Qualitative Observations of Mean Flow Field

Josh · July 7, 2010, 23:53

Hello all -

I'm running an unsteady simulation in CFX. I know how to extract and average quantitative data, e.g., Cp, in order to study the mean flow field. However, I don't know (nor can I find an explanation of) how to study the mean flow field of an unsteady simulation qualitatively, e.g., produce a contour plot of the average vorticity.

One way, I suppose, would be to find an instantaneous moment that resembles an average. For example, monitor the lift force, then choose an instantaneous moment between the peaks and valleys of the lift fluctuations and use that moment as your "average" flow field for contours, streamlines, etc.

I'm wondering if anyone has any better ideas. All suggestions are welcome and very much appreciated!

Thanks!

ghorrocks · July 8, 2010, 08:43

If you have the averages of velocities you can calculate derivatives of it and get vorticity from there. Never tried it but hopefully it works. Also have a think about whether the vorticity of an averaged velocity field equals the averaged vorticity - I will leave that one up to you!

Your suggestion sounds like the moving window running average approach I used to a while back to convert a LES simulation into bulk and turbulent velocities. You have to be careful the window is the right size to be big enough to average "turbulence" away, but small enough to leave the "bulk" flow. This is a pretty standard thing in LES modelling.

Josh · July 8, 2010, 09:43

Hi Glenn -

I like the idea of the moving window running average approach. However, I'm using URANS modelling with the SST and SST-SAS models, not LES (well, you could argue that SST-SAS is a hybrid URANS-LES model, but anyway...)

Is this (moving window running average approach) possible using the CFX post-processor? I hate to ask, but could you refer me to a source that explains how to go about doing this? I understand the basic quantitative theory behind a moving average, but I'm wondering how to plot, say, an averaged velocity contour.

Thanks!

ghorrocks · July 8, 2010, 18:47

My PhD thesis is where I did this stuff. Have a look in the square piston model, the bit where I do the LES model.
http://hdl.handle.net/2100/248

Josh · July 9, 2010, 16:52

Thanks, Glenn.

For those interested in what he's talking about, refer to section 6.9.6 of his thesis. However, this data is still quantitative. I was hoping there was a way to obtain an average contour of pressure, for example, over time.

July 7, 2010, 23:53	Qualitative Observations of Mean Flow Field	#1
Josh Senior Member Joshua Counsil Join Date: Jul 2009 Location: Halifax, Nova Scotia, Canada Posts: 366 Rep Power: 17	Hello all - I'm running an unsteady simulation in CFX. I know how to extract and average quantitative data, e.g., Cp, in order to study the mean flow field. However, I don't know (nor can I find an explanation of) how to study the mean flow field of an unsteady simulation qualitatively, e.g., produce a contour plot of the average vorticity. One way, I suppose, would be to find an instantaneous moment that resembles an average. For example, monitor the lift force, then choose an instantaneous moment between the peaks and valleys of the lift fluctuations and use that moment as your "average" flow field for contours, streamlines, etc. I'm wondering if anyone has any better ideas. All suggestions are welcome and very much appreciated! Thanks!

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Flow meter Design	CD adapco Group Marketing	Siemens	3	June 21, 2011 08:33
velocity of massless particles and velocity of flow field in same coordinate	payam_IUST	FLUENT	0	October 18, 2009 23:24
rerun with same flow field, different fluid props	Tim	Phoenics	0	March 3, 2004 14:51
fluid flow fundas	ram	Main CFD Forum	5	June 17, 2000 21:31
computation about flow around a yawed cone	Tylor Xie	Main CFD Forum	0	June 9, 1999 07:33

July 8, 2010, 08:43		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,700 Rep Power: 143	If you have the averages of velocities you can calculate derivatives of it and get vorticity from there. Never tried it but hopefully it works. Also have a think about whether the vorticity of an averaged velocity field equals the averaged vorticity - I will leave that one up to you! Your suggestion sounds like the moving window running average approach I used to a while back to convert a LES simulation into bulk and turbulent velocities. You have to be careful the window is the right size to be big enough to average "turbulence" away, but small enough to leave the "bulk" flow. This is a pretty standard thing in LES modelling.

July 8, 2010, 09:43		#3
Josh Senior Member Joshua Counsil Join Date: Jul 2009 Location: Halifax, Nova Scotia, Canada Posts: 366 Rep Power: 17	Hi Glenn - I like the idea of the moving window running average approach. However, I'm using URANS modelling with the SST and SST-SAS models, not LES (well, you could argue that SST-SAS is a hybrid URANS-LES model, but anyway...) Is this (moving window running average approach) possible using the CFX post-processor? I hate to ask, but could you refer me to a source that explains how to go about doing this? I understand the basic quantitative theory behind a moving average, but I'm wondering how to plot, say, an averaged velocity contour. Thanks!

July 8, 2010, 18:47		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,700 Rep Power: 143	My PhD thesis is where I did this stuff. Have a look in the square piston model, the bit where I do the LES model. http://hdl.handle.net/2100/248

July 9, 2010, 16:52		#5
Josh Senior Member Joshua Counsil Join Date: Jul 2009 Location: Halifax, Nova Scotia, Canada Posts: 366 Rep Power: 17	Thanks, Glenn. For those interested in what he's talking about, refer to section 6.9.6 of his thesis. However, this data is still quantitative. I was hoping there was a way to obtain an average contour of pressure, for example, over time.