Automatic Control System for Thermal Comfort Based on Predicted Mean Vote and Energy Saving

For human-centered automation, this study presents a wireless sensor network using predicted mean vote (PMV) as a thermal comfort index around occupants in buildings. The network automatically controls air conditioning by means of changing temperature settings in air conditioners. Interior devices of air conditioners thus do not have to be replaced. An adaptive neurofuzzy inference system and a particle swarm algorithm are adopted for solving a nonlinear multivariable inverse PMV model so as to determine thermal comfort temperatures. In solving inverse PMV models, the particle swarm algorithm is more accurate than ANFIS according to computational results. Based on the comfort temperature, this study utilizes feedforward–feedback control and digital self-tuning control, respectively, to satisfy thermal comfort. The control methods are validated by experimental results. Compared with conventional fixed temperature settings, the present control methods effectively maintain the PMV value within the range of ${\pm} 0.5$ and energy is saved more than 30% in this study.