Fault Detection and Classification in Electrical Power Distribution Network Through Machine Learning

This article used Support Vector Machines (SVM), Random Forest (RF), Decision Tree (DT), Extra Trees (ET), and Extreme Gradient Boosting (XGBoost) for fault detection and classification in electrical power distribution network. The study covers the detection and classification of different fault conditions like line to ground fault, double line fault, double to ground fault and three phase faults along with normal condition (no fault). The experimental results demonstrated that ensemble methods, particularly RF, XGBoost, and ET, significantly outperformed the individual models, achieving high accuracy rates exceeding 99%. The ET model emerged as the best performer, showcasing exceptional precision, recall, and F1-scores, with very few misclassifications as evidenced by the confusion matrices.