Neural Sequenced Active Fault Management for Resilient Microgrids

Neural sequenced active fault management (NSAFM) is devised to maintain the microgrids' reliable operation and also to properly control microgrids' and renewable energy's sequence current under balanced or unbalanced faults. The main contributions include 1) a neural sequenced control framework for microgrids with fault ride-through capability; 2) an optimization-based sequenced AFM formulated to regulate the sequence current of renewable energy under unbalanced faults; 3) a learning-based sequenced AFM control algorithm, which transfers computation from online optimization to offline training. The deployable neural sequenced AFM scheme is thoroughly verified on a microgrid with a single-phase-to-ground fault using hardware-in-the-loop (HIL) in a Real-Time Digital Simulator (RTDS) environment, and the experimental results show that the proposed method can significantly improve system resilience regarding the fault current contribution.