Research On The Intelligent Control Strategy Of Flatness For Tandem Cold Mill

Flatness is an important target for quality of the strip steel. And the flatness control is the key technology of modern rolling mills with high precision of flatness. The intelligent realization of flatness pattern recognition and control technology is the leading research subject all over the world in the control of modern rolling mills. Taking the intelligent flatness control system of tandem cold rolling mill as the subject, the author has done a lot of theoretical research and simulation analysis on the flatness pattern recognition, control strategy of executing machine as well as the integrated flatness control system.The flatness pattern recognition is the key to the flatness control. Analyzing the problems existing in the traditional method and the network method in flatness pattern recognition, the author has designed a BRE-GA-BP network based on the binary encoding combined with real number encoding, which has been used in the foundation of the intelligent recognition model of the flatness. There are only 3 inputs and 3 outputs in this model, and each internal layer of the network has a definite physics meaning. The flatness pattern recognition model designed above has a quick recognition speed and high precision. It is a simple and useful method for the flatness pattern recognition.Because of the non-linear character, time-variability, strong coupling and the severe interference of the flatness executing machine, it is hard to get a good control effect using the classical control method and modern control method based on the empirical model. Thus, the author designed an IMC (internal model control) strategy based on dynamic network and minimum-variance self-turning control. This control strategy can identify the model and the inverse model of the system dynamically, so it has little dependence of the system model and can overcome the disadvantages brought by the changes of the model as well as kinds of nonlinear factors. Meanwhile, regulating function of the executing machine of the flatness control system is made full use of. And the dynamic performance of the system is improved.Taking the locale data as basis, the neural networks of flatness controller module and flatness calculating module has been founded. And based on the intelligent recognition method of flatness and the control strategy of flatness executing machine, the author has founded the integrated intelligent flatness control system of the tandem cold mill. And after some simplification, the simulation block diagram of integrated intelligent flatness control system was founded under the simulink environment in MATLAB, and the corresponding simulation analysis was done. The result of the simulation shows that the control effect of this system is very good and the precision of flatness can be controlled in the range of±5I. Besides, it can better adapt to the changes of system model and interference of the environment. In a word, it is an effective control method for the flatness of tandem cold mill.