Data symbolization, derived from the study of symbolic dynamics, involves discretization of measurement data to aid in observing and characterizing temporal patterns. In this study, symbolization-based methods are developed for analysis of time series from experimental engineering systems to test hypotheses concerning stationarity, temporal reversibility, and synchronization. Stationarity is examined in the context of process control and dynamical state matching; temporal reversibility, in the context of model discrimination and selection of control schemes (linear versus nonlinear); and synchronization, in the context of modes of interactions between system components. Statistical significance is estimated using the method of surrogate data with Monte Carlo probabilities.
Please email comments about this page to firstname.lastname@example.org.
This page's URL is http://www-chaos.engr.utk.edu/abs/abs-finney-phd.html.
Updated: 2001-12-28 ceaf