Multi-Kernel Neural Network Sliding Mode Control for Permanent Magnet Linear Synchronous Motors

In this paper, a multi-kernel neural network sliding mode control (MNNSMC) method is proposed to enhance the position tracking performance and disturbance suppression capability of permanent magnet linear synchronous motors (PMLSMs). The designed MNNSMC strategy consists two parts, with dynamic boundary layer (MNN). former, utilized guarantee that variable converges manifold asymptotically. latter, disturbances are introduced into design kernel functions further approximate compensate for internal external disturbances, such as parameter variations, positioning force, friction, un-modeled nonlinearities PMLSMs. stability analyzed proved based on Lyapunov Theorem. Experiments conducted PMLSM servo drive system validate effectiveness presented method, results show scheme has significant improvement performance, capability, robustness compared an existing method.