Fault Analysis On Transmission Lines Using Artificial Neural Network

Abstract

In power distribution technique it is essential to minimize transients, line voltage dips and spikes which are present due to the occurrence of fault detection. As a fault occurs in the power system, the necessary steps are taken to remove the fault in transmission line using relays & circuit breakers conventionally. But, if this fault occurrence is predicted in advance, then we can maintain a better voltage profile and quality power to consumers. We introduce a neural network in the power system, which can learn and therefore be trained to find solutions, recognize patterns and classify data. One can use a control methodology in Artificial Neural Network (A.N.N) which can classify & predict the future events. The magnitude of control variables based on previously acquired samples is taken and these values are used to recognize the type of abnormal event that may occur on the network. An analysis of the learning & generalization characteristics of elements in power system is presented using Neural Network (NN) toolbox in MATLAB.