Soft sensor of chemical processes with large numbers of input parameters using auto-associative hierarchical neural network

doi:10.1016/j.cjche.2014.10.004

Abstract

Abstract: To explore the problems of monitoring chemical processes with large numbers of input parameters, a method based on Auto-associative Hierarchical Neural Network (AHNN) is proposed. AHNN focuses on dealing with datasets in high-dimension. AHNNs consist of two parts: groups of subnets based on well trained Autoassociative Neural Networks (AANNs) and a main net. The subnets play an important role on the performance of AHNN. A simple but effective method of designing the subnets is developed in this paper. In this method, the subnets are designed according to the classification of the data attributes. For getting the classification, an effective method called Extension Data Attributes Classification (EDAC) is adopted. Soft sensor using AHNN based on EDAC (EDAC-AHNN) is introduced. As a case study, the production data of Purified Terephthalic Acid (PTA) solvent system are selected to examine the proposed model. The results of the EDAC-AHNN model are compared with the experimental data extracted from the literature, which shows the efficiency of the proposed model.

Key words: Soft sensor, Auto-associative hierarchical neural network, Purified terephthalic acid solvent system, Matter-element

摘要： To explore the problems of monitoring chemical processes with large numbers of input parameters, a method based on Auto-associative Hierarchical Neural Network (AHNN) is proposed. AHNN focuses on dealing with datasets in high-dimension. AHNNs consist of two parts: groups of subnets based on well trained Autoassociative Neural Networks (AANNs) and a main net. The subnets play an important role on the performance of AHNN. A simple but effective method of designing the subnets is developed in this paper. In this method, the subnets are designed according to the classification of the data attributes. For getting the classification, an effective method called Extension Data Attributes Classification (EDAC) is adopted. Soft sensor using AHNN based on EDAC (EDAC-AHNN) is introduced. As a case study, the production data of Purified Terephthalic Acid (PTA) solvent system are selected to examine the proposed model. The results of the EDAC-AHNN model are compared with the experimental data extracted from the literature, which shows the efficiency of the proposed model.

关键词: Soft sensor, Auto-associative hierarchical neural network, Purified terephthalic acid solvent system, Matter-element

Yanlin He, Yuan Xu, Zhiqiang Geng, Qunxiong Zhu . Soft sensor of chemical processes with large numbers of input parameters using auto-associative hierarchical neural network[J]. Chin.J.Chem.Eng., 2015, 23(1): 138-145.

References

[1] S.R. Ponnuswamy, S.L. Shah, C.A. Kiparissides, Computer optimal control of batch polymerization reactors, Ind. Eng. Chem. Res. 50 (11) (1987) 2229-2236.

[2] J.R. Banga, K.J. Versyck, J.F. VanImpe, Computation of optimal identification experiments for nonlinear dynamic process models: A stochastic global optimization approach, Ind. Eng. Chem. Res. 41 (10) (2002) 2425-2430.

[3] J.S. Chang, S.C. Lu, Y.L. Chiu, Dynamic modeling of batch polymerization reactors via the hybrid neural-network rate-function approach, Chem. Eng. J. 130 (1) (2007) 19-28.

[4] C. Shene, C. Diez, S. Bravo, Neural networks for the prediction of the state of Zymomonas mobilis CP4 batch fermentations, Comput. Chem. Eng. 23 (8) (1999) 1097-1108.

[5] C. Shene, C. Diez, S. Bravo, S. Alasheh, F.S. Mjalli, H.E. Alfadala, Forecasting influent-effluent wastewater treatment plants using time series analysis and artificial neural network, Comput. Chem. Eng. 23 (2007) 1097-1108.

[6] D.E. Rumelhart, G.E. Hintont, R.J. Williams, Learning representations by backpropagating errors, Nature 323 (9) (1986) 533-636.

[7] J. Thinault, V.V. Breusegem, A. Cheruy, On-line prediction of fermentation variables using neural networks, Biotechnol. Bioeng. 36 (10) (1990) 1041-1048.

[8] D.C. Psichogios, L.H. Ungar, A hybrid neural network-first principles approach to process modeling, AICHE J. 38 (10) (1992) 1499-1511.

[9] I.M. Galván, J.M. Zaldívar, H. Hernández, E. Molga, The use of neural networks for fitting complex kinetic data, Comput. Chem. Eng. 20 (12) (1996) 1451-1465.

[10] J.F. Pollard, M.R. Broussard, D.B. Garrison, K.Y. San, Process identification using neural networks, Comput. Chem. Eng. 16 (4) (1992) 253-270.

[11] W. Bian, D. Tao, Max-min distance analysis by using sequential SDP relaxation for dimension reduction, IEEE Trans. Pattern Anal. Mach. Intell. 33 (5) (2011) 1037-1050.

[12] H.W. Wang, R. Guan, J.J. Wu, CIPCA: complete-information-based principal component analysis for interval-valued data, Neurocomputing 86 (1) (2012) 158-169.

[13] A. Namphol, S.H. Chin, M. Arozullah, Image compression with a hierarchical neural network, IEEE Trans. Aerosp. Electron. Syst. 32 (1996) 326-338.

[14] H. Bourlard, Y. Kamp, Auto-association by multilayer perceptrons and singular value decomposition, Biol. Cybern. 59 (4) (1988) 291-294.

[15] P. Baldi, K. Hornik, Neural networks and principal component analysis: Learning from examples without local minima, Neural Netw. 2 (1) (1989) 53-58.

[16] Y. Cai, Y. Xu, Q.X. Zhu, Data filtering method and application based on autoassociative neural network, Comput. Appl. Chem. 26 (5) (2009) 673-676.

[17] M.A. Aly, A.F. Atiya, Novelmethods for the feature subset ensembles approach, Int. J. Artif. Intell. Mach. Learn. 6 (4) (2006) 21-27.

[18] L. Rokach, O. Maimon, Feature set decomposition for decision trees, Intell. Data Anal. 9 (2) (2005) 131-158.

[19] L. Rokach, O. Maimon, Decision-tree instance-space decomposition with grouped gain-ratio, Inf. Sci. 177 (17) (2007) 3592-3612.

[20] X.Y. Zheng, Y. Xu, Q.X. Zhu, S.W. Peng, Data attributes decomposition-based hierarchical neural network, Intell. Comput. Intell. Syst. (ICIS) 1 (2010) 343-347.

[21] D.X. Qiu, A comparative study of the K-means algorithm and the normal mixture model for clustering: bivariate homoscedastic case, J. Stat. Plan. Infer. 140 (7) (2010) 1701-1711.

[22] W. Cai, Extension management engineering and applications, Int. J. Oper. Quant. Manage. 5 (1) (1999) 59-72.

[23] W. Cai, Extension theory and its application, Chin. Sci. Bull. 44 (17) (1999) 1538-1548.

[24] Y. Xu, Q.X. Zhu, A new extension theory-based production operation method in industrial process, Chin. J. Chem. Eng. 21 (1) (2013) 44-54.

[25] M.H. Wang, C.P. Hung, Extension neural network and its applications, Neural Netw. 16 (5) (2003) 779-784.

[26] M.H. Wang, Extension neural network-type 2 and its applications, IEEE Trans. Neural Networks 16 (6) (2005) 1352-1361.

[27] M.H. Wang, Extension neural network-type 3, Lect. Notes Comput. Sci 3496 (1) (2005) 503-508.

[28] A.K. Giri, R.K. Patel, S.S. Mahapatra, Artificial neural network (ANN) approach for modelling of arsenic (III) biosorption fromaqueous solution by living cells of Bacillus cereus biomass, Chem. Eng. J. 178 (2011) 15-25.

[29] Y. Xu, Q.X. Zhu, Research and implementation of decreasing the acetic acid consumption in purified terephthalic acid solvent system, Chin. J. Chem. Eng. 16 (4) (2008) 650-655.

[30] K.A. Al-Shayji, Modeling, Simulation and Optimization of Large-scale Commercial Desalination Plant(PhD Thesis) Virginia Polytechnic Institute and State University, VA, USA, 1998.