A real-time fault detection and classification algorithm for transmission line faults based on MODWT during power swing

Abstract

An exploration of the faulty signal of three-phase overhead transmission lines can provide versatile information about the presence of fault during power swing (PS) and its type in a specific time domain to accustom protective relaying scheme for real-time application. In this paper, an illustrative feature extraction technique, maximal overlap discrete wavelet transform (MODWT), has been used to extract the attributes from the faulty signals during PS condition. The standard deviated values of the MODWT coefficients of the current signals only have been used as input features for fault detection and classification in an Indian power transmission network. The proposed scheme is based on a fault triangle with different fault planes. The performance of the proposed scheme has been assessed with real-time field fault data using “Wavewin” software environment and further appraised with other existing schemes too. Moreover, the proposed scheme is not affected by zero sequence mutual impedance between the two inductively coupled parallel circuits and independent of communication latency affect because it employs one terminal data only.