» Download Article

Authors
Mats Nikus; Mikko Vermasvuori; Nikolai Vatanski; Sirkka-Liisa Jamsa-Jounela

Digital Object Identifier (DOI)
10.3182/20060830-2-SF-4903.00040

Page Numbers:
226-231

Index Terms
fault detection,monitoring,support vector machines,classification,regression,dearomatization process

Abstract
The aim of the work presented in this paper is to assess the ability of support vector machines (SVM) for detecting measurement faults. Two different support vector machine approaches for detecting faults are tested and compared to neural networks. The first method is based on a SVM regression model together with an analysis of the residuals whereas the second method is based on a SVM classifier. The methods were applied to a rigorous first principles based dynamic simulator of a dearomatization process.

References

[1] Chang C.-C. and C.-J. Lin (2001), Libsvm: a
    library for support vector machines, SVM
    software on the internet, available at
    http://www.csie.ntu.edu.tw/~cjlin/libsvm.
[2] Jemwa G. T. and C. Aldrich (2005), Monitoring of an
    industrial liquid-liquid extraction system with
    kernel-based methods, Hydrometallurgy, 78, 41-51.
[3] Hinkley, D. V. (1971), Inference about the change-point
     from cumulative sum tests, Biometrika, 58,
    509-523.
[4] Kulkarni A., Jayaraman V. K. and B. D. Kulkarni
    (2005), Kowledge incorporated support vector
    machines to detect faults in the Tennessee Eastman
    process, Computers & Chemical Engineering, 29,
    2128-2133.
[5] Platt J. C. (1998), Fast training of support vector
    machines using sequential minimal optimization,
    in: Advances in Kernel Methods -Support Vector
    Learning, (B. Schölkopf, C. J. C. Burges, and A. J.
    Smola (Ed)) 1st Edition, MIT Press, Cambridge,
    Massachusetts, 185-208.
[6] Pöyhönen, S., A. Arkkio, P. Jover and H. Hyötyniemi
    (2005), Coupling pairwise support vector machines
    for fault classification, Control Engineering
    Practice, 13, 759-769.
[7] Ribeiro B. (2005), Support vector machines for
    quality monitoring in a plastic injection moulding
    process, IEEE Transactions on systems, man, and
    cybernetics-Part C: Applications and reviews, 35,
    401-410.
[8] Saxén B. and H. Saxén, NNDT - A neural network
    development tool - User's guide, Technical Report
    94-8, Heat Eng. Lab, Åbo Akademi University,
    Åbo, Finland, 1994.
[9] Vapnik V. N. (1998), Statistical learning theory,
    John Wiley & Sons, New York, NY.
[10] Venkatasubramanian V., R. Rengaswamy, K. Yin,
     S. N. Kavuri (2003), A review of process fault
     detection and diagnosis Part I: Quantitative model-based
      methods, Computers & Chemical
     Engineering, 27, 293-311.
[11] Vermasvuori M. T., N. Vatanski and S-L JämsäJounela
      (2005), Data-based fault detection of the
     online analysers in a dearomatisation process, 1st
     Workshop on Networked Control System
     and Fault Tolerant Control, Ajaccio, France,
     October 6-7th, 2005 (www.strepnecst.org/1st_workshop/Vermasvuori.pdf)
     
[12] Wang H. Q., Zhang P and S. X. Ding (2006), Kernel
     and SVM based theory for industrial process
     modelling and fault diagnosis. EU project NeCST
     internal presentation, Paris, March 2006.