Fault diagnostic method of power transformers based on hybrid genetic algorithm evolving wavelet neural network

• Published in: IET electric power applications Vol. 2; no. 1; pp. 71 - 76
• Main Authors: Pan, C., Chen, W., Yun, Y.
• Format: Journal Article
• Language: English
• Published: London Institution of engineering and technology 01.01.2008; The Institution of Engineering & Technology
• Subjects: Applied sciences; Diagnostic software; Electric power generation; Electrical engineering. Electrical power engineering; Exact sciences and technology; Fault diagnosis; Genetic algorithms; Networks; Neural networks; Power electronics, power supplies; Testing. Reliability. Quality control; Transformers; Transformers and inductors; Wavelet; Learning; Backpropagation algorithm; Wavelet transformation; Genetic algorithm; High precision; Power electronics; Neural network; Power transformer; Fault diagnostic
• ISSN: 1751-8660; 1751-8679
• DOI: 10.1049/iet-epa:20070302

The main drawbacks of a back propagation algorithm of wavelet neural network (WNN), commonly used in fault diagnosis of power transformersm are the optimal procedure is easily stacked into the local minima and cases that strictly demand initial value. A fault diagnostic method is presented based on a real-encoded hybrid genetic algorithm evolving a WNN, which can be used to optimise the structure and the parameters of WNN instead of humans in the same training process. Through the process, compromise is satisfactorily made among network complexity, convergence and generalisation ability. A number of examples show that the method proposed has good classifying capability for single- and multiple-fault samples of power transformers as well as high fault diagnostic accuracy.