Fault Diagnosis in Power Transmission Line using Decision Tree and Random Forest Classifier

Abstract

Fault diagnosis includes both detection of the type of fault that has occurred as well as locating the fault on a vast stretch of transmission network. There exists wide scope of application of Artificial Intelligence (AI) and Machine Learning (ML) based pattern recognition techniques in fault detection. The method proposed in this paper attempts to correctly identify the location of fault, type of fault and the faulty phase of transmission line in the form of an encoded 13-bit binary number. This number is then decoded and appropriate remedial measures are taken to isolate the fault. In this paper, Decision Tree (DT) and Random Forest based classification techniques have been proposed for fault diagnosis in a radial transmission line. Python platform has been used to simulate the proposed fault classifiers and implemented on a developed SIMULINK model of a 100 km long radial transmission line. The proposed method delivered fault recognition accuracy in the range of 95-100% under different fault scenarios. © 2022 IEEE.