Abstract:

A model predictive control (MPC) method for the grid-connected inverter based on adaptive error correction is proposed to address the problem that the grid-connected inverter based on the finite control set model predictive control (FCS-MPC) experience prediction current errors, and the mismatch of filter inductance parameters can further exacerbate these errors. Firstly, a model of the grid-connected inverter based on FCS-MPC is established, and the causes of the prediction current errors are studied. Then, an error correction link is added to the current prediction model. By using the extended Kalman filter (EKF) to identify the filter inductance parameters in real time, the MPC reference current can be adaptively corrected when the inductance parameters are mismatched, and the harmonic components of the grid-side current can be suppressed. Finally, the effectiveness of the proposed method is verified by simulations.

Key words: finite control set model predictive control (FCS-MPC), grid-connected inverter, adaptive error correction, inductance parameter identification