A new single-ended artificial neural network-based protection scheme for shunt faults in six-phase transmission line

Abstract

Summary In this paper, modular concept of artificial neural network (ANN)-based technique is introduced to identify and locate all type of shunt faults (120) in a six-phase transmission line. The proposed algorithm is composed of two stages. In the first stage, ANN-based algorithm has been developed to detect and classify all possible types of shunt faults within one cycle time. Then, the second stage dispenses the location of shunt faults. Total 11 numbers of modular ANNs have been developed in both the stages which are based on the fundamental components of voltage and current signals at the sending end of transmission line only. For validation of the proposed scheme, simulation studies have been carried out on a six-phase transmission system. The test results of ANN-based fault detector/classifier and locator indicate that the proposed algorithm correctly detects/classifies and locates the shunt faults. The results demonstrate high speed, reliability and suitability of the proposed technique and its adaptability to changing system conditions viz. fault type, fault inception angle, fault location, high fault resistance, short circuit capacity of source and its X/R ratio. Even the cases of slight variations in system frequency, generated voltage and initial power flow angle are also taken into account. Copyright © 2014 John Wiley & Sons, Ltd.