Sequential recovery of cyber-physical power systems with distributed generation

In this paper, a sequential recovery framework is constructed for post-disaster cyber-physical power systems from the viewpoint of the theory of complex networks. Grounded in the power flow analysis pattern of cyber- physical power systems, the framework integrates the recovery sequence, the access of distributed generation, and the cascading failures during restoration. Furthermore, the deep Q-network algorithm is utilized to determine the most effective sequence for repairing malfunctioning parts, thereby maximizing the recovery of the power that was lost. The effects of distributed generation node quantity ratio and permeability on system recovery performance are analyzed on the cyber-physical power system test cases generated grounded in IEEE 30-bus system and IEEE 118-bus system. Meanwhile, the applicability and superiority of the proposed recovery strategy are verified by comparison with alternative strategies.