Predictive Control Algorithm for Energy-Efficient Electric Vehicles Thermal Management System

For the near future, the automotive industry is moving towards electrification to reduce greenhouse gas emissions and make mobility environmentally sustainable. In this context, minimising energy consumption becomes a crucial aspect, both to reduce waste and to extend the range of electric vehicles. The aim of this work is to design a Model Predictive Control (MPC) algorithm capable of minimising the energy consumption of the vehicle’s thermal management system, while maintaining cabin and battery temperatures within target limits. The study is conducted in the Matlab/Simulink environment using a baseline model developed by MathWorks. The proposed control strategy focuses on Battery Thermal Management (BTM) and Heating, Ventilation, and Air Conditioning (HVAC) systems, which are integrated and operate on a vapour-compression refrigerant cycle, requiring dedicated on-board components. A two-layer control strategy manages the operation of these components. In particular, a predictive controller regulates the compressor power and a defined cooling split factor, which divides cooling capacity between the chiller (for BTM) and the evaporator (for HVAC). Subsequently, the angular speeds of the compressor, blower, battery pump, and fan are regulated via a second-layer control based on PID controllers; meanwhile, on/off commands are employed for the control of other parts, such as the motor pump, cabin heater, radiator bypass, valves for the chiller bypass and the radiator bypass. The designed predictive controller was tested over different driving cycles, showing a significant reduction in energy consumption while notably improving temperature tracking performance. Specifically, over the UDDS cycle, the energy consumption of the thermal system was reduced by 45.5% compared to the baseline control. This pronounced achievement corresponds to a still meaningful 4.9% decrease in overall vehicle energy consumption, which accounts for both thermal and powertrain demands. Under the tested conditions, thermal management represents 6.4% of total vehicle energy. In electric vehicles, where the energy required for thermal control can reach the same order of magnitude as that required for traction, the proposed MPC approach could lead to even greater reductions, resulting in a significant increase in range. These results highlight the potential of model-based predictive control for thermal management to improve energy efficiency and extend the driving range of future electric vehicles.