CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Vizualisation Sofware for Grid-Free methods (http://www.cfd-online.com/Forums/main/782-vizualisation-sofware-grid-free-methods.html)

ZinedineKhatir May 4, 1999 06:46

Vizualisation Sofware for Grid-Free methods
 
Hi Everybody,

I'd hopwe that somebody would be able to help me or at least guide me in the right way. I'am currently invloved in Turbulence study using a Grid-Free method known as Vortex Method. The problem I'm encountering now concerns the Post Processing of my Results. Well I am looking for a Vizualisation Tool which is capable of handling Grid-Free Methods. I would like to vizualise computed quantities such as Velocity and Perturbation velocity, vorticity and perturbation vorticy, Pressure fields , etc ... everybody knows that representing data is one important feature in CFD as well as designing alogorithms.

So if anybody can give any suggestions, it will very much appreciated.

Thanks a lot for help.

Zinedine.

Adrin Gharakhani May 4, 1999 15:17

Re: Vizualisation Sofware for Grid-Free methods
 
Hi Zinedine,

There are basically no visualization softwares that specifically target grid-free CFD, mainly because the bulk of CFD is carried out by grid-based methods. With the fast growth of mesh-free finite elements this may change.

You have three options:

1: setup a "background" grid system, evaluate the properties that you want on the grids, and use standard tools to visualize your data. (the grids can be just a uniform distributions, because you need not worry about accuracy and resolution at the post-proccesing stage)

2: use you vortex element locations as nodes for Delaunay triangulation. Now you have an unstructured grid system and again you can use standard software to visualize data

3: this is my favorite approach but you have to be careful with interpretation of the results: use a package that can visualize particles and you can color-code (and add vectors, etc.) to the vortex elements (particles). This is the least time consuming approach and requires no additional storage for post-processing. In this respect the vortex element method becomes the CFD tool which requires the least amount of memory and disk space of all the CFD's out there. Again the issue is interpretation. As you known, vortex elements concentrate in regions with vorticity, so there will be no elements in essentially potential flow regions. You just have to mentally realize that it implies a (fairly) uniform velocity profile there.

I use the public domain Visual3 which was developed at MIT. You can use it for 2D as well. It is a very powerful tool but you need to write your own little code to read data, etc. Since it's not a commercial package it's not GUI based per se. I'm not sure what the out-of-US restriction may be. Just search for Visual3 on the internet and you'll find it.

Adrin Gharakhani

ZinedineKhatir May 5, 1999 05:21

Re: Vizualisation Sofware for Grid-Free methods
 
Thank you very much Adrin.

I will let you know how it goes.

Zinedine.

S.N.Jayasinghe May 5, 1999 07:24

Re: Vizualisation Sofware for Grid-Free methods
 
I would like to help you if you could send me a copy of your software. As i will test for compatability firstly and send you a copy.

Regards

Suwan

ZinedineKhatir May 5, 1999 08:31

Re: Vizualisation Sofware for Grid-Free methods
 
Hi Suwan,

Thank you for your help. Well my code is still under improvement. It is written in Fortran. I have the output DATA for the velocities, vorticities, and positions in Files. I can choose the format I desire. I am using Vortex Element Method combined with BEM.

Hope this helps and look forward to hearing form you.

Regrads,

Zinedine.

Adrin Gharakhani May 6, 1999 14:28

Re: Vizualisation Sofware for Grid-Free methods
 
I have found that for 3D problems color coding velocity vectors with helicity density can be tremendously helpful with visualizing "3D effects". If the velocity vector lengths are a function of the speed then there is no need to color code the vectors by speed! Instead we can color them with helicity density which is dotproduct(velocity,vorticity)/|velocity|.|vorticity|, where |.| means magnitude. So this quantity is the cosine of the angle between the velocity and vorticity vectors and varies from -1 to 1.

So ... If -1 is blue and 1 is red, then 0 is green. If the flow is essentially 2D then your vectors will be green because in 2D flows vorticity and velocity vectors are normal to each other. However, in 3D the angle is no more 90 degrees. So regions of red would imply that vorticity and velocity are in phase (or aligned in the same direction). Blue means they are out of phase.

Right away you can see where three-dimensionality of the flow develops which is (normally) an indication of where vorticity begins to stretch. Contour/surface plots of pressure while helpful do not tell you about 3D effects right away.

Adrin Gharakhani

ZinedineKhatir May 7, 1999 03:59

Re: Vizualisation Sofware for Grid-Free methods
 
hi Adrin,

Thank for this comment. I have to tell you that I haven't looked at Visual3 yet, as I have just downloaded the User's file form the WWW. So I can tell than I don't have a clue of how it works. Maybe you could help in that matter !! Cheers,

Zinedine.



All times are GMT -4. The time now is 03:01.