Predictive control for reliable microgrid energy management under uncertainties

This paper proposes the basic elements of an approach to reliable microgrid energy management based on receding horizon control. The microgrid considered for exemplification is connected to an external grid via a transformer and contains a local consumer, a renewable generator (wind turbine) and a storage facility (battery). In order to minimize costs, a scheduling of the battery usage needs to be determined. To this end, a predictive control framework is proposed here, which allows to naturally take into consideration cost values, power consumption and generation profiles, as well as functioning constraints under uncertainty due to variations in the environment (wind speed), to failures and subsequent repairs of the generator. The proposed approach is validated through simulation results using real numerical data for a reliability test system.