Abstract:
Electrical fault detection is one of the most critical aspects of maintaining grid resilience and reducing downtime in power systems. However there is a scarcity of adequ...Show MoreMetadata
Abstract:
Electrical fault detection is one of the most critical aspects of maintaining grid resilience and reducing downtime in power systems. However there is a scarcity of adequet systems to detect these anomalies and therefore this research employs modern tools, including machine learning and data analytics, to develop a system capable of distinguishing between anomalies such as equipment breakdowns or short circuits and normal functioning through the analysis of various electrical parameters. In this study, a dataset consisting of Line Current and Line Voltage of different phases was analyzed by applying feature scaling post which different machine learning algorithms such as Decision Tree, Random Forest, MLP, Logistic Regression, and SVM were used to detect and classify the electrical faults that may occur. From the research it was found that KNN outperformed all the other algorithms and managed to achieve the highest overall accuracy as well as highest class-based accuracies as well.
Published in: 2024 IEEE Third International Conference on Power Electronics, Intelligent Control and Energy Systems (ICPEICES)
Date of Conference: 26-28 April 2024
Date Added to IEEE Xplore: 23 October 2024
ISBN Information: