The High- Precision Optimized Adaptive Neural Backstepping Control for PMLSM under High-Amplitude Trajectories and Severe Load Shocks

Abstract

This paper proposes an adaptive neural backstepping control strategy for Permanent Magnet Linear Synchronous Motors (PMLSMs) operating under high-amplitude trajectories and severe external disturbances. The control scheme integrates a nonlinear backstepping framework with a Radial Basis Function Neural Network (RBFNN) to online approximate lumped uncertainties and load variations. To ensure smooth disturbance estimation and avoid high-frequency oscillations, the neural parameters are systematically optimized, and a σ-leakage mechanism is employed to prevent neural weight drift. Lyapunov-based analysis guarantees Uniformly Ultimately Bounded (UUB) stability of the closed-loop system. MATLAB simulation results demonstrate micrometer-level tracking accuracy and rapid disturbance rejection under a 0.4 m reference trajectory and a 25 N load disturbance, confirming the robustness and effectiveness of the proposed approach for high-precision motion control applications.