A novel intelligent protection scheme for high impedance fault detection in distribution feeder

Abstract

This paper presents an analysis and simulation methodology to discuss the possible impacts of high impedance fault (HIF) in distribution feeder. Its object is to establish protection recommendations for both insulated and un-insulated overhead distribution feeder. As ground impedance and arc are nonlinear in nature, the feeder fallen on the ground with high impedance will cause a nonlinear fault current. However, the conventional ground fault protection schemes and algorithms have difficulty in HIF recognition. The wavelet transform (WT) technique has been successfully applied in many fields. The properties of scaling and translation of WT can be used to identify the low frequency components in stable signal and high frequency components in transient signal. Hence, this paper proposed a novel technique to effectively discriminate between the HIF and the switch operations in distribution feeder by combining a preprocessing module based on discrete wavelet transform (DWTs) with a neural networks (NN). The DWTs is firstly applied to extract of distinctive features of the voltage and current signals at the supply side of feeder and transform into a series of detailed and approximated wavelet components. Then, the coefficients of variation of the wavelet components are calculated. This information is introduced to training NN for identifying an HIF from the switches operations. The simulated results clearly show that the proposed technique can accurately identify the HIF in insulated and un-insulated overhead distribution feeder.