A novel online prediction method for vehicle cabin temperature and passenger thermal sensation

Abstract

Vehicle cabin temperature and passenger thermal sensation are essential indicators for the vehicle thermal management system. However, the temperature distribution in the vehicle cabin is nonlinear, and the passenger thermal sensation is subjective. In this study, an online prediction method for advanced control is proposed. The cabin temperature is predicted based on the physics-data hybrid-driven model. Passenger comfort is predicted based on the human–machine interaction technology. The results show that the root mean squared error of the temperature prediction model is less than 0.5 °C, and the prediction accuracy of passenger thermal sensation is more than 91.7 %. The method proposed in this paper has broad applicability, and only a small amount of data is needed to complete the prediction process.