Advertisement

Neural Prediction of Product Quality Based on Pilot Paper Machine Process Measurements

  • Paavo Nieminen
  • Tommi Kärkkäinen
  • Kari Luostarinen
  • Jukka Muhonen
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6593)

Abstract

We describe a multilayer perceptron model to predict the laboratory measurements of paper quality using the instantaneous state of the papermaking production process. Actual industrial data from a pilot paper machine was used. The final model met its goal accuracy 95.7% of the time at best (tensile index quality) and 66.7% at worst (beta formation). We anticipate usage possibilities in lowering machine prototyping expenses, and possibly in quality control at production sites.

Keywords

MLP Prediction Paper quality Pilot paper machine 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Leiviskä, K. (ed.): Papermaking Science and Technology, 2nd edn. Process and Maintenance Management, vol. 14. Paperi ja Puu Oy (2009)Google Scholar
  2. 2.
    Ribeiro, B., Dourado, A., Costa, E.: Industrial kiln multivariable control: MNN and RBFNN approaches. In: ICANNGA 1995, pp. 408–411. Springer, Heidelberg (1995)Google Scholar
  3. 3.
    Ribeiro, B.: Prediction of the lime availability on an industrial kiln by neural networks. In: IJCNN 1998, vol. 3, pp. 1987–1991. IEEE, Los Alamitos (1998)Google Scholar
  4. 4.
    Rajesh, K., Ray, A.K.: Artificial neural network for solving paper industry problems: A review. Journal of Scientific & Industrial Research 65, 565–573 (2006)Google Scholar
  5. 5.
    Edwards, P., Murray, A., Papadopoulos, G., Wallace, A., Barnard, J., Smith, G.: The application of neural networks to the papermaking industry. IEEE Transactions on Neural Networks 10(6), 1456–1464 (1999)CrossRefGoogle Scholar
  6. 6.
    Wang, F., Sanguansintukul, S., Lursinsap, C.: Curl forecasting for paper quality in papermaking industry. In: Asia Simulation Conference – 7th International Conference on System Simulation and Scientific Computing, pp. 1079–1084 (2008)Google Scholar
  7. 7.
    Lampinen, J., Taipale, O.: Optimization and simulation of quality properties in paper machine with neural networks. In: ICNN 1994, pp. 3812–3815. IEEE, Los Alamitos (1994)Google Scholar
  8. 8.
    Kärkkäinen, T.: MLP in layer-wise form with applications to weight decay. Neural Computation 14(6), 1451–1480 (2002)CrossRefzbMATHGoogle Scholar
  9. 9.
    Kärkkäinen, T., Heikkola, E.: Robust formulations for training multilayer perceptrons. Neural Computation 16(4), 837–862 (2004)CrossRefzbMATHGoogle Scholar
  10. 10.
    Nieminen, P., Kärkkäinen, T.: Ideas about a regularized MLP classifier by means of weight decay stepping. In: Kolehmainen, M., Toivanen, P., Beliczynski, B. (eds.) ICANNGA 2009. LNCS, vol. 5495, pp. 32–41. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  11. 11.
    Nieminen, P., Kärkkäinen, T.: Comparison of MLP cost functions to dodge mislabeled training data. In: IJCNN 2010. IEEE, Los Alamitos (2010)Google Scholar
  12. 12.
    Jolliffe, I.T.: Principal Component Analysis, 2nd edn. Springer, New York (2002)zbMATHGoogle Scholar
  13. 13.
    Haykin, S.: Neural Networks: A Comprehensive Foundation, 2nd edn. Prentice Hall, New Jersey (1999)zbMATHGoogle Scholar
  14. 14.
    Hagan, M.T., Menhaj, M.B.: Training feedforward networks with the Marquardt algorithm. IEEE Transactions on Neural Networks 5(6), 989–993 (1994)CrossRefGoogle Scholar
  15. 15.
    Kobyashi, M., Zamani, A., Ozawa, S., Abe, S.: Reducing computations in incremental learning for feedforward neural network with long-term memory. In: IJCNN 2001, pp. 1989–1994. IEEE, Los Alamitos (2001)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Paavo Nieminen
    • 1
  • Tommi Kärkkäinen
    • 1
  • Kari Luostarinen
    • 2
  • Jukka Muhonen
    • 2
  1. 1.Department of Mathematical Information TechnologyUniversity of JyväskyläFinland
  2. 2.Metso PaperJyväskyläFinland

Personalised recommendations