Predictive control strategies for electric vehicles thermal management system

Optimization of the thermal management in an electric vehicle (EV) is a crucial challenge aimed to improve the overall energy efficiency and ensuring optimal operating conditions for both battery system and passengers comfort. The Battery Thermal Management (BTM) system and the Heating, Ventilation and Air Conditioning (HVAC) system are among the most energy-demanding systems in an EV, significantly impacting the total driving range. Made these considerations, the thesis focuses on the development and implementation, in Matlab/Simulink environment, of an adaptive Model Predictive Control (MPC) strategy to enhance the energy efficiency of these systems while maintaining both battery and cabin temperatures within the desired limits. The work is structured in five chapters. The first one presents a deep analysis and reverse engineering of the Simscape model developed by MathWorks and used for simulation, providing a detailed description of components, specially the ones directly affected by the developed AMPC controller, having then clear the reference model used for evaluating the performances of the different control strategies. The second chapter introduces the initial control logic, which is based on reactive and PID strategies, and lays the theoretical fundamentals for MPC. In the third chapter is presented the hearth of this thesis work, describing first of all the two control configurations explored, one prioritizing the battery and the other the cabin temperature regulation, then analyzing relatives predictive models build and controls configuration. This dual-priority control allows the system to dynamically adjust control objectives based on current operating conditions. In the fourth chapter is presented an extensive performance evaluation through the simulations results, in which is demonstrated that the proposed adaptive MPC strategy achieves a reduction in cooling system energy consumption of 28% on the UDDS test cycle and 18% on the WLTC cycle compared to baseline control logic, without compromising the thermal comfort or battery safety. This translates into an overall 2.7% reduction in total vehicle energy consumption under the tested condition of 26°C environmental temperature, which paves the way for even greater percentage savings in more extreme environmental conditions, where the cooling system has a greater impact on the total consumption of the vehicle. Finally, the fifth chapter discusses the conclusions and potential future developments, underlining how these simulations confirm that adaptive MPC offers a significant advantage in therms of energy efficiency thanks to the prediction of the system’s states evolution allowing for the calculation of the optimal control sequence, which convert into the ability for an EV to cover greater mileage with the same battery capacity.