Servo control method based on neural network and disturbance observation

In this paper, the method of improving the performance of permanent magnet synchronous motor in the presence of disturbance and friction is studied. First, collected data are used to train BP neural network to get an accurate friction model. Friction model is used to compensate the friction. Considering the influence of friction over-compensation or less-compensation and external disturbance, the disturbance observer is used to compensate the disturbance. Finally, the simulation analysis of the proposed compensation method shows that the proposed method based on the neural network and the disturbance observer can improve the position and velocity tracking accuracy.