CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   Wildland surface fire representation in OpenFOAM (http://www.cfd-online.com/Forums/openfoam/76797-wildland-surface-fire-representation-openfoam.html)

 Bufacchi June 4, 2010 12:07

Wildland surface fire representation in OpenFOAM

Hello all.

I'm working on my PhD thesis at University of Sao Paulo in Brazil. It is about wildland surface fire (vegetation is up to 1.8 meter tall). I decided to use OpenFOAM and not any other commercial tool. I'm using OpenFOAM version 1.6 on OpenSuse 11.1 (64 bits platform). As a beginner, I read the user's and programmer's guides, I went through the tutorials on the user's guide, I read many posts on the OpenFOAM forums and searched the src directory for information on the codes and their usage.

The physical model for a wildland surface fire is divided into a vegetation region and the atmosphere just above it. The vegetation is heated, dehydrates (looses water), undergoes pyrolysis (where reacting components are produced), and oxidates (surface burning). The reacting pyrolysis components reacts on the atmosphere. There are convection and radiation heat transfer to the air and to the vegetation surrounding the fire. The relative importance of each heat transfer type depends on the ambient wind. After burning, the vegetation becomes coal and ash.

The mathematical model I envision is a two domain (or region), where the vegetation is a porous media. As the air velocity is small, I plan to use a low Mach number approach (Navier-Stokes derived equations), which eliminates the acoustic waves (the spatially nonuniform portion of the pressure appears only in the momentum equation). Then, density varies with temperature, but not with pressure. Turbulence is everywhere and LES is used to model it. Buoyant flow rules the hot air (due to the fire) spread. The simulation is all transient.

To represent the model above, I was thinking to use chtMultiRegionFoam, replacing the solid region by a porous media code. The question that arises is the OpenFOAM code to use (or begin with and then change), because the porous media has to shrink as fire moves downstream consuming vegetation that becomes coal and ash.

I would appreciate comments and hints that can guide me through OpenFOAM in order to represent the model described above, and what codes to use or start with.