Composite Vector Model Predictive Control With Time-Varying Control Period for PMSM Drives

Abstract

In this article, a composite vector model predictive control (MPC) with a time-varying control period for the permanent magnet synchronous motor (PMSM) drives is proposed. First, the conventional vector selection modes in the MPC method are analyzed. Then, a novel selection mode that selects two nonzero-voltage vectors as optimal voltage vector and has a time-varying control period is developed. Different from the conventional methods, this proposed method can adjust the control period to adapt to the reference voltage vector that is outside the linear modulation range and improve the steady-state control performance. Compared with the conventional three-vector method, this method has the advantages of reducing the switching frequency and suppressing the common-mode voltage due to the reduction use of the zero-voltage vector. Finally, experimental comparison results of different methods validate the effectiveness of the proposed method.