Model predictive control for thermal energy storage and thermal comfort optimization of building demand response in smart grids

摘要

Demand response (DR) can effectively manage electricity use to improve the efficiency and reliability of power grids. Shutting down part of operating chillers directly in central air-conditioning systems can meet the urgent power reduction needs of grids. But during the special events of fast DR, how to optimally control the active cold storage considering the indoor environment of buildings and the needs of grids at the same time is rarely addressed. A model predictive control (MPC) approach, with the features of shrunk prediction horizon, self-correction and simple parameter determination of embedded models, is therefore developed to optimize the operation of a central air-conditioning system integrated with cold storage during fast DR events. The chiller power demand and cooling discharging rate of the storage are optimized to maximize the building power reduction and meanwhile to ensure the acceptable indoor environment. Case studies are conducted to test and validate the proposed method. Results show that the proposed MPC approach can effectively handle the optimal controls of cold storage during DR events for required power reduction and acceptable indoor environment. Due to the feedback mechanism of MPC, the control performance is not negatively influenced by the simplified parameter identification of models, which will be convenient for real applications. While achieving the expected building power reduction for the power grid, the indoor environment is effectively improved in the DR events using the MPC and the maximum indoor temperature is reduced significantly without extra energy consumed.