Extended-control-set model-free predictive current control for surface-mounted permanent magnet synchronous motors

Model predictive current control (MPCC) has been widely used in the high-performance control of permanent magnet synchronous motors (PMSMs). However, MPCC exhibits low robustness to motor parameter variations, while the conventional finite-control-set MPCC (FCS-MPCC) based on 8 voltage vectors always leads to large steady-state current fluctuations. To solve these problems, this article proposes an extended-control-set model-free predictive current control (ECS-MFPCC) algorithm based on an ultralocal model (ULM). The proposed scheme uses an extended state observer (ESO) to observe lumped perturbations, and does not depend on accurate motor parameters, resulting in very strong robustness. Additionally, utilizing virtual vector synthesis theory, the number of voltage vectors in the finite-control-set is extended from 8 to 44, which can significantly reduce the current tracking errors and fluctuations without increasing the computational complexity due to the target-sector pre-selecting technique. Moreover, a voltage correction technique is presented for the ESO to minimize the observation error of lumped perturbations. Finally, the effectiveness of the proposed method is verified through simulation and experimental results.