The Prediction Of Masteel CSP Hut Rolled Strip Based On Data Mining

In this paper, performance prediction, the CSP technology and data mining techniqueshave been studied with the aim of finding an efficient data mining algorithms to study the largeamounts of data accumulated in the steel production process, and through the relations of thecontent of chemical elements and process parameters and final mechanical properties of stripproducts to establish a predictive model, which control and forecast the product process and toprovide guidance on the optimization of production process.This paper selects CSP hot rolled steel plates Q235as research subjects, respectively, fromits rolling, cooling, and coiling the production process to extract data to establish the appropriatedata mart,and on this basis, the use of SPSS and Matlab to establish the regression predictionmodel and BP neural network model.And the following results:the content of carbon,manganese, silicon,open rolling temperature,finishing rolling temperature, cooling startingtemperature,cooling finish temperature,coiling temperature,the original thickness and thefinished thickness are the factors of affecting the final mechanical properties ofthe Q235steel,and they are used as the input parameters of the model; the output parameters ofthe model is the tensile strength,yield strength and elongation (tensile strength is thestudy); BP neural network prediction model because of its strong nonlinear mapping ability,theywere able to achieve the desired requirements,the BP neural network model able to predict thetensile strength of error control within20MPA and98%of the predicted sample error rate ofless than5%, better than the regression model is able to predict the mechanical properties ofsteel materials such as tensile strength,etc.,nd its high accuracy, with certain promotionalvalueï¼›With the use of the trained BP neural network to research the effect of content ofchemical elements and process parameters on the tensile strength,and found that the content ofchemical elements has a positive effect on the tensile strength,and the process parameters have anegative effect on the tensile strength,the law can provide guidance on the theMasteel Q235steel production process optimization.