Study on the application of reinforcement learning in the operation optimization of HVAC system
reinforcement learning, HVAC system, operation optimization, control strategy, VAV system, energy saving
Supervisory control can be used to optimize the HVAC system operation and achieve building energy conservation, while reinforcement learning (RL) is considered as a promising model-free supervisory control method. In this paper, we apply RL algorithm to the operation optimization of air-conditioning (AC) system and propose an innovative RL-based model-free control strategy combining rule-based and RL-based control algorithm as well as complete application process. We use a variable air volume (VAV) air-conditioning system for a single-storey office building as a case study to validate the optimization performance of the RL-based controller. We select control strategies with the rule-based control controller (RBC) and proportional-integral-derivative (PID) controller respectively as the reference cases. The results show that, for the air supply of single zone, the RL controller performs the best in terms of both non-comfortable time and energy costs of AC system after one-year exploration learning. The total energy consumption of AC system reduced by 7.7% and 4.7%, respectively compared with RBC and PID strategies. For the air supply of multi-zone, the performance of RL controller begins to outperform the reference strategies after two-year exploration learning and two-year buffer stage. From the seventh year on, RL controller performs much better in terms of both non-comfortable time and operating costs of AC system, while the operating cost of AC system is reduced by 2.7% to 4.6% compared with the reference strategies. In addition, RL controller is more suitable for small-scale operation optimization problems.
Tsinghua University Press
Xiaolei Yuan, Yiqun Pan, Jianrong Yang, Weitong Wang, Zhizhong Huang. Study on the application of reinforcement learning in the operation optimization of HVAC system. Building Simulation 2021, 14(1): 75-87.