Modeling error compensation in FCS-MPC of a three-phase inverter

The optimal control action of the traditional finite control set model predictive control (FCS-MPC) of a three-phase inverter is selected with the supposition that the designed prediction model of three-phase inverter is consistent with the real system, and the influence of modeling errors is neglected. But in fact, due to the existence of modeling errors, the selected optimal control action is no longer optimal when it is active. In this paper, the three-phase inverter finite control set model predictive control with modeling error compensation (FCS-MPCMEC) is proposed, in this scheme the output of the three-phase inverter prediction model at current control instant is compensated by the modeling error of the last control instant, and the closed-loop control of FCS-MPC for the three-phase inverter is achieved. The control performance of the three-phase inverter in the conditions of no load, resistive-inductive load, nonlinear load and load step are analyzed by the comparison simulations between the FCS-MPC scheme and the FCS-MPCMEC scheme for four types of modeling error. The results verify that the FCS-MPCMEC scheme excels the traditional FCS-MPC scheme on the three-phase inverter output voltage tracking control with the modeling error above.