Automatic Fault Identification in Sensor Networks Based on Probabilistic Modeling

  • Stavros NtalampirasEmail author
  • Georgios Giannopoulos
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8985)


This work proposes a mechanism able to automatically categorize different types of faults occurring in critical infrastructures and especially water distribution networks. The mechanism models the relationship exhibited among the sensor datastreams based on the assumption that its pattern alters depending on the fault type. The first phase includes linear time invariant modeling which outputs a parameters vector. At the second phase the evolution of the parameter vectors is captured via hidden Markov modeling. The methodology is applied on data coming from the water distribution network of the city of Barcelona. The corpus contains a vast amount of data representative of nine network states. The nominal is included for enabling fault detection. The achieved classification rates are quite encouraging and the system is practical.


Linear time invariant modeling Hidden Markov model Fault diagnosis Critical infrastructure protection 



The authors would like to thank Prof. Joseba Quevedo and Dr. Miquel A. Cuguero of the Advanced Control System group ( of the Polytechnic University of Catalonia for their useful and constructive insights regarding the dataset.


  1. 1.
    Quevedo, J., Alippi, C., Cuguero, M.A., Ntalampiras, S., Puig, V., Roveri, M., Garcia, D.: Temporal/spatial model-based fault diagnosis vs. hidden Markov models change detection method: application to the Barcelona water network. In: 2013 21st Mediterranean Conference on Control Automation, pp. 394–400 (2013)Google Scholar
  2. 2.
    Bashan, A., Berezin, Y., Buldyrev, S.V., Havlin, S.: The extreme vulnerability of interdependent spatially embedded networks. Nat Phys. 9, 667–672 (2013)CrossRefGoogle Scholar
  3. 3.
    Sobahni-Tehrani, E.: Fault Detection, Isolation, and Identification for Nonlinear Systems Using a Hybrid Approach. Concordia University, Canada (2008)Google Scholar
  4. 4.
    Zampino, E.J.: The extreme vulnerability of interdependent spatially embedded networks. In: Proceedings of Annual of Reliability and Maintainability Symposium, pp. 16–22 (2001)Google Scholar
  5. 5.
    Crosetti, P.A.: Fault tree analysis with probability evaluation. IEEE Trans. Nucl. Sci. 18(1), 465–471 (1971)CrossRefGoogle Scholar
  6. 6.
    Dexter, A.L.: Fuzzy model based fault diagnosis. IEE Proc. Control Theor. Appl. 142(6), 545–550 (1995)CrossRefzbMATHGoogle Scholar
  7. 7.
    Mendonsa, L.F., Sousa, J.M.C., Sa da Costa, J.M.G.: Fault detection and isolation of industrial processes using optimized fuzzy models. In: The 14th IEEE International Conference on Fuzzy Systems, pp. 851–856 (2005)Google Scholar
  8. 8.
    Cybenko, G.: Approximation by superpositions of a sigmoidal function. Math. Control, Signals, Syst. 2, 303–314 (1989)MathSciNetCrossRefzbMATHGoogle Scholar
  9. 9.
    Li, B., Chow, M.-Y., Tipsuwan, Y., Hung, J.C.: Neural-network-based motor rolling bearing fault diagnosis. IEEE Trans. Ind. Electron. 47(5), 1060–1069 (2000)CrossRefGoogle Scholar
  10. 10.
    Patton, R.J., Chen, J., Siew, T.M.: Fault diagnosis in nonlinear dynamic systems via neural networks. In: International Conference on Control, vol. 2, pp. 1346–1351 (1994)Google Scholar
  11. 11.
    Yu, D., Shields, D.N., Daley, S.: A hybrid fault diagnosis approach using neural networks. Neural Comput. Appl. 4(1), 21–26 (1996)CrossRefGoogle Scholar
  12. 12.
    Ming, Y., et al.: A Hybrid Approach for Fault Diagnosis based on Multilevel Flow Models and Artificial Neural Network, vol. 2. IEEE Computer Society (2006)Google Scholar
  13. 13.
    Duan, H., Xiufen, Y., Ma, G.: Novel hybrid approach for fault diagnosis in 3-dof flight simulator based on BP neural network and ant colony algorithm. In: IEEE Swarm Intelligence Symposium, pp. 371–374 (2007)Google Scholar
  14. 14.
    Bonissone, P.P., Xue, F., Subbu, R.: Fast meta-models for local fusion of multiple predictive models. Appl. Soft Comput. 11(2), 1529–1539 (2011)CrossRefGoogle Scholar
  15. 15.
    Quevedo, J., et al.: Validation and reconstruction of flow meter data in the Barcelona water distribution network. Control Eng. Pract. 18(6), 640–651 (2010)MathSciNetCrossRefGoogle Scholar
  16. 16.
    Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.H.: The WEKA data mining software: an update. ACM SIGKDD Explor. Newsl. 11(1), 10–18 (2009)CrossRefGoogle Scholar
  17. 17.
    Funahashi, K., Nakamura, Y.: Approximation of dynamical systems by continuous time recurrent neural networks. Neural Networks 6(6), 801–806 (1993)CrossRefGoogle Scholar
  18. 18.
    Potamitis, I.: Single channel enumeration and recognition of an unknown and time-varying number of sound sources. In: 16th European Signal Processing Conference, Laussane, pp. 25–29, August 2008Google Scholar

Copyright information

© European Union 2016

Authors and Affiliations

  1. 1.Joint Research Center, European CommissionIspra, VareseItaly

Personalised recommendations