring to the power industry.

Author Biography

Loi Lei Lai graduated from Aston University in Birmingham with a BSc and a PhD. He was awarded a DSc by City University London. He is also an honrorary graduate of City University. In 1984, he was appointed Senior Lecturer at Staffordshire Polytechnic. From 1986 to 1987, he was a Royal Academy of Engineering Industrial Fellow to both GEC Alsthom Turbine Generators Ltd and the Engineering research Centre. He is currently Head of Energy Systems Group and Chair in Electrical Engineering at City University London. In the last decade, Professor Lai has authored/co-authored 200 technical publications. He has also written a book entitled Intelligent System Applications in Power Engineering - Evolutionary Programming and Neural Networks and, in 2001, edited the book Power System Restructuring and Deregulation - Trading, Performance and Information Technology, both published by John Wiley & Sons, Ltd. He was award the IEEE Third Millennium Medal and won the IEEE Power Engineering Society, United Kingdom and Republic of Ireland (UKRI), chapter, Outstanding Engineer Award in 2003. In 1995, he received a high-quality paper prize from the International Association of Desalination, USA and in 2006 he was awarded a Prize paper by the IEEE Power Generation Committee. He is a Fellow of the IEEE and the IET (Institution of Engineering and Technology).
Among his professional activities, he is a Founder and was the Conference Chairman of the international Conference on Power Utility Deregulation, Restructuring man of the International Conference on Power Utility Deregulation, Restructuring and Power Technologies (DRPT) 2000, co-sponsored by the IEEE (now IET) and Power Technologies (DRPT) 2000, co-sponsored by the IEE (now IET) and IEEE. He reviews grant proposals regularly for the EPSRC, Australian Research Council and Hong Kong research Grant Council. In 2001, he was invited by the Hong Kong Institution of Engineers to be Chairman of an Accreditation Visit Team to accredit the BEng (Hons) degree in Electrical Engineering. Since 2005, Professor Lai has been invited as a judge for the Power/Energy Category in the IET Innovation in Engineering Awards. He was also Student Recruitment Office of the IEEE UKRI Section Executive Committee. He is a member of the Intelligent Systems Subcommittee in Power System Analysis, Computing and Economic Committee, IEEE Power Engineering Society; a Member of the Executive Team of the Power Trading and Control Technical and Professional Network, IET; an Editor of the IEE Proceedings - Generation, Distribution and Generation (now IET Generation, Distribution and Generation); an Editorial Board Member of the International Journal of Electrical Power & Energy Systems published by Elsevier Science Ltd, UK; International Advisor, Hong Kong Institution of Engineers (HKIE) Transactions and an Editorial Board Member of the European Transactions on Electrical Power published by John Wiley & Sons, Ltd. He was a research Professor at Tokyo Metropolitan University, is also Visiting professor at Southeast University Nanjing and Guest Professor at Fudan University, Shanghai. He has also been invited to deliver keynote addresses and plenary speeches to several major international conferences sponsored by the IET and IEEE.

Preface

Acknowledgments

xix

About the author

xxi

Object-Oriented Analysis, Design and Development of Artificial Neural Networks

(36)

Introduction

(5)

ANN Architectures

(1)

Feedforward Multilayer ANNs

(1)

Structure of ANNs

(4)

Object-Oriented Design (OOD)

(2)

Object-Oriented Analysis

(1)

Finding Objects

(1)

Organising Objects

(1)

Describing Object Interaction

(1)

Defining Operations on Objects

(1)

Object-Oriented Design

(1)

Implementation

(1)

Object-Oriented Testing

(1)

OOD Approach to Artificial Neural Networks

(4)

Identifying Objects

(1)

The Data Dictionary

(3)

Learning in ANNs

(12)

Data Representation in ANNs

(1)

What ANNs Learn

(1)

Learning Algorithms in ANNs

(1)

Unsupervised Learning Networks

(1)

Clustering Networks

(1)

Vector Quantisation

(1)

Supervised Learning

(1)

Decision-Based Supervised Learning

(1)

Approximation-Based Supervised Learning

(2)

Backpropagation Learning Algorithm for Multilayer Perceptron Networks

(1)

Supervised Learning Parameters

(2)

Speeding up Supervised Learning

(1)

Analysis, Design and Implementation of ANNs

(5)

Software Architecture

(1)

System Development

(1)

Software Life-Cycle Models

(1)

The Waterfall Life-Cycle Model

(1)

The Evolutionary Life-Cycle Model

(1)

Prototyping

(1)

Phases of Software Development

(1)

Object-Oriented Analysis and Design of Feedforward ANNs (FANNs)

(5)

FANNs

(1)

Organising the Objects

(1)

Describing Object Interactions

(1)

Defining Operations on Objects

(1)

ANN Systems Design

(2)

Implementation and Testing

(1)

Conclusions

(1)

References

(2)

Evolutionary Computation

(32)

Introduction

(1)

Genetic Algorithms (GAS)

(5)

Features of GAS

(4)

Object-Oriented Analysis of GAS

(3)

Identifying Objects in the GA Domain

(1)

The Data Dictionary

(1)

Discovering Object Operations

(1)

Object-Oriented GA Design

(12)

Object Design

(3)

Implementation

(1)

Object-Oriented Testing

(8)

Evolutionary Programming (EP)

(4)

Features of EP

(1)

Two Simple Algorithms for EP

(1)

Classical EP

(1)

Adaptive EP with β

(2)

Object-Oriented Analysis, Design and Implementation of EP

(5)

Object-Oriented Analysis of EP

(1)

The Data Dictionary

(1)

Developing Object Operations

(1)

Object-Oriented EP Design

(1)

Object Design

(3)

Implementation

(1)

Object-Oriented Testing

(1)

References

(2)

Hybrid Evolutionary Algorithms and Artificial Neural Networks

(14)

Introduction

(1)

Why Develop EANNs?

(1)

EANN Methods

(1)

The Evolution of EANN Weights

(2)

Representation of Connection Weights

(1)

The Evolution of an EANN Architecture

(1)

The Evolution of EANN Training Algorithms

(2)

The EANN Framework

(1)

OO GA-ANN Hybridisation

(2)

OO EP-ANN Hybridisation

(1)

Conclusions

(1)

References

(2)

An Evolutionary Programming Approach to Reactive Power Planning

(26)

Introduction

(2)

Problem Formulation

(3)

Evolutionary Programming (EP)

(3)

Initialisation

(1)

Statistics

(1)

Mutation

(1)

Competition

(1)

Inner Loop Convergence Criterion

(1)

Outer Loop Convergence Criterion

(2)

Numerical Results

(6)

Initial Power Flow Results

(2)

EP Optimal Results

(2)

Comparison with the BFGS Method

(1)

Practical Results

(5)

Conclusions

102

(1)

Acknowledgments

102

(1)

References

103

(1)

Appendix

104

(5)

Optimal Reactive Power Dispatch Using Evolutionary Programming

109

(8)

Introduction

109

(1)

Problem Formulation

110

(1)

Evolutionary Programming (EP)

111

(1)

Initialisation

112

(1)

Numerical Results

112

(3)

Initial Condition

112

(2)

Optimal Results

114

(1)

Conclusions

115

(1)

Acknowledgements

115

(1)

References

115

(2)

Application of Evolutionary Programming to Transmission Network Planning

117

(10)

Introduction

117

(1)

Problem Formulation

118

(3)

Evolutionary Programming

121

(1)

Initialisation

121

(1)

Numerical Results

121

(5)

Conclusions

126

(1)

Acknowledgments

126

(1)

References

126

(1)

Application of Evolutionary Programming to Generator Parameter Estimation

127

(14)

Introduction

127

(1)

The Generator Model in a Power System

128

(3)

The EP Parameter Estimation Algorithm

131

(2)

Initialisation

132

(1)

Integration

132

(1)

Mutation

132

(1)

Results

133

(4)

Simulation Results

134

(1)

Micromachine Identification Results

135

(2)

Conclusions

137

(1)

Acknowledgments

137

(1)

References

137

(1)

Appendix

138

(3)

Evolutionary Programming for Economic Dispatch of Units with Non-Smooth Input-Output Characteristic Functions

141

(8)

Introduction

141

(1)

Problem Formulation

142

(3)

Valve Point Effect

143

(1)

Objective Function

144

(1)

Evolutionary Programming (EP)

145

(1)

Initialisation

145

(1)

Numerical Results

145

(2)

Conclusions

147

(1)

Acknowledgements

147

(1)

References

147

(2)

Power Flow Control in Facts Using Evolutionary Programming

149

(10)

Introduction

149

(1)

Facts Model

150

(2)

Phase Shifter

150

(2)

Series Compensator

152

(1)

Problem Formulation

152

(2)

Evolutionary Programming (EP)

154

(1)

Initialisation

154

(1)

Numerical Results

154

(2)

Conclusions

156

(1)

Acknowledgments

156

(1)

References

156

(3)

Multi-Time-Interval Scheduling for Daily Operation of a Two-Co-Generation System with Evolutionary Programming

159

(12)

Introduction

159

(2)

Modelling of Multi-Co-Generation Systems

161

(3)

Evolutionary Programming (EP)

164

(1)

Initialisation

165

(1)

Case Study

165

(3)

Initial Result

166

(1)

Optimal Result

167

(1)

Conclusions

168

(1)

Acknowledgments

168

(1)

References

168

(3)

Application of Evolutionary Programming to Fault Section Estimation

171

(12)

Introduction

171

(1)

Problem Formulation

172

(1)

Mathematical Model and Fitness Function

173

(2)

Evolutionary Programming (EP)

175

(1)

Initialisation

175

(1)

Genetic Algorithms (GAS)

175

(1)

EP VS. GA

176

(1)

Case Study

176

(4)

Conclusions

180

(1)

Acknowledgments

181

(1)

References

181

(2)

Neural Networks for Fault Diagnosis in HVDC Systems

183

(12)

Introduction

183

(1)

Fault Diagnosis

184

(1)

Pattern Classification

184

(1)

Feature Extraction

184

(1)

Classification Procedure

185

(1)

HVDC Systems

185

(5)

Rectifier Current Control

187

(1)

Inverter CEA Controller

188

(1)

Firing Pulse Detectors

188

(1)

Non-linearities in Components

189

(1)

Converter Faults

189

(1)

Test Cases

190

(1)

Conclusions

191

(2)

References

193

(2)

An ANN Approach to the Diagnosis of Transformer Faults

195

(14)

Introduction

195

(1)

Dissolved Gas Analysis

196

(1)

Key Gas Method

196

(1)

Ratio Methods

197

(1)

IEC Codes for Dissolved Gas Analysis

197

(2)

Fuzzy Sets of IEC Gas Ratio Codes

199

(2)

Fuzzy Sets of Key Gases

201

(1)

Results

201

(1)

The Artificial Neural Network

202

(1)

The ANN Approach to Fault Diagnosis

203

(1)

The ANN for Major Fault Type Diagnosis

203

(1)

The ANN for Cellulose Condition Detection

204

(1)

ANN Diagnosis Results

204

(2)

Discussion

206

(1)

Conclusions

206

(1)

References

206

(3)

Real-Time Frequency and Harmonic Evaluation Using Artificial Neural Networks

209

(11)

Introduction

209

(2)

Non-Linear Least Squares

211

(1)

NN Formulation

212

(2)

The Dynamic Gradient System

212

(1)

Neural Network Implementation

213

(1)

Model Validation by Simulation

214

(1)

Real Application of the System

215

(3)

Conclusions

218

(1)

Acknowledgments

219

(1)

References

220

(15)

Artificial Neural Network Applications in Digital Distance Relay

221

(1)

Introduction

221

(1)

The Adverse Effect of Fault Resistance

222

(1)

An Adaptive Relaying Scheme

222

(2)

An Ideal Trip Characteristic

224

(1)

Use of ANNS in Identifying Trip Region

224

(4)

Design of an ANN

225

(1)

Pre-Fault Setting for On-Line Training and Testing

225

(1)

Trip Decision Making

226

(1)

NN Training Result

226

(1)

Testing Patterns and Results

227

(1)

Effect of Power Flow Changes on the Ideal Operating Region

228

(3)

Decrease in Power Flow

228

(1)

Increase in Power Flow

229

(1)

Increase in Active and Reactive Power Flows

230

(1)

Proposed Tripping Scheme

231

(1)

Off-Line Training of the NN

232

(1)

Conclusions

232

(1)

References

232

(3)

Application of Artificial Neural Networks to Transient Stability Assessment

235

(10)

Introduction

235

(1)

Transient Stability

236

(1)

Critical Clearing Time (CCT)

236

(1)

Methods of Fast Assessment of CCT

237

(1)

Neural Network (NN) Application

237

(2)

Training Patterns

237

(1)

Training Features

238

(1)

Pre-Conditioning of Training Patterns

239

(1)

Performance of the ANN

239

(2)

Training

240

(1)

Testing

241

(1)

Discussion

241

(1)

Conclusions

242

(1)

References

242

(3)

Application of Neural Networks and Evolutionary Programming to Short-Term Load Forecasting

245

(8)

Introduction

245

(1)

Load Forecasting with ANNS

246

(2)

Simulation Results

248

(2)

The NN Approach

248

(1)

The EP-ANN Approach

249

(1)

Conclusions

250

(1)

References

250

(3)

Select Bibliography

253

(8)

Index

261

Supplemental Materials

What is included with this book?

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.