Study On Optimization Of Flatness Control Models For Six High Tandem Cold Rolling Mill

This paper was based on the strip shape optimization project of 1720mm tandem cold rolling mill(TCM) in Lianyuan Iron & Steel Corporation and the 1800mm TCM project of Tangshan Iron & Steel Corporation(TISCO). Flatness control systems of six high TCM have been thoroughly studied. Aiming at the defects of the two flatness control systems, shape preset model of thin strip cold rolling, flatness feedforward control model and flatness feedback control model which base on optimization theory have been built. Good flatness control ability and reliable control stability of these models were proven by industry experiments and online application. The main works are shown as follows:(1) Flatness control systems of 1720mm TCM in Lianyuan Iron & Steel Corporation and 1800mm TCM in TISCO have been studied and analyzed. Defects of the two flatness control systems were pointed out.(2) Rolls elastic deformation model of six high mill was developed by using modified influence function method. In this model, the effects of work roll edge contact out of the strip width on work roll’s load distribution and deformation have been considered in rolling thin strip. Work roll thermal camber has been calculated based on the measured temperature of roll’s surface. The total rolls deformation model of six high mill was developed by combining rolls elastic deformation model and work roll thermal camber model. The results of calculated data were in good agreement with the measured value. This verified that the theory model is reliable.(3) Flatness control preset model of thin strip was put forward. The relationship between optimum intermediate roll position and strip width, specific rolling force, work roll diameter and work roll crown has been discussed, and the intermediate roll position preset model was set up. The effects of strip width, specific rolling force, intermediate roll position, entry strip thickness, strip crown, roll diameter and roll crown on optimum bending force have been analyzed, and the bending force preset model was built up. (4) Effects of main factors on the gains in global flatness feedforward control model were discussed. Multidimensional lookup table that considered strip width, intermediate roll position, roll diameter and roll crown was created after many gain values of difference rolling conditions have been calculated. In practical application, linear interpolation method was employed to give the gain values.(5) An optimum flatness feedforward control model was put forward which intergrated work roll bender and intermediate roll bender to eliminate roll load disturbances. The roll load disturbance was transformed to the transverse specific force distribution, different regulate effects on specific force distribution in roll gap of work roll beneder and intermediate roll bender have been considered, and optimum corrections of the bending actuators were calculated by using a global approach.(6) A flatness feedback control model base on multivariable optimization theory and an emulsion multizone roll cooling model which using dichotomy method were built up. Flatness error was transformed to the variation of specific force distribution in roll gap, The flatness control cost function was solved by Frank-Wolfe algorithm and then the optimum corrections in flatness control actuators’regulatory range could be got. In this model, the flatness control abilities and effects among the flatness control actuators have been taken into account. In the emulsion multizone roll cooling control model, the cooling sprays control state decided by the sprays open ratio after the residual flatness error has been sorted in descending order by employing dichotomy method.(7) Industry experiments were carried out to validate and analyse the flatness control results of these models that have been built up in this paper. The industry experiments and practical application results showed that the precision of shape preset control was enhanced, the flatness control effects of feedforward control and feedback control were improved, and good strip shape was obtained by applying these models.The present paper was focused on the optimization of flatness control models for six high TCM. It can provide great help in theory and practical application to the development of flatness control system of six high TCM.