Fault Classification and Detection in Transmission Lines Using ANN

Abstract

The ANN is a potent tool for the classification and detection of the faults on the transmission line due to pattern recognition ability. This paper centers on progressing an ANN to classify and detect a fault on power transmission line. The study employs feed-forward ANN with back-propagation algorithm in evolving the fault detector & classifier. Simulation and modelling of transmission lines was done in MATLAB using simpowersystems Toolbox. The instant currents and voltage numeric are taken after training the fault classifier & detector. Simulation results have been delivered to determine the effectiveness of the technologically advanced intellectual detector & classifier for fault. The performance of the classifier & detector is assessed by the Mean Square Error (MSE) and confusion matrix. The detector attained a tolerable MSE of 8.5571e-7 and accuracy of 100%. The classifier attained a tolerable MSE of 0.63035 and accuracy of 88%, presenting that the performance of the technologically advanced intellectual classifier detector is acceptable.