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

As the electric arc furnace (EAF)operating conditions is difficult to directly measured, resulting in the EAF not easly to determine the various operating conditions. In view of these problems, based on the characteristics of the EAF ferroalloy smelting, this method about the use of multi-scale decomposition of the nuclear functions of the least square support vector machines (LS-SVM) is proposed to predict the operating conditions of the EAF, this method can keep the general approximation ability of curve, meanwhile it improves the approximation ability in the local area model. The experiment results show that it has superior characteristics in the respect of prediction precision and prediction speed and the operating conditions can be predicted accurately and timely, both being economic of production cost and increasing.