|January 13, 2006, 07:02||
I am going to implement ALE (Arbitrary Lagrangian Eulerian) formulation into my existing 2D FVM NS solver to solve flow past a heaving/plunging airfoil. I've finished modifying most of my code to include the mesh velocity. However I have some enquiries:
1. my original simulation was flow past an airfoil with inlet u=1. Is it possible to verify my new code by moving the mesh instead to the left ie inlet u=0, mesh vel=-1(right positive)? in that case, will the BC at airfoil surface be u=-1 or 0? I'm thinking it should be 0...
So is this possible and will the results (u,v,p) be the same as the original?
2. in the papers by Hirt, Margolin, they mention sometimes about "rezoning"? Can someone shed some light on what this term means? Is it applicable to my above case, or only for deforming meshes?
3. most ALE formulation of the NS eqns just added an additional "relative mesh velocity" to the convective terms. however, in a paper by Schulz "Unsteady Flow Structure Interaction for Incompressible Flows Using Deformable Hybrid Grids" JCP 143, p569, the author has added another "moving grid source term" given by -(du(mesh)x+dv(mesh)/dy)*[0 u v](transpose). Is it because it is a RANS (reynolds avg) eqns?
thank you very much
|Thread||Thread Starter||Forum||Replies||Last Post|
|Mesh Movement - (ALE) Equations||Araz||CFX||1||February 21, 2011 01:27|
|OpenFOAM-1.6 install cookbook||MadsR||OpenFOAM Installation||372||November 20, 2010 12:57|
|Question abt Overset & ALE||Ben||Main CFD Forum||0||June 18, 2007 12:06|
|Fractional step in ALE form - possible?||Wee||Main CFD Forum||2||October 3, 2006 10:20|
|ALE||tomosada||CD-adapco||0||December 13, 2004 09:35|