An Intelligent Approach of Fault Classification and Localization of a Power Transmission Line

This paper depicts an approach with an artificial neural network (ANN) for classification and localization of faults on the power transmission line. A huge number of transmission and distribution lines are used for the transportation of electrical energy in the power system. Moreover, developing a completely reliable system is not possible within the economic and technical limitations. So, there always remains a probability of faults in the transmission lines. When occurs, it is aimed to locate and classify the fault to restore the sound condition of the transmission line. For analyzing the transmission line fault, this paper adopts the fault-tolerant neural network-based method as it can process incomplete and noisy data. This approach can deal with nonlinear problems and can carry out the prediction if trained. Feedforward neural networks along with the backpropagation algorithm are used each of the three phases of the transmission line for the fault localization process. The proposed method uses the instantaneous measure of the fault current to return the fault type and the distance from the experimental end. The modeling of the power system and the development of the neural network for this method have been conducted in the MATLAB/Simulink environment. The simulation results illustrated in this paper manifests that the proposed model is promising in performance.