A TRANSIENT-BASED FAULT LOCATOR USING MATHEMATICAL MORPHOLOGY AND WAVELET ANALYSIS

摘要

A novel integrated morphology-wavelet filter algorithm used for transient-based fault locator is proposed. This algorithm aims to highlight the characteristics associated with the system faults, while eliminating possible random and pulse noises aliased in the fault transient. The travelling wave-based fault locator is given as an example to illustrate the availability of the proposed algorithm. By use of the preset filter implemented with mathematical morphology transform, the wavefront of the travelling wave can be extracted from the original fault signal distorted by a variety of noises. Furthermore, a high-sensitive singularity-detecting algorithm using wavelet transform is utilized to identify and locate the wavefront of the travelling wave with high accuracy. With the above integrated filter algorithm, various types of protections based on travelling wave or transient will have higher security and reliability. The effectiveness of this algorithm is testified with the simulations based on ATP program. This integrated morphology-wavelet filter algorithm has much less computational complexity than conventional low-pass filter and multiresolution waveform transform, which makes it possible to extend it to implement the transient-based protection.