From video to computation of biological fluid–structure interaction problems

This work deals with the techniques necessary to obtain a purely Eulerian procedure to conduct CFD simulations of biological systems with moving boundary flow phenomena. Eulerian approaches obviate difficulties associated with mesh generation to describe or fit flow meshes to body surfaces. The chal...

Full description

Saved in:
Bibliographic Details
Published inTheoretical and computational fluid dynamics Vol. 30; no. 1-2; pp. 41 - 66
Main Authors Dillard, Seth I., Buchholz, James H. J., Udaykumar, H. S.
Format Journal Article
LanguageEnglish
Published Berlin/Heidelberg Springer Berlin Heidelberg 01.04.2016
Springer
Springer Nature B.V
Subjects
Analysis
Biological
Biological systems
Boundaries
Boundary conditions
Boundary layer
Classical and Continuum Physics
Computation
Computational fluid dynamics
Computational Science and Engineering
Dynamical systems
Eels
Engineering
Engineering Fluid Dynamics
Finite element method
Fluid dynamics
Fluid mechanics
Fluid-structure interaction
Hydrodynamics
Image processing
Mesh generation
Original Article
Swimming
United States
Cartesian grid methods
Image-based modeling
Morphing
Level sets
Moving boundaries
Optical flow
Online AccessGet full text

Cover

More Information
Summary:This work deals with the techniques necessary to obtain a purely Eulerian procedure to conduct CFD simulations of biological systems with moving boundary flow phenomena. Eulerian approaches obviate difficulties associated with mesh generation to describe or fit flow meshes to body surfaces. The challenges associated with constructing embedded boundary information, body motions and applying boundary conditions on the moving bodies for flow computation are addressed in the work. The overall approach is applied to the study of a fluid–structure interaction problem, i.e., the hydrodynamics of swimming of an American eel, where the motion of the eel is derived from video imaging. It is shown that some first-blush approaches do not work, and therefore, careful consideration of appropriate techniques to connect moving images to flow simulations is necessary and forms the main contribution of the paper. A combination of level set-based active contour segmentation with optical flow and image morphing is shown to enable the image-to-computation process.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0935-4964
1432-2250
DOI:10.1007/s00162-015-0358-5