Neuro-Fuzzy Control Of Thickness For5-Stand Mill

The accuracy of thickness is one the most important quality. Bad accuracy of thickness can cause a bad stamping performance of cold rolling strip. And there are strict requirements of stamping performance for auto sheet and household appliances sheet. So, the improvement of accuracy of strip thickness in deep processing of cold rolling strip is urgent. Atfer many years’ development, the Automatic Gauge Control (AGC) system has been developped, but the accuracy of thickness is always the subject to improve.In this paper, though the study of control system of pickling line tandem cold mill with six rolls in one of thefive stands which was introduced by Tianjin Ansteel Tiantie Cold Rolling Sheet Ltd.,Co. Composition and functions were analyzed. Design concept of AGC system’s program and the SIROLL solution based on SIEMENS all in one automation platform has been mastered. In addition, a Neuro-Fuzzy Controller with an integrator (FNNC-I) was designed to improve the accuracy of thickness.First of all, the base of the thickness control system was mainly introduced. After introduced the elastic equation of rolling mill, stiffness coefficient of stand, control mode of thickness control system, hydraulic gap control and continuous variable crown, the reasons cause variation in thickness and compensations for variation in thickness variation in thickness were analyzed emphatically.After that, the theory of neural network and fuzzy Control, the design of FNNC-I controller were demonstrated. And the FNNC-I controller was applied to thickness control of tandem cold mill. Because of nonlinear, large time delay characteristics of tandem cold mill thickness control, it is difficult to control thickness within a small tolerance for PID controller, whose parameters are set only for one stable situation. According to these characteristics, based on the analysis of thickness control theory, a fuzzy neural network controller (FNNC) with simple structure was designed, which is realized by a BP network with2-5-1structure parallelled with a integrator. The simulation results show that the performance of system was improved and it was better than that of PID controller.Finally, the realisation of SIROLL solution based on SIEMENS PCS7to tandem cold mill was introduced. It include hardware configuration of controller, programming, cofiguration of communication, configuration of HMI and so on. Eventially the TDC(Technology and Drive Control) control system based on PROFIBUS field bus and industrial ethernet was nrealized. It met the command of cold rolling process control with real-time,security and stability and built a solid fundation for the further research for cold rolling thickness control system.