System Identification Using Delaunay Tessellation of Self-Organizing Maps

  • Janos Abonyi
  • Ferenc Szeifert
Part of the Advances in Soft Computing book series (AINSC, volume 19)


The Self-Organizing Map (SOM) is a vector quantization method which places prototype vectors on a regular low-dimensional grid in an ordered fashion. A new method to obtain piecewise linear models of dynamic processes is presented. The operating regimes of the local linear models are obtained by the Delaunay tessellation of the codebook of the SOM. The proposed technique is demonstrated by means of the identification of a pH process.


Voronoi Diagram Delaunay Triangulation Mean Square Prediction Error Best Match Unit Delaunay Tessellation 
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.
    T. Kohonen, The self-organizing map, Proceedings of the IEEE 78 (9) (1990) 1464–1480.CrossRefGoogle Scholar
  2. 2.
    J. Abonyi, R. Babuska, F. Szeifert, Fuzzy modeling with multivariate membership functions: Gray-box identification and control design, IEEE Trans. on Systems, Man, and Cybernetics, Part B. 31(5) (2001) 755–767.Google Scholar
  3. 3.
    J. Principe, L. Wang, M. Motter, Local dynamic modleing with Self-Organizing Maps and applications to nonlinear system identification and control, Proceedings of the IEEE 86 (11) (1998) 2241–2258.CrossRefGoogle Scholar
  4. 4.
    D. Cubanski, D. Cyganski, Multivariable classification through adaptive Delaunay-based c° spline approximation, IEEE Transactions on Pattern Analysis and Machine Intelligence 17 (1995) 403–417.CrossRefGoogle Scholar
  5. 5.
    T. Ullrich, H. Tolle, Delaunay-based local model networks for nonlinear system identification, in: IASTED International Conference Applied Modelling and Simulation, Banff, Canada, 1997.Google Scholar
  6. 6.
    M. Brown, T. Ullrich, Comparison of node insertion algorithms for delaunay networks, in: Proceedings of IMACS 2nd MATHMOD Conference, Wienna, 1997, pp. 445–780.Google Scholar
  7. 7.
    S. M. Omohundro, The Delaunay triangulation and function learning, in: Technical Report 90–001, International Computer Science Institute Berkeley, California, 1989.Google Scholar
  8. 8.
    N. Bhat, T. McAvoy, Determining model structure for neural models by network stripping, Computers and Chemical Engineering 16 (1992) 271–281.CrossRefGoogle Scholar
  9. 9.
    J. Abonyi, R. Babuska, H. Verbruggen, F. Szeifert, Incorporating prior knowledge in fuzzy model identification, International Journal of Systems Science 31 (5) (2000) 657–667.MATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Janos Abonyi
    • 1
  • Ferenc Szeifert
    • 1
  1. 1.Department of Process EngineeringUniversity of VeszpremHungary

Personalised recommendations