Integrated Input Training Neural Network PCA and RBF for Chemical Process Modelling

Many applications of Principal Component Analysis (PCA) can be found in dimensionality reduction. But linear PCA method is not well suitable for nonlinear chemical process. A nonlinear PCA (NLPCA) method based on input training neural network (ITNN) is proposed for the nonlinear system modeling. Contrasting to the auto-associative neural network (ANN), ITNN has less hidden layers and faster training speed. Moreover, The ITNN is combined with RBF Neural Network (RBFNN) to model the yields of ethylene and propylene in the naphtha pyrolysis system. From the practical application, ITNN-PCA combined with RBFNN is an effective method of nonlinear chemical process modeling.