A higher?precision air conditioning load model serves as a crucial foundation for developing and implementing effective air conditioning control strategies, which is conducive to reducing electricity consumption and saving power costs. Firstly, by analyzing the impact of building structure, indoor and outdoor environment, and meteorological factors, a grey?box model is constructed for predicting air conditioning loads.. This model consists of a third?order equivalent thermal parameter model and a second?order equivalent moisture resistance model. Subsequently, the optimization objective function is established by minimizing the error between the indoor temperature and humidity output from the model and the measured temperature and humidity. Then, a parameter identification method based on the particle swarm optimization (PSO) algorithm is proposed and employed to obtain the crucial parameters of the grey?box model. Experimental studies demonstrate that the identified equivalent thermal resistance and moisture resistance models accurately reflect the indoor temperature and humidity distribution and variation characteristics, thus possessing practical application value in predicting air conditioning loads.