A Novel Multi-Stage Economic Dispatch of Microgrid Based on Consensus Protocol of Multi-Agent System

Based on the multi-agent system, this paper presents a novel real-time optimization strategy for microgrid, in which distributed generations (DGs) are parallel-connected through droop control. The economic dispatch consists of three stages: 1). A first-order consensus protocol is adopted to obtain the optimal active power regardless of the voltage and frequency regulation. 2). To track the reference, a novel secondary-order consensus protocol is proposed. 3). To optimize the system frequency and voltage, another secondary-order consensus protocol is used to control the droop parameters adaptively. In this paper, the consensus protocol at each stage is fully distributed, in which each agent only requires its neighbors' information. Moreover, the system has globally asymptotic stability proved by the extended Lyapunov theorem. Lastly, simulation cases are studied to validate the proposed strategy.