A Distributed Mixed-Integer Framework to Stochastic Optimal Microgrid Control

Abstract

In this paper, we consider distributed control of microgrids composed of storages, generators, renewable energy sources, critical and controllable loads. After recalling a Mixed-Integer Linear Programming (MILP) formulation of the optimal control problem associated to the microgrid, we show that this optimization problem can be recast as a distributed optimization problem. Following the approaches proposed in the literature, we formulate a distributed, two-stage stochastic optimization problem associated to the microgrid and solve this problem using a distributed approach for MILPs. The resulting algorithm is able to compute high-quality feasible solutions to the two-stage stochastic problem, for which we provide a theoretical upper bound on the constraint violation. Numerical computations show the efficacy of the proposed solution.