TIME-SERIES ANALYSIS FOR SYSTEM-IDENTIFICATION AND DIAGNOSTICSстатья
Информация о цитировании статьи получена из
Web of Science
Статья опубликована в журнале из списка Web of Science и/или Scopus
Дата последнего поиска статьи во внешних источниках: 27 мая 2015 г.
Аннотация:The problem of construction of dynamical system phase space and estimation of attractor dimension from time series is considered. Two techniques for improving the phase space construction according to Takens theorem are compared: the method proposed by Broomhead and King and another method based on the works of Neymark and co-workers. The second method is shown to be rather efficient for investigation of systems with essentially inhomogeneous attractors. An iterative filtering algorithm that allows the extraction of deterministic components of experimental data is proposed. The applcation of these methods to prediction of the time series, model fitting and attractor characteristics calculation is also considered. The use of quantitative characteristics of chaotic motion for dynamical system diagnostics is discussed.