Gating Artificial Neural Network Based Soft Sensor

  • Petr Kadlec
  • Bogdan Gabrys
Part of the Studies in Computational Intelligence book series (SCI, volume 134)


This work proposes a novel approach to Soft Sensor modelling, where the Soft Sensor is built by a set of experts which are artificial neural networks with randomly generated topology. For each of the experts a meta neural network is trained, the gating Artificial Neural Network. The role of the gating network is to learn the performance of the experts in dependency on the input data samples. The final prediction of the Soft Sensor is a weighted sum of the individual experts predictions. The proposed meta-learning method is evaluated on two different process industry data sets.


Hide Unit Soft Sensor Correlation Base Feature Selection Principle Component Regression Gating Network 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Jolliffe, I.T.: Principal Component Analysis. Springer, Heidelberg (2002)zbMATHGoogle Scholar
  2. 2.
    Bishop, C.M.: Neural Networks for Pattern Recognition. Oxford University Press, Oxford (1995)Google Scholar
  3. 3.
    Jang, J.S.R., Sun, C.T., Mizutani, E.: Neuro-fuzzy and soft computing. Prentice Hall Upper Saddle River, Englewood Cliffs (1997)Google Scholar
  4. 4.
    Angelov, P.P., Filev, D.P.: Flexible models with evolving structure. International Journal of Intelligent Systems 19(4), 327–340 (2004)zbMATHCrossRefGoogle Scholar
  5. 5.
    Angelov, P.P., Filev, D.P.: An approach to online identification of takagi-sugeno fuzzy models. Systems, Man and Cybernetics, Part B, IEEE Transactions on 34(1), 484–498 (2004)CrossRefGoogle Scholar
  6. 6.
    Kasabov, N.K., Song, Q.: Denfis: dynamic evolving neural-fuzzy inference system and itsapplication for time-series prediction. Fuzzy Systems, IEEE Transactions on 10(2), 144–154 (2002)CrossRefGoogle Scholar
  7. 7.
    Gabrys, B., Bargiela, A.: Neural networks based decision support in presence of uncertainties. Journal of Water Resources Planning and Management 125(5), 272–280 (1999)CrossRefGoogle Scholar
  8. 8.
    Champagne, M., Dudzic, M., Inc, T., Temiscaming, Q.: Industrial use of multivariate statistical analysis for process monitoring and control. In: American Control Conference, 2002 Proceedings of the 2002, vol. 1 (2002)Google Scholar
  9. 9.
    Li, W., Yue, H.H., Valle-Cervantes, S., Qin, S.J.: Recursive pca for adaptive process monitoring. Journal of Process Control 10(5), 471–486 (2000)CrossRefGoogle Scholar
  10. 10.
    Fortuna, L.: Soft Sensors for Monitoring and Control of Industrial Processes. Springer, Heidelberg (2007)zbMATHGoogle Scholar
  11. 11.
    Dunia, R., Qin, J., Edgar, T.F., McAvoy, T.J.: Sensor fault identification and reconstruction using principal component analysis. In: Proceedings of the 13th Triennial World Congress, pp. 259–264 (1996)Google Scholar
  12. 12.
    Dunia, R., Qin, S.J.: Joint diagnosis of process and sensor faults using principal component analysis. Control Engineering Practice 6(4), 457–469 (1998)CrossRefGoogle Scholar
  13. 13.
    Amazouz, M., Pantea, R.: Use of multivariate data analysis for lumber drying process monitoring and fault detection. In: Crone, S.F., Stahlbock, S.L., R. (eds.) International Conference on Data Mining, pp. 329–332 (2006)Google Scholar
  14. 14.
    Macias, J.J., Zhou, P.X.: A method for predicting quality of the crude oil distillation. In: Evolving Fuzzy Systems, 2006 International Symposium on, pp. 214–220 (2006)Google Scholar
  15. 15.
    Jordan, M.I., Barto, A.G.: Task decomposition through competition in a modular connectionist architecture: The what and where vision tasks. Cognitive Science 15, 219–250 (1991)CrossRefGoogle Scholar
  16. 16.
    Jacobs, R.: Adaptive mixtures of local experts. Neural Computation 3(1), 79–87 (1991)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Petr Kadlec
    • 1
  • Bogdan Gabrys
    • 1
  1. 1.Computational Intelligence Research GroupBournemouth University Fern Barrow, PooleUnited Kingdom

Personalised recommendations