Description
Partial Discharge Classification in Solid Insulation
Hybrid Genetic-Algorithm and Neural Network Approach
Mohd Jamil, Mohamad Kamarol; Rosle, Norfadilah; Rohani, Mohamad Nur Khairul Hafizi
ISBN 978-981-92-6639-5 · Springer · X, 145 p. 54 illus., 48 illus. in color. ·
Description / Beschreibung
This book highlights an intelligent framework for detecting and classifying partial discharge in solid insulation, a quiet but persistent threat to the reliability of high-voltage power systems. It presents the complete diagnostic pipeline, from capturing discharge signals with a custom, low-cost Rogowski coil sensor to characterising their phase-resolved patterns and distilling them into informative features through statistical analysis and principal component analysis. At its core lies a hybrid classifier that pairs a feedforward backpropagation neural network with a genetic algorithm, improving accuracy, efficiency, and generalisation beyond conventional methods. Bridging high-voltage insulation engineering and machine learning, the book explains specialist concepts in accessible terms while preserving technical rigour, offering researchers, graduate students, and practising engineers a practical route toward smarter insulation diagnostics.
Contents / Inhaltsverzeichnis
Introduction.- Partial Discharge Detection Methods and Rogowski Coil Sensing in Solid Insulation.- Partial Discharge Feature Extraction and Pattern Recognition Techniques for Cable Insulation Classification.- Machine Learning and Optimisation Approaches for Partial Discharge Classification in Cable Insulation.- Experimental Procedure and Hybrid Classification Framework for Partial Discharge Detection in Solid Insulation.- Partial Discharge Pulse Measurement, Pattern Analysis, and Statistical Characterisation of Solid Insulation Defects.- Performance Evaluation of Standard and Genetic-Algorithm Enhanced Neural Networks for Partial Discharge Classification.- Summary and Conclusions.
Contributors / Mitwirkende
Ir. Dr. Mohamad Kamarol Mohd Jamil obtained his D.Eng. from Kyushu Institute of Technology, Japan. His research interests include the insulation properties in oil palm, solid dielectric material, insulation properties of environmentally benign gas, and PD detection technique for insulation diagnosis of power apparatus and electrical machine. He is also involved in temperature rise and short-circuit electromagnetic study of busbar system and HVDC system.
Dr. Norfadilah Rosle received her Ph.D. in Electrical Engineering from Universiti Sains Malaysia, Penang. Her research interest is partial discharge detection and measurement on solid insulation, signal processing and artificial intelligence. She is now working as a senior lecturer in Universiti Malaysia Perlis, Malaysia.
Dr. Mohamad Nur Khairul Hafizi Rohani obtained his Ph.D from the University Malaysia Perlis. His research interest is partial discharge detection and measurement on solid and liquid insulation, signal and image processing and partial discharge sensor development for high voltage equipment applications.
Features / Besonderheiten
Discusses specialist concepts in accessible terms while preserving technical rigour Presents complete diagnostic pipeline, from capturing discharge signals to characterising their phase-resolved patterns Highlights an intelligent framework for detecting and classifying partial discharge in solid insulation
