Accurate numerical simulation of moisture front in porous material

TL;DR

This paper enhances numerical simulations of moisture transfer in porous materials by incorporating advection, introducing efficient schemes, and demonstrating improved accuracy over purely diffusive models through comparison with experimental data.

Contribution

It proposes the inclusion of moisture advection in the governing equations and introduces the Scharfetter-Gummel scheme for improved numerical simulation accuracy.

Findings

Scharfetter-Gummel scheme outperforms Crank-Nicolson in accuracy and speed.

Including advection improves the match between simulations and experimental data.

The scheme is well-balanced and asymptotically preserved.

Abstract

When comparing measurements to numerical simulations of moisture transfer through porous materials a rush of the experimental moisture front is commonly observed in several works shown in the literature, with transient models that consider only the diffusion process. Thus, to overcome the discrepancies between the experimental and the numerical models, this paper proposes to include the moisture advection transfer in the governing equation. To solve the advection-diffusion differential equation, it is first proposed two efficient numerical schemes and their efficiencies are investigated for both linear and nonlinear cases. The first scheme, Scharfetter-Gummel (SG), presents a Courant-Friedrichs-Lewy (CFL) condition but is more accurate and faster than the second scheme, the well-known Crank-Nicolson approach. Furthermore, the SG scheme has the advantages of being well-balanced and asymptotically preserved. Then, to conclude, results of the convective moisture transfer problem obtained with the SG numerical scheme are compared to experimental data from the literature. The inclusion of an advective term in the model may clearly lead to better results than purely diffusive models.