Enhancing Electrical Fault Detection in Transmission Lines Through Machine Learning

The purpose of this study is to apply machine learning techniques to improve the detection of electrical irregularities in transmission lines. The major objective is to provide a reliable framework that reliably identifies and categorises different electrical anomalies, improving power distribution networks' reliability and efficiency. The models were trained and assessed using a range of machine learning methods, such as Support Vector Machines (SVM), Random Forest Classifiers, and Decision Tree Classifiers, XGB Classifier, and Logistic Regression. The dataset replicated real-world electrical transmission line issues. The major conclusions demonstrate that every model-aside from SVM maintains a high level of accuracy when evaluated on fresh data and that every model has a training accuracy that is always high, frequently surpassing 100%. Evaluation criteria like recall, F1-score, and accuracy show how well the ensemble techniques Decision Tree, Random Forest, and XGB Classifier identify electrical defects. According to the study, ensemble techniques can enhance fault identification in power distribution networks when used with logistic regression. The efficacy and interpretability of logistic regression are noteworthy. This research provides practical solutions to enhance the resilience and dependability of electrical transmission networks while also advancing our understanding of machine learning applications for problem identification.