Fuzzy Approximation-Based Model-Free Predictive Control for Permanent Magnet Synchronous Motor Drives

Conventional model predictive control (MPC) is highly vulnerable to motor parameter variations. Meanwhile, existing parameter-based MPC schemes are often constrained by the accuracy of model reconstruction. To overcome these limitations, this article proposes a model-free predictive control (MFPC) strategy based on a fuzzy approximation method for a permanent magnet synchronous motor (PMSM). Leveraging the exceptional nonlinear mapping capability of fuzzy approximation, the proposed strategy approximates the autoregressive term within a structurally simple first-order autoregressive model with exogenous input (ARX). This significantly enhances model reconstruction accuracy. Furthermore, discrete-time Lyapunov stability analysis rigorously demonstrates that the estimation errors of the internal states under the proposed control scheme are uniformly ultimately bounded (UUB). Finally, experimental results reveal that the proposed MFPC strategy achieves superior steady-state current quality while ensuring excellent dynamic performance, effectively validating the advantages of the proposed method.