Research Papers

Tennessee Eastman Process Diagnosis Based on Dynamic Classification With SVDD

[+] Author and Article Information
Foued Theljani

National Engineering School of Tunis,
University of Tunis El Manar,
BP 37,
Le Belvedere 1002, Tunisia
e-mail: foued.theljani@enit.rnu.tn

Kaouther Laabidi, Moufida Ksouri

National Engineering School of Tunis,
University of Tunis El Manar,
BP 37,
Le Belvedere 1002, Tunisia

Salah Zidi

Lille 1 University - Science and Technology,
Villeneuve d'Ascq 59650,
Lille, France

1Corresponding author.

Contributed by the Dynamic Systems Division of ASME for publication in the JOURNAL OF DYNAMIC SYSTEMS, MEASUREMENT, AND CONTROL. Manuscript received October 29, 2013; final manuscript received April 22, 2015; published online June 4, 2015. Assoc. Editor: Srinivasa M. Salapaka.

J. Dyn. Sys., Meas., Control 137(9), 091006 (Sep 01, 2015) (10 pages) Paper No: DS-13-1421; doi: 10.1115/1.4030429 History: Received October 29, 2013; Revised April 22, 2015; Online June 04, 2015

The support vector domain description (SVDD) is an efficient kernel method inspired from the SV machine (SVM) by Vapnik. It is commonly used for one-classification problems or novelty detection. The training algorithm solves a constrained convex quadratic programming (QP) problem. This assumes prior dense sampling (offline training) and it requires large memory and enormous amounts of training time. In this paper, we propose a fast SVDD dedicated for multiclassification problems. The proposed classifier deals with stationary as well as nonstationary (NS) data. The principle is based on the dynamic removal/insertion of informations according to adequate rules. To ensure the rapidity of convergence, the algorithm considers in each run a limited frame of samples for the training process. These samples are selected according to some approximations based on Karush–Kuhn–Tucker (KKT) conditions. An additional merge mechanism is proposed to avoid local optima drawbacks and improve performances. The developed method is assessed on some synthetic data to prove its effectiveness. Afterward, it is employed to solve a diagnosis problem and faults detection. We considered for this purpose a real industrial plant consisting in Tennessee Eastman process (TEP).

Copyright © 2015 by ASME
Your Session has timed out. Please sign back in to continue.


Tax, D., and Duin, R., 1999, “Support Vector Domain Description,” Pattern Recognit. Lett., 20(11--13), pp. 1191–1199. [CrossRef]
Tax, D., and Duin, R., 2004, “Support Vector Data Description,” Mach. Learn., 54(1), pp. 45–66. [CrossRef]
Hua, X., and Ding, S., 2011, “Incremental Learning Algorithm for Support Vector Data Description,” J. Software, 6(7), pp. 1166–1173. [CrossRef]
Tavakkoli, A., Nicolescu, M., and Bebis, G., 2008, “Incremental SVDD Training: Improving Efficiency of Background Modeling in Videos,” International Conference on Signal and Image Processing, Kailua-Kona, HI, pp. 1–6.
Osuna, E., Freund, R., and Girosi, F., 1997, “Improved Training Algorithm for Support Vector Machines,” Neural Networks in Signal Processing (IEEE NNSP'97), Amelia Island, FL, Sep. 24–26, pp. 276–285. [CrossRef]
Jianhua, X., 2010, “Constructing a Fast Algorithm for Multi-Label Classification With Support Vector Data Description,” 2010 IEEE International Conference on Granular Computing (GRC’10), IEEE Computer Society, Washington, DC, pp. 817–821. [CrossRef]
Vapnik, V. N., 1998, Statistical Learning Theory, Wiley, New York.
Camci, F., and Chinnam, R. B., 2008, “General Support Vector Representation Machine for One-Class Classification of Non-Stationary Classes,” Pattern Recognit., 41(10), pp. 3021–3034. [CrossRef]
Cortes, C., and Vapnik, V., 1995, “Support Vector Network,” Mach. Learn., 20(3), pp. 273–297. [CrossRef]
Friedman, J., 1996, “Another Approach to Polychotomous Classification,” Technical Report, Department of Statistics, Stanford University, Stanford, CA.
Platt, J. C., Cristianini, N., and Shawe-Taylor, J., 2000, “Large Margin DAGs for Multiclass Classification,” Neural Information Processing Systems Conference, Vol. 12, pp. 547–553.
Sachs, A., Thiel, C., and Schwenker, F., 2006, “One-Class Support-Vector Machines for the Classification of Bioacoustic Time Series,” Int. J. Artif. Intell. Mach. Learn., 6(4), pp. 29–34. [CrossRef]
Amadou, B. H., Lecoeuche, S., and Maouche, S., 2005, “AUDyC Neural Network Using a New Gaussian Densities Merge Mechanism,” 7th International Conference on Adaptive and Natural Computing Algorithms (ICANNGA'05), Coimbra, Portugal, pp. 155–158. [CrossRef]
Downs, J. J., and Vogel, E. F., 1993, “A Plant-Wide Industrial Process Control Problem,” Comput. Chem. Eng., 17(3), pp. 245–255. [CrossRef]
Kano, M., Nagao, K., Hasebe, S., Hashimoto, I., Ohno, H., Strauss, R., and Bakshi, B., 2002, “Comparison of Multivariate Statistical Process Monitoring Methods With Applications to the Eastman Challenge Problem,” Comput. Chem. Eng., 26(2), pp. 161–174. [CrossRef]
Lee, J.-M., Yoo, C., and Lee, I.-B., 2004, “Statistical Monitoring of Dynamic Processes Based on Dynamic Independent Component Analysis,” Chem. Eng. Sci., 59(14), pp. 2995–3006. [CrossRef]
Kruger, U., Zhou, Y., and Irwin, G., 2004, “Improved Principal Component Monitoring of Large-Scale Processes,” J. Process Control, 14(8), pp. 879–888. [CrossRef]
Xu, Y., Ye, L.-L., and Zhu, Q.-X., 2015, “A New DROS-Extreme Learning Machine With Differential Vector-KPCA Approach for Real-Time Fault Recognition of Nonlinear Processes,” ASME J. Dyn. Syst. Meas. Control, 137(5), p. 051011. [CrossRef]
Lyman, P., and Georgakis, C., 1995, “Plant-Wide Control of the Tennessee Eastman Problem,” Comput. Chem. Eng., 19(3), pp. 321–331. [CrossRef]
Verron, S., Tiplica, T., and Kobi, A., 2008, “Fault Detection and Identification With a New Feature Selection Based on Mutual Information,” J. Process Control, 18(5), pp. 479–490. [CrossRef]
Diehl, C. P., 2003, “SVM Incremental Learning, Adaptation and Optimization,” International Joint Conference on Neural Networks, Vol. 4, pp. 2685–2690. [CrossRef]
Ge, Z. Q., Yang, C. J., and Song, Z. H., 2009, “Improved Kernel PCA-Based Monitoring Approach for Nonlinear Processes,” Chem. Eng. Sci., 64(9), pp. 2245–2255. [CrossRef]
Zhang, R., Wang, W. J., and Ma, Y. C., 2010, “Approximations of the Standard Principal Components Analysis and Kernel PCA,” Expert Syst. Appl., 37(9), pp. 6531–6537. [CrossRef]
Chen, J., and Liu, K.-C., 2002, “On-Line Batch Process Monitoring Using Dynamic PCA and Dynamic PLS Models,” Chem. Eng. Sci., 57(1), pp. 63–75. [CrossRef]
Li, S., and Elbestawi, M. A., 2007, “Tool Condition Monitoring in Machining by Fuzzy Neural Networks,” ASME J. Dyn. Syst., Meas., Control, 188(4), pp. 665–672. [CrossRef]
Michael, D.-J., Jr., William, H. W., and Marion, R. R., 2008, “A One-Sided MEWMA Chart for Health Surveillance,” Qual. Reliab. Eng. Int., 24(5), pp. 503–518. [CrossRef]
Bersimis, S., Psarakis, S., and Panaretos, J., 2007, “Multivariate Statistical Process Control Charts: An Overview,” Qual. Reliab. Eng. Int., 23(5), pp. 517–543. [CrossRef]
Zhang, G., Li, N., and Li, S., 2011, “A Modified Multivariate EWMA Control Chart for Monitoring Process Small Shifts,” International Conference on Modeling, Identification and Control, Shanghai, China, pp. 26–29.


Grahic Jump Location
Fig. 1

Novel detection for drifting Gaussian distribution

Grahic Jump Location
Fig. 2

Novel detection for drifting banana distribution

Grahic Jump Location
Fig. 3

Cardinality variation of the training set and the SVs set associated to the Gaussian distribution

Grahic Jump Location
Fig. 4

Cardinality variation of the training set and the SVs set associated to the banana distribution

Grahic Jump Location
Fig. 5

Two drifting clusters: (a) without merge mechanism and (b) with merge mechanism

Grahic Jump Location
Fig. 6

A diagram of the TEP simulator

Grahic Jump Location
Fig. 7

Detection provided by NS-SVDD based on the computation of the decision function score



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In