Evaluation of Effective Thermal Conductivities of Porous Textile Composites

TL;DR

This paper presents a multi-scale homogenization method to estimate the effective thermal conductivities of porous textile composites, validated against experimental data, highlighting the approach's accuracy and potential for complex material analysis.

Contribution

The study introduces a combined multi-scale homogenization and image analysis approach for accurately predicting thermal conductivities of porous composites.

Findings

Good agreement between computational and experimental thermal conductivities.

Transient flow solution provides detailed heat transfer insights.

Method effectively captures geometrical complexity of real composites.

Abstract

An uncoupled multi-scale homogenization approach is used to estimate the effective thermal conductivities of plain weave C/C composites with a high degree of porosity. The geometrical complexity of the material system on individual scales is taken into account through the construction of a suitable representative volume element (RVE), a periodic unit cell, exploiting the information provided by the image analysis of a real composite system on every scale. Two different solution procedures are examined. The first one draws on the classical first order homogenization technique assuming steady state conditions and periodic distribution of the fluctuation part of the temperature field. The second approach is concerned with the solution of a transient flow problem. Although more complex, the latter approach allows for a detailed simulation of heat transfer in the porous system. Effective thermal conductivities of the laminate derived from both approaches through a consistent homogenization on individual scales are then compared with those obtained experimentally. A reasonably close agreement between individual results then promotes the use of the proposed multi-scale computational approach combined with the image analysis of real material systems.