Improved double-vector model predictive control to reduce current THD for ANPC-5L inverters

The classical model predictive control (MPC) usually aims to minimize the current error at the end of the control period. This method is equivalent to minimizing the current total harmonic distortion (THD) when the control period tends to be infinitely small. However, due to the limitations of hardware in practice, the optimization effect of current THD is not obvious. Aiming at the above problem, an improved double-vector model predictive control method is proposed for the single-phase active-neutral-point-clamped five-level (ANPC-5L) inverter. Two voltage vectors in each control cycle are used in this method to reduce the current THD by minimizing the area enclosed by the actual and reference current. In addition, a capacitor voltage control method is designed for the problem of multiple control objectives coupling in this structure. In this method, an idea of hierarchy is designed for voltage jump and switching transition limits, neutral point (NP) voltage, and flying capacitor (FC) voltage, achieving the effect of weightless factor and good dynamic and static performance. Finally, the effectiveness of the proposed method is verified by simulation and experimental results.