Research On The Fault Signals Recognition In Power Systems Based On Wavelet Theory

In the past few years, with the rapid development and application of computer, integrated circuit, signal process, communication and optical fiber technology, power system reliability requirements increasingly demanding. Power system fault diagnosis is the important guarantee of fast and accurate processing power system fault, but also guarantees the safe and stable operation of the power system.The fault diagnosis method based on wavelet analysis theory is an important branch of this field. While in recent years, the principal wavelet analysis method of different forms has been widely used and achieved good results.The main research work is as follows:First of all, from the angle of the engineering application, three methods of the signal de-noising and gives the signals de-noising in the specific application were researched and realized. From the detection of signal singularity, the basic theory of the signal singularity detection of modulus maxima theory was given. Based on two different types of singularities, the specific detection signals were given.Secondly, the meaning of detecting partial discharge signals of GIS was introduced. According to the characteristics of partial discharge signals, based on wavelet packet transform of feature extraction was put forward. According to the wavelet packet decomposition of the tree and features extraction, a new feature vector extraction method was proposed that different characteristic, take different dimension reduction method and draw the characteristic data distribution chart. In order to GIS in three common partial discharge signals were identified, the BP neural network model was designed and three kinds of fault signal for each group of30network were trained and15group of fault signals were tested. Based on Matlab simulation software results that the method of based on wavelet transform and neural network of PD signal diagnosis recognition is feasible.Finally, the summary of research about this paper and opinions on fault detection and diagnosis research in power system in future are given.The computer simulation results the correctness and validity of the analyses and syntheses methods in this paper were proved, which indicates that the research has its practical application.