Decision Support for Power System Restoration: Mitigating Risks of Dynamic Phenomena

Enhancing existing power system restoration procedures is crucial as renewable energy sources, grid-connected converters, and underground cables become more prevalent. Faster power system restoration is key to achieving higher power system resilience. This paper proposes a systematic approach to re-energize the grid using a Markov Decision Process, while taking into account transient and dynamic phenomena during power system restoration. By quantifying the risks associated with frequency stability, resonant temporary overvoltages, and converter stability, our method informs operators of problematic states that would lead to critical operating points and ultimately set up the restoration process to fail. Furthermore, it calculates an optimal set of actions from a given state, considering these dynamic factors. The assigned costs for all states during power system restoration guide decision-making, ensuring system stability and power balance.