An incremental scenario approach for building energy management with uncertain occupancy

Arman Karshenas,Kostas Margellos,Simone Garatti

DOI: https://doi.org/10.48550/arXiv.1911.06733

2019-11-16

Abstract:We deal with the problem of energy management in buildings subject to uncertain occupancy. To this end, we formulate this as a finite horizon optimization program and optimize with respect to the windows' blinds position, radiator and cooling flux. Aiming at a schedule which is robust with respect to uncertain occupancy levels while avoiding imposing arbitrary assumptions on the underlying probability distribution of the uncertainty, we follow a data driven paradigm. In particular, we apply an incremental scenario approach methodology that has been recently proposed in the literature to our energy management formulation. To demonstrate the efficacy of the proposed implementation we provide a detailed numerical analysis on a stylized building and compare it with respect to a deterministic design and the standard scenario approach typically encountered in the literature. We show that our schedule is not agnostic with respect to uncertainty as deterministic approaches, while it requires fewer scenarios with respect to the standard scenario approach, thus resulting in a less conservative performance.

Systems and Control,Optimization and Control