A New Protection Detection Technique for High Impedance Fault Using Neural Network

Abstract

The paper presents the application of Neural Network technique as a pattern recognition to high impedance faults (HIFs). The relay is based on a novel low-frequency (3rd and 5th harmonic feature diagnostic vector). The currents and voltages are used as a featured extracted signals for fault discrimination. The focus of this paper is to design a robust ANN-based relay, which can determine the high impedance low current faults on distribution radial electrical systems. A variety of faults and system conditions have been simulated to evaluate the reliability and sensitivity of the proposed technique.