MACHINE LEARNING FOR FAULT DETECTION IN THREE-PHASE TRANSMISSION LINES AND ELECTRIC MACHINES

Abstract

This research explores machine learning algorithms for fault detection and classification in electrical machines and three-phase transmission lines. Using MATLAB Simulink, fault scenarios were simulated to generate datasets, which were preprocessed and divided into training, validation, and testing sets. Algorithms including Decision Trees, XGBoost, k-Nearest Neighbors (KNN), and Random Forest were evaluated. The best-performing model was integrated with Simulink for real-time fault detection. Results indicate that KNN and Random Forest outperform other methods in accurately identifying and classifying faults, enhancing system reliability and reducing downtime.