Research On Mallat Algorithm And Its Application In Fault Location Estimation Of Power Systems

Our society today is completely dependent on reliable power delivery. The expectation of high quality, uninterrupted power has been constantly growing because of the proliferation of infrastructure across China. Interruptions to electrical power have become highly costly as they impact industrial processes, information systems, and security requirements. In order to minimize system downtime and increase system security, an accurate fault location technique with short diagnostic latency is desired.This topic is a part of the project Research and Realization of Fault Location of Power Transmission Lines funded by Shaanxi Electric Power Corporation, Northwest China Grid. This paper studies fault location methods for different lines and different fault types. In addition, improvements are made based on existing methods and simulation models are built accordingly in order to verify the accuracy with which the improved approach can locate the fault distance.In the proposed approach, we first apply lifting wavelet transform to remove noise from fault signals and wavelet transform singularity analysis to measure the signals so that fault features can be extracted. The fault features are then analyzed with wavelet transform. To determine the exact times when the heads of traveling wave arrive at fault locators, wavelet transform modulus maxima theory is used. It should be noted that the impact of wave velocity on the calculation of fault locations was neglected.In this paper, we compare the proposed fault location approach with traditional methods through modeling and simulations in Matlab/Simulink environment. The results presented demonstrate the effectiveness of the proposed method for correct fault diagnosis and fault location on power systems, which suggests a potential to meet the industry’s requirements for general implementation.