HVAC Scheduling under Data Uncertainties: A Distributionally Robust Approach (2103.05850v1)

Abstract: The heating, ventilation and air condition (HVAC) system consumes the most energy in commercial buildings, consisting over 60% of total energy usage in the U.S. Flexible HVAC system setpoint scheduling could potentially save building energy costs. This paper first studies deterministic optimization, robust optimization, and stochastic optimization to minimize the daily operation cost with constraints of indoor air temperature comfort and mechanic operating requirement. Considering the uncertainties from ambient temperature, a Wasserstein metric-based distributionally robust optimization (DRO) method is proposed to enhance the robustness of the optimal schedule against the uncertainty of probabilistic prediction errors. The schedule is optimized under the worst-case distribution within an ambiguity set defined by the Wasserstein metric. The proposed DRO method is initially formulated as a two-stage problem and then reformulated into a tractable mixed-integer linear programming (MILP) form. The paper evaluates the feasibility and optimality of the optimized schedules for a real commercial building. The numerical results indicate that the costs of the proposed DRO method are up to 6.6% lower compared with conventional techniques of optimization under uncertainties. They also provide granular risk-benefit options for decision making in demand response programs.