Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Civil-Comp Proceedings
ISSN 1759-3433 CCP: 85
PROCEEDINGS OF THE FIFTEENTH UK CONFERENCE OF THE ASSOCIATION OF COMPUTATIONAL MECHANICS IN ENGINEERING Edited by: B.H.V. Topping
Paper 28
Three-Dimensional Imaging Techniques for Material Characterisation V. Bui Xuan1, P. Young1, G. Tabor1 and G. Limbert2
1School of Engineering, Computer Science and Mathematics, University of Exeter, United Kingdom
V. Bui Xuan, P. Young, G. Tabor, G. Limbert, "Three-Dimensional Imaging Techniques for Material Characterisation", in B.H.V. Topping, (Editor), "Proceedings of the Fifteenth UK Conference of the Association of Computational Mechanics in Engineering", Civil-Comp Press, Stirlingshire, UK, Paper 28, 2007. doi:10.4203/ccp.85.28
Keywords: non destructive testing, finite element, finite volume, image based meshing, material characterisation, 3D imaging.
Summary
New image-based techniques have been developed allowing for robust and accurate meshing of objects which are obtained via volume scanning modalities (such as MRI, CT and XMT). These approaches are valuable tools for reconstructing the geometry of complex objects, and enable a whole new class of problems to be tackled effectively using physics based simulation tools. In particular these techniques have opened up exciting new avenues in materials research, particularly in materials characterization and non destructive testing.
A selection of case studies will demonstrate the potential of these new tools for characterising the behaviour of microstructures (e.g. foams, composites, etc.), and illustrate the ability to model topologically complex problems with a high degree of accuracy robustly and straightforwardly. In particular the use of these meshing techniques as the basis for both structural and hydro-dynamic analysis of architectures of extremely complex topology is illustrated by exploring the behaviour of auxetic foam based on high resolution image data. Finally cooling of a mince pie in a light breeze is modelled as an example of coupled fluid flow-thermal analysis. Automated mesh construction from 3D imaging is one of the most interesting new developments in computational mechanics. Because of their inherent geometrical complexity, the construction of even generic models of typical micro-structures is time-consuming and potentially inaccurate. Most previous approaches to this problem have concentrated on generating surface representations for the outer boundary and have left the mesh generation as a separate step. In contrast, our approach has been to generate the mesh directly from 3D images as part of the image analysis process. This approach leads to simpler and more robust multi-part meshes which are suitable for modelling using a range of different physics based simulation methods including coupled fluid structure interaction (FSI) problems. In addition, the accuracy of models generated is contingent only on the accuracy of the original imaging data providing the most faithful and realistic models possible from the image data. The present study illustrates the ease with which FE and FV models of microstructures of arbitrarily complex topology created based on 3D tomographic images (MicroCT) can be straightforwardly generated and used to carry out (large deformation) compression simulations using ABAQUS and flow analyses using FLUENT. This shows the potential of image based meshing techniques coupled with physics based simulation tools to provide accurate, fast, robust and practical tools to explore the bulk structural properties of foams and composites materials based on micro-structural data. Future studies will explore comparing quantitative data obtained from experimental, analytical and numerical analyses.
purchase the full-text of this paper (price £20)
go to the previous paper |
|