Model Predictive Control of Two-Stage Rectifier Based on Linear Extended State Observer

This paper presents a model predictive control (MPC) method of two-stage rectifier based on linear extended state observer (LESO). The control performance of conventional MPC is easy to be affected under parameter mismatches and external perturbation condition. To solve this problem, the LESOs are designed to observe the system lumped disturbances. Then, predictive model would be compensated by the observed values to improve the accuracy in one period. The cost function of MPC is evaluated by the method of deadbeat control. Furthermore, to reduce the current ripple, the carrier phase shifting pulse width modulation (CPS-PWM) is applied. The simulation results validate the strong robustness and better dynamic performance in two-stage rectifier compared with conventional method.