A Model Predictive Current Control for Interior PMSM Based on Least Squares Parameter Adaptive Feedback Correction

The model predictive current control (MPCC) of an interior permanent magnet synchronous machine (IPMSM) requires an accurate motor parameter model to predict future currents and achieve high control performance. However, the inductance parameters of an IPMSM are easily affected by factors such as magnetic field saturation, leading to large current prediction errors, high current ripple, and poor stability. Therefore, an MPCC strategy for an IPMSM based on parameter adaptive feedback correction is proposed. First, based on the mathematical model of the IPMSM in the synchronous rotary coordinate, the cross-coupling relationship between the dq-axis inductance deviations and the current prediction error is derived to form an explicit prediction error model. Then, the influence of the d-axis and q-axis inductance parameter deviations of the IPMSM on the current prediction error is discussed in detail. Next, based on the established mathematical model of the prediction error, the recursive least squares scheme is adopted to identify the d-axis and q-axis deviations of the inductance parameters online. Finally, unlike conventional open-loop RLS correction, a PI-based closed-loop correction loop is designed that feeds the prediction error back to adjust the inductance deviations, thereby forcing the prediction error toward zero while inherently compensating for inverter dead-time effects. Simulations and experiments were conducted, and the results show that the proposed scheme greatly improves the accuracy of current prediction and inductance parameter estimation, and enhances robustness against parameter mismatch and dead-time disturbances. The key novelty lies in the PI-feedback-driven RLS closed-loop structure that simultaneously achieves error elimination and dead-time compensation.