The Application Of AFS Fuzzy Classification Method In Laminar Cooling Process On The Runout Table

In order to improve the metallurgical properties of the hot rolled slab, manufacturers usually use the laminar cooling system on the runout table to cool the slab from the austenitic finishing temperature down to the ferrite cooling temperature according to the slab thickness and the produced steel grades. Since the laminar cooling system has synthetic complexities such as highly nonlinear, time-varying parameters and variations of operating conditions, which result in that the traditional feedback control method can not be used directly, making it difficult to process modeling and control effectively. Laminar cooling process for hot rolled strip is the last working procedure for finish product, so it is critical to the microstructure and properties of the strip.This research was a part of the 973 project "Research on theory and algorithm of real-time intelligent control for complex industrial processes"(No.2002CB312201-06) supported by the China National Key Basic Research and Development Program, In the framework of AFS theory, research on key issues in the modeling and control of the laminar cooling process on the runout table is conducted, and two new approaches are proposed in this thesis.1 In the frame of AFS theory, a survey of the AFS theory's application in the extraction of fuzzy rule is given. First, the approach calculate fuzzy closeness degrees between input features and evaluation feature, and reduct redundant features, fuzzy logic cluster in the frame of AFS theory, converse extraction of fuzzy rule; then reduction of fuzzy rule and get the better fuzzy rule's set .The extraction of fuzzy rule based on AFS fuzzy theory is used in the feedback controller of laminar cooling process on the runout table and gets the primary result by matlab.2 On the basis of AFS theory and closeness degrees, comprehensive evaluation matrix is constructed and weighting set of fuzzy comprehensive evaluation method is optimized. The method of fuzzy comprehensive evaluation is used in the feedback and feedforward controller of laminar cooling process on the runout table, and the paper also gives the result of the simulink result. In the end of the paper, the simulation of simulation platform for optimal control of laminar cooling system is given: not only reduce the difference in temperature, but also has the better control effect.