Implementation of a neural network based fault classifier for power system protection

This paper presents a novel technique for classifying power system faults and hence performing the function of a protective relay. An artificial neural network (ANN) model is developed and trained, using a database consisting of the results of transient and steady state performance of the power system, to capture all of the salient features and form a single relay module. The trained module is tested on a prototype power system, and found to give reasonably accurate results for classification of faults.