The prediction model of silicon content in hot metal based on LS-SVR optimized by estimation distributed algorithm

Accurate prediction of silicon content in hot metal is very helpful for operation of blast furnace. A prediction model of silicon content in hot metal based on least square support vector regression (LS-SVR) is proposed in this paper. As the parameters of LS-SVR have great impact on prediction results, an estimation of distribution algorithm (EDA) is presented to optimize the parameters. The verifying result of practical data shows that the proposed algorithm can optimize LS-SVM parameters, which makes the prediction model has good efficiency.