Model Predictive Control Based on Discrete Space Vector Modulation for Single-Stage Dual-DC-Port Inverter

Model predictive control (MPC) provides an at tractive solution for the single-stage dual-dc-port inverter (SDI) due to its capabilities to deal with multiobjective control under unbalanced dc-port voltages. However, the existing MPC for SDI uses only one of the basic voltage vectors throughout the control period, which increases the dc-port power ripple and reduces current quality. To address this issue, the candidate voltage vectors are extended from 27 to 117 based on unevenly distributed basic voltage vectors, which reduces the dc-port power ripple and improves power quality due to extra control degree of freedom being provided. To reduce the computational burden caused by extended voltage vectors under unbalanced port voltages, the space vector diagram is divided into asymmetrical quadrangle sub-sectors, where the candidate voltage vectors required to be iterated are reduced by 82%. On this basis, an MPC based on discrete space vector modulation (SVM) is proposed for the SDI, enabling simultaneous grid current tracking and flexible power regulation. The comprehensive comparison experimental tests are conducted to verify the effectiveness and advantages of the proposed MPC based on discrete SVM.