Robust Model-Predictive Control for a Compound Active-Clamp Three-Phase Soft-Switching PFC Converter Under Unbalanced Grid Condition

Two main problems, including unbalanced grid input and circuit parameters uncertainty in the three-phase soft-switching converter control system, are considered in this paper. First, the constant-switching-frequency model-predictive control (CF-MPC) is improved for application in compound active-clamp zero-voltage-switching three-phase power factor correction converter. Second, a robust-constant-frequency model-predictive control (RCF-MPC) method is proposed for current loop; the novel contributions are: 1) the negative-sequence current component is introduced into the cost function of the model-predictive control to decrease the influence of unbalanced grid input; and 2) a robust item is added into the predictive model to reduce the influence of the circuit parameter uncertainty. Third, a robust variable structure controller is proposed for voltage loop considering the capacitor and load uncertainty. The simulation and experiment results show that, compared with the original CF-MPC method and the traditional proportional-integral control method, the proposed RCF-MPC method decreases the total harmonic distortion under the unbalanced grid condition. Meanwhile, the voltage loop robust variable structure controller has better transient and steady-state performance with respect to the load perturbation.