Optimal Thermal Management for Electrified Vehicles: a model predictive control approach based on route information

In the last decade, as a consequence of the widespread environmental consciousness, vehicle electrification demand is growing quickly. One of the most critical devices of electrical vehicles is the Thermal Management System (TMS) whose accurate design plays a crucial role to guarantee adequate battery performance, vehicle safety, and passengers’ comfort. The aim of the thesis is to propose a Model Predictive Control (MPC)-based approach to optimally regulate the behavior of both the battery thermal management system (BTMS) and the cabin air-conditioning (AC) system, with the purpose of simultaneous battery lifespan maximization and energy consumption minimization. The proposed control structure is based on two separate devices: (i) the optimal reference generator and (ii) the MPC feedback loop. By exploiting information on future vehicle missions, provided through the Vehicle-to-Everything (V2X) connectivity, the reference generator computes the optimal thermal trajectory of the system over a sufficiently large prediction horizon. Such an optimal trajectory is then injected into the MPC-feedback loop which acts on the Compact Refrigeration Unit (CRU) and the valves of the refrigeration circuit in order to guarantee reference tracking, disturbance attenuation, and satisfaction of hard constraints on the main physical variables involved in the problem. The effectiveness of the proposed solution is shown through several simulation tests performed by applying the designed control systems to an accurate mathematical model of the real plant developed by the researchers of the Department of Energy (DENERG) of Politecnico di Torino. The obtained results highlighted the impact of prediction horizon length on the TMS performance. By increasing the prediction horizon, significant improvements in terms of energy savings occur.