Dual Closed-Loop Model-Free Adaptive Predictive Current Control for Permanent Magnet Synchronous Motors

For permanent magnet synchronous motor (PMSM) systems exhibiting strong coupling, parameter uncertainties, and dynamic load disturbances, this paper proposes a Model-Free Adaptive Control with Multi-Vector Predictive Current Control (MFAC-MPCC) framework, integrating Compact Form Dynamic Linearization MFAC (CFDL-MFAC) for the speed loop and triple-vector model predictive current control (MPCC) for the current loop. The hybrid architecture decouples adaptive speed regulation from high-precision current tracking, achieving robust performance without reliance on motor parameters. Key advantages include: (1) Model-free operation via CFDL-MFAC, dynamically linearizing the speed loop using only I/O data; (2) Enhanced disturbance rejection through real-time adaptive tuning; (3) Superior current tracking via triple-vector MPCC, minimizing steady-state errors and harmonic distortions. Simulation comparisons with traditional PI control and CFDL-MFAC approach validate the effectiveness and superiority of the proposed MFAC-MPCC scheme.