Adaptive Neural Network Control for Permanent Magnet Synchronous Linear Motor with State Constraints and Input Nonlinearities

An adaptive controller based on neural network is proposed for the problems of model uncertainty, state constraints and input nonlinearities (such as nonlinear electromagnetic drive force / input limitation) in permanent magnet synchronous linear motor(PMSLM) . Specifically, in order to reduce the noise sensitivity and further improve the tracking accuracy, the desired signal that only depends on the reference trajectory is used to replace the measurement signal. Then, the neural network is designed to approach the unknown model and the nonlinear function online, and the approximation error is processed by constructing continuous control. In addition, an obstacle Lyapunov function is constructed to ensure that the state of the system always satisfies the constraints during the operation process; through strict theoretical analysis, it is proved that the tracking performance meets the requirements. Finally, simulation experiments verify the effectiveness and robustness of the proposed controller. © 2023 Northeastern University. All rights reserved.