Optimal Operation of Interconnected Home-Microgrids with Flexible Thermal Loads: A Comparison of Decentralized, Centralized, and Hierarchical-distributed Model Predictive Control

This paper presents the formulation and comparison of three different model predictive control coordination strategies for a group of interconnected home-microgrids. First, a decentralized coordination strategy is taken as a reference case, followed by a proposed centralized coordination. Lastly, by means of dual decomposition, a hierarchical-distributed strategy is derived from it. The simulation results indicate that the proposed centralized formulation is able to improve the electrical power balancing among systems, with a reduction in extreme power peaks at the point of common coupling. The hierarchical-distributed strategy also shows an improvement in power profile compared to the decentralized strategy, while solving the global problem in a distributed way. The trade-off among local balancing with peak power reduction, energy bill, and storage losses has to be further investigated.