Unsteady simulation grid velocity

[nbully79]

Hi all!
I am doing a 2D unsteady CFD simulation for an oscillating airfoil on a moving grid. I have a structured grid and I know the grid locations at all time instants.
But, I am not able to figure out ""how to calculate the grid velocity for any cell?
Also, how do I use this to "implement Geometric Conservation Law"? I couldn't find any description about grid speed, maybe its too simple, but any help will be appreciated.
Thanks,

 

[btrueblood]

Look on the NPARC website. The original NPARC had a moving grid option (and I believe the WIND code also allows for it; there are bound to be papers written on the topic referenced there.

 

[nbully79]

Thanks btrueblood! I have looked at some of the papers, but all of them are very condensed and they just write that grid velocities need to be obtained from the modified grid and previous grid data, but none of them show how to. Maybe it's very obvious to everybody but it beats me. I will really appreciate any help in this matter.
Thanks again,

 

[btrueblood]

Hmm. In my simplistic mind, the method would be to simply go point-by-point through the grid, and calculate the displacement, then divide by the time step. This gives you the nodal velocities. The cell velocity would then be the average of the velocities of the nodes bounding the cell.

You may want to try and contact some of the authors, and ask them directly (I have had success in the past). It helps if you are an NPARC user, but they will still talk to you if not. Look on the web page for "tech support" contacts.

You don't mention what code you are using, is it a commercial code, and if so, can you request tech support from them? If you are using a "home-built" code, then I would suggest trying to get a copy of NPARC/Wind via the website; it's basically free for US citizens, scholars, and companies. You get the source code (mostly written in Fortran) and can thus delve in and find the methodology they use.