Control-Oriented Model for Thermal Energy Management of Battery Electric Vehicles

This paper presents a control-oriented thermal model for a novel heating, ventilation, and air conditioning (HVAC) system with a heat pump designed specifically for battery electric vehicles (BEVs). The model is validated through high-fidelity simulations from GT-SUITE, achieving a root mean square error of $1.54 \,\%$ for cabin air temperature and $0.70 \,\%$ for battery temperature. The model is used for exploring energy reduction strategies. Optimal results demonstrate a $5.7 \,\%$ decrease in energy consumption during a cool-down scenario in high ambient temperatures. The usage of the model is showcased in a cabin thermal pre-conditioning problem where the vehicle is not occupied for an interval of $600 \,\mathrm{s}$ between two drives. Results show that the ability to predict occupancy and optimally manage cabin temperature could reduce energy usage by $12.9 \,\%$, compared to maintaining cabin temperature at the same level. Another example examines the impact of predicted data uncertainty and the result shows that it is more energy-efficient to maintain cabin thermal comfort instead of turning off the HVAC system when the likelihood of passenger arrival is uncertain and exceeds approximately $60 \,\%$ of the predicted non-occupied interval.