Robust Model Predictive Current Control of PMSM Based on Nonlinear Extended State Observer

The control performance of traditional model predictive control (MPC) usually deteriorates dramatically with the permanent magnet synchronous motor (PMSM) parameter changes in operation, leading to current harmonics and torque ripple. This article proposes a robust model predictive current control (MPCC) method based on a nonlinear extended state observer (NESO) to resolve this problem. The NESO is used to observe the current disturbance part to avoid the influence of motor parameters on the current model. Because this method is not sensitive to changes in motor parameters, the system is robust and has a low current harmonic. Second, the ESO designed in this article based on the nonlinear function has a lower observation error. In addition, this article uses the fixed coefficient method to simplify the design process. A more accurate parameter design method is proposed based on the bounded analysis of observation error. At the same time, the root locus diagram proves that NESO contributes to the stability of the current loop. Finally, hardware and simulation experiments proved that the robust MPCC based on NESO has lower harmonic content and better dynamic performance.