A new method for arcing fault location using discrete wavelet transform and wavelet networks

摘要

This paper presents a new method for arcing (transient) fault location by combining discrete wavelet transform (DWT) and wavelet networks (WNs). The single-end relaying point voltage and current waveforms (captured with a sampling rate of 2.5 kHz) are de-noised by DWT to obtain a higher signal-to-noise ratio. The feature vectors contain the magnitudes of the three phase post-fault voltage and current fundamental frequency phasors obtained by DWT. The feature vectors for different arcing fault conditions are applied to WN for training and testing purposes. Although the prime aim of the proposed method is arcing fault location, it can also be applied for permanent faults. The accuracy of the proposed method has been tested and verified under various fault conditions on a 400 kV overhead line using the Electro-Magnetic Transient Program (EMTP) software package. Performance of the new method has also been compared with several existing methods under similar fault conditions, and the test results show better accuracy of the proposed method. Copyright © 2011 John Wiley & Sons, Ltd.