Artificial Intelligence Applications in Electrical Transmission and Distribution Systems Protection

✍ Scribed by Almoataz Y. Abdelaziz (editor), Shady Hossam Eldeen Abdel Aleem (editor), Anamika Yadav (editor)

• Publisher: CRC Press
• Year: 2021
• Tongue: English
• Leaves: 513
• Edition: 1
• Category: Library

✦ Synopsis

Artificial intelligence (AI) can successfully help in solving real-world problems in power transmission and distribution systems because AI-based schemes are fast, adaptive, and robust and are applicable without any knowledge of the system parameters. This book considers the application of AI methods for the protection of different types and topologies of transmission and distribution lines. It explains the latest pattern-recognition-based methods as applicable to detection, classification, and location of a fault in the transmission and distribution lines, and to manage smart power systems including all the pertinent aspects.

FEATURES

• Provides essential insight on uses of different AI techniques for pattern recognition, classification, prediction, and estimation, exclusive to power system protection issues
• Presents an introduction to enhanced electricity system analysis using decision-making tools
• Covers AI applications in different protective relaying functions
• Discusses issues and challenges in the protection of transmission and distribution systems
• Includes a dedicated chapter on case studies and applications

This book is aimed at graduate students, researchers, and professionals in electrical power system protection, stability, and smart grids.

✦ Table of Contents

Cover
Half Title
Title Page
Copyright Page
Contents
Preface
Editors
Contributors
1. Application of Metaheuristic Algorithms in Various Aspects of Electrical Transmission and Systems Protection
1.1 Introduction
1.2 Mathematical Representation of Optimization Problem
1.3 Metaheuristic Algorithms
1.4 Optimal Relay Coordination
1.4.1 Formulation of Relay Coordination Problem
1.4.2 Illustrative Example
1.4.3 State of Research in Optimal Relay Coordination
1.5 Optimal PMU Placement
1.5.1 Formulation of PMU Placement Problem
1.5.2 Illustrative Example
1.5.3 State of Research in Problem of PMU Placement
1.6 Estimation of Fault Section on Distribution Network
1.6.1 Formulation of Fault Section Estimation Problem as an Optimization Problem
1.6.2 Illustrative Example
1.6.3 State of Research in Fault Section Estimation
1.7 Estimation of Fault Location on Transmission Lines
1.7.1 Formulation of Fault Location Estimation Problem as an Optimization Problem
1.7.2 Illustrative Example
1.7.3 State of Research in Fault Location Estimation
1.8 Conclusion
References
2. AI-Based Scheme for the Protection of Power Systems Networks Due to Incorporation of Distributed Generations
2.1 Introduction to Distributed Generation (DG)
2.1.1 What Is Distributed Generating (DG)?
2.1.2 Advantages of DG over Conventional Power Generation
2.1.3 Applications of DG
2.2 Impact of Integration of Distributed Generation on the Power System
2.3 Problems during DG Interconnection
2.3.1 Operating (Economic) Issues
2.3.2 Technical Issues
2.3.3 Protection/Safety Issues
2.4 Islanding (Formation of Electrical Island)
2.4.1 Power Quality Issue
2.4.2 Personnel Safety
2.4.3 Out of Synchronism Reclose
2.5 Islanding Detection
2.5.1 Remote Method
2.5.2 Active Islanding Detection Method
2.5.3 Passive Islanding Detection Method
2.5.4 Hybrid Method of Islanding Detection
2.6 Application of Artificial Intelligence for Islanding Detection
2.6.1 Fuzzy Logic
2.6.2 Artificial Neural Network (ANN)
2.6.3 Machine Learning Classifier
2.7 Case Study of Classifier (Machine Learning)-Based Islanding Detection
2.7.1 Relevance Vector Machine
2.7.2 Simulation and Test Cases
2.7.3 Feature Vector Formation
2.7.4 Training of RVM Classifier
2.7.5 Result and Discussion
2.8 Protection Miscoordination due to DG Interconnection
2.8.1 Issue of Protection Miscoordination
2.8.2 Application of AI Technique for Restoration of Protection Coordination
2.9 Summary
References
3. An Intelligent Scheme for Classification of Shunt Faults Including Atypical Faults in Double-Circuit Transmission Line
3.1 Introduction
3.2 Description of an Indian Power System Network
3.3 Ensemble Tree Classifier (ETC) Model for Classification of CSFs, CCFs, and EVFs
3.3.1 Designing of Exclusive Data Sets
3.3.2 Discrete Wavelet Transform (DWT)
3.3.3 Bagged Decision Tree
3.3.4 Boosted Decision Tree
3.3.5 Training/Validation of Proposed ETC Model
3.4 Comparative Assessment of Proposed ETC Model Based Classifier Modules
3.5 Relative Assessment of Proposed Scheme with Other AI Technique-Based Fault Classification Schemes
3.6 Effect of Variation in Sampling Rate on Performance of Proposed Classification Scheme
3.7 Conclusion
Acknowledgments
References
4. An Artificial Intelligence-Based Detection and Classification of Faults on Transmission Lines
4.1 Introduction
4.2 The Basic Concepts of Distance Protection
4.2.1 Causes of Current Increase upon Fault Occurrence
4.2.2 Causes of Faults
4.2.3 Types of Faults
4.2.4 Sources of Errors in Detection and Classification of Faults
4.2.5 Distance Relay MHO Characteristic
4.3 AI-Based Fault Diagnosis System
4.3.1 Training Data for Artificial Neural Network: (Input/Target) Pairs
4.3.2 Feed Forward Artificial Neural Network
4.3.2.1 Multi-Layer Perceptron Neural Network
4.3.2.2 Radial Basis Function Network
4.3.2.3 Chebyshev Neural Network
4.3.2.4 Probabilistic Neural Network as a Detailed Example of FFNN
4.3.3 Support Vector Machine as an Example of ML
4.3.4 Convolution Neural Network as an Example of DL
4.4 Conclusion
References
5. Intelligent Fault Location Schemes for Modern Power Systems
5.1 Introduction
5.2 Conventional Fault Location Review
5.2.1 Traveling Wave-Based Fault Locators
5.2.2 Impedance Measurement-Based Fault Locators
5.2.3 Requirements for Fault Location Process
5.3 AI-Based Fault Location Schemes
5.3.1 ANN-Based Fault Location Computation
5.3.2 FL-Based Fault Location Computation
5.3.3 GA-Based Fault Location Computation
5.3.4 WT-Based Fault Location Computation
5.4 Recent Trends in Distribution Network and Smart Grid Requirements
5.5 Smart Fault Location Techniques
5.5.1 Fault Indicators
5.5.2 Distributed Smart Meters
5.5.3 IoT for Data Collections
5.5.4 Unmanned Aerial Vehicles (Drones)
5.6 Concluding Remarks
References
6. An Integrated Approach for Fault Detection, Classification and Location in Medium Voltage Underground Cables
6.1 Introduction
6.2 Autoregressive Modeling
6.3 Extreme Learning Machine
6.3.1 Training Extreme Learning Machine
6.4 Integrated Approach of the Protection Scheme
6.5 Test System
6.5.1 Simulation Parameters for Training and Testing
6.6 Fault Detection
6.7 Fault Classification
6.8 Fault Location
6.9 Results and Discussion
6.9.1 Comparative Evaluation
6.10 Summary
References
7. A New High Impedance Fault Detection Technique Using Deep Learning Neural Network
7.1 Introduction
7.2 Fault Model
7.3 The Proposed Deep Learning Approach
7.4 The Simulated Experiments and Discussions
7.5 Case Study
7.6 Conclusions
Appendix
References
8. AI-Based Scheme for the Protection of Multi-Terminal Transmission Lines
8.1 Introduction to Multi-Terminal Transmission Line
8.2 Need of a Multi-Terminal Transmission Line
8.2.1 Benefits of a Multi-Terminal Transmission Line
8.2.2 Limitations of a Multi-Terminal Transmission Line
8.2.3 Protection and Other Technical Issues with Multi-Terminal Transmission Line
8.3 Conventional Protection Schemes
8.3.1 Distance Protection Scheme
8.3.2 Current Differential Scheme
8.4 Advanced Multi-End Protection Schemes
8.4.1 Synchronized and Unsynchronized Measurement-Based Schemes
8.4.2 Fundamental and Transient Frequency-Based Schemes
8.4.2.1 Fundamental Frequency-Based Schemes
8.4.2.2 Transient Frequency-Based Schemes
8.5 AI or Knowledge-Based Schemes
8.5.1 ANN-Based Schemes
8.5.2 Fuzzy Interference Systems
8.5.3 Support Vector Machine-Based Schemes
8.6 Adaptive Protection Schemes
8.7 Conclusion
References
9. Data Mining-Based Protection Methodologies for Series Compensated Transmission Network
9.1 Introduction
9.2 Relaying Challenges in Series Compensated Transmission Network
9.2.1 Under- and Overreaching of Relays
9.2.2 Current and Voltage Inversion
9.2.3 Precarious Operation of MOV
9.2.4 Harmonics and Transients
9.3 Data Mining-Based Protection Mechanism
9.3.1 DWT and Non-Parametric ML (KNN) Based Fault Events Classification Scheme
9.3.2 DWT and Non-Parametric ML (SVM) Based Fault Events Classification Scheme
9.3.3 DWT and Non-Parametric ML (PNN) Based Fault Events Classification Scheme
9.4 Feasibility and Competency Analysis
9.4.1 Transforming Fault Events Identification
9.5 Summary
Appendix
References
10. AI-Based Protective Relaying Schemes for Transmission Line Compensated with FACTS Devices
10.1 Introduction
10.2 FACTS Technology
10.3 Protection Issues with FACTS Technology Integration
10.4 Overview of AI
10.5 AI-Based Application in FACTS-Compensated Transmission Line Protection
10.5.1 Training Data Collection and Processing
10.5.2 Training Algorithms
10.6 Conclusion and Perspectives
References
11. AI-Based PMUs Allocation for Protecting Transmission Lines
11.1 Introduction
11.2 Basics of PMUs and WAMS
11.2.1 Basic PMU Structure
11.2.2 PMU Placement Rules
11.2.3 PMU Placement Problem Formulation
11.2.3.1 Case #1: Base case
11.2.3.2 Case #2: Considering ZIBs
11.2.3.3 Case #3: Loss of a Single PMU
11.2.3.4 Case #4: Single Line Outage
11.3 Conventional Mathematical Techniques for PMUs Allocation
11.3.1 Exhaustive Search
11.3.2 Integer Programming
11.3.3 Integer Quadratic Programming
11.4 AI Application to PMUs Allocation
11.5 Case Study
11.5.1 IEEE 14-Bus System
11.5.1.1 Case #1: Base Case
11.5.1.2 Case #2: Considering ZIBs
11.5.1.3 Case #3: Loss of a Single PMU
11.5.1.4 Case #4: Single Line Outage
11.5.2 IEEE 30-Bus System
11.5.2.1 Case #1: Base Case
11.5.2.2 Case #2: Considering ZIBs
11.5.2.3 Case #3: Loss of a Single PMU
11.5.2.4 Case #4: Single Line Outage
11.6 Application of PMUs in Protecting Transmission Lines
References
12. An Expert System for Optimal Coordination of Directional Overcurrent Relays in Meshed Networks
12.1 Introduction
12.2 Importance of the ES and Its Objectives
12.3 Problem Formulation of the Optimal Coordination of DOCR
12.4 Structure of the Introduced ES
12.4.1 The Mechanism by Which the Introduced ES Work
12.5 An ES for Optimal Coordination of DOCR
12.5.1 Optimal Coordination Facts
12.5.2 Optimal Coordination Rules
12.6 Verification of the Introduced ES
12.6.1 IEEE 3-Bus Test System
12.6.2 The 8-Bus Test System
12.6.3 The IEEE 5-Bus Test System
12.7 Conclusion
References
13. Optimal Overcurrent Relay Coordination Considering Standard and Non-Standard Characteristics
13.1 Introduction
13.1.1 Methods for Coordination of DOCRs
13.2 DOCRs Coordination Problem
13.2.1 Boundaries of the Coordination Problem
13.2.1.1 Limits on Relay Characteristics
13.2.1.1.1 Limits on Pickup Current Setting
13.2.1.1.2 Limits on TDS
13.2.1.1.3 Parameters of Characteristics Relay Curve
13.2.1.2 Boundaries on DOCRs Coordination
13.3 Recent Optimization Techniques
13.3.1 WCA and MWCA
13.3.1.1 Conventional WCA
13.3.1.2 MWCA Algorithm
13.3.2 MFO and IMFO Algorithms
13.3.2.1 The MFO Algorithm
13.3.2.2 The IMFO Algorithm
13.4 Results and Discussion
13.4.1 Description of Test Systems
13.4.1.1 The Nine-Bus Network
13.4.1.2 The 15-Bus Test System
13.4.2 Formulated the Coordination Problem Using Standard-CRC
13.4.2.1 Using MWCA for Solving the Coordination Problem
13.4.2.1.1 Case 1: Nine-Bus Network
13.4.2.1.2 Case 2: 15-Bus Network
13.4.3 Solving the Problem of Coordination with Conventional CRC and Non- Conventional CRC
13.4.3.1 Scenario 1: Using Conventional CRC in Solving the Problem of Coordination
13.4.3.1.1 Nine-Bus system
13.4.3.1.2 15-Bus Network
13.4.4 Scenario 2: Using Non-Conventional CRC in Solving the Problem of Coordination
13.4.4.1 Nine-Bus Network
13.4.4.2 15-Bus Network
13.5 Conclusions
References
14. Artificial Intelligence Applications in DC Microgrid Protection
14.1 Introduction
14.2 Technical Considerations of DC Microgrid Protection
14.2.1 DC Fault Current Characteristics
14.2.1.1 Analysis of the First Stage of the Fault Current
14.2.1.2 Analysis of the Second Stage of the Fault Current
14.2.2 Technical Issues
14.2.2.1 Equipment Fault-Tolerant
14.2.2.2 Grounding System
14.2.2.3 DC Protective Devices
14.2.2.4 Protection Algorithm Capabilities
14.3 DC Microgrid Protection Approaches
14.4 AI-Based Approaches Effectiveness Investigation
14.4.1 WT Principles
14.4.2 Feature Extraction
14.4.3 Feature Extraction Results
14.4.4 Pattern Recognition with ANN
14.4.5 Classification Results
14.4 Conclusion
References
15. Soft Computing-Based DC-Link Voltage Control Technique for SAPF in Harmonic and Reactive Power Compensation
15.1 Introduction
15.2 System Topology of SAPF
15.3 Reference Generation Techniques for SAPF System
15.3.1 Hybrid Control Approach Based Synchronous Reference Frame Method for Active Filter Design (HSRF)
15.4 Design of Proposed Fuzzy Logic Controller in SAPF System
15.5 Proposed Controller Design Technique for Switching Pattern Generation in SAPF System
15.6 Simulation Results for Harmonic Compensation Using SAPF
15.7 Experimental Results
15.8 Conclusions
References
16. Artificial Intelligence Application for HVDC Protection
16.1 Introduction
16.1.1 Protection Tools Based on Artificial Intelligence
16.1.1.1 Generation
16.1.1.2 Description
16.1.1.3 Decision Making
16.2 Overview of HVDC Technology
16.3 HVDC Protection
16.3.1 DC Fault Phenomena
16.3.2 Multi-Terminal HVDC Protection
16.4 AI-Based Fault Detection
16.5 AI-Based Fault Classification
16.6 Al-Based Fault Location
16.7 AI-Based Commutation Failure (CF) Identification
16.8 Discussion
16.9 Conclusion
References
17. Intelligent Schemes for Fault Detection, Classification, and Location in HVDC Systems
17.1 Introduction
17.2 An Overview of HVDC Systems
17.2.1 CSC-HVDC Systems
17.2.2 VSC-HVDC Systems
17.2.3 Requirements and Challenges
17.3 Fault Detection and Classification in CSC-HVDC Systems
17.3.1 Input Features
17.3.2 Learning Algorithms/Models
17.4 Fault Location in CSC-HVDC Systems
17.4.1 Input Features
17.4.2 Learning Algorithms/Models
17.5 Fault Detection and Classification in VSC-HVDC Systems
17.5.1 Input Features
17.5.2 Learning Algorithms/Models
17.6 Fault Location in VSC-HVDC Systems
17.6.1 Input Features
17.6.2 Learning Algorithms/Models
17.7 Considerations for Practical Implementations
17.7.1 Implementation Costs
17.7.2 Unseen New Cases
17.7.3 High-Resistance Faults
17.7.4 Temporary Arc Faults
17.7.5 Fault Locations Very Close to Line Terminals
17.7.6 Operation of Adjacent Circuit Breakers
17.7.7 Lightning Disturbances
17.7.8 Measurement Noises/Errors
17.7.9 Inaccurate Line Parameters
17.7.10 Communication Delay, Disturbance, and Failure
17.7.11 Time Synchronization Errors
17.8 Conclusion
References
18. Fault Classification and Location in MT-HVDC Systems Based on Machine Learning
18.1 Introduction
18.2 Machine Learning-Based Fault Diagnostic Technique
18.2.1 Support Vector Machines
18.2.2 Feature Extraction and Selection
18.3 DC Faults in MT-HVDC Systems
18.4 Voltage Source Converters
18.5 Control System of Voltage Source Converters
18.6 Control of MT-HVDC System
18.7 MT-HVDC Test System and Simulation Results
18.7.1 DC Voltage Analysis
18.7.2 Frequency-Based Analysis
18.7.3 Machine Learning Algorithm
18.8 Conclusion
Acknowledgement
References
Index