CASPIAN SEA LEVEL PREDICTION USING ARTIFICIAL NEURAL NETWORK AND EMPIRICAL MODE DECOMPOSITION
https://doi.org/10.24057/2071-9388-2010-3-4-25-31
Abstract
This paper demonstrates the possibility of using nonlinear modeling for prediction of the Caspian Sea level. Phase space geometry of the of a model can be reconstructed by the embedology methods. Dynamical invariants, such as the Lyapunov exponents, the Kaplan-Yorke dimension, and the prediction horizon were estimated from reconstruction. Fractal and multifractal analyses were carried out for various time intervals of the Caspian Sea level and multifractal spectra were calculated. Then, historical data resolution was improved with the help of fractal approximation. The EMD method was used to reduce noise of the time series. Global nonlinear predictions were made with the help of Artificial Neural Network for combinations of different empirical modes.
About the Authors
Nikolai MakarenkoRussian Federation
Leading scientist, Pulkovo Astronomical Observatory, Saint-Petersburg, Russia
Institute of Mathematics, Pushkin str. 200010, Almaty, Kazakhstan
Lyailya Karimova
Kazakhstan
Principal scientist, Institute of Mathematics, Pushkin str. 200010, Almaty, Kazakhstan
Olga Kruglun
Kazakhstan
Senior scientist, Institute of Mathematics, Pushkin str. 200010, Almaty, Kazakhstan
References
1. Barnsley, M.F. (2000) Fractals Everywhere. Hawley Rising, 550 p.
2. Bishop, Ch.M., (2006) Pattern Recognition and Machine Learning. Springer, 738 p.
3. Cochran,W.O., Hart J.C., Patrick J. F. (1998) On Approximating Rough Curves With Fractal
4. Functions. Proceedings of the Graphics Interface 1998 Conference, June 18–20, 1998,
5. Vancouver, BC, Canada. pp. 65–72.
6. Flandrin, P., Gonçalves, P. (2004) Empirical mode decompositions as data-driven wavelet-like
7. expansions. Inter. Journ. of Wavelets, Multiresolution and Information Processing. Vol. 2, pp. 1–20.
8. Golitsyn, G.S. (1995) Changes in the Caspian Sea Level as aProblem for Diagnosing and
9. Predicting Regional Climatic Changes. Izv. RAN, Fiz. Atm. Okeana Vol. 31, pp. 385–391. (in
10. Russian).
11. Halsey, T.C., Jensen, M.H., Kadanoff, L.P., Procaccia, I., Schraiman, B.I. (1968) Fractal measures and
12. their singularities: the characterizations of strange sets. Phys. Rev. A., Vol. 33, pp. 1141–1151.
13. Huang, N.E., Shen, Zh., Long, St.R., and et al. (1998) The empirical mode decomposition
14. and the Hilbert spectrum for nonlinear and non-stationary time series analysis Proc. R. Soc.
15. Lond. A., Vol. 454, pp. 903–995.
16. Karimova, L., Mukhamejanova, S., Makarenko, N. (2003) Fractal geometry methods and
17. neurocomputing for Caspian Sea level forecasting. EGS – AGU – EUG Joint Assembly,
18. Abstracts from the meeting held in Nice, France, 6–11 April 2003, abstract #186.
19. Karimova, L.M., Kuandykov, Y.B., Makarenko, N.G., Novak, M.M., Helama, S., (2007) Fractal
20. and topological dynamics for the analysis of paleoclimatic records. Physica A: Statistical
21. Mechanics and its Applications, N 373, pp. 737–746.
22. Kozhevnikova, I.A., Shveikina, V.I. (2008) Nonlinear dynamics of level in the Caspian Sea.
23. Water Resources, Vol. 35, pp. 297–304.
24. Makarenko, N.G., Karimova, L.M., Kyandykov, Y.B., Novak, M.M. (2004) Nonlinear Dynamics
25. and Prediction of the Caspian Sea Level. Thinking in Patterns, M. M. Novak (ed.). World
26. Scientific, pp. 91–102.
27. Makarenko, N.G. (2003) Embedology and Neural Networks, Moscow. Lectures on
28. neuroinformatics. “ Neuroinformatics-2003”, part. 1, pp. 86–148, (in Russian).
29. Poggio, T., Girosi, F. A (1989) Theory of networks for approximation and learning. MIT AI
30. Lab. Technical Report, Memo No. 1140, Paper No. 31.
31. Riedi, R. H., Scheuring, I., (1997) Conditional and Relative Multifractal Spectra. Fractals, N 5,
32. pp. 153–168.
33. Sauer, T., Yorke, J.A., Casdagli, M. (1991) Embedology. J. Statist. Phys., Vol. 65, pp. 579–616.
34. Schaw, R. (1981) Strange attractors, chaotic behavior, and information flow. Z. Naturforsch,
35. Vol. 36a, pp. 80–112.
Review
For citations:
Makarenko N., Karimova L., Kruglun O. CASPIAN SEA LEVEL PREDICTION USING ARTIFICIAL NEURAL NETWORK AND EMPIRICAL MODE DECOMPOSITION. GEOGRAPHY, ENVIRONMENT, SUSTAINABILITY. 2010;3(4):25-31. https://doi.org/10.24057/2071-9388-2010-3-4-25-31
Views:
755
Email this article 