Hardenability prediction of gear steel in refining process

Hardenability prediction is very difficult in the steel refining process. Based on the idea that the accuracy of model can be significantly improved by combining several sub-models, a multiple support vector machine(MSVM) based hardenability prediction model is proposed in this paper. The influence factors of hardenability are analysised to determines the number of sub-model and the input variables of the sub-model. In order to improve the precision and generalization capability of the prediction model, genetic algorithm (GA) is adopted to optimize the parameters of MSVM. The simulation results demonstrate the efficiency of the method.