Visual Monitoring of Industrial Operation States Based on Kernel Fisher Vector and Self-organizing Map Networks

  • Wei-Peng Lu
  • Xue-Feng YanEmail author


As industrial process becomes increasingly complicated and the correlation between industrial process variables tends to exhibit strong nonlinear characteristics, how to effectively and visually monitor industrial operation states is challenging. Amethod based on kernel Fisher vector and self-organizing map networks (KFV-SOM) is proposed to improve the visualization of process monitoring. InKFV-SOM, kernel Fisher discriminant analysis is first employed to map data into high-dimensional space by using a nonlinear function, and the optimal Fisher feature vector, which can represent industrial operation states fittingly, is extracted. Thatis, the normal state and different kinds of faults can be distinguished well in the Fisher feature vector space. Thetopological structure of the Fisher feature vector space is then visualized intuitively on the two-dimensional output map of self-organizing map (SOM) with the Fisher feature vector as the input of the SOM network. Thus, the KFV-SOM caneffectively realize the visualization of monitoring. Continuousstirred tank reactor process is applied to illustrate the capability of KFV-SOM. Resultshows that KFV-SOM caneffectively visualize monitoring, and it is better in showing the operation states of normal state and different kinds of faults on the output map of the SOM networkthan SOM, SOM integratedwith principal component analysis, SOM integratedwith correlative component analysis, SOM integratedwith Fisher discriminant analysis, and SOM integratedwith canonical variable analysis.


Continuous stirred tank reactor process industrial operation state monitoring kernel Fisher discriminant analysis self-organizing map 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    D. H. Zhou and Y. Y. Hu, “Fault diagnosis techniques for dynamic systems,” Acta Automatica Sinica, vol. 35, no. 6, pp. 748–758, July 2009.CrossRefGoogle Scholar
  2. [2]
    A. T. James, O. P. Gandhi, and S. G. Deshmukh, “Fault diagnosis of automobile systems using fault tree based on digraph modeling,” International Journal of System Assurance Engineering and Management, vol. 9, no. 2, pp. 494–508, April 2018.CrossRefGoogle Scholar
  3. [3]
    Y. Zhuand, L. Geng, “Research on SDG faultdiagnosis of ocean shipping boiler system based on fuzzy granular computing under data fusion,” Polish Maritime Research, vol. 25, no. s2, pp. 92–97, September 2018.CrossRefGoogle Scholar
  4. [4]
    W. P. Wagner, “Trends in expert system development: a longitudinal content analysis of over thirty years of expert system case studies,” Expert Systems with Applications, vol. 76, pp. 85–96, June 2017.CrossRefGoogle Scholar
  5. [5]
    H. Ma, H. J. Liang, H. J. Ma, and Q. Zhou, “Nussbaum gain adaptive backstepping control of non-linear strict-feedback systems with unmodeled dynamics and unknown dead zone,” International Journal of Robust and Nonlinear Control, vol. 28, no. 17, pp. 5326–5343, September 2018.MathSciNetCrossRefzbMATHGoogle Scholar
  6. [6]
    Y. H. Zhang, H. J. Liang, H. Ma, Q. Zhou, and Z. D. Yu, “Distributed adaptive consensus tracking control for nonlinear multi-agent systems with state constraints,” Applied Mathematics and Computation, vol. 326, pp. 16–32, June 2018.MathSciNetCrossRefGoogle Scholar
  7. [7]
    Y. H. Zhang, J. Sun, H. J. Liang, and H. Y. Li, “Event-triggered adaptive tracking control for multi-agent systems with unknown disturbances,” IEEE Transactionson Cybernetics, pp. 1–12, September 2018. DOI: 10.1109/TCYB.2018.2869084Google Scholar
  8. [8]
    X. H. Chang and G. H. Yang, “Nonfragile filtering of continuous-time fuzzy systems,” IEEE Transactionson Signal Processing, vol. 59, no. 4, pp. 1528–1538, April 2011.MathSciNetCrossRefzbMATHGoogle Scholar
  9. [9]
    M. Q. Shen and D. Ye, “Improved fuzzy control design for nonlinear Markovian-jump systems with incomplete transition descriptions,” Fuzzy Sets and Systems, vol. 217, pp. 80–95, April 2013.MathSciNetCrossRefzbMATHGoogle Scholar
  10. [10]
    J. L. Liu, S. M. Fei, E. G. Tian, and G. Zhou, “Co-design of event generator and filtering for a class of TS fuzzysystems with stochastic sensor faults,” Fuzzy Sets and Systems, vol. 273, pp. 124–140, August 2015.MathSciNetCrossRefzbMATHGoogle Scholar
  11. [11]
    C. Yang, T. Teng, B. Xu, Z. Li, J. Na, and C. Y. Su, “Global adaptive tracking control of robot manipulators using neural networks with finite-time learning convergence,” International Journal of Control Automation & Systems, vol. 15, no. 11, pp. 1916–1924, August 2017.CrossRefGoogle Scholar
  12. [12]
    M. Hamadacheand D. Lee, “Principal component analysis based signal-to-noise ratio improvement for inchoate faulty signals: application to ball bearing fault detection,” International Journal of Control Automation & Systems, vol. 15, no. 2, pp. 506–517, April 2017.CrossRefGoogle Scholar
  13. [13]
    S. E. Calce, H. K. Kurki, D. A. Weston, and L. Gould, “Principal component analysis in the evaluation of osteoarthritis,” American Journal of Physical Anthropology, vol. 162, no. 3, pp. 476–490, March 2017.CrossRefGoogle Scholar
  14. [14]
    B. Wang, H. Pan, and W. Yang, “Robust bearing degradation assessment method based on improved CVA,” IET ScienceMeasurement & Technology, vol. 11, no. 5, pp. 637–645, July 2017.CrossRefGoogle Scholar
  15. [15]
    D. Borek, R. Bromberg, J. Hattne, and Z. Otwinowski, “Real-space analysis of radiation-induced specific changes with independent component analysis,” Journal of Synchrotron Radiation, vol. 25, no. 2, pp. 451–467, February 2018.CrossRefGoogle Scholar
  16. [16]
    H. Wang, X. Lu, Z. Hu, and W. Zheng, “Fisher discriminant analysis with Ll-norm,” IEEE Transactionson Cybernetics, vol. 44, no. 6, pp. 828–842, July 2017.CrossRefGoogle Scholar
  17. [17]
    A. F. Silva, M. C. Sarraguça, P. R. Ribeiro, A. O. Santos, B. T. De, and J. A. Lopes, “Statistical process control of cocrystallization processes: a comparison between OPLS andPLS,” International Journal of Pharmaceutics, vol. 520, no. 1-2, pp. 29–38, March 2017.CrossRefGoogle Scholar
  18. [18]
    J. W. Tang and X. F. Yan, “Neural network modeling relationship between inputs and state mapping plane obtained by FDAt-SNE forvisual industrial process monitoring,” Applied Soft Computing, vol. 60, pp. 577–590, November, 2017.CrossRefGoogle Scholar
  19. [19]
    F. An, X. Zhang, L. Chen, and H. J. Mattausch, “A memory-based modular architecture for SOM and LVQ withdynamic configuration,” IEEE Transactionson Multi-scale Computing Systems, vol. 2, no. 4, pp. 234–241, October 2016.CrossRefGoogle Scholar
  20. [20]
    T. Chopraand J. Vajpai, “Classification of faults in damadics benchmark process control system using self-organizing maps,” International Journal of Soft Computing & Engineering, vol. 1, no. 3, pp. 22312307, March 2011.Google Scholar
  21. [21]
    F. Corona, M. Mulas, R. Baratti, and J. A. Romagnoli, “On the topological modeling and analysis of industrial process data using the SOM,” Computers & Chemical Engineering, vol. 34, no. 12, pp. 2022–2032, December 2010.CrossRefGoogle Scholar
  22. [22]
    H. Y. Yu, F. Khan, V. Garaniya, and A. Ahmad, “Self-organizing map based fault diagnosis technique for non-gaussian processes,” Industrial & Engineering Chemistry Research, vol. 53, no. 21, pp. 8831–8843, May 2014.CrossRefGoogle Scholar
  23. [23]
    J. Q. Hu, L. B. Zhang, and W. Liang, “Dynamic degradation observer for bearing fault by MTSSOM system,” Mechanical Systems and Signal Processing, vol. 36, no. 2, pp. 385–400, April 2013.CrossRefGoogle Scholar
  24. [24]
    Z. G. Feng and T. Xu, “Comparison of SOM andPCA-SOM infault diagnosis of ground-testing bed,” Procedia Engineering, vol. 15, no. 1, pp. 1271–1276, December 2011.CrossRefGoogle Scholar
  25. [25]
    X. Y. Chen and X. F. Yan, “Using improved self-organizing map for fault diagnosis in chemical industry process,” Chemical Engineering Research and Design, vol. 90, no. 12, pp. 2262–2277, December 2012.CrossRefGoogle Scholar
  26. [26]
    X. Y. Chen and X. F. Yan, “Fault diagnosis in chemical process based on self-organizing map integrated with Fisher discriminant analysis,” Chinese Journal of Chemical Engineering, vol. 21, no. 4, pp. 382–387, July 2013.CrossRefGoogle Scholar
  27. [27]
    Y Song, Q. C. Jiang, X. F. Yan, and M. J. Guo, “A multi-SOM withcanonical variate analysis for chemical process monitoring and fault diagnosis,” Journal of Chemical Engineering of Japan, vol. 47, no. 1, pp. 40–51, January 2014.CrossRefGoogle Scholar
  28. [28]
    W. Chuand X. Li, “On the asymptotic behavior of the kernel function in the generalized langevin equation: a one-dimensional lattice model,” Journal of Statistical Physics, vol. 170, no. 2, pp. 378–398, November 2018.MathSciNetCrossRefGoogle Scholar
  29. [29]
    H. Ishibashiand T. Furukawa, “Hierarchical tensor SOM networkfor multilevel-multigroup analysis,” Neural Process Letter, no. 1, pp. 1–15, June 2017.Google Scholar
  30. [30]
    G. Abaei, A. Selamat, and H. Fujita, “An empirical study based on semi-supervised hybrid self-organizing map for software fault prediction,” Knowledge-based Systems, vol. 74, pp. 28–39, January 2015.CrossRefGoogle Scholar
  31. [31]
    Y Yang, W. Sheng, Y Han, and X. Ma, “Multi-beam pattern synthesis algorithm based on kernel principal component analysis and semi-definite relaxation,” IET Communications, vol. 12, no. 1, pp. 82–95, January 2018.CrossRefGoogle Scholar
  32. [32]
    J. D. A. Santosand G. A. Barreto, “An outlier-robust kernel RLS algorithmfor nonlinear system identification,” Nonlinear Dynam, vol. 90, no. 3, pp. 1707–1726, Novembe 2017.MathSciNetCrossRefGoogle Scholar
  33. [33]
    N. D. Hoang and D. T. Bui, “Predicting earthquake-induced soil liquefaction based on a hybridization of kernel Fisher discriminant analysis and a least squares support vector machine: a multi-dataset study,” Bulletin of Engineering Geology & the Environment, vol. 77, no. 1, pp. 191–204, February 2018.CrossRefGoogle Scholar
  34. [34]
    S. Mika, G. Ratsch, J. Weston, B. Scholkopf, and K. R. Mullers, “Fisher discriminant analysis with kernels,” Neural Networks for Signal Processing IX: Proceedings of the IEEE SignalProcessing Society Workshop, pp. 41–48, 1999.Google Scholar
  35. [35]
    X. Zhang, W. Yan, X. Zhao, and H. Shao, “Nonlinear biological batch process monitoring and fault identification based on kernel Fisher discriminant analysis,” Process Biochemistry, vol. 42, no. 8, pp. 1200–1210, August 2007.CrossRefGoogle Scholar
  36. [36]
    L. Cao and Y. Wang, “Fault-tolerant control for nonlinear systems with multiple intermittent faults and time-varying delays,” International Journal of Control Automation & Systems, vol 16, no. 2, pp. 609–621, March 2018.CrossRefGoogle Scholar

Copyright information

© ICROS, KIEE and Springer 2019

Authors and Affiliations

  1. 1.Wei-Peng Lu is with the School of Information Science and EngineeringEast China University of Science and TechnologyShanghaiPeople’s Republic of China
  2. 2.Xue-Feng Yan is with the Key Laboratory of Advanced Control and Optimization for Chemical Processes of Ministry of EducationEast China University of Science and TechnologyShanghaiPeople’s Republic of China

Personalised recommendations