A Hybrid Intelligent Control System Based on PMV Optimization for Thermal Comfort in Smart Buildings

With the fast development of human society, on one hand, environmental issues have drawn incomparable attention, so energy efficiency plays a significant role in smart buildings; on the other hand, spending more and more time in buildings leads occupants constantly to improve the quality of life there. Hence, how to manage devices in buildings with the aid of advanced technologies to save energy while increase comfort level is a subject of uttermost importance. This paper presents a hybrid intelligent control system, which is based on the optimization of the predicted mean vote, for thermal comfort in smart buildings. In this system, the predicted mean vote is adopted as the objective function and after employing particle swarm optimization the near-optimal temperature preference is set to a proportional-integral-derivative controller to regulate the indoor air temperature. In order to validate the system design, a series of computer simulations are conducted. The results indicate the proposed system can both provide better thermal comfort and consume less energy comparing with the other two intelligent methods: fuzzy logic control and reinforcement learning control.