Optimal energy management in smart sustainable buildings – A chance-constrained model predictive control approach

Recent European environmental directives, the prevalent consumer desire to minimize electricity costs, and the grid-driven need for flexible buildings all lead to a common outcome: the smart sustainable building (SSB). Coordinated by their building energy management systems (BEMS), SSBs steer their operation towards monetary gains for their owners, and flexibility for grid operators. Another key feature is their sustainability, expressed by the mandatory nearly-zero-energy (nZE) mandate, i.e., balancing yearly energy consumption and on-site renewable energy production. In this paper, we present a generic and comprehensive (in terms of device composition) BEMS framework for SSBs. Aside from operating cost minimization, the BEMS is additionally tasked with overseeing the SSB's environmental profile, ensuring that the nZE mandate is not jeopardized in the pursuit of monetary gains. This is achieved through a novel adaptive sustainability criterion. The inherent uncertainties of solar irradiance and ambient temperature are reflected on the occupants’ thermal comfort, the relevant limitation being cast as chance constraints. The overall mixed-integer linear programming (MILP) problem is solved through model predictive control (MPC). The main contributions lie in the joint consideration of a) a comprehensive devices set, b) weather forecast uncertainties, and c) the employment of the novel adaptive sustainability criterion. The proposed framework is validated in a nigh-exhaustive case study and evaluated with respect to cost management and ability to manage the SSB's nZE status.