Destabilizing Attack and Robust Defense for Inverter-Based Microgrids by Adversarial Deep Reinforcement Learning

The droop controllers of inverter-based resources (IBRs) can be adjustable by grid operators to facilitate regulation services. Considering the increasing integration IBRs at power distribution level systems like microgrids, cyber security is becoming a major concern. This paper investigates data-driven destabilizing attack and robust defense strategy based on adversarial deep reinforcement learning for microgrids. Firstly, full-order high-fidelity model reduced-order small-signal typical microgrids are recapitulated. Then control gains analyzed, which reveals its impact system stability. Finally, problems formulated as Markov decision process (MDP) MDP (AMDP). solved twin delayed deterministic policy gradient (TD3) algorithm find least effort path obtain corresponding strategy. simulation studies conducted in an microgrid with 4 IEEE 123-bus 10 evaluate proposed method.