Thickness-tension Unsteady State On-line Prediction And Coordinative Optimization In Tandem Cold Rolling

Rolling process parameters are continuous changingin the unsteady state rolling process.The acceleration and deceleration rolling is a typical unsteady state rolling.In this paper,in order to deal with the problem of thickness and tension control in acceleration and deceleration process during cold rolling,the rolling parameters online prediction and thickness-tension multivariable optimization control strategy were investigated.The main research contents are as follows:(1)The main parameters of the deformation zone in cold rolling and their calculation methods were analyzed.The rolling force model,deformation resistance model,friction coefficient model and forward slip model which were suitable for online application were given.And the mass flow equation,mill spring equation,tension equation and other basic equations of cold rolling process were studied.(2)Using the analysis method of support vector machine regression in machine learning algorithm,the prediction accuracy of conventional SVM and PSO-SVM with different parameter optimization principles was compared.Based on the mathematical model of rolling theory,a simulation model of thickness and tension deviation about the change of roll gap and speed ratio during the process of acceleration and deceleration was established.The regression results of different speed segments were analyzed and the corresponding regression coefficients were calculated.(3)Considering the coupling relationship between exit thickness and interstand tension as well as the influence of interference,a single stand incremental model was deduced,which mainly included the rolling force increment equation,the thickness increment equation,the forward slip rate increment equation and the interstand tension increment equation,etc.,and the inertia segment of actuator was given.Based on the derived incremental equations,the coefficient matrix of the fifth stand state space expression was put forward,and the model validation was done..(4)According to ILQ theory and the state space model previous determined,a thickness-tension control system based on the coordinated and optimization control of roll gap and roll speed was established.The traditional PI control model and the ILQ optimal control model are respectively established.The control effects following the change of setting are analyzed.The anti-interference ability of the controller was tested by the change of roll speed,roll gap and roll eccentricity.The control effects were compared when model mismatched.Compared with the traditional control system,the thickness-tension coordination and optimization control system based on ILQ theory has better control effect and anti-interference ability.