Soft-sensing for multiple models of carbon content of fly ash based on SVM fusion theory

Because the accuracy of model could be significantly improved by combining multiple models, a support vector machine (SVM) fusion modeling approach was proposed to build the soft sensor model. The model is built based on the time series data and the sub-model is built based on least square SVM algorithm in every sub-space. In order to minimize the severe correlation among sub-models, and improve the accuracy and robustness of the model, the sub-models are combined by SVM algorithm. In view of the problems that current power plant boiler ash carbon measurement methods are time-lag and low accuracy, we build a model using the method. The procedure of simulation and theoretical analysis indicate that the proposed method is effective.