New hybrid model predicting solution to final sulfur content

Desulfurization of molten steel is a primary task in ferrous metallurgy. So the prediction of final sulfur content is an important step. According to the problem that some key parameters in prediction process are hardly to be obtained, this paper proposed a hybrid method. In this method, first, use the method which integrates AdaBoost and LS-SVM to obtain the key parameters. Then put them into the mechanism model and get the values of final sulfur content. Thus, the problem of parameters can be solved by it. At the same time, it can overcome the disadvantage that intelligent method depends on data lack of technical guidance. From the simulation results, this method can meet the production requirement; the hit frequency had reached 80%.