Prediction error analysis of finite-control-set model predictive current control for PMSMs

Abstract

For permanent magnet synchronous motor (PMSM) drive systems adopting finite-control-set model predictive current control (FCS-MPCC) strategy, the parameters used in prediction model may not be the same as the actual values during the operation, especially for d-and q-axis stator inductances. Such parameter mismatches can result in prediction error for FCS-MPCC and further deteriorate the steady-state control performance of motor system. In order to solve this problem, the prediction error was firstly defined as an evaluation index and its mathematical model was then derived. The theoretical analysis shows that the same positive deviation and negative deviation of d- or q-axis stator inductance can lead to different effects on the prediction error of d- or q-axis current, especially the negative one. Finally, the experimental results verify the theoretical analysis of prediction error. Besides, the corresponding current tracking errors are also analyzed. It provides effective guidelines for reduction of prediction error and improvement of control performance for model predictive control.