Research On Decoulping Decentralized Control For The Speed Tension System Of Reversible Cold Strip Rolling Mill

Reversible cold strip rolling mill is a special equipment for producing ordinary carbon steel,low alloy steel and other strip products.In the actual rolling production process,the left coiler,the main mill and the right coiler of the reversing cold strip mill are flexibly connected through the strip.The speed and tension system of reversible cold strip rolling mill is multivariable,nonlinear,strongly coupled.uncertain and slow time-varying characteristics,together with constraints such as maximum acceleration and deceleration of the main drive and strong interference of heavy loads during the rolling process,pose new challenges to the analysis and control of the rolling mill system.This project focuses on the decoupling and decentralized control of the speed and tension system of reversible cold strip rolling mill.The main research work is as follows:First of all,combining with the rolling production process of cold rolled strip,based on Hooke's law,relevant rolling theory of cold rolled strip,motor dynamic equation,rigid body rotation law and other theoretical knowledge,the mechanism model of the speed tension system is deduced.Considering the perturbation of the friction coefficient and the armature resistance in the speed subsystem of the main rolling mill and the perturbation of the rolling moment,an adaptive robust control method based on the global integral sliding mode surface is proposed to effectively improve the tracking control accuracy of the system.Secondly,based on the diagonal matrix decoupling network and backstepping method,the static decoupling and decentralized control of the speed tension system is studied.The static decoupling of the speed tension system is realized by the diagonal matrix decoupling network;the controllers of the various subsystems are designed by combining the backstepping control and the command filter,and the neural network disturbance observer is constructed to observe uncertain items of the system,which effectively weakens the coupling between the speed and tension and enhances the robust stability of system.Then,based on DFL theory and dynamic surface backstepping method,dynamic decoupling and decentralized control of the speed tension system is studied.The DFL theory is used to realize the dynamic decoupling and linearization of the speed tension system.Combine backstepping control and dynamic surface control to complete the design of the decoupled linear subsystem controllers,and use fuzzy adaptive method to estimate the matching uncertainties in the designed controller,which improve the dynamic and static performance of the system effectively.Finally,the coupling term in the speed tension system is regarded as an external disturbance,and a control method based on nonlinear disturbance observer,dynamic surface backstepping control and neural network adaptive approximation is proposed.This method constructs a nonlinear disturbance observer to observe the unmatched uncertainties of the system,and combines the backstepping control with the dynamic surface control to complete the design of the speed and tension system controller.Neural network adaptive method is used to approximate the matching uncertain items of the system,and the estimated value of the output is introduced into the designed controller for compensation.The speed tension system realizes effective decoupling and decentralized control.