Derating of automotive drive systems using model predictive control

Temperature monitoring and derating of electric vehicle (EV) drive trains is of key importance to increase power and torque utilization while preventing thermal damages. Derating by static regulations and standard closed-loop controllers (e.g. PI-type) is not optimal since conservative safety margins have to be chosen to prevent temperature constraint violations. In contrast, model predictive control (MPC) is able to fully utilize the overload capacity of a given drive train and to operate within defined temperature limits. The main contributions of this work is to close the gap between the optimal motion control of EVs and the derating of electrical drive trains. For this goal, simplified vehicle, motor and track models are introduced since the time constants of the mechanical car model are significantly smaller compared to the thermal motor dynamics. A simulative study of a permanent magnet synchronous motor based drive train and the SFTP-US06 drive cycle illustrate the proposed MPC-based derating.