CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > Fidelity CFD

NLH reconstruct flow in time

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   December 3, 2016, 03:00
Default NLH reconstruct flow in time
  #1
New Member
 
Join Date: Nov 2013
Posts: 14
Rep Power: 12
Tegegn Dejene is on a distinguished road
I did NLH simulation and when I reconstruct the flow in time for NLH Basic or Multi-rank with 1 rank cases and the output variables selected like harmonics of static quantity,harmonics of total quantities, harmonics of velocity and deterministic stresses are not available on QAP quantities list on CFView;only few quantities are such as static pressure,static temperature etc.
Any one can suggest the possible solution to this issue. For your information, when I do local reconstruction and spectral analysis,all output variables selected in computed variables including harmonics variables are available on QAP quantities of CFView.
Tegegn Dejene is offline   Reply With Quote

Old   December 3, 2016, 17:35
Default
  #2
Senior Member
 
DarylMusashi's Avatar
 
Holger Dietrich
Join Date: Apr 2011
Location: Germany
Posts: 174
Rep Power: 14
DarylMusashi is on a distinguished road
Dear Tegegn,

if you are opening the .run file you open the time-averaged solution of the NLH computation. This file contains all the selected variables from the output tab, including the harmonics of the flow variables.

What you did with this file is the "reconstruction in time". This means you transform the computed harmonics into time domain. All flow variables of your defined time steps are computed on basis of these harmonics. This is the reason why the computed harmonics are no longer explicitly selectable in the reconstructed solution - they are transformed to "normal" flow variables and are included in them.
DarylMusashi is offline   Reply With Quote

Old   December 5, 2016, 02:27
Default
  #3
New Member
 
Join Date: Nov 2013
Posts: 14
Rep Power: 12
Tegegn Dejene is on a distinguished road
Thanks a lot Mr. Holger and I appreciate for your fast response and continuous support to understand NLH simulation. I have one more query : now I want display the contours of deterministic stress at span wise and/or stream wise plane from unst.run files after reconstruction how can I do it?
Tegegn Dejene is offline   Reply With Quote

Old   December 6, 2016, 16:29
Default
  #4
Senior Member
 
DarylMusashi's Avatar
 
Holger Dietrich
Join Date: Apr 2011
Location: Germany
Posts: 174
Rep Power: 14
DarylMusashi is on a distinguished road
Dear Tegegn,

in CFView please go to File -> Turbomachinery. The upper left is the meridional averaged view, the lower left one is the blade to blade (spanwise) view. Please click the window of interest with the left mouse, so that it is marked with a red border.
At the middle left part (Quantities) you can select the quantity you are interested in, for example the deterministic stresses. Below this, under the "Representations" tab you may select "Contours and Iso Values". There you can select to plot a "strip contour" (please see attached picture).
By right-clicking the color bar at the right you can change the number of stripes by selecting "Color Control".

In the lower left window a spanwise cut at 50% span is created as default. If you want another span position cut please select the lower left window first (red marked border). Directly under the "surface" tab on the left you can select "Blade to Blade" and define the cut position S. S = 0.5 is the default at 50% span, as mentioned. After hitting "Apply" please don't forget to "Save" this cut plane.

Please have a closer look at the two attached pictures.
Attached Images
File Type: jpg strip_contour.jpg (74.9 KB, 6 views)
File Type: jpg Spanwise_Cuts.jpg (90.8 KB, 14 views)
DarylMusashi is offline   Reply With Quote

Old   December 7, 2016, 00:11
Default
  #5
New Member
 
Join Date: Nov 2013
Posts: 14
Rep Power: 12
Tegegn Dejene is on a distinguished road
Dear Holger

Thank you again for the reply;how to make contours at different span is clear for me but my question is the deterministic stress(average of product density and velocity into velocity) is not available in quantity tab. I tried using define new quantity but I couldn't. I made contours for other output variables which are available in quantity tab;the problem is for deterministic stress and vorticity ;they are not available to select.
Tegegn Dejene is offline   Reply With Quote

Old   December 7, 2016, 17:26
Default
  #6
Senior Member
 
DarylMusashi's Avatar
 
Holger Dietrich
Join Date: Apr 2011
Location: Germany
Posts: 174
Rep Power: 14
DarylMusashi is on a distinguished road
Dear Tegegn,
the deterministic stresses and the vorticity are only available for the time-averaged solution of the NLH computation (opening the .run file). I will try to get an explanation for this, soon.
DarylMusashi is offline   Reply With Quote

Old   December 15, 2016, 15:27
Default
  #7
Senior Member
 
DarylMusashi's Avatar
 
Holger Dietrich
Join Date: Apr 2011
Location: Germany
Posts: 174
Rep Power: 14
DarylMusashi is on a distinguished road
Dear Tegegn,

thank you for providing the interesting paper "Investigation of the unsteady flow of a transonic compressor using the nonlinear harmonic method" by Peng Zhang, Bo Liu, Lei Shi and Jun Li" (http://journals.sagepub.com/doi/pdf/...54410015606940) about NUMECA's NLH method.

To sum up the authors investigated, that the NLH method predicts the unsteady behavior of the machine to their complete satisfaction. One big advantage: With the NLH method your computational time is reduced to less than 10% in comparison to a "classical" unsteady simulation.

To come back to your initial question: In this paper the authors compared the time-averaged NLH solution with an unsteady, time-averaged solution. For comparison they inspected the calculated deterministic stresses, as they are an indicator of unsteadiness in the flow field. The most relevant statements, which outline that they compared time-averaged solutions, seem to be for me:

"The deterministic stresses appear in the NLH method through time averaging of periodic unsteadiness due to the relative blade motion."
"Because of the non-linearity, the deterministic stress terms appear in the time-averaged equation in the same way as Reynolds stress terms appear in the Reynolds averaged equations."
"And the deterministic stress can be directly computed from these harmonic values, which can account for the unsteady effects on the time-averaged flow."

I hope it is clearer for you now, why the deterministic stresses are only available in the time-averaged solution of the NLH computation?

Kind regards
Holger
DarylMusashi is offline   Reply With Quote

Old   December 16, 2016, 00:48
Default
  #8
New Member
 
Join Date: Nov 2013
Posts: 14
Rep Power: 12
Tegegn Dejene is on a distinguished road
Dear Holger

I appreciate and thanks for brief explanation about deterministic stress in NLH simulation.

when come to my point about post processing the deterministic stress in FINE/Turbo of NLH method, deterministic stress which is the avg (rho*u')u' or avg(rho*v')v' etc is still not readily available in quantity tab of time averaged flow. May be can we calculate the deterministic stress for each harmonics in time averaged solution and average for all harmonics by using the option define new quantity in quantity tab?
Tegegn Dejene is offline   Reply With Quote

Old   December 16, 2016, 12:24
Default
  #9
Senior Member
 
DarylMusashi's Avatar
 
Holger Dietrich
Join Date: Apr 2011
Location: Germany
Posts: 174
Rep Power: 14
DarylMusashi is on a distinguished road
Dear Tegegn,
I think manually defining the deterministic stresses is not possible, because each deterministic stress component is associated to a specific harmonic.

In FINE/Turbo >> Outputs Tab >> Computed Variables >> Harmonics please select all your deterministic stress components you are interested in as output variables. Then you can easily post-process each component and its related harmonics.
DarylMusashi is offline   Reply With Quote

Old   December 17, 2016, 00:00
Default
  #10
New Member
 
Join Date: Nov 2013
Posts: 14
Rep Power: 12
Tegegn Dejene is on a distinguished road
Dear Holger

I am not talking about defining defining deterministic stress in output tab of computed variable under harmonics option;I did that before run. It is done and now my question is post processing deterministic stress in CFView. The deterministic stress those selected in output tab are not available .run file,in short the deterministic stress is not available to select in quantity tab of CFView after solution converged.

Thanks
Tegegn Dejene is offline   Reply With Quote

Old   December 17, 2016, 14:23
Default
  #11
Senior Member
 
DarylMusashi's Avatar
 
Holger Dietrich
Join Date: Apr 2011
Location: Germany
Posts: 174
Rep Power: 14
DarylMusashi is on a distinguished road
Dear Tegegn,
unfortunately I cannot confirm this behaviour on my side. I tested it with the flow configuration "NLH - basic" and "NLH - multirank 1 and 2".

I activated the deterministic stresses in the output tab (screenshot 2). Please take not that I used a Linux machine, so the activated boxes are just yellow. On a Windows machine the activated boxes have checkmarks.

After a few iterations I stopped the simulation and used the CFView button on top (or you can load the .run file in CFView) to load the time-averaged solution of the computation.

In CFView in the left tab all selected quantities (and the deterministic stresses) are available.

May I ask which version of FINE/Turbo you are using?
Attached Images
File Type: jpg 1_Flow_Model.JPG (181.1 KB, 17 views)
File Type: jpg 2_Computed_Variables.jpg (127.2 KB, 14 views)
File Type: jpg 3_CFView.jpg (122.9 KB, 18 views)
DarylMusashi is offline   Reply With Quote

Old   December 18, 2016, 23:49
Default
  #12
New Member
 
Join Date: Nov 2013
Posts: 14
Rep Power: 12
Tegegn Dejene is on a distinguished road
Dear Holger

Great appreciation to you kind help and support. I think the issue may be related with the version of FINE/Turbo I am using. It is 9.2.

Thanks
Tegegn Dejene is offline   Reply With Quote

Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
[Other] Contribution a new utility: refine wall layer mesh based on yPlus field lakeat OpenFOAM Community Contributions 58 December 23, 2021 02:36
High Courant Number @ icoFoam Artex85 OpenFOAM Running, Solving & CFD 11 February 16, 2017 13:40
Stuck in a Rut- interDyMFoam! xoitx OpenFOAM Running, Solving & CFD 14 March 25, 2016 07:09
Help for the small implementation in turbulence model shipman OpenFOAM Programming & Development 25 March 19, 2014 10:08
pisoFoam with k-epsilon turb blows up - Some questions Heroic OpenFOAM Running, Solving & CFD 26 December 17, 2012 03:34


All times are GMT -4. The time now is 21:05.