Simulation-based policy improvement for power management in buildings

The technology advances in distributed power generation, co-generation of cooling, heat and power, storage devices and micro grid provide the opportunity to better manage the power in buildings in order to save power, and to reduce the CO2 emission. Due to the complexity of such a multi-energy system, simulation is usually the only faithful way to accurately describe the system dynamics and for performance evaluation. However, simulation is usually time-consuming and provides only noisy observations. Thus finding the optimal power management policy is nontrivial. In this paper, a joint schedule problem is considered to schedule solar power, wind power, combined cooling, heating, and power generation (CCHP), battery, and high temperature chiller in order to satisfy the load on electricity, sensible heat load, and latent heat load in buildings with the minimal average cost. The rollout method is applied to improve from given base policies through simulations. Numerical results show that the method obtains policies better than the base policies.