Model Predictive Control of a New MRAS Observer for PMSM

In response to the issues of inaccurate rotor angle observation, large speed estimation errors, and weak load disturbance rejection capability in the traditional model reference adaptive system (MRAS) for permanent magnet synchronous motors (PMSM), this paper proposes a fast super-twisting algorithm adaptive observer (FSTA-AO) based on the model predictive direct speed control (MPDSC) without weighting model prediction. First, replacing the traditional PI adaptive rate with the fast super-twisting algorithm has improved the system’s precision in observing the rotor angle and estimating the speed. Additionally, an extended sliding mode load torque observer is designed to provide real-time feedback of the observed load torque value to the model predictive direct speed control, thereby enhancing the system’s disturbance rejection capability against load disturbances. Finally, simulation results indicate that in the control method of MPDSC, compared to traditional MRAS, the MRAS based on FSTA-AO exhibits superior observation performance and dynamic characteristics.