Work

Morphological and thermal characterization of a porous geopolymer and simulation of heat transfer by the monte carlo method formulated in path spacE

Year: 2026
Type: conference-paper
Abstract: This article proposes a methodology that combines experimental characterization and advanced numerical modeling to determine the apparent thermal conductivity of a porous geopolymer with a complex microstructure. Heat transfer is modeled by accounting for the coupling between conduction and radiation within a real three-dimensional geometry obtained from X-ray microtomography images, without adopting geometric simplifications or conventional volumetric meshing. First, macroscopic experimental measurements are carried out to estimate the thermophysical properties of the material. In parallel, the tomographic images are processed to reconstruct the three-dimensional microstructure of the porous medium and to extract morphological parameters. Secondly, the coupled conduction-radiation heat transfer problem is solved using an innovative statistical approach based on a path-space formulation. A numerical hot-plate experiment is replicated in order to evaluate the apparent thermal conductivity of the equivalent homogenized medium. Beyond the individual experimental and numerical steps, the main contribution of this work lies in the proposal of an integrated framework that links experimental characterization, morphological analysis, and numerical simulation to enable the estimation of macroscopic thermal properties from real microstructures. The results highlight the major impact of the porous microstructure and the size of the simulated domain on the estimated apparent thermal conductivity. One of the key results of this study shows that the computational cost depends only on the dimension along the z-axis, parallel to the imposed heat flux, which allows the strong heterogeneity of the porous medium in the x y plane to be taken into account. This approach therefore proves to be efficient for the thermal analysis of complex porous media, due to the insensitivity of the computational cost to the dimensions perpendicular to the imposed heat flux.
Source: HAL AMU
Language: English
Open Access status: green
OpenAlex ID: W7214568673
FWCI:
Cites:
Cited by:
Related to: 10