Soft sensor for ratio of soda to aluminate based on PCA-RBF multiple network

Abstract

Based on principal component analysis, a multiple neural network was proposed. The principal component analysis was firstly used to reorganize the input variables and eliminate the correlativity. Then the reorganized variables were divided into 2 groups according to the original information and 2 corresponding neural networks were established. A radial basis function network was used to depict the relationship between the output variables and the first group input variables which contain main original information. An other single-layer neural network model was used to compensate the error between the output of radial basis function network and the actual output variables. At last, The multiple network was used as soft sensor for the ratio of soda to aluminate in the process of high-pressure digestion of alumina. Simulation of industry application data shows that the prediction error of the model is less than 3%, and the model has good generalization ability.

Keywords

principal component analysis / multiple neural network / soft sensor / ratio of soda to aluminate / generalization ability

References

[1] LiY, SundararajnN, SaratchandranP. Neurocontroller design for nonlinear fighter aircraft maneuver using fully tuned RBF networks[J]. Automatica, 2001, 37(8): 1293-1301

[2] ZhouPei-ling, TaoXiao-li, FuZhong-qian, et al.. Precipitation predicting based on improved radial basis function networks[J]. Mini-micro System, 2001, 22(2): 244-246(in Chinese)

[3] LiuMei-qin, ShenYi, LiaoXiao-xin. Application of a class of new RBF neural networks to modeling nonlinear systems[J]. Control and Decision, 2001, 16(3): 277-281(in Chinese)

[4] ShinM, GoelA L. Empirical data modeling in software engineering using radial basis function[J]. IEEE Transactions on Software Engineering, 2000, 26(6): 567-576

[5] ErhardR. Application of Bayesian trained RBF networks to nonlinear time-series modeling [J]. Signal Processing, 2003, 83(7): 1393-1410

[6] RojasI, PomaresH, BernierJ L, et al.. Time series analysis using normalized PG-RBF network with regression weights[J]. Neurocomputing, 2002, 42(1): 267-285

[7] SchmidhuberJ. Discovering neural nets with low Kolmogorov complexity and high generalization capability [J]. Neural Networks, 1997, 10(5): 857-873

[8] LiJin, ZhangZi-ping, ChenHong-fang. The PCA method in neural networks[J]. Journal of China University of Science and Technology, 2001, 31(1): 40-44(in Chinese)

[9] DanS. Training radial basis neural networks with the extended Kalman filter[J]. Neurocomputing, 2002, 48(2): 455-475

[10] ChenS, CowanC F N, GrantP M. Orthogonal least squares learning algorithm for radial basis function networks[J]. IEEE Transactions on Neural Networks, 1991, 2(2): 302-309

[11] SherstinskyA, PicardR W. On the efficiency of orthogonal least squares training method for radial basis function networks[J]. IEEE Transactions on Neural Networks, 1996, 7(1): 195-200

[12] ChangE S, YangH H, BösS. Orthogonal least squares learning algorithm with local adaptation process for the radial basis function networks[J]. IEEE Signal Processing Letters, 1996, 3(8): 253-255

[13] MusaviM T, BryantR J, QiaoM, et al.. Mouse chromosome classification by radial basis function network with fast orthogonal search[J]. Neural Networks, 1998, 11(4): 769-777

[14] DrioliC, RocchessoD. Orthogonal least squares algorithm for the approximation of a map and its derivatives with a RBF network[J]. Signal Processing, 2003, 83(2): 283-296