The load-influencing factors and variation patterns of train air-conditioning systems in plateau environments are extremely complex, leading to long-term operation under load mismatch and seriously affecting the energy-saving control effect. This study focuses on the predictive control of train air-conditioning systems in plateau environments. Using neural network technology, a load prediction model for the train air-conditioning system is established. Then, a predictive control method for the train air-conditioning system is established based on fuzzy PID control. Based on joint simulation in AMESim and Simulink, the energy consumption of the air-conditioning system is simulated under predictive control mode, and the impact of advanced time in predictive control on the control effectiveness of the air-conditioning system is analyzed. The results demonstrate that incorporating a temperature feedback correction loop into the predictive control mode can enhance the control effectiveness of the air-conditioning system. Compared with the feedback control mode, the daily energy-saving rate of the air-conditioning system under predictive control reached 13.44%. In predictive control mode, the temperature fluctuation amplitude inside the carriage is reduced, resulting in a more comfortable interior environment. This research result can provide a reference for developing energy-saving control strategies for train air-conditioning systems in complex environments, particularly in plateau areas.
Teng et al. (Wed,) studied this question.