Dimensionality Reduction with Input Training Neural Network and Its Application in Chemical Process Modelling

• SYSTEM ENGINEERING • Previous Articles Next Articles

Dimensionality Reduction with Input Training Neural Network and Its Application in Chemical Process Modelling

ZHUQunxiong; LIChengfei

School of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China

Received:1900-01-01 Revised:1900-01-01 Online:2006-10-28 Published:2006-10-28
Contact: ZHU Qunxiong

输入训练神经网络的维数约简算法及其在化工过程建模中的应用

朱群雄; 李澄非

School of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China

通讯作者: 朱群雄

Abstract

Abstract: Many applications of principal component analysis (PCA) can be found in dimensionality reduction. But linear PCA method is not well suitable for nonlinear chemical processes. A new PCA method based on improved input training neural network (IT-NN) is proposed for the nonlinear system modelling in this paper. Momentum factor and adaptive learning rate are introduced into learning algorithm to improve the training speed of IT-NN. Contrasting to the auto-associative neural network (ANN), IT-NN has less hidden layers and higher training speed. The effectiveness is illustrated through a comparison of IT-NN with linear PCA and ANN with experiments. Moreover, the IT-NN is combined with RBF neural network (RBF-NN) to model the yields of ethylene and propyl-ene in the naphtha pyrolysis system. From the illustrative example and practical application, IT-NN combined with RBF-NN is an effective method of nonlinear chemical process modelling.

Key words: chemical process modelling, input training neural network, nonlinear principal component analysis, naphtha pyrolysis

摘要： Many applications of principal component analysis (PCA) can be found in dimensionality reduction. But linear PCA method is not well suitable for nonlinear chemical processes. A new PCA method based on improved input training neural network (IT-NN) is proposed for the nonlinear system modelling in this paper. Momentum factor and adaptive learning rate are introduced into learning algorithm to improve the training speed of IT-NN. Contrasting to the auto-associative neural network (ANN), IT-NN has less hidden layers and higher training speed. The effectiveness is illustrated through a comparison of IT-NN with linear PCA and ANN with experiments. Moreover, the IT-NN is combined with RBF neural network (RBF-NN) to model the yields of ethylene and propyl-ene in the naphtha pyrolysis system. From the illustrative example and practical application, IT-NN combined with RBF-NN is an effective method of nonlinear chemical process modelling.

关键词: chemical process modelling;input training neural network;nonlinear principal component analysis;naphtha pyrolysis

ZHUQunxiong, LIChengfei. Dimensionality Reduction with Input Training Neural Network and Its Application in Chemical Process Modelling[J]. .

朱群雄, 李澄非. 输入训练神经网络的维数约简算法及其在化工过程建模中的应用[J]. CIESC Journal.

References

1    Jackson, J.E., “Principal components and factor analysis (Ⅰ) Principal components”, J.Quality Technol., 12,  201—203(1980).
2    Dong, D., McAvoy, T.J., “Nonlinear principal component analysis based on principal curve and neural networks”, Comput. Chem. Eng., 20(1), 65—78(1996).
3    Kramer,M.A. “Nonlinear principal component analysis using autoassociative neural networks”, AIChE J., 37(2), 233—243(1991).
4    Tan, S., Mavrovouniotis, M.L., “Reducing data dimen-sionality through optimizing neural network inputs”, AIChE J., 41(6), 1471—1480(1995).
5    Jia, F., Martin, E.B., Morris, A.J., “Non-linear principal components analysis for process fault detection”, Com-put. Chem. Eng., 22, (S1), S851—S854(1998).
6    Shao, R., Jia, F., Martin, E.B., Morris, A.J., “Wavelets and non-linear principal components analysis for process monitoring”, Control Eng. Prac., 7, 865—879(1999).
7    Geng, Z.Q., Zhu, Q.X., “Multiscale nonlinear principle component analysis(NLPCA) and　its　application for chemical process monitoring”, Ind. Eng. Chem. Res., 44, 3585—3593(2005).
8    Hu, T., Du, H.B., Yao, P.J., “Prediction of process trends based on neural networks”, Chin. J. Chem. Eng., 10(3), 286—289(2002).
9    Hastie, T.J., Stuetzle, W., “Principal curves”, J. Amer. Stat. Assoc., 84, 502—516(1989).
10    Hertz, J., Krogh, A., Palmer, R.G., Introduction to the Theory of Neural Computation, Wesley Publisher Comp., Redwood City, CA, 198(1991).
11    Kumar, P., Kunzru, D., “Modeling of naphtha pyrolysis”, Ind. Eng. Chem. Process Des. Dev, 24, 774—782(1985).
12    Yang, E.F., Zhou, Q., Hu, Y.F., Xu, Y.M., “A soft-sensing approach to on-line predict the yields of industrial pyro-lysis furnace based on PCA-RBF neural networks”, J. Syst. Simulation, 13(Suppl), 194—197( 2001). (in Chi-nese)