Multivector Model Predictive Power Control of Three-Phase Rectifiers With Reduced Power Ripples Under Nonideal Grid Conditions

Model predictive power control (MPPC) is a promising control scheme for bidirectional ac-dc rectifiers. However, the control performance of conventional MPPC (C-MPPC) deteriorates under nonideal grid conditions. Furthermore, the one-switching-vector-per-control-interval characteristic of C-MPPC leads to high ripples of control variables. To improve the steady-state performance of rectifiers under nonideal grid voltage conditions, a multivector MPPC (MV-MPPC) scheme is proposed. The proposed method presents a constant switching frequency and better steady-state control performance without increasing its sampling frequency. By selecting two active vectors and one zero vector, the range of optimal vector for active and reactive power regulation can be extended from fixed phase and magnitude to arbitrary phase and magnitude. Incorporated with a second-order generalized integrator for obtaining the quadrature component of grid voltage, the proposed method is applicable to the grid conditions where low-order harmonics exist. The preliminary calculation is adopted to avoid repetitive computation of the predicted values in the implementation of MV-MPPC, which reduces the calculation burden of the proposed scheme. A thorough experimental evaluation of the proposed scheme with the C-MPPC and duty-optimal MPPC has been conducted to validate the superiority of the proposed MV-MPPC solution.