Application of ANN to power system fault analysis

Abstract

This paper presents the computer architecture development using Artificial Neural Network (ANN) as an approach for predicting fault in a large interconnected transmission system. Transmission line faults can be classified using the bus voltage and line fault current. Monitoring the performance of these two factors are very useful for power system protection devices. The ANN is designed to be incorporated with a matrix based software tool MATLAB Version 6.0, which deals with fault diagnosis in power system. In MATLAB software modules, the balanced and unbalanced fault can be simulated. The data generated from this software are to be used as training and testing sets in the Neural Ware Simulator.