부산대학교 도서관

Model predictive current control (MPCC) is widely used and more popular control method for a permanent magnet synchronous motor (PMSM) drive. It is less complex and simple to understand as compared with other predictive control methods. MPCC involves the evaluation of a simpler cost function for selecting an optimal voltage vector (VV), which is based on the minimization of error between reference and sensed currents. For a two-level inverter with limited number of VV, MPCC gives poor steady-state performance as a single VV is applied over one sampling interval. In this article, for improving the steady-state performance of PMSM, an improved MPCC is proposed. In this proposed MPCC method, an active VV along with a zero VV is applied similar to the duty cycle control (DCC); however, the timings of active VV and zero VV are selected to decrease the q -axis current ripples. The method introduced in this article is based on the calculation of q -axis current slopes to calculate optimal timing for different VVs. The method introduced in this article is compared with conventional MPCC (C-MPCC) and MPCC with DCC. Simulation and experimentation is carried out for comparison and to confirm the effectiveness of introduced MPCC.