The endpoint prediction of electric arc furnace based on least squares support vector machines

In order to overcome the problem that the least square support vector machines (LS-SVM) using Gaussian kernel cannot approximate arbitrary signal with multi-scale, a scaling ker-nel for LS-SVM is proposed. LS-SVM can be used simultaneously to approximate to the target function and improve the effectiveness of generalization and approximation in the local area model. The LS-SVM with scaling kernel can approximate arbitrary signal with multi-scale, and the proposed algorithm is promising in application since only one free parameter is adjusted for optimization. Based on the characteristics of electric arc furnace(EAF) smelting, this method about the use of multi-scale decomposition of the nuclear functions of the LS-SVM is proposed to predict the endpoint of EAF,test results of an actual power system show that it has better local approximation and generalization capabilities when appropriate numbers and parameters of the LS-SVM are chosen.