Discussion Groups

Simulation Mechanical and Multiphysics

Reply
Valued Mentor
712 Posts
59 Kudos
Registered: ‎08-30-2007

CFD for Air Flow Modeling

4399 Views, 4 Replies
04-27-2010 01:40 PM
I serendipitously stumbled upon this Algor stuff just recently.
Does Algor CFD work well for simulating wind through buildings (exteriors) and over piles of dirt?
We have clients wanting us to do wind analysis in certain areas. I'm downloading the SimulationExpress program now, but I can't tell from the descriptions if it includes the CFD package. Guess I'll find out in a few minutes. Edited by: hardin@cdgengineers.com on Apr 27, 2010 3:39 PM
Employee
259 Posts
26 Kudos
Registered: ‎06-14-2010

Re: CFD for Air Flow Modeling

06-14-2010 12:31 PM in reply to: hardin

Wind through buildings:  Yes,  this is a typical external flow, in modeling, a box needs to be added to enclose the building, just image the how wind tunnel works in reality.

 

Flow over piles of dirt: No.   This is a process of multiphase flow which invloves are the air and moving of the dirt particles. AlgoR CFD does not support multiphase flow at this point.

Jianhui Xie, Ph.D
Principal Engineer
MFG-Digital Simulation
New Member
2 Posts
0 Kudos
Registered: ‎02-06-2011

Re: CFD for Air Flow Modeling

02-14-2011 07:46 PM in reply to: Joey.X

Hi Jianhui Xie,

 

Are there any tutorials on how to perform an external flow analysis in Algor? I cant find any in the Algor help/tutorial. I am a new Algor user and any help would be great.

 

Thanks in advance

 

Faisal

Employee
37 Posts
1 Kudo
Registered: ‎05-14-2008

Re: CFD for Air Flow Modeling

02-15-2011 07:50 AM in reply to: hardin

Hi Hardin,

 

Algor CFD works very well for internal and external air flow.

 

I've done two examples of such problems, you can see the results here:

 

http://www.youtube.com/watch?v=32F2QsFY03A

 

http://www.youtube.com/watch?v=puhAKLkqP9Q&feature=related

 

If you want to simulate wind over piles of settled dust (that is that the pile is considered rigid for a given wind flow), that is possible being a classical external flow problem, however if you want to simulate the dirt particles as well than its not possible with Algor CFD.

 

As far as I know Simulation express does NOT have CFD feature.

 

Regards,

Shakeel

New Member
1 Posts
0 Kudos
Registered: ‎02-15-2011

Re: CFD for Air Flow Modeling

02-16-2011 08:18 PM in reply to: fag2

Hi, Faisal,

 

Please see attached Algor CFD model for external flow. It's flow through 2-D submarine with 10 degree attack angle, it's an easy 2-D model but illustrate many concepts in CFD external flow modeling. What I did was doing 2-D sketch in CAD and import 2D geometry (fluid domain only) to Autodesk Algor simulation.

 

This model is a full model with ready result, as Autodesk Algor user, you can open and see all the pre-simulated result. Also you can play around to change BCs, meshing, change to unsteady analsysis fo transient effects etc.

 

Basically the modeling for extensive flow is like putting the obstacles in wind tunnel or water tunnel. External flow

And below are some other hints of external flow modeling for your information.

(a) Modeling in fluid domain only. If you make models from solid parts/assembly in CAD, you need generate fluid domains either by Boolean operation in CAD or fluid general generation in Algor.  Note that Algor allows multiple fluid parts for complicated models.

(b) Result accuracy (selected major factors)

      1) Eliminate/reduce wall effect, this is a complicated topic, refer experimental aerodynamics for more details. Since the simulated obstacles is always put inside an enclosure (like wind/water tunnels), the size of the enclosure will affect result accuracy. Ideally the larger enclosure, the less effect it will be, but it brings more expensive computational cost (under same mesh size). In attached model, the symmetrical velocity BCs were specified on the top and bottom wall, which have less wall effect than complete wall.

     2)  Distance to inlet and outlet, it is suggested to set proper distance from the obstacles, the specified constant velocity or pressure at inlet or outlet will impact the whole flow pattern in incompressible flow.

     3) Mesh size. Accurate CFD result needs mesh study for solution dependency on mesh.

(c)  Solution robustness

     Set fixed velocity inlet and  zero pressure outlet (or Algor's inlet/outlet at outlet) gives best convergence combination, and this is close the condition in wind tunnel experiment.

     There are many considerations for solution robustness and convergence.  Sorry I can't put here for lengthy.

 

Jianhui Xie

 

 

 

Jianhui Xie , Ph.D
Sr. Software Engineer
Autodesk, Inc.
285 Kappa Drive
Pittsburgh, PA 15238
.....