Adaptive Model-Free Predictive Current Control for SPMSM Drives With Optimal Virtual Vector Modulation

This article proposes an adaptive model-free predictive current control (MFPCC) method with optimal virtual vector modulation (OVVM) for surface-mounted permanent magnet synchronous motors. First, an ultralocal motor model based on the rotating coordinate system is established by using an adaptive extended state observer (AESO). Second, an adaptive law is devised to achieve a more balanced mitigation of disturbance and noise interference and a better transient and steady-state performance. Simultaneously, an integrated parameter estimator is designed based on the AESO, allowing for real-time adjustments of control coefficients within the ultralocal model. This dynamic calibration enhances the accuracy of model-free predictive current control and facilitates the realization of system-wide self-adaptation to handle intricate and dynamic working scenarios effectively. Third, an extended voltage vector control set based on virtual vectors is constructed to improve the accuracy of voltage selection. On this basis, apply the OVVM method to generate symmetrical pulsewidth modulation waveforms to further reduce current harmonics and improve steady-state performance. Finally, the experimental results have verified the effectiveness of the proposed MFPCC method.