Predicting the overall response of an orthogonal 3D woven composite using simulated and tomography-derived geometry

Effects of two meso-scale geometry generation approaches on predicting effective elastic properties and thermal conductivity of an orthogonal 3D woven composite are studied in this paper. The two approaches used are fabric mechanics simulation (DFMA software, Kansas State University) and direct processing of X-ray microtomography (mu CT) data. Two models are created in DFMA using two different sets of cross-sectional areas as input nominal (based on the initial weave pattern) and adjusted (informed by volume fraction measurements obtained from microscopy data). In addition, one conformal mesh and three voxel mesh models are created from mu CT data. Experimental measurements of transverse Young's moduli are used to evaluate the accuracy of the predicted results. In each case, a unit cell with in-plane periodic boundary conditions is modeled, which has not been previously done in the case of mu CT-based models. The effect of high frequency oscillations in local material orientations imparted by a wavy centerline (artifact of mu CT image processing) on the elastic and thermal conductivity properties is studied. The differences in volume fractions and shapes of bundles of fibers (tows) between mu CT-based and DFMA-based models are also investigated to determine sensitivity of effective thermo-mechanical properties to each of these factors.