Humans in the Loop: A Stochastic Predictive Approach to Building Energy Management in the Presence of Unpredictable Users

Abstract

Efficient building energy management has attracted a great deal of academic interest with significant potential energy savings to be envisaged. Social scientists strive to achieve these savings by employing behavior-based approaches, while engineers investigate control strategies for the efficient operation of the building devices. This work can be seen as a first step towards bridging these two approaches by proposing a control scheme that encapsulates building occupant behavior into the energy management system. In particular, the occupants willingness to tolerate comfort bound violations is modeled as a random measurable uncertainty and incorporated into the building energy management system through disturbance feedback control policies. The respective optimal control problem is formulated as a mixed-integer stochastic optimization problem, and a computationally tractable approximation of it is derived by restricting the disturbance feedback control policies to admit an affine structure. An extensive numerical study verifies that the proposed approach can significantly reduce the energy consumption of the buildings.