Single-ended fault location for transmission lines using traveling wave and multilayer perceptron network

Abstract

Transmission lines are subjected to many kind of fault. Therefore fault location scheme is needed to determine the exact location of fault. This paper proposed a method for estimating transmission line fault location system using traveling wave method and artificial neural network (ANN). The method starts by decomposing the current signal from the faulted phase using DWT technique. Then, time fault measurement is extracted from the decomposed signal. Finally, a type of ANN called multilayer perceptron network (MLP) is used to locate the fault location. The proposed method is benchmark against the existing method and evaluated by using 5 measurement indexes; Coefficient of Determination (R2) as well as four error measures-Percentage Prediction Error (PPE), Mean Square Error (MSE), Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE). The results shows that integration of traveling wave and ANN absolutely could improve the performance especially for fault occurs in short distance.