Fault Detection in AC Microgrid Using Gaussian Naive Bayes: A Probabilistic Approach
Rudranarayan Pradhan, Monali Nayak, G. Venkat Shyam Sunder, Baidhar Hembram, Monalisa Biswal
Abstract
Microgrids are emerging as a dependable choice to improve the dependability and resilience of decentralized energy systems. They play a critical role in advancing modern smart grid technologies. Integrating renewable energy resources with microgrids ensures a secure energy supply for consumers. Microgrids offer several benefits, including minimized transmission losses, reduced carbon emissions, and enhanced system reliability. To achieve an uninterrupted power supply system, it is crucial for the power system to quickly identify various types of faults to prevent complete blackouts. Conventional protection solutions are inadequate for microgrids (MGs) because they struggle with bidirectional current flow, small fault currents at the inverter interface of MGs, and the variability in their operational methods. Therefore, nowadays machine learning deep learning algorithms have been integrated with microgrid protection systems to ensure the reliable and efficient operation of these complex energy systems. This work proposes an intelligent method for fault detection for microgrids using the Gaussian Naive Bayes algorithm, which is a combination of Naive Bayes and Gaussian distribution.