Research Of Flatness Characterization Method And System Lag Compensation For Cold Rolled Strip

As one of the most important quality indexes of cold rolled strip,the related technology of flatness has always been a difficult and hot spot in the field of cold rolling.Since flatness control is a complex and comprehensive subject,this paper only focuses on the current flatness characterization and evaluation methods and the time-delay compensation of flatness control system.Based on the flatness control process of a 1450 mm 5-H reversible UCM cold mill,the following two aspects of research are carried out by combining theoretical research,experimental verification and industrial practice:(1)On-line visualized characterization and evaluation method of strip flatnessThe method of parabolic blending curve is used to interpolate and fit the strip flatness measurements along the width direction,and the sonogram relationship between the expanded strip flatness data and the color values is established,and based on the actual distribution of flatness defects in the width direction of cold rolled strip,the drawing model of flatness sonogram of cold rolled strip is established.In order to verify the accuracy of the model,a comparative analysis and model optimization were carried out based on actual production data and physical flatness distribution.On this basis,the strip flatness sonogram control is developed and integrated into the strip flatness control system of 1450 mm 5-H reversible UCM cold mill.The real-time drawing of strip flatness sonogram can reflect the details of strip flatness distribution and its changing trend in any area of the whole roll,and realize on-line fine monitoring of strip flatness quality.(2)Predictive Compensation Control Strategy for Flatness Delay and Optimal Tuning Method of Predictive Model ParametersSmith predictor is introduced into the modeling process of flatness control system,and a compensation control model of flatness lag based on self-tuning Smith predictor is established.The effects of Smith predictive PID control and conventional PID control on dynamic performance and robustness of the system under different influence factors such as time lag parameters and model parameters are compared and analyzed,and the corresponding control strategies are put forward.To solve the mismatch problem of prediction model,a self-tuning regulator based on minimum variance is formed by combining minimum variance control with recursive least squares estimation,which can be used to modify system parameters on-line in real time.In addition,considering the advantages of Smith predictor and fuzzy control,a model of flatness lag compensation control based on Smith predictor-fuzzy PID is established by combining Smith predictive PID control with fuzzy control,which provides a theoretical basis for further improving the robustness and adaptive ability of flatness closed-loop control system.