Design Method of Model Predictive Control for Development of Current Control System of PMSM

This paper describes the design method for Model Predictive Control (MPC) that is suitable for current control systems of Permanent Magnet Synchronous Motor (PMSM). A current control system based on MPC predicts future current behaviors for considerable input voltage and determines the optimum input voltage by evaluating these behaviors. Conventional MPC sets the cost function, making it effective for several control purposes. However, the suitability of the search space of conventional MPC has not been discussed and its achievable performance was limited. In this paper, we consider an appropriate search space for current control system of PMSM by referring to existing modulation methods and inverter operation. Then, we propose new MPC that satisfies that search space and show that the proposed MPC can usage from a linear region to a square-wave region by the appropriate search space. c 2020 The Institute of Electrical Engineers of Japan.