Model Predictive Position Control of Tubular Permanent Magnet Linear Synchronous Motor for Precision Positioning Based on Neural Network Model Reference Adaptive Disturbance Observer

To improve the dynamic performance of position tracking in permanent magnet synchronous linear motors, a model predictive position control method based on disturbance observer is proposed. Firstly, a novel neural network enhanced model reference adaptive observer is designed to estimate the lumped disturbance of the system. Taking the estimated disturbance as a new state variable, it is explicitly embedded in the framework of model prediction, which realizes the online estimation and compensation of disturbance, and effectively solves the deterioration of control performance caused by inaccurate system parameters and unknown disturbance in model prediction method. The increment of the control input is used as the input of the prediction equation, which makes the control input smoother and avoids drastic changes. The adaptive gain of the observer is designed by Lyapunov theory and the stability of the system is analyzed. A large number of experiments and analysis are carried out on the tubular permanent magnet linear synchronous motor platform, which proves the effectiveness of the proposed method.