Research On Coiling Tension Control System Based On BP Neural Network

With the continuous development of the steel industry, the demand of hot rolled strip is increasing, tension control directly impacts on product’s quality, so the coiler tension control system is now the focus of the researched at home and aboard Because coiling tension control system is a complex, nonlinear and time-varying system, making the coiling tension control system has become a complicated control problem. This paper will studies this system. The main contents are as follows:Firstly, the paper puts forward background, purpose and significance, then the history and status quo are reviewed, some intelligent control methods used in the coiling tension control system are analyzed in recent years and the paper presents the development tendency of neural network applied in this system.Secondly, elaborates the coiling tension control system equipments and processes. By analyzing the mode of coiling tension control, the composite tension control mode is selected. Through analysis of winding characteristics, it shows that tension has relationship with speed of motor. Then study the torque of coiling tension control system, tension also relates the magnetic field flux and armature current, using the maximum torque method to keep tension stable. Through detailed study of the principles of coiling tension control system, the three-loop control system mode is determined. Through the analysis of tension part and DC motor, designed the current, speed and tension controllers and established the coiling tension control system model, then analyzed the inertia torque and friction torque compensation and roll diameter calculation methods.At last, because roll diameter of control tension system is increasing, cause the flywheel torque increases, so the motor electrical constants will changes. For coiler tension system complexity, nonlinear and time-varying characteristics, study the BP neural network structures and algorithms and use coiling tension control system PID control and BP neural network PID control to simulation. Verification of the BP neural network PID control feasibility. The results also show that BP neural network PID control system reduces the overshoots, improves anti-jamming performance. Therefore, it is suitable for coiling tension control system.