Gossip-Based Economic Dispatch Algorithm of Grid-Connected Microgrids Over Unbalanced Directed Networks

A gossip-based distributed economic dispatch (ED) algorithm of grid-connected microgrids is introduced in this paper to optimize distributed generation costs despite random communication link failures. The algorithm solves the ED problem under asynchronous and synchronous directed communication within the framework of a multi-agent system. It employs a leader-following consensus approach to achieve distributed ED, ensuring supply-demand balance and adherence to generator constraints. The convergence analysis proves the algorithm's ability to reach mean square consensus and almost sure consensus if the expectation of the switching topologies is strongly connected, which is a milder condition than uniformly jointly strongly connected or balanced switching topologies. The specific proof holds for any initial value, indicating the proposed algorithm has no requirement in initialization. Simulations validate the algorithm's robustness against communication disruptions and its effectiveness in plug-and-play and maintaining economic operations.