Abstract: In response to the demand for wide speed domain operation of permanent magnet synchronous motors （PMSM） in rail transit traction systems, an improved model predictive weak field control （MPFWC） strategy with full speed domain partitioning is proposed, which significantly improves the control accuracy and dynamic stability of the system in a wide speed domain. The incremental prediction equation combined with delay compensation technology effectively suppresses the impact of permanent magnet flux attenuation caused by high-temperature demagnetization on control performance. The optimization strategy of PI controller integrating fuzzy logic, which dynamically optimizes PI parameters through fuzzy adaptive adjustment, significantly improves the torque output accuracy and anti-interference performance of the system in a wide speed range. The simulation results show that for the torque ripple performance in the power reduction region, the torque ripple amplitude is suppressed to 0.35 N·m, and the current fluctuation amplitude is reduced by 40%. At the same time, the fuzzy PI controller reduces the speed overshoot from 5% to 2.5% and shortens the steady-state response time by 10%. The research results provide new theoretical basis and technical implementation path for the design of high reliability rail transit traction systems.