This paper presents a novel adaptable energy management system for smart buildings. In this framework we model the energy consumption of a living unit, and its energy exchange with the surroundings. We explicitly consider the impact of the outside environment and design features such as building orientation, automatic shading, and double facade. We formulate this problem as a nonlinear optimization model in which the living unit minimizes a performance function subject to the energy flows from and toward the unit as well as the building-specific features. It is solved using off-the-shelf solvers. We present computational experiments to validate the proposed approach, considering different objective functions and several building configurations. The experiments show that our approach enhances the unit's performance and also provides demand flexibility for the grid.
Technical report, School of Mathematics, University of Edinburgh, 2020.