Abstract: Human thermal comfort is influenced by various factors such as environmental temperature and humidity. Currently, most air conditioners on the market primarily focus on temperature control, with few systems capable of simultaneously controlling both temperature and humidity. To address this, analyzed the primary factors influencing human thermal comfort and the principles of air conditioner temperature and humidity control technology. Historical data from air conditioners operating in steady-state conditions were extracted, and a BP neural network model was used to establish the mapping relationship between indoor and outdoor temperature and humidity and compressor frequency and indoor fan speed. Enabled the selection of the optimal temperature and humidity control combination that ensures human thermal comfort while minimizing energy consumption, thereby enabling real-time control of target temperature and humidity levels. Experimental results show that, while maintaining the same level of human thermal comfort, simultaneous optimization control of indoor temperature and humidity can reduce air conditioner energy consumption by up to 12.3%. Provides guidance for the development of air conditioning systems based on big data for thermal comfort control.

Key words: Thermal comfort, Temperature and humidity, Energy saving, Neural network