Open Access

ARTICLE

Adaptive Multi-Scale Computational Modeling of Composite Materials

P. Raghavan¹, S. Ghosh²

1 Graduate Research Associate
2 Professor, Department of Mechanical Engineering, The Ohio State University, Columbus, Ohio, USA

Computer Modeling in Engineering & Sciences 2004, 5(2), 151-170. https://doi.org/10.3970/cmes.2004.005.151

Abstract

This paper presents an adaptive multi-level computational model that combines a conventional displacement based finite element model with a microstructural Voronoi cell finite element model for multi-scale analysis of composite structures with non-uniform microstructural heterogeneities as obtained from optical or scanning electron micrographs. Three levels of hierarchy, with different resolutions, are introduced in this model to overcome shortcomings posed by modeling and discretization errors. Among the three levels are: (a) level-0 of pure macroscopic analysis; (b) level-1 of macro-micro coupled modeling, used for signaling the switch over from macroscopic analyses to pure microscopic analyses; and (c) level-2 regions of pure microscopic modeling. The adaptive Voronoi cell finite element model is utilized effectively for analysis of extended microstructural regions with high efficiency and accuracy. Identification of statistically equivalent RVE (SERVE) for evaluating the effective properties are made through the use of correlation functions for different variables. Upon determination of SERVE's for actual microstructures, numerical examples of a composite plate and a composite laminate are solved to demonstrate the ability of the multi-scale computational model in analyzing complex heterogeneous structures.

---

Keywords

---