Due to the integration of volatile renewable energy and random energy consumption in the building grid, uncertainties have become a big concern for the operation of energy-efficient buildings. To minimize energy usage expenses under uncertainty, it is necessary to determine the optimal power production for a building from various energy sources, including the electric grid, battery, and combined heat and power with a boiler unit. This article presents a multi-stage mixed-integer stochastic programming model for optimal operation of energy-efficient building systems considering controllable electric and thermal loads. Taking into account the randomness of non-controllable electric and thermal loads, as well as solar power generations through the multi-stage scenario tree, the operation of energy-efficient buildings will be more robust against changes in uncertain variables. With information of uncertainties updated hourly, the rolling scheduling method is introduced to determine an adaptive power output of electric grid, charging/discharging status of the battery, and operation of combined heat and power with a boiler unit. The simulation results offer a set of adaptive decision solutions within the scheduling horizon.
