Research On The Intelligent Model Of Selective Cooling Control Of Work Roll For Cold Rolling

During the cold rolling process,the thermal expansion distribution of the work rolls directly affects the shape of the roll gap,which in turn affects the flatness distribution of the strip.Therefore,the selective cooling control of the work roll is an important link in the closed-loop feedback control of shape and is a necessary means to control local sheet shape defects.Due to the time-varying,nonlinear,and strong coupling characteristics of the work roll selective cooling control system,it is difficult to establish an accurate mathematical model for the selective cooling control of the work roll.To improve the control accuracy and anti-interference ability of the work roll selective cooling control system,this paper applies intelligent control algorithms to the model for the selective cooling control of the work roll.The main contents of this paper are as follows.First of all,in view of the accuracy of the selective cooling control of the work roll,the historical data during the cold rolling process are collected,and the support vector machine regression algorithm is applied to the selective cooling control model of the work roll with the environmental factors such as the gap power,reduction rate,and coolant temperature as the input variables,so as to predict the minimum coolant output.In this process,the differential evolution grey wolf algorithm is used to optimize the parameters in the model.The simulation results show that the minimum coolant output model based on support vector machine regression has high prediction accuracy.Secondly,in view of the anti-interference ability of the control system for the selective cooling of the work roll,the control model for the selective cooling of the cold rolling work roll is established based on the fuzzy support vector machine technology with the flatness residual,the flatness residual change rate and the flatness deviation changing with space as the input variables,so as to obtain the coolant output of each nozzle.The performance comparison with the PID and fuzzy controllers shows that the controller based on the fuzzy support vector machine has a stronger anti-interference ability.Finally,the minimum coolant output model based on support vector machine regression is combined with the selective work roll cooling control model based on fuzzy support vector machine to obtain the selective work roll cooling total control model.The control effect of the selective work roll cooling total control model is verified using the selective work roll cooling control system of a 1450 mm five-stand tandem cold rolling mill in a certain steel plant as the experimental object.The experimental results indicate that the overall control model for selective work roll cooling can improve the control accuracy and anti-interference ability of the selective work roll cooling control system.