Hierarchical Model Predictive Control for Energy Efficient Buildings with Multi-Energy Storage Systems

Although the potential role of energy storage to support integration of Renewable Energy Sources (RES) and help meet the challenging decarbonization and energy targets, is well recognized, there is still little understanding of impacts and synergies of Thermal Energy Storage (TES) and Electric Energy Storage (EES) systems as underpinning viable solution for bringing flexibility and support the grid by providing demand-side services. In this study, a robust hierarchical Model Predictive Control (MPC) approach for the energy management of commercial buildings with multi-energy systems, in particular, TES and EES systems, is proposed. The proposed control framework integrates cost-saving, demand response, environmental aspects. The proposed control architecture is hierarchical to better deal with the complexity of the energy management problem. At the higher level, optimal trajectories are scheduled by taking into account the fluctuation of the electricity tariffs and longer prediction horizons, while, at the lower level, the regulator is responsible for a robust reference tracking, which takes the uncertainty into account. A data-driven approach for modeling the building's thermal dynamics and the storage systems are adopted. Numerical results carried on a university building verify economic benefits, promising control performance, and robustness of the proposed strategy.