Future Trends in Fault Detection Strategies for DC Microgrid

As the reliance on DC microgrids (DCMG) increases in modern energy systems; effective fault detection strategies become paramount to ensuring reliability and efficiency. Current methodologies, including traditional statistical approaches and machine learning techniques, have laid the groundwork for monitoring and managing faults. Unlike traditional AC systems, DCMGs encounter unique challenges, including low inertia and rapid dynamic changes, which can render conventional protection methods inadequate. Additionally, the discharge currents from filter capacitors in converter-based DC systems demand swift fault detection to ensure reliability. This paper introduces an innovative fault detection strategy that employs a moving average of fault currents. We validate the proposed method through a series of test scenarios on a 510V, 25kW low-voltage DC microgrid simulated in MATLAB/SIMULINK. The findings demonstrate that our approach not only accelerates fault detection but also enhances sensitivity and security. Moreover, comparative analysis shows that this technique performs robustly in noisy environments, surpassing existing derivative-based methods.