Assessing the Performance of Machine Learning Based Fault Detection in Microgrid Applications: A Comparative Analysis

Abstract

As renewable energy sources are integrated more frequently and decentralised power plants in modern power systems, microgrids have emerged as a promising solution to enhance energy resilience and efficiency. However, the dynamic and decentralized nature of microgrids presents unique challenges, particularly in the realm of detection of faults and management. This article proposes a comprehensive framework for the detection of faults tailored specifically for microgrid applications. The framework incorporates advanced sensing technologies, machine learning algorithms, and distributed control strategies to effectively identify and localize faults within the microgrid infrastructure. By leveraging real-time data analytics and intelligent decision-making processes, the proposed solution aims to minimize downtime, optimize energy flow, and enhance overall system reliability. Case studies and simulation results exhibit the effectiveness and scalability of the proposed error detection solutions based on Artificial Neural Networks (ANN) across various microgrid scenarios.  Finally, future research directions and potential areas for innovation in fault detection solutions for microgrids are identified, aiming to foster continued advancements in the field and facilitate the deployment of reliable and resilient microgrid systems.