Chaotic Systems Predictability Using Neuro-Fuzzy Systems and Neural Networks with Bred Vectors

  • Pettras Leonardo Bueno dos Santos
  • Haroldo Fraga de Campos Velho
  • Rosangela Cintra
  • Sandra Sandri
Chapter
Part of the Studies in Fuzziness and Soft Computing book series (STUDFUZZ, volume 317)

Abstract

The predictability of the behavior of chaotic systems is of great importance because many real-world phenomena have some type of chaotic regime. In chaotic systems, small changes in the initial conditions can lead to very different results from the original system trajectory. The prediction of chaotic systems behavior is usually very difficult, particularly in practical applications in which initial conditions are obtained by measurement instruments, very often subject to acquisition errors. Here we use “bred vectors” methodology to generate pairs of input/output that are then used to train Neural Networks and Neuro-Fuzzy Systems. We apply the approach to predict regime change for Lorenz strange attractors and the nonlinear coupled three-waves problem from solar physics.

Keywords

Chaotic systems Bred vectors Lorenz attractor Three-waves problem Neural networks Neuro-fuzzy systems 

Notes

Acknowledgment

The authors thank FAPESP and CNPq for research support.

References

  1. 1.
    Lorenz, E.N.: Deterministic non-periodic flow. J. Atmos. Sci. 20(130–141), 1963 (1963)Google Scholar
  2. 2.
    Toth, Z., Kalnay, E.: Ensemble forecasting at NCEP and the breeding method. Mon. Weather Rev. 126, 3292–3302 (1997)Google Scholar
  3. 3.
    Cintra, R.S., Campos Velho, H.F.: Predictability for a chaotic solar plasma system. In: Symposia: Advanced Modeling on Computational Fluid Dynamics (2008)Google Scholar
  4. 4.
    Evans, E., Bathi, N.K., Kinney, J., Pann, L., Peña, M., Yang, S.-C., Kalnay, E., Hansen, J.: Rise undergraduates find that regime changes in Lorenz’s model are predictable. Bull. Amer. Meteor. Soc. 85, 521–524 (2004)Google Scholar
  5. 5.
    Pasini, A., Pelino, V.: Can we estimate atmospheric predictability by performance of neural network forecasting? The toy case studies of unforced and forced lorenz models. In: Proceedings of CIMSA’2005 (2005)Google Scholar
  6. 6.
    Shing, J., Jang, R.: ANFIS: adaptive-network-based fuzzy inference systems. IEEE Trans. Syst. Man Cybern. 23 714–723 (1993)Google Scholar
  7. 7.
    Haykin, S.: Neural Networks: A Comprehensive Foundation. Prentice Hall, New Jersey (1999)MATHGoogle Scholar
  8. 8.
    Chian, A.C.-L., Lopes, S.R., Alves, M.V.: Nonlinear excitation of Langmuir and Alfvén waves by auroral whistler waves in the planetary magnetosphere. Astron. Astrophys. 288, 981–984 (1994)Google Scholar
  9. 9.
    Guégan, D., Leroux, J.: Predicting chaos with Lyapunov exponents: zero plays no role in forecasting chaotic systems. In: Tlelo-Cuautle, E. (Ed.) Chaotic Systems. InTech. http://www.intechopen.com/articles/show/title/predicting-chaos-with-lyapunov-exponents-zero-plays-no-role-in-forecasting-chaotic-systems Accesseed 04 mar 2011
  10. 10.
    Alligood, K.T., Sauer, T.D., Yorke, J.A.: Chaos: An Introduction to Dynamical Systems. Springer, New York (1996)Google Scholar
  11. 11.
    Kalnay, E.: Atmospheric Modeling, Data Assimilation and Predictability. Cambridge University Press, Cambridge (2005)Google Scholar
  12. 12.
    Kalnay, E., Peña, M., Yang, S.-C., Cai, M.:Breeding and predictability in coupled Lorenz models. In: Proceedings of the ECMWF (2002)Google Scholar
  13. 13.
    Meunier, C., Bussac, M.N., Laval, G.: Intermittency at the onset of stochasticity in nonlinear resonant coupling processes. Physica D4 236–243 (1982)Google Scholar
  14. 14.
    Lopes, S.R., Chian, A.-L.: A coherent nonlinear theory of auroral Langmuir-Alfvén-whistler (LAW) events in the planetary magnetosphere. s 365, 669–676 (1996)Google Scholar
  15. 15.
    WEKA Ð Waikato Environment for Knowledge Analysis. http://www.cs.waikato.ac.nz/ml/weka/ Acessed 25 jul 2012
  16. 16.
    Gomide, F., Pedrycz, W.: An Introduction to Fuzzy Sets. MIT Press, Cambridge (1998)Google Scholar
  17. 17.
    Lin, C.-T., Lee, C.S.G.: Neural Fuzzy Systems: A Neuro-Fuzzy Synergism to Intelligent Systems. Prentice Hall, Upper Saddle River (1996)Google Scholar
  18. 18.
    dos Santos, P.L.B., Sandri, S., de Campos Velho, H.F.: Previsibilidade em sistemas ca—óticos utilizando o sistema Neuro-Difuso ANFIS (in Portuguese). In: II Congresso Brasileiro de Sistemas Fuzzy (CBSF), pp. 113–126. Natal, Brazil (2012)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Pettras Leonardo Bueno dos Santos
    • 1
  • Haroldo Fraga de Campos Velho
    • 1
  • Rosangela Cintra
    • 1
  • Sandra Sandri
    • 1
  1. 1.The Brazilian National Institute for Space Research (INPE)São José dos CamposBrazil

Personalised recommendations