| The rising demands from such as auto industry, household electrical appliance industry and so on for the quality of cold rolled strip increasingly demands good strip flatness. The quality of flatness is becoming much more important and has been evaluated as one of the most important targets for cold rolled strip. Therefore, the study of flatness control for clod rolled strip has always been the focus in the cold rolling automation. For purpose of the development of flatness control system with high precision and can meet the requirement of flatness control for cold rolled strip, corresponding studies have been made in this paper. The studies mainly consists of flatness measurement, flatness preset system, closed-loop feedback flatness control, feed forward flatness control as well as crucial models for closed-loop feedback flatness control. The developed flatness control system has been applied to a1250mm6-H reversible UCM cold mill. Applications show that the flatness control system is capable to obtain good flatness. The main work is illuminated as follows:(1) The characteristics and configuration of the ABB magnetoelastic shapemeter roll and the BFI piezoelectric shapemeter roll which represented the mainstream application has been studied. The measured flatness with different wrap angle has been calculated. Meanwhile, in order to obtain accurate flatness measurement, corresponding method and models have been developed and validated by rolling applications. The developed models for flatness measurement mainly consist of calibration and filter calculation of measured flatness signal, thickness distribution across strip, the correction of measured radial force in edge of strip, the interpolation of radial force in dummy segments.(2) The priorities for different flatness actuators have been set according to their capacity on flatness control and dynamic characteristics in rolling process. Moreover, the corresponding preset calculation process of the flatness actuators with the aid of influence function has been described. The flatness target curve based on crown control, flatness control as well as requirements of subsequent processing has been developed. Dynamic compensation models such as coil shape, geometric error of shapemeter roll installation position, and temperature difference across strip, thickness drop in strip edge have been developed in accordance with the rolling process and equipment condition. At the same time, additional manual curves have been set to increase flexibility to flatness control. On this basis, the flatness target curve setting software has been developed combined with the actual production condition, and the models have been validated by actual rolling processes.(3) Based on efficiency factors of flatness actuators, the optimal multi-variable flatness control algorithm with the aid of constrained least square method has developed. Meanwhile, a self-learning model of actuator efficiency factors has been established to obtain accurate efficiency factors matrixes of flatness actuators. With the constrained least square method, the flatness control system calculates the adjustments of the actuators dependent on flatness deviation in the relay manner. Subsequently, the adjustments will be limited to their adjustable range and compensated by dynamic gains. In addition, the influence of measurement delay on control system has been analyzed to find a method to compensate this influence resulting from measurement delay. The Smith predictor has been introduced to the flatness control system for delay compensation. Based on the compensation method, the flatness system has been developed and simulated. Two different modes of closed-loop flatness control have been developed by analysis of the simulation results. The developed control modes are dependent on rolling speed and both of them have two loops. One loop named the optimal control algorithm is to calculate the adjustments of actuators during each control cycle. It severs as the outer loop for both of the two control modes. During high rolling speed, the inner loop will be designed with a conventional proportional-integral-derivative (PID) controller, named high speed control mode. During low rolling speed, it will be switched to a PID controller integrated with the Smith predictor, named low speed control mode.(4) Efficiencies of work roll bending, intermediate roll bending as well as rolling force have been analyzed to establish the scheme of feed forward flatness control. The developed feed forward flatness control system uses work roll bending and intermediate roll bending to compensate the influence of rolling force fluctuation. Based on the efficiencies of the three actuators, the optimal algorithm for feed forward flatness control has been developed with the aid of constrained least square method.(5) A series of crucial models for closed-loop feedback flatness control system have been developed according to the theoretical analysis and actual rolling experiments and validated in actual rolling processes. The developed models includes shifting speed control of intermediate roll, substitution control of work roll bending, asymmetrical roll bending control as well as fuzzy selective work roll cooling control.(6) Hardware platform of the flatness control system has been configured and corresponding software has been developed. Based on the configured platform and developed software, models and schemes of flatness control have been tested offline. In order to test the actual result, flatness of both normal thickness gauge and ultra thin strips during rolling processes has been evaluated. Dynamic responses and adjustments during each control cycle of flatness actuators as well as the resulting flatness deviation change have been analyzed.(7) The results of the study are practical for flatness control of cold rolling mill. Currently, flatness control system for cold rolling mill has been developed and applied to the revamp of a1250mm6-H reversible UCM cold mill, based on the strategies and models. Applications show that the study has certain significance to the flatness control system revamp of existing mill and the flatness control system development of upcoming new mill. |
PDF Full Text Request