Neuro wavelet algortihm for detecting high impedance faults in extra high voltage transmission systems

Abstract

High impedance faults are not easy to be measured and detected by convetional relay protection. This paper proposed simualtions studies for detection high impedance fault in extra high voltage transmission line (EVT). The fault simulations based on simplified 2 diodes model. Current signal from the measurement is processed using discrete wavelet transform type haar wavelet to obtain coefficient detail. The output of discrete wavelet transform will be used for pattern recognition based on an backpropagation neural networks algorithm. The fault is modified to distribution system for EVT. The Characteristics of the proposed scheme are analyzed by comprehensive studies and the result clearly explain that it can accurately detect high impedance fault in the EVT with varies condition.