Multi-agent system architecture for enhanced resiliency in autonomous microgrids

This paper proposes a hierarchical Multi-Agent System (MAS) based controller architecture for enhanced resiliency during the controller failure condition. A self-healing scheme is proposed for fault detection and recovery during the controller failure. The proposed scheme combines both sentinel agent approach and autonomous workload distribution technique. Further, an acquisitive optimization strategy is implemented for energy management during the power generation failure. Faster response is achieved by programming the agents with a Peer-to-Peer (P2P) interchangeable sender-receiver communication strategy. The proposed controller architecture has been verified by considering a power system case study.