A System Identification Method Based on Multi-layer Perception and Model Extraction

  • Chang Hu
  • Li Cao
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3174)


Artificial Neural Networks (ANNs) have provided an interesting and labor-saving approach for system identification. However, ANNs fall short when an explicit model is needed. In this paper, a method of getting the explicit model by extracting it from a trained ANN is proposed. To identify a system, a Multi-Layer Perceptron (MLP) is constructed, trained and a polynomial model is extracted from the trained network. This method is tested in the experiments and shows its capability for system identification, compared with the Least Squared method.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    McLoone, S.: Neural Network Identification: a Survey of Gradient Based Methods; Optimisation in Control: Methods and Applications (Ref. No. 1998/521), IEE Colloquium on, November 10, 4/1–4/4 (1998)Google Scholar
  2. 2.
    Setiono, R., Leow, W.K., Zurada, J.M.: Extraction of Rules from Artificial Neural Networks for Nonlinear Regression. IEEE Transactions on Neural Networks 13(3), 564–577 (2002)CrossRefGoogle Scholar
  3. 3.
    Stubberud, A., Wabgaonkar, H., Stubberud, S.: A Neural-Network-Based System Identification Technique. In: Proceedings of the 30th IEEE Conference on Decision and Control, December 11-13, vol. 1, pp. 869–870 (1991)Google Scholar
  4. 4.
    Muller, C., Mangeas, M.: Neural Networks and Times Series Forecasting: a Theoretical Approach, Systems, Man and Cybernetics. In: Proceedings of International Conference on Systems Engineering in the Service of Humans, October 17-20, vol. 2, pp. 590–594 (1993)Google Scholar
  5. 5.
    Xiao, D.Y., Fang, C.Z.: Process Identification, August 1988, pp. 160–162. Tsinghua University Press, Beijing (1988)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Chang Hu
    • 1
  • Li Cao
    • 1
  1. 1.Department of AutomationTsinghua UniversityBeijingP.R.China

Personalised recommendations