Multiscale analysis on the anisotropic thermal conduction of laminated fabrics by finite element method

• A “Two-scale finite element method” based on virtual fiber models containing random fiber distribution and fiber twist is proposed. • Effects of fiber numbers and diameters on the thermal conductivity of twist yarns and quartz ply-fabrics are predicted and analyzed. • Detailed heat flux distribution and temperature distribution between fibers and the air are simulated and analyzed. • The isotropic thermal conductivity of the yarn cross-section and the anisotropic thermal conductivity of quartz fabrics are illustrated. In this paper, we present a novel “Two-scale finite element method” (tFEM) based on virtual fiber models to investigate the heat conduction behavior of stacked quartz woven fabrics and predict their anisotropic thermal conductivity. Considering the random fiber distribution and the twisted characteristic of yarns, the yarn-scale model was established. Furthermore, the single-layer fabric was composed of interwoven virtual yarns, which are stacked to be the multi-laminate fabric-scale model. Both the yarn-scale model and the fabric-scale model were combined with air matrix to form the two-phase composite model. Hot-Disk thermal constant analyzer was used to measure the anisotropic thermal conductivity of yarns and woven fabrics. Excellent agreement between simulations and experiments is obtained, which indicates the multiscale finite element models in this paper is accurate. The innovations of the study are that not only the effects of fiber numbers and diameters on the thermal conductivity of twist yarns and quartz ply-fabrics are analyzed, but also the detailed heat flux distribution and temperature distribution between fibers and the air are simulated and analyzed. Moreover, the isotropic thermal conductivity of the yarn radial direction and the anisotropic thermal conductivity of quartz fabrics are illustrated.