Detection and classification of faults on six phase transmission line using ANN

Abstract

The demand of electrical energy is continuously increasing. High phase order transmission system is a viable alternative due to increasing costs of right of way. Six phase transmission lines can carry more power for same phase to phase voltage with the same right of way economically. Protection of six phase transmission lines has been a very challenging task. Earlier simulations of six phase transmission line have been done in PSCAD/EMTDC software. In this paper MATLAB <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> software and its associated Simulink <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> and Simpowersystem <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> toolboxes have been used to simulate the six phase transmission line. Application of Artificial Neural Network for protection of six phase transmission line against ground faults is presented, which the authors believe has not been reported earlier. Fundamental components of six phase voltages and currents have been used as inputs for training of the artificial neural network for detection and classification of faulted phase using neural network toolbox of MATLAB <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> . A sample 765 kV system of 60 km length has been selected for study. The study takes into account the effect of variation in fault inception angle, fault distance location and fault resistance. The results indicate the suitability of proposed technique and its adaptability to changing system conditions.