Predicting physical variables in time-delay embedding
- Department of Physics and Marine Physical Laboratory, Scripps Institution of Oceanography, University of California, San Diego, Mail Code 0402, La Jolla, California 92093-0402 (United States)
- U.S. Naval Research Laboratory, Code 6341, Washington, D.C. 20375-5000 (United States)
- Institute for Nonlinear Science, University of California, San Diego, Mail Code 0402, La Jolla, California 92093-0402 (United States)
In time-delay reconstruction of chaotic attractors we can accurately predict the short-term future behavior of the observed variable [ital x]([ital t])=[ital x]([ital n])=[ital x]([ital t][sub 0]+[tau][sub [ital s]n]) without prior knowledge of the equations of motion by building local or global models in the state space. In many cases we also want to predict variables other than the one which is observed and require methods for determining models to predict these variables in the same space. We present a method which takes measurements of two variables [ital x]([ital n]) [ital and] [ital z]([ital n]) and builds models for the determination of [ital z]([ital n]) in the phase-space made out of the [ital x]([ital n]) and its time lags. Similarly we show that one may produce models for [ital x]([ital n]) in the [ital z]([ital n]) space, except where special symmetries prevent this, such as in the familiar Lorenz model. Our algorithm involves building local polynomial models in the reconstructed phase space of the observed variable of low order (linear or quadratic) which approximate the function [ital z]([ital n])=[ital F]([bold x]([ital n])) where [bold x]([ital n]) is a vector constructed from a sequence of values of observed variables in a time delay fashion. We train the models on a partial data set of measured values of both [ital x]([ital n]) and [ital z]([ital n]) and then predict the [ital z]([ital n]) in a recovery set of observations of [ital x]([ital n]) alone. In all of our analyses we assume that the observed data alone are available to us and that we possess no knowledge of the dynamical equations. We test this method on the numerically generated data set from the Lorenz model and also on a number of experimental data sets from electronic circuits.
- DOE Contract Number:
- FG03-90ER14138
- OSTI ID:
- 5238997
- Journal Information:
- Physical Review. E, Statistical Physics, Plasmas, Fluids, and Related Interdisciplinary Topics; (United States), Journal Name: Physical Review. E, Statistical Physics, Plasmas, Fluids, and Related Interdisciplinary Topics; (United States) Vol. 49:3; ISSN PLEEE8; ISSN 1063-651X
- Country of Publication:
- United States
- Language:
- English
Similar Records
Local false nearest neighbors and dynamical dimensions from observed chaotic data
Nested strange attractors in spatiotemporal chaotic systems