Stribeck Model-Based Cascade-Free MPC with Disturbance Observer for PMSM Servo System

In permanent magnet synchronous motor (PMSM) servo systems, the traditional cascade control exhibits compromised dynamic tracking accuracy under friction and disturbance, and requires complex tuning. In response, this paper proposes a composite control strategy based on disturbance compensation. First, a nonlinear PMSM system model is established by combining the Stribeck friction model and a lumped disturbance containing friction modeling error. Subsequently, an enhanced firefly algorithm is employed to identify the static parameters of the Stribeck model globally. Second, a generalized control input decoupling strategy is introduced to eliminate nonlinear variable coupling, while a single-loop incremental predictive control architecture is developed. Finally, a nonlinear disturbance observer is designed to provide real-time estimation of lumped disturbances, which are then incorporated into the PMSM prediction model through a feedforward compensation mechanism. Simulation results demonstrate that the proposed composite control strategy exhibits faster transient response and enhanced disturbance rejection capabilities than cascade control.