Five Stand Cold Rolling Mill Thickness Control System Research And Application

Thickness accuracy is an important quality indicators of the strip products, with the rolling theory and control theory and its application in the rolling process, the strip thickness of the product accuracy has been greatly improved, But the actual system application process, there are many specific technical problems to be solved, Therefore looking for is not dependent on the system model, the control effect to meet the needs of the actual production of a new control method is necessary. The neural network method has been widely used in recent years in the field of automatic control, it does not rely on accurate system model, the neural network combining with the conventional control method is rolling hot research field of automatic control.This paper first describes the cold-rolled sheet with domestic and foreign basic profiles, combined with the needs and development of cold-rolled sheet and strip rolling technology, and analysis of its core technology-AGC (Automatci Guage Conrtol AGC), discussed the existing thickness of the basic principles of automatic control systems and control.Several typical AGC control model, a program to reduce and eliminate the thickness fluctuations.Through in-depth analysis of the new flow AGC systems, combined with the Shougang steel company moved to cold-rolled1450mm cold rolling mill of the five-rack pickling project design requirements, changes in the disturbance and control the amount of the entire plant operation state the law.Tandem cold rolling process for a comprehensive analysis of static and dynamic characteristics, respectively, five-stand tandem cold rolling static, dynamic model, the model considers the entrance incoming thickness, roll gap, roll speed, friction coefficient, the deformation resistancecold rolling mill thickness of each rack export. On this basis, the injury caused by the disturbance or control the amount of finished thickness changes in the quantitative analysis.Strip during the rolling process, there are many uncertainties and slow time-varying factors, for the same control parameters of PID control can not adapt to this change.Address this issue, the PI control method based on BP neural network, BP neural network to adjust the controller parameters online. Of the proposed control method, a simulation study. Design and implement a five-stand cold rolling mill thickness control system. Simulation and system operation results show that, through adaptive neural network to adjust the controller parameters, the thickness of the automatic control system design with satisfactory control.