Abstract
Black box reconstruction is both the most difficult and the most tempting modelling problem when any prior information about an appropriate model structure is lacking. An intriguing thing is that a model capable of reproducing an observed behaviour or predicting further evolution should be obtained only from an observed time series, i.e. “from nothing” at first sight. Chances for a success are not large. Even more so, a “good” model would become a valuable tool to characterise an object and understand its dynamics. Lack of prior information causes one to utilise universal model structures, e.g. artificial neural networks, radial basis functions and algebraic polynomials are included in the right-hand sides of dynamical model equations. Such models are often multi-dimensional and involve quite many free parameters.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
The set M is a compact smooth manifold and the quantity d is its topological dimension (Sect. 10.1.1). There are generalisations of the theorem to the case of non-smooth sets M and fractal dimension d, which are beyond the scope of our discussion (Sauer et al., 1991). We note that the set M mentioned in the theorems does not inevitably correspond to the motion on an attractor. For instance, let an attractor be a limit cycle C “reeled” on a torus M. If one is interested only in the description of an established periodic motion on the cycle C, then it is sufficient to use \(D>2\) model variables for reconstruction according to Takens’ theorems. If one needs to describe motions on the entire torus M, including transient processes, then it is necessary to use \(D>4\) variables. In practice, one often has a single realisation corresponding to established dynamics. Therefore, one usually speaks of the reconstruction of an attractor.
- 2.
An equivalent description of a motion on a limit cycle is assured for \(D=3\) even if the cycle “lives” in an infinite-dimensional phase space.
- 3.
If an object is a map \({\textbf{y}}(t_{n+1})=\textbf{F}(\textbf{y}(t_{n}))\), then an evolution operator \(\Phi_{\Delta t}(\textbf{y}(t_{0}))\) is just the function F. If an object is a set of ODEs \({\mathrm{d}}\textbf{y}/{\mathrm{d}}t=\textbf{F}(\textbf{y}(t))\), then the function \(\Phi_{t}(\textbf{y}(t_{0}))\) is the result of the integration of the ODEs over a time interval of length t. If an original system is given by a partial differential equation \(\partial\textbf{y}/\partial t=\textbf{F}(\textbf{y},\partial\textbf{y}/\partial\textbf{r},\partial^{2}\textbf{y}/\partial\textbf{r}^{2},\ldots)\), where r is a spatial coordinate, then y is a vector belonging to an infinite-dimensional space of functions y(r) and Φ t is an operator acting in that space.
- 4.
It means that for the same segments \([\eta(t),\eta(t+\tau),\ldots,\eta(t+(D-1)\tau)]\) encountered at different time instants t, one observes the same continuation (i.e. the same future). It gives a justification to the predictive method of analogues applied already by E. Lorenz. The method is based on the search of the time series segments, which “resemble” a current segment, in the past and subsequent usage of a combination of their “futures” as a forecast. In a modern formulation, it is realised with local models (Sect. 10.2.1).
- 5.
For \(l=1\) and a small sampling interval Δ t, a reconstructed phase orbit stretches along the main diagonal, since it appears that \(x_{1}\approx x_{2}\approx \ldots\approx x_{D}\).
References
Abarbanel, H.D.I., Brown, R., Kadtke, J.B.: Prediction and system identification in chaotic nonlinear systems: time series with broadband spectra. Phys. Lett. A. 138, 401–408 (1989)
Anosov, O.L., Butkovsky, O.Ya., Kravtsov, Yu.A.: Limits of predictability for linear autoregression models. J. Communications Technol. Electron. 40(12), 1866–1873, (in Russian) (1995)
Bär, M., Hegger, R., Kantz, H.: Fitting partial differential equations to space-time dynamics. Phys. Rev. E. 59, 337–343 (1999)
Bezruchko, B.P., Smirnov, D.A.: Constructing nonautonomous differential equations from a time series. Phys. Rev. E. 63, 016207, (2001)
Broomhead, D.S., King, G.P.: Extracting qualitative dynamics from experimental data. Physica D. 20, 217–236 (1986)
Broomhead, D.S., Lowe, D.: Multivariable functional interpolation and adaptive networks. Complex Syst. 2, 321–355 (1988)
Brown, R., Rulkov, N.F., Tracy, E.R.: Modeling and synchronizing chaotic systems from experimental data. Phys. Lett. A. 194, 71–76 (1994)
Cao, L., Mees, A.I., Judd, K.: Dynamics from multivariate time series. Phys. D. 121, 75–88 (1998)
Casdagli, M.: Nonlinear prediction of chaotic time series. Phys. D. 35, 335–356 (1989)
Casdagli, M., Eubank, S., Farmer, J.D., Gibson, J.: State space reconstruction in presence of noise. Phys. D. 51, 52–98 (1991)
Cellucci, C.J., Albano, A.M., Rapp, P.E.: Comparative study of embedding methods. Phys. Rev. E. 67, 066210 (2003)
Eckmann, J.P., Ruelle, D.: Ergodic theory of chaos and strange attractors. Rev. Mod. Phys. 57, 617–656 (1985)
Farmer, J.D., Sidorowich, J.J.: Predicting chaotic time series. Phys. Rev. Lett. 59, 845–848 (1987)
Fraser, A.M., Swinney, H.L.: Independent coordinates for strange attractors from mutual information. Phys. Rev. A. 33, 1131–1140 (1986)
Gerschenfeld, N.A., Weigend, A.S. (eds.): Time Series Prediction: Forecasting the Future and Understanding the Past. SFI Studies in the Science of Complexity, Proc. V. XV. Addison-Wesley, New York (1993)
Gibson, J.F., Farmer, J.D., Casdagli, M., Eubank, S.: An analytic approach to practical state space reconstruction. Phys. D. 57, 1–30 (1992)
Giona, M., Lentini, F., Cimagalli, V.: Functional reconstruction and local prediction of chaotic time series. Phys. Rev. E. 44, 3496–3502 (1991)
Gliklikh Yu.E.: What a smooth manifold is. Soros Educ. J. 11, 155–159, (in Russian) (1998)
Gorban’ A.N.: Multivariable functions and neural networks. Soros Educational J. No. 12, 105–112, (in Russian) (1998)
Gouesbet, G.: Reconstruction of the vector fields of continuous dynamical systems from scalar time series. Phys. Rev. A. 43, 5321–5331 (1991)
Gouesbet, G., Letellier, C.: Global vector-field approximation by using a multivariate polynomial \(L_{2}\) approximation on nets. Phys. Rev. E. 49, 4955–4972 (1994)
Gouesbet, G., Meunier-Guttin-Cluzel, S., Ménard, O.: Global reconstructions of equations of motion from data series, and validation techniques, a review. In: Gouesbet, G., Meunier-Guttin-Cluzel, S., Ménard, O. (eds.) Chaos and Its Reconstructions, pp. 1–160. Nova Science Publishers, New York, (2003)
Grassberger, P., Procaccia, I.: Measuring the strangeness of strange attractors. Phys. D. 9, 189–208 (1983)
Gribkov, D.A., Gribkova, V.V., Kravtsov Yu.A., et al. Reconstructing structure of dynamical system from time series. J. Commun. Technol. Electron. 39(2), 269–277, (in Russian) (1994)
Hubner, U., Weiss, C.-O., Abraham, N.B., Tang, D.: Lorenz-like chaos in \(\mathit{NH}_{3}\) – FIR lasers (data set A). In: Gerschenfeld, N.A., Weigend, A.S. (eds.) Time Series Prediction: Forecasting the Future and Understanding the Past. SFI Studies in the Science of Complexity, Proc. V. XV, pp. 73–104. Addison-Wesley, Reading, MA (1993)
Janson, N.B., Pavlov, A.N., Anishchenko, V.S.: One method for restoring inhomogeneous attractors. Int. J. Bif. Chaos. 8, 825–833 (1998)
Johnson, L.W., Riess, R.D.: Numerical Analysis, 2nd edn. Addison-Wesley, Reading, MA (1982)
Judd, K., Mees, A.I. On selecting models for nonlinear time series. Phys. D. 82, 426–444 (1995)
Judd, K., Mees, A.I.: Embedding as a modeling problem. Phys. D. 120, 273–286 (1998)
Judd, K., Small, M.: Towards long-term prediction. Phys. D. 136, 31–44 (2000)
Kalitkin, N.N.: Numerical Methods. Nauka, Moscow, (in Russian) (1978)
Kantz, H., Schreiber, T.: Nonlinear Time Series Analysis. Cambridge University Press, Cambridge (1997)
Kartsev, V.P.: Adventures of Great Equations. Znaniye, Moscow, (in Russian) (1976)
Kennel, M.B., Brown, R., Abarbanel, H.D.I.: Determining embedding dimension for phase-space reconstruction using a geometrical construction. Phys. Rev. A. 45, 3403–3411 (1992)
Kravtsov Yu.A.: Randomness, determinacy, predictability. Phys. Uspekhi. 158(1), 93–115, (in Russian) (1989)
Kugiumtzis, D., Lingjaerde, O.C., Christophersen, N.: Regularized local linear prediction of chaotic time series. Phys. D. 112, 344–360 (1998)
Landa, P.S., Rosenblum, M.G.: Comparison of methods for construction of phase space and determination of attractor dimension from experimental data. Tech. Phys. 59(11), 1–8, (in Russian) (1989)
Letellier, C., Aguirre, L.A.: Investigating nonlinear dynamics from time series: The influence of symmetries and the choice of observables. Chaos. 12, 549–558 (2002)
Letellier, C., Le Sceller, L., Gouesbet, G., et al.: Recovering deterministic behavior from experimental time series in mixing reactor. AIChE J. 43(9), 2194–2202 (1997)
Letellier, C., Le Sceller, L., Maréchal, E., et al.: Global vector field reconstruction from a chaotic experimental signal in copper electrodissolution. Phys. Rev. E. 51, 4262–4266 (1995)
Letellier, C., Maquet, J., Labro, H., et al. Analyzing chaotic behavior in a Belousov-Zhabotinskyi reaction by using a global vector field reconstruction. J. Phys. Chem. 102, 10265–10273 (1998a)
Letellier, C., Macquet, J., Le Sceller, L., et al. On the non-equivalence of observables in phase space reconstructions from recorded time series. J. Phys. A: Math. Gen. 31, 7913–7927 (1998b)
Liebert, W., Schuster, H.G.: Proper choice the of time delay for the analysis of chaotic time series. Phys. Lett. A. 142, 107–111 (1989)
Makarenko, N.G.: Fractals, attractors, neural networks and so forth. Procs. IV All-Russian Conf. “Neuroinformatics-2002”. Part 2, pp. 121–169. Moscow, (in Russian) (2002)
Makarenko, N.G.: Embedology and Neuro-Prediction. Procs. V All-Russian Conf. “Neuroinformatics-2003”. Part 1, pp. 86–148. Moscow, (in Russian) (2003)
Malinetsky, G.G., Potapov, A.B.: Contemporary Problems of Nonlinear Dynamics. Editorial URSS, Moscow, (in Russian) (2000)
Parlitz, U., Mayer-Kress, G.: Predicting low-dimensional spatiotemporal dynamics using discrete wavelet transforms. Phys. Rev. E. 51, R2709–R271 (1995)
Parlitz, U., Merkwirth, C.: Prediction of spatiotemporal time series based on reconstructed local states. Phys. Rev. Lett. 84, 1890–1893 (2000)
Pavlov, A.N., Janson, N.B., Anishchenko, V.S.: Application of statistical methods to solve the problem of global reconstruction. Tech. Phys. Lett. 23(4), 297–299 (1997)
Pavlov, A.N., Janson, N.B., Anishchenko, V.S.: Reconstruction of dynamical systems. J. Commun. Technol. Electron. 44(9), 999–1014 (1999)
Press, W.H., Flannery, B.P., Teukolsky, S.A., Vetterling, W.T.: Numerical Recipes in C. Cambridge University Press, Cambridge (1988)
Rulkov, N.F., Sushchik, M.M., Tsimring, L.S., Abarbanel, H.D.I.: Generalized synchronization of chaos in directionally coupled chaotic systems. Phys. Rev. E. 51, 980–994 (1995)
Samarsky, A.A.: Introduction to Numerical Methods. Nauka, Moscow, (in Russian) (1982)
Sauer, T., Yorke, J.A., Casdagli, M.: Embedology. J. Stat. Phys. 65(3–4), 579–616 (1991)
Sauer, T.: Time series prediction by using delay coordinate embedding. in Time Series Prediction: Forecasting the Future and Understanding the Past. Eds. N.A. Gerschenfeld, A.S. Weigend. SFI Studies in the Science of Complexity, Proc. V. XV, pp. 175–193. Addison-Wesley (1993)
Schroer, C., Sauer, T., Ott, E., Yorke, J.: Predicting chaos most of the time from embeddings with self-intersections. Phys. Rev. Lett. 80, 1410–1413 (1998)
Sitz, A., Kurths, J., Voss, H.U.: Identification of nonlinear spatiotemporal systems via partitioned filtering. Phys. Rev. E. 68, 016202 (2003)
Sitz, A., Schwartz, U., Kurths, J., Voss, H.U.: Estimation of parameters and unobserved components for nonlinear systems from noisy time series. Phys. Rev. E. 66, 016210 (2002)
Small, M., Judd, K., Mees, A.: Modelling continuous processes from data. Phys. Rev. E. 65, 046704 (2002)
Small, M., Judd, K.: Comparisons of new nonlinear modeling techniques with applications to infant respiration. Physica, D. 117, 283–298 (1998)
Small, M., Tse, C.K.: Optimal embedding: A modelling paradigm. Phys. D. 194, 283–296 (2004)
Smirnov, D.A., Bezruchko, B.P., Seleznev Ye.P.: Choice of dynamical variables for global reconstruction of model equations from time series. Phys. Rev. E. 65, 026205 (2002)
Smirnov, D.A., Bezruchko, B.P.: Nonlinear dynamical models from chaotic time series: methods and applications. In: Winterhalder, M., Schelter, B., Timmer, J. (eds.) Handbook of Time Series Analysis, pp. 181–212. Wiley-VCH, Berlin (2006)
Smith, L.A.: Identification and prediction of low-dimensional dynamics. Phys. D. 58, 50–76 (1992)
Smith, L.A.: Maintenance of uncertainty. Proc. Int. School of Physics “Enrico Fermi”, Course CXXXIII, pp. 177–246. Italian Physical Society, Bologna, (1997). Available at http://www.maths.ox.ac.uk/∼lenny
Stark, J., Broomhead, D.S., Davies, M., Huke, J.: Takens embedding theorem for forced and stochastic systems. Nonlinear Analysis. Theory, Methods, and Applications. Proc. 2nd Congress on Nonlinear Analysis. Elsevier Science Ltd., 30(8), 5303–5314 (1997)
Stoer, J., Bulirsch, R.: Introduction to Numerical Analysis. Springer, New York (1980)
Takens, F.: Detecting strange attractors in turbulence. Lect. Notes Math. 898, 366–381 (1981)
Timmer, J.: Parameter estimation in nonlinear stochastic differential equations. Chaos, Solitons Fractals 11, 2571–2578 (2000)
Tong, H.: Nonlinear Time Series Analysis: a Dynamical System Approach. Oxford University Press, Oxford (1990)
Voss, H.U., Bünner, M., Abel, M.: Identification of continuous, spatiotemporal systems. Phys. Rev. E. 57, 2820–2823 (1998)
Wan, E.A.: Time series prediction by using a connectionist network with internal delay lines. In: Gerschenfeld, N.A., Weigend, A.S. (eds.) Time Series Prediction: Forecasting the Future and Understanding the Past. SFI Studies in the Science of Complexity, Proc. V. XV, pp. 195–217. Addison-Wesley (1993)
Yule, G.U.: On a method of investigating periodicities in disturbed series, with special reference to Wolfer’s sunspot numbers. Phil. Trans. R. Soc. Lond. A. 226, 267–298 (1927)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Bezruchko, B.P., Smirnov, D.A. (2010). Model Equations: “Black Box” Reconstruction. In: Extracting Knowledge From Time Series. Springer Series in Synergetics. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-12601-7_10
Download citation
DOI: https://doi.org/10.1007/978-3-642-12601-7_10
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-12600-0
Online ISBN: 978-3-642-12601-7
eBook Packages: Physics and AstronomyPhysics and Astronomy (R0)