Thermal adaptive behavior and thermal comfort for occupants in multi-person offices with air-conditioning systems

A better understanding of the adaptive behavior of office users is very important for creating comfortable indoor thermal environment and working healthily and efficiently. This study considers a multi-person office with air conditioning systems in Guangzhou, China as a research object. The relationship between users' adaptive behaviors and thermal satisfaction and influences of environmental parameters and gender on the probabilities of adaptive behaviors were studied. The logical regression method was used to conduct a probabilistic prediction analysis of adaptive behaviors in cold and hot environment. The study shows that user only accepts an indoor temperature change of approximately 1 °C higher than the neutral temperature in hot environment. Under the same thermal sensation vote, the indoor temperature when the user's adaptive behaviors are generated is approximately 1 °C higher (in hot environment) or lower (in cold environment) than that when no action occurs. The indoor temperature and globe temperature significantly affect the occurrence probabilities of adaptive behaviors (P < 0.005). The overall regression accuracy is higher than 70%. The characteristic coefficients of the adaptive probabilistic prediction model of cold and hot behaviors (using indoor temperature data) are 26 °C and 28 °C, which are the critical temperatures of indoor thermal satisfaction and dissatisfaction, respectively. Men engage in hot adaptive behaviors when the indoor temperature is approximately 22 °C, which is around 3 °C lower than women.