Density-Induced Support Vector Data Description for Fault Detection on Tennessee Eastman Process

  • Yangtao Xue
  • Li ZhangEmail author
  • Bangjun Wang
  • Baige Tang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11304)


Fault detection can be taken as a behavior of detecting abnormal data in process data. Support vector data description (SVDD) has been successfully used for fault detection. Although density-induced support vector data description (D-SVDD) can give a better description of target data by introducing relative density degrees than SVDD, the problem of an additional parameter selection hinders the application of D-SVDD, which has a great influence on the performance of D-SVDD. This paper bounds this additional parameter for D-SVDD and applies D-SVDD to fault detection on TE process monitoring. Experiment shows D-SVDD is promising.


D-SVDD Fault detection TE process 



This work was supported in part by the National Natural Science Foundation of China under Grant No. 61373093, by the Natural Science Foundation of Jiangsu Province of China under Grant Nos. BK20140008, and by the Soochow Scholar Project.


  1. 1.
    Downs, J.J., Vogel, E.F.: A plant-wide industrial process control problem. Comput. Chem. Eng. 17(3), 245–255 (1993)CrossRefGoogle Scholar
  2. 2.
    Ge, Z., Song, Z., Gao, F.: Review of recent research on data-based process monitoring. Ind. Eng. Chem. Res. 52(10), 3543–3562 (2013)CrossRefGoogle Scholar
  3. 3.
    Lee, K., Kim, D.W., Lee, D., Lee, K.H.: Improving support vector data description using local density degree. Pattern Recognit. 38(10), 1768–1771 (2005)CrossRefGoogle Scholar
  4. 4.
    Lee, K., Kim, D.W., Lee, K.H., Lee, D.: Density-induced support vector data description. IEEE Trans. Neural Netw. 18(1), 284–289 (2007)CrossRefGoogle Scholar
  5. 5.
    Mahadevan, S., Shah, S.L.: Fault detection and diagnosis in process data using one-class support vector machines. J. Process. Control. 19(10), 1627–1639 (2009)CrossRefGoogle Scholar
  6. 6.
    Qin, S.J.: Survey on data-driven industrial process monitoring and diagnosis. Annu. Rev. Control 36(2), 220–234 (2012)CrossRefGoogle Scholar
  7. 7.
    Schölkopf, B., Platt, J.C., Shawe-Taylor, J., Smola, A.J., Williamson, R.C.: Estimating the support of a high-dimensional distribution. Neural Comput. 13(7), 1443–1471 (2001)CrossRefGoogle Scholar
  8. 8.
    Tax, D.M., Duin, R.P.: Support vector domain description. Pattern Recognit. Lett. 20(11), 1191–1199 (1999)CrossRefGoogle Scholar
  9. 9.
    Tax, D.M., Duin, R.P.: Support vector data description. Mach. Learn. 54(1), 45–66 (2004)CrossRefGoogle Scholar
  10. 10.
    Xiao, Y., Wang, H., Zhang, L., Xu, W.: Two methods of selecting gaussian kernel parameters for one-class SVM and their application to fault detection. Knowl. Based Syst. 59, 75–84 (2014)CrossRefGoogle Scholar
  11. 11.
    Yin, S., Ding, S.X., Haghani, A., Hao, H., Zhang, P.: A comparison study of basic data-driven fault diagnosis and process monitoring methods on the benchmark tennessee eastman process. J. Process. Control. 22(9), 1567–1581 (2012)CrossRefGoogle Scholar
  12. 12.
    Yin, S., Ding, S.X., Xie, X., Luo, H.: A review on basic data-driven approaches for industrial process monitoring. IEEE Trans. Ind. Electron. 61(11), 6418–6428 (2014)CrossRefGoogle Scholar
  13. 13.
    Zhang, K., Qian, K., Chai, Y., Li, Y., Liu, J.: Research on fault diagnosis of tennessee eastman process based on KPCA and SVM. In: 2014 Seventh International Symposium on Computational Intelligence and Design (ISCID), vol. 1, pp. 490–495 (2014)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  • Yangtao Xue
    • 1
  • Li Zhang
    • 1
    Email author
  • Bangjun Wang
    • 1
  • Baige Tang
    • 1
  1. 1.School of Computer Science and Technology and Joint International Research Laboratory of Machine Learning and Neuromorphic ComputingSoochow UniversitySuzhouChina

Personalised recommendations