Stochastic Predictive Control of Multi-Microgrid Systems

Abstract

This paper presents a stochastic predictive control algorithm for a number of microgrids connected to the same distribution system. Each microgrid includes a variety of distributed resources such as wind turbine, photo voltaic units, energy storage devices and loads. Considering the uncertainty of loads and renewable-based distributed resources, the power management problem is formulated in the framework of stochastic control. The microgrids operation are coupled through a joint probabilistic constraint which requires the power flow from utility to each microgrid limits to a pre-specified range. Utilizing probabilistic distribution function of uncertain variables, the deterministic counterpart of the problem is derived and a close-optimal solution of the problem is achieved. The Monte-Carlo simulation analysis is used to justify the robustness characteristics of the solution.