Application Research Of Multiwavelet Theory In Transmission Line Fault Location

In recent years with the rapid development of power system, it put higher demand for high-voltage transmission line fault location’s rapidity and accuracy as the system voltage level is getting higher and higher, transmission capacity is growing bigger and bigger and transmission line is getting longer and longer. Fault location based on transient traveling wave has a rapid development as the fault location of transmission line in traveling wave method are not subject to failure transition resistance, line structure, the system impedance, the transformer ratio and other factors.High-voltage transmission line short-circuit fault will produce a transient traveling wave signal. The transient traveling wave signal is a non-stationary signals which is mutant and singular. It contains a wealth of fault information which was sent close to the speed of light along the transmission line to the substation at both ends of the transmission line. Correctly identified and fault extracted signal is the premise of achieving fault locating quickly and accurately. For the feature of fault transient traveling wave, whether it is with a simple frequency analysis or with a simple time-domain analysis can not accurately describe the transient traveling wave which changes non-smoothly. Wavelet transform has good frequency localization properties and can automatically adjust the time frequency window based on the frequency changes of the signal. Therefore using the wavelet transform can effectively detect the singularity characteristics of the transient traveling wave signal.This article first briefly discusses the importance and research status of transmission line fault location and then introduces the basic theory of the wavelet transform and multi-wavelet transform for the defects in the fault location. The paper first give the multi-wavelet decomposition of the fault signal collected and then use the multi-wavelet compromise de-noising to obtain fault signal after de-noising. According to the singularity detection principle of multi-wavelet transform, we use modulus maxima point to get the time the traveling wave head reaching both ends of the line and then get accurate fault location by double-ended method. The experimental results show that this method can not only effectively remove the influence of noise but also able to meet the requirements of the accuracy of fault locating.