Model predictive control based optimal torque distribution strategy for a compound electric vehicle

Abstract

This research explores energy-efficient torque distribution strategies for a compound electric vehicle (EV), whose propulsion system employs a single induction motor (IM) at the front wheel axle and two permanent magnet synchronous motors (PMSM) in the rear wheels. In addition, it is our research goal to utilize the information and communication technology (ICT) to enhance the energy efficiency of the compound EV. The model predictive control (MPC) strategy is proposed with the aim to utilize up-to-date road and traffic information through ICT to keep optimizing the torque distribution of the compound EV while satisfying driver's torque demands and constraints due to vehicle dynamics, safety and actuator limitations, etc. The performance of the proposed MPC-based torque distribution scheme is evaluated through the hardware-in-the-loop simulations (HiLS). Experimental results show that the MPC strategy can outperform alternative solutions including the dynamic programming approach. Moreover, the FTP-72 driving cycle tests reveal that the proposed compound EV consumes less energy than existing EVs with a single induction motor by ~6% in urban driving conditions.