Locating high-impedance fault in a smart distribution system using wavelet entropy and hybrid self-organizing mapping network

Abstract

Owing to small fault current, it is generally hard to detect high-impedance faults (HIFs) caused by downed conductors in power systems using traditional protection relays. The energized downed conductor may be likely to impose a safety risk to both technicians and the public. This paper presents a novel method for identifying the faulted line (feeder) section using limited measurement facilities in a smart distribution system. The discrete wavelet transform is used to extract the features of transients caused by HIFs. The signal entropies attained by wavelet coefficients serve as inputs to the hybrid self-organizing mapping neural network for locating the HIF line section. The simulation results obtained from an 18-busbar distribution system show the applicability of the proposed method.