Research On Intelligent Methods Of Flatness Recognition · Prediction And Control Simulation

Flatness is the important quality indexes of rolled plate and strip, and flatness control is the key technology of modern times high precise rolling mill. As far as the technique is concerned, the key scientific problem in the world is pattern recognition and the realization of shape intelligent control. Flatness pattern recognition is the key technology of shape control. A new flatness pattern recognition method is put forward, equivalent fuzzy neural network model, based on analyzing tradition shape pattern-recognition methods and intelligent methods. This is a new method that gets over many problems, such as, the different topology network is needed to accomplished recognition assignment when the width is changing, the large learn assignment, slow convergence and local minimal in the network, and so on. Fuzzy neural network model selects Lerande orthodoxy polynomial as the basic flatness pattern mode and it have only 3-input and 3-output. The network not only has a simple structure but also every node has definite physical significance, it has higher discriminating precision. During the rolling process, roll gap takes to involve the great non-line factors, such as bending force, rolling force, initial roll curve, fray roll curve, worn roll curve and other initial conditions, and strong coupling character, this article established the flatness intelligent prediction model based on Elman dynamic recursion network algorithms to apply flatness online prediction for enhance the steady state performance and real time performance of the system in shape control rolling process. There are some problems such as a great deal of nonlinear factor, time varying character and strong coincidence in the course of skip steel rolling. So it is difficult to set up the precise mathematic model for shape control system. This article established intelligent shape coupling control model based on PID neural network multivariable control system. Then simulate the shape intelligent control system using some real data of a certain steel company, and results indicate that shape control system has good quality. Selecting research on flatness pattern recognition method and intelligent control system for cold strip mill as research task, not only has the research important significant for developing shape control and control theory, but also has practical sense and applying value for flatness intelligent control technology, it has certain engineering guide significance.