AI-Enhanced Wide-Area Traveling Wave-Based Fault Location Method for Transmission Networks With a Unified Power Flow Controller

This article proposes an artificial intelligence (AI)-based fault location method for transmission lines that uses the arrival times of voltage traveling waves. First, a decision tree (DT) is employed to identify the faulty line based on the initial wave arrival time (WAT) at each substation. Then, a multilayer perceptron (MLP) neural network is used to estimate the fault location along the identified line. Both the DT and MLP models are trained under various fault conditions, including different fault types and fault resistances. To improve robustness, cases involving incorrect or unavailable WAT data are also included in the training process. The proposed method is evaluated on a 10-bus, 500 kV test network containing a flexible AC transmission system (FACTS) device using PSCAD and MATLAB. The obtained results show that the proposed method can accurately identify a faulty line and estimate the fault location, even when the WATs of some buses are incorrect or unavailable. These findings indicate that the proposed method is robust and well suited for wide-area fault location in complex transmission networks where conventional methods may be less effective.