Imperial smelting furnace fault prediction model based on hammerstein model using least squares support vector machines

In this paper, the Hammerstein fault prediction modeling based on least squares support vector machines (LS-SVM) is presented for the prediction the key parameters of the imperial smelting furnace (ISF). ISF is a nonlinear, multi-input and multi-output (MIMO) system that is difficult to model by the classical methods. Due to the particularly simple structure of the Hammerstein model and the generalization performance of LS-SVM, a Hammerstein model using LS-SVM is built and applied to the ISF. The simulation research shows this model adapts well to the change of parameters, provides accurate prediction and is with desirable application value.