Cargando…

A Semi-implicit Treatment of Porous Media in Steady-State CFD

There are many situations in computational fluid dynamics which require the definition of source terms in the Navier–Stokes equations. These source terms not only allow to model the physics of interest but also have a strong impact on the reliability, stability, and convergence of the numerics invol...

Descripción completa

Detalles Bibliográficos
Autores principales: Domaingo, Andreas, Langmayr, Daniel, Somogyi, Bence, Almbauer, Raimund
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Springer Netherlands 2016
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4923709/
https://www.ncbi.nlm.nih.gov/pubmed/27429500
http://dx.doi.org/10.1007/s11242-016-0657-3
Descripción
Sumario:There are many situations in computational fluid dynamics which require the definition of source terms in the Navier–Stokes equations. These source terms not only allow to model the physics of interest but also have a strong impact on the reliability, stability, and convergence of the numerics involved. Therefore, sophisticated numerical approaches exist for the description of such source terms. In this paper, we focus on the source terms present in the Navier–Stokes or Euler equations due to porous media—in particular the Darcy–Forchheimer equation. We introduce a method for the numerical treatment of the source term which is independent of the spatial discretization and based on linearization. In this description, the source term is treated in a fully implicit way whereas the other flow variables can be computed in an implicit or explicit manner. This leads to a more robust description in comparison with a fully explicit approach. The method is well suited to be combined with coarse-grid-CFD on Cartesian grids, which makes it especially favorable for accelerated solution of coupled 1D–3D problems. To demonstrate the applicability and robustness of the proposed method, a proof-of-concept example in 1D, as well as more complex examples in 2D and 3D, is presented.