Distributed Optimization for Energy Allocation in Mictrogrid Power System

In this paper, a continuous-time distributed algorithms is proposed to solve the energy allocation problem in microgrid power system. Projected output feedback method is employed in the approach to deal with the local capacity constraints. Each agent in the multi-agent network has its own local objective function, local hound constraint and can access the information from the agent nearby. The agents determine the optimal power generation in the responsible area to minimize the sum of local generation cost function while reaching the global supply and demand balance. Also, the distributed method is applied to test on a six-agent system and the simulation result demonstrates that the algorithm proposed can successfully obtain the optimal power generation and the auxiliary variables of each generator agent can always reach consensus.