Prediction of linear and non-linear behavior of 3D woven composite using mesoscopic voxel models reconstructed from X-ray micro-tomography

Abstract Mesoscopic representative volume cells (RVCs), reflecting the internal structure of a 3-Dimensional (3D) orthogonal weave composite, were reconstructed in this work from the X-ray micro-tomography (μCT) images. In comparison with the conventional idealized modelling strategy, the proposed voxel models reproduces successfully the varied yarn cross-section and the non-fully-symmetric undulated yarn path. It is demonstared that local details of the yarn geometry have small influence on the homogenized elastic properties but play key role in predicting the damage and failure process. Continuum damage mechanics models were formulated and implemented in the user subroutine UMAT of ABAQUS/Standard. The strength of the composite based on the model generated from the µCT images shows better agreement with the experimental results compared to the idealized one. The statistic distribution of damage variable is employed to describe the overall damage intensity when the peak stress value is reached. The quantitative analysis shows that the reproduced unevenness and the initial imperfections of the yarns are prone to cause and accumulate damage.