Premium
Morphology based domain partitioning of multi‐phase materials: a preprocessor for multi‐scale modelling
Author(s) -
Valiveti D. M.,
Ghosh Somnath
Publication year - 2006
Publication title -
international journal for numerical methods in engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.421
H-Index - 168
eISSN - 1097-0207
pISSN - 0029-5981
DOI - 10.1002/nme.1834
Subject(s) - domain (mathematical analysis) , scale (ratio) , characterization (materials science) , preprocessor , representation (politics) , resolution (logic) , phase (matter) , computer science , biological system , algorithm , mathematical morphology , materials science , image (mathematics) , image processing , artificial intelligence , nanotechnology , mathematics , chemistry , mathematical analysis , physics , organic chemistry , quantum mechanics , politics , political science , law , biology
This paper develops a microstructural morphology‐based domain partitioning method (MDP) as a comprehensive pre‐processor for multi‐scale simulation of heterogeneous multi‐phase materials. The MDP method systematically creates a multi‐scale image simulation–characterization methodology to enable domain partitioning that can delineate homogenizable regions from those where explicit representation of phases is necessary for analysis. The methods are strictly based on geometric features of the material morphology and do not use any mechanical response functions. The first step in this development simulates high‐resolution microstructural information from low resolution images of the domain and only a limited high resolution micrographs from optical or scanning electron microscopy. The second step uses quantitative characterization of these high resolution images with, e.g. phase distribution functions, to create effective metrics that can relate microstructural features to the material's physical behaviour. The third step invokes domain partitioning to demarcate regions corresponding to different length scales in a concurrent multi‐scale model. Partitioning criteria are defined in terms of descriptors of microstructural characteristics and these are used to adaptively create multi‐level domain partitions. The method developed is tested on a micrograph of a cast aluminium alloy A356. Copyright © 2006 John Wiley & Sons, Ltd.