Design tension controller of unwinding system based on BP neural network

In the printing process, it mainly depends on the stability of the systematic tension control to ensure the quality of print. Meanwhile, systematic tension is mainly created from unwinding part, from which it easy to cause fluctuation and variation of tension. Hence, how to maintain the stability of the tension in unwinding system have become a key problem and a hot issue to tension control system. In this paper, a PID controller based on the BP neural network, which is used on the unwinding tension control system, is proposed, and the neural network control theory and the PID control algorithm are adapted according to the nonlinearity and time-variation of the unwinding system. Meanwhile, this controller can self-adjust PID parameters online and has a better adaptive capacity and overcome the shortcoming of the traditional PID controller which can not assure optimal control performance for system in any working condition. Finally, this kind of controller is verified to have the ideal performance by simulation.