Chaotic time series forecasting based on SVM for silicon content in hot metal

A chaotic time series forecasting model based on support vector machine(SVM) for silicon content in hot metal is proposed which combines the support vector machine and chaotic forecasting theory. The original silicon content time series is reconstructed to a high dimension space through the skills of state space reconstruction. The training sample and testing sample are obtained based on the states in the state space, and then the support vector machine theory is used for forecasting. The simulation results show that the proposed model has better curve fitting and higher forecasting accuracy compared to that of RBF, AOLM and Volterra adaptive model. The hit rate reaches 88% in successive 100 heats in test set in the range of [Si] 0.1%. It seems promising and determinant in providing the experts with the right tools for the prediction in this difficult problem, and it can satisfy the requirements of on-line prediction of silicon content in hot metal. It develops the theory and method for silicon content forecasting in hot metal.