A Self-learning intelligent passenger vehicle comfort cooling system control strategy

This paper establishes a coupled thermal model of the air conditioning system and cabin for electric vehicles, which considers the effects of solar radiation, vehicle speed, and the external environment on the heat exchanged with the cabin. An intelligent air conditioning system control strategy that can learn passengers’ thermal comfort preferences is proposed. This strategy predicts the preferred predicted mean vote of passengers by learning their thermal comfort preferences, then converts it into a target temperature for the air conditioning system. The performance of the proposed strategy is compared with that of conventional on-off and fuzzy PID controllers. In a simulated hot environment, the proposed control algorithm directly and automatically decreased the cabin temperature to the passengers’ preferred temperature without any manual adjustment. It can also maintain stable passenger PMVs and corresponding temperatures regardless of solar radiation, vehicle speed, and external environmental changes. The proposed strategy can improve the thermal comfort of passengers, compared with the on-off and fuzzy PID controllers; moreover, it consumes less energy. In a simulated driving cycle, its energy consumption was 31.8% less than that of the on-off controller and 10% less than that of the fuzzy PID controller, and its COP was respectively 20.4% and 18.7% more than those of on-off controller and the fuzzy PID controller. Therefore, the proposed strategy can make vehicles, especially electric ones, more efficient.