Files in this item

FilesDescriptionFormat

application/pdf

application/pdf9522195.pdf (4MB)Restricted to U of Illinois
(no description provided)PDF

Description

Title:Vector-valued multidimensional signal processing and analysis in the context of fluid flows
Author(s):Zhong, Jialin
Doctoral Committee Chair(s):Huang, Thomas S.; Adrian, Ronald J.
Department / Program:Electrical and Computer Engineering
Discipline:Electrical Engineering
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):Engineering, Electronics and Electrical
Physics, Fluid and Plasma
Computer Science
Abstract:This thesis addresses two major issues in the processing and analysis of the velocity fields of fluid obtained either from experiment or from computational fluid dynamics. The first issue is recovering the velocity field from random samples in the context of the Particle Tracking Velocimetry (PTV) or Particle Imaging Velocimetry (PIV) experiment. The theory of recovering vector-valued signals from random point processes is generalized from those for scalar-valued ones. Two interpolation methods are developed based on the physics of fluids. A physically constrained optimal interpolation method is developed when the velocity field is modeled as a random field. A robust one-step interpolation method is developed when the field is modeled as a deterministic function. The second issue in this thesis is the analysis of vortex structures in turbulent fluid flows. A framework for identifying regions of vortices is established, which contains vortex structure modeling, pointwise linear approximation of flow fields, local fluid motion classification, and vortex structure extraction. The model defined in this work is a generic one, with emphasis on the global properties of vortex structures. It is shown that through this pointwise linear approximation, a flow field can be segmented into regions of different topological natures. Local fluid motion is classified by the extended critical point model or by the second invariant of the local deformation tensor. The relationship between these two types of classifiers is explicitly connected. It is also shown, that under the invariant and monotonic criteria, the second invariant II is sufficient to classify the motion into dominating rotational or dominating irrotational motion. The regions of vortex structures are extracted by assimilating spatial points of the same class of fluid motion.
Issue Date:1994
Type:Text
Language:English
URI:http://hdl.handle.net/2142/22094
Rights Information:Copyright 1994 Zhong, Jialin
Date Available in IDEALS:2011-05-07
Identifier in Online Catalog:AAI9522195
OCLC Identifier:(UMI)AAI9522195


This item appears in the following Collection(s)

Item Statistics