Key points are not available for this paper at this time.
Integrating advanced control strategies is essential in reducing energy cost, optimizing interaction with the electric grid, and enhancing indoor environmental quality in buildings. Enhancing energy flexibility in the building demand profile is essential for the safe and efficient operation of smart grids. This paper presents a grid-interactive model predictive control methodology to improve energy flexibility and maintain indoor environmental quality in school buildings. The proposed model predictive control framework employs data-driven grey-box thermal network models for classrooms with convective heating systems. The methodology is implemented in an all-electric school building in Montreal, Canada, during very cold winter days. Two control scenarios are investigated and compared: 1) a reference scenario using a proportional-integral controller and business as usual thermostat setpoints and 2) a flexible control scenario using model predictive control. Both scenarios were tested under two conditions: with and without occupants in classrooms. Four classrooms operated with proportional-integral controller and usual setpoint profiles were considered reference cases, while another four classrooms employed model predictive control as flexible cases. Results showed that the model predictive control increased energy flexibility by 36% in unoccupied conditions and 61% in occupied conditions while reducing energy consumption by 25% and 63%, respectively. Throughout both scenarios, the model predictive control maintained satisfactory thermal comfort and indoor air quality. This approach is scalable and transferable to other institutional or mid-size commercial buildings.
Morovat et al. (Mon,) studied this question.