Research On Modeling And Decoupling Control Of Hydraulic Looper System In Hot Strip Mill

Aiming at the problem that traditional decoupling controller design relies on accurate mathematical model,this paper designs two kinds of decoupling control systems with good robustness,and carries out simulation research combined with linear and nonlinear models.Based on the physical relationship of the working state of the hydraulic looper,the linear coupling model of the hydraulic looper system is obtained by combining the relationship of the tension variation and the conservation of the angular momentum of the output torque of the hydraulic cylinder.According to strictbeck friction equation,a nonlinear model of hydraulic looper system considering friction factor is established.The neural network is used to establish the non analytical inverse model,which solves the problem of poor robustness of the analytical inverse controller and reduces the dependence of the inverse control on the mathematical model.The results show that the neural network ?-order inverse decoupling control system weakens the coupling of the system,enhances its anti-interference ability compared with the diagonal decoupling control,but the system response time is long.The simulation results based on the nonlinear model show that the neural network inverse system can eliminate the nonlinear coupling term,effectively reduce the coupling between looper height and strip tension,and has good robustness.The nonlinear function of active disturbance rejection controller is improved,and the improved active disturbance rejection control system is designed by using this function,which is applied to the hydraulic looper control system.Simulation results based on linear model unit step and process parameter perturbation show that the control algorithm eliminates the coupling effect between hydraulic looper height and strip tension,and speeds up the response speed of the system.Based on the nonlinear system model,the simulation results show that the improved ADRC system solves the coupling of hydraulic looper system better than the neural network ?-order inverse decoupling control system,and is more suitable for this kind of strong coupling and nonlinear controlled object.Figure 40;Table 5;Reference 50...