Design Controller Based on RBF Neural Network for the Rewinding Tension System

The control precision of the rewinding tension system is an important index for the roll-to-roll gravure printing machine. However, the PID control based on the common fixed parameters cannot solve the problems of strong coupling, multi-interference and time-varying parameters of the tension system, which severely limits the improvement of the rewinding tension control precision. Therefore, a PID feedforward control based on RBF neural network is proposed in this paper, to solve the key problems that restrict the improvement of control precision of the rewinding tension system. Firstly, a feedforward controller is designed for the disturbances existing in the rewinding tension system. Then, PID parameters online self-tuning method is designed based on RBF network. Finally, the performance of the designed controller is verified by simulation in MATLAB. The simulation results show that the proposed control strategy has good anti-interference and decoupling performance, and realizes the online self-tuning of PID parameters.