Optimal control of PMSMs using model predictive control with integrator

Permanent magnet synchronous motors (PMSM) have been receiving much attention due to their high efficiency. Usually, an inverter is used to drive the PMSM. In conventional current control systems, the inverter is regarded as an ideal amplifier. However, there exist various constraints such as voltage saturation in current transient response and output nonlinearity in overmodulation. In addition, an output of the inverter is limited to discrete values in a precise sense. Thus, the system has a nature of a hybrid dynamical system. In this paper, we propose a novel model predictive control (MPC) based current control system to improve performance in cases where the inverter cannot provide the appropriate output. In MPC, a steady state error often occurs due to the modeling error. In order to overcome the problem, we extend the model in which the accumulation of the steady state error is included. The effectiveness of the proposed method is shown through simulations and experiments.