An Internal Model Control Method Based on Reversible Residual Neural Network Model Applied to Turntable Servo Systems

To meet the special requirements of high dynamic and performance for turntable servo systems, an internal model control (IMC) method based on reversible residual neural network model is proposed in this paper. In order to improve the accuracy and modeling efficiency of the forward and inverse models in the internal model controller and ensure the correlation between these two models, a reversible residual neural network model based on the forward and inverse dynamic characteristics of turntable servo systems is proposed. Through the special design of the network structure, the forward and inverse models can be obtained by one-way training. Then, the forward and inverse models are applied to the traditional IMC algorithm framework, the IMC method based on the reversible network model is obtained, and the stability of the control system is proved. The simulation results show that the proposed IMC method has better nonlinear compensation ability, tracking effect, and dynamic performance than the traditional IMC method. In addition, the design process of the proposed internal model controller is simpler and more practical. The method described in this paper has important value and significance for the intelligent development of traditional servo control.