A high-performance model predictive torque control concept for induction machines for electric vehicle applications

Induction machines are widely used in electric vehicles due to their high reliability and low costs. Controlling these machines to meet the high-performance demands presents a significant challenge since they are often operated at high speed and within operating ranges where magnetic saturation plays a significant role. Furthermore, specific motor parameters are not accurately known or vary during operation, e.g., due to temperature changes. Therefore, there is still a demand for control strategies to meet these demands systematically. This paper proposes a novel control strategy combining a model predictive control (MPC) concept with a fast feedback controller and a nonlinear observer. The proposed MPC strategy is based on a magnetic nonlinear model and allows for a long prediction horizon. It features high torque dynamics while ensuring energy optimality in the steady state. The results also show excellent performance for high rotational speeds and the operation at the system limits, outperforming state-of-the-art control concepts. •Control of induction machine reaching high torque dynamics and efficiency.•Systematic consideration of voltage and current constraints of a saturated machine.•Verification on an industrial test bench.•Comparison with a state-of-the-art control system.