Skip to main content
SpringerLink
Log in

A hybrid neural network model for predicting silicon content

Modell eines hybriden neuronalen Netzes zur Voraussage des Siliziumgehalts

  • Reviewed Original Papers
  • Published:
e & i Elektrotechnik und Informationstechnik Aims and scope Submit manuscript

Abstract

The silicon content of the hot metal in the blast furnace ironmaking process normally reflects the thermal state of the furnace and affects the fuel rate. In this paper a hybrid neural network model is proposed to predict the silicon contentn steps ahead. A time-delay neural network, which has self-loops to represent dynamics, is adopted here. The learning procedure of this network has been divided into two states. A BP algorithm with forgetting factor is first introduced to find the appropriate structure of the network. The temporal difference (TD) method with forgetting factor is then used forn-step-ahead prediction. The results show that the method can perform satisfactoryn-step-ahead prediction and is suited for implementation.

Zusammenfassung

Der Siliziumgehalt der heißen Eisenschmelze im Schmelzofen spiegelt normalerweise den thermischen Zustand des Ofens wider und beeinflusst die Feuerung. Dieser Beitrag behandelt ein hybrides neuronales Netz, das den Siliziumgehalt mitn-Schritten voraussagt. Dabei wird ein zeitverzögertes neuronales Netz verwendet, das innere Schleifen zur Darstellung der Dynamik beinhaltet. Sein Lernverfahren wurde zweistufig geteilt: Zunächst wird der Backpropagation-Algorithmus mit Vergessensfaktor vorgestellt, um den geeigneten Netzaufbau zu finden; sodann wird die Methode zur zeitlichen Differenzierung mit Vergessensfaktor zurn-Schritt-Voraussage verwendet. Die Ergebnisse zeigen, dass die Methoden-Schritt-Voraussagen zufriedenstellend erfüllt und für die Anwendung geeignet ist.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Bilsari, C., Saten, H.: Application of feed-forward neural networks for predicting silicon content in pig iron from blast furnace. Åbo Akademi, Åbo, Finland.

  2. Gaungliang, Z., Jitang, M., Bjorkman, B.: A neural network model for predicting the silicon content of the hot metal at no. 2 blast furnace of SSAB LULEA. Ironmaking Conference Proceedings (1996), pp. 211–221.

  3. Lapedes, A., Farber, R.: How neural networks work. Neural Information Processing System (1988), pp. 442–456.

  4. Narendra, K. S., Parthasarathy, K.: Identification and control of dynamical systems using neural network. IEEE Trans. Neural Network 1 (1990), pp. 4–47.

    Article  Google Scholar 

  5. Hansen, J. V., Nelson, R. D.: Neural network and traditional time series methods: A synergistic combination in state economic forecast. IEEE Trans. Neural Network 4 (1997), pp. 863–873.

    Article  Google Scholar 

  6. Waibel, A.: Modular construction of time-delay neural networks for speech recognition. Neural Computation 1 (1989), pp. 39–46.

    Article  Google Scholar 

  7. Williams, R. J., Zipser, D.: Gradient-based learning algorithm for recurrent network. In: Chauvin, Y., Rumelhart, D. E.: Back propagation: theory, architecture, and application, London: Erlbaum. 1995.

    Google Scholar 

  8. Elman, J. L.: Finding structure in time series prediction. Cognitive Science 14 (1990), pp. 179–211.

    Article  Google Scholar 

  9. Mozer, M. C.: Neural net architectures for temporal sequence processing times series prediction: forecasting the future and understanding the past, MA: Addison-Wesley. 1994.

    Google Scholar 

  10. Masumi, I.: Prediction of time series by a structural learning of neural network, fuzzy sets and system. 82 (1996), pp. 167–176.

  11. Masumi, L.: Learning of modular structured network. Artificial Intelligence 75 (1995), pp. 51–62.

    Article  Google Scholar 

  12. Russell, N. T., Bakker, H. H. C.: Modular modeling of an evaporator for long-range prediction. Expert System 11 (1997), pp. 347–355.

    Google Scholar 

  13. Sutton, R. S.: Learning to predict by the methods of temporal difference. Machine Learning 3 (1983), pp. 9–44.

    Google Scholar 

  14. Tesauro, G.: Practical issues in temporal difference learning. Machine Learning 3–4 (1992), pp. 257–277.

    MATH  Google Scholar 

  15. Sungzoon Cho, Yongjung Cho et al.: Reliable roll force prediction in cold mill using multiple neural network. IEEE Trans. Neural Network 8 (1997), 4, pp. 874–882.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. School of Information Science and Engineering, Northeastern University, Wenhua Street, PRC-110006, Shenyang, P. R. China

    Y. Wang (Ph. D. Candidate), J. Zhou (Professor) & S. Wang (Professor)

Authors
  1. Y. Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. J. Zhou
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. S. Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

About this article

Cite this article

Wang, Y., Zhou, J. & Wang, S. A hybrid neural network model for predicting silicon content. Elektrotech. Inftech. 117, 18–23 (2000). https://doi.org/10.1007/BF03161394

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF03161394

Keywords

Schlüsselwörter

Search

Navigation