|
[Sponsors] |
December 4, 2014, 19:26 |
PODSolver : Proper Orthogonal Decomposition
|
#1 |
Member
Join Date: Feb 2014
Posts: 63
Rep Power: 12 |
Hi,
I would like to analyse my LES simulation results using PODsolver in foam-extend-3.1. On tutorial cases there are eigen values of the scalar T (TPOD1) and TPODeconstruct field, in 0 directory. So my problem : Is there a way in OpenFOAM to generate these files, or if not how do i generate those files for initial condition. Thank you. |
|
January 12, 2015, 07:22 |
POD - foam extend
|
#2 |
Member
Gowain
Join Date: May 2014
Location: Ireland
Posts: 86
Rep Power: 12 |
Hi
I am looking to POD for reduced order modeling of aircraft compartments - I have created a code with matlab that works for RANs simulations, however I would like to investigate the POD tool in foam extend. I am also having problems getting it to work - if you have had any subsequent success please let me know. regards gowain |
|
January 12, 2015, 13:45 |
|
#3 |
Member
Join Date: Feb 2014
Posts: 63
Rep Power: 12 |
I am also stuck getting it to work.
However i know there are two other utilities called scalarSnapshots and vectorSnapshots to create initial snapshots. However beyond that i am stuck on identifying POD modes on different time steps. By the way did you decompose your OpenFOAM unstructured velocity field data using a POD matlab code? |
|
May 18, 2016, 11:37 |
|
#4 |
Member
Ali
Join Date: Aug 2011
Location: Milwaukee
Posts: 34
Rep Power: 15 |
Hello everyone,
I know this is a kinda old post, but I was wondering if you guys figured out how to generate the snapshots? I am trying to use the 1DPODsin example and generate the snapshots. I removed the excising snapshots and ran scalarSnaoshots and I get this error Error in `scalarSnapshots': double free or corruption (out): 0x00007ffe8729c050 *** I am also wondering if you guys aware of any development for POD on NS? Thanks |
|
May 19, 2016, 05:25 |
Pod
|
#5 |
Member
Gowain
Join Date: May 2014
Location: Ireland
Posts: 86
Rep Power: 12 |
Hi guys,
I have abandoned the built in POD with openF0am long ago, have moved to MATLAB and ANSYS where I have written my own codes and a unique framework to bypass a Galerkin projection. So reduced order modelling with the POD is my expertise, I am a researcher, if there is anything I can help you out with let me know the problem I would be happy to help. |
|
May 19, 2016, 10:46 |
|
#6 |
Member
Ali
Join Date: Aug 2011
Location: Milwaukee
Posts: 34
Rep Power: 15 |
Hey Gowain,
That's awesome! ok first of all congrats for pulling this out. After that. I have to write my codes in open source tools and now my next step is POD. I was wondering if there is anyway that you can share the code then I can write a version using OpenFOAM and Octave? If not, do you have any publication, report or anything that I can start developing based on your work? Thanks Ali |
|
December 21, 2016, 07:27 |
|
#7 |
New Member
Ahmet
Join Date: Mar 2011
Posts: 10
Rep Power: 15 |
Is there any progress in this issue?
I am getting the error @alib022 on tutorial files in foam-extend-3.2 too. |
|
January 11, 2017, 04:42 |
POD-Galerkin methods inside OF
|
#8 |
New Member
Giovanni Stabile
Join Date: Sep 2014
Location: Trieste (Italy) - Florence (Italy)
Posts: 14
Rep Power: 12 |
I have successufully implemented POD-Galerkin methods inside foam-extend 3.2 and OpenFoam 2.4.0. With OpenFoam foam-extend I used its own eigenvalue solver with OpenFoam 2.4.0 I linked OpenFoam with Eigen that is a fast and light c++ eigenvalue solver. If you need any other tip I can provide you some more informations. Soon I will be able also to provide you some articles about the topic. At the moment they are still under review.
|
|
January 11, 2017, 14:37 |
|
#9 |
New Member
Ahmet
Join Date: Mar 2011
Posts: 10
Rep Power: 15 |
I really want to know anything about POD or your related studies @giovastabile. Maybe you would get first cite to your paper
|
|
January 12, 2017, 01:46 |
|
#10 | |
Member
Ali
Join Date: Aug 2011
Location: Milwaukee
Posts: 34
Rep Power: 15 |
Quote:
@giovastabile, congratulations Please share your paper, possibly your code here. I am still actively using POD, and your code will be a big help. Looking forward to read your paper. |
||
January 12, 2017, 04:23 |
Paper and Code
|
#11 |
New Member
Giovanni Stabile
Join Date: Sep 2014
Location: Trieste (Italy) - Florence (Italy)
Posts: 14
Rep Power: 12 |
Regarding the paper I'll be able to share it in few days. We will soon share also the code but it will take some months. All the documentation needs to be properly written.
|
|
January 13, 2017, 02:30 |
|
#12 |
Member
Ali
Join Date: Aug 2011
Location: Milwaukee
Posts: 34
Rep Power: 15 |
||
January 13, 2017, 04:30 |
Paper
|
#13 |
New Member
Giovanni Stabile
Join Date: Sep 2014
Location: Trieste (Italy) - Florence (Italy)
Posts: 14
Rep Power: 12 |
Below you can find my paper. If you need explanation or tip about the technical implementation in OpenFoam please let me know. Also if you have any comment don't hesitate to contact me.
BR, Giovanni The link is: https://arxiv.org/abs/1701.03424 |
|
January 14, 2017, 23:48 |
|
#14 |
Senior Member
Pete Bachant
Join Date: Jun 2012
Location: Boston, MA
Posts: 173
Rep Power: 14 |
If you put it up now, others may help you write the documentation!
|
|
January 16, 2017, 06:41 |
Release of the C++ routines
|
#15 |
New Member
Giovanni Stabile
Join Date: Sep 2014
Location: Trieste (Italy) - Florence (Italy)
Posts: 14
Rep Power: 12 |
I'm sorry but I cannot release the routines without a proper documentation and a quality check. I'm sorry but I'm working within an European research project and there some conditions that need to be fulfilled before releasing the software. I hope that in some weeks I'll be able to release at least the part to compute the POD modes and later the part for the generation of the Reduced Order Model.
|
|
June 14, 2017, 12:00 |
|
#16 |
New Member
M. Salman Siddiqui
Join Date: Mar 2015
Posts: 16
Rep Power: 11 |
Have you consider for the release of your codes yet?
|
|
June 20, 2017, 05:08 |
|
#17 |
New Member
Giovanni Stabile
Join Date: Sep 2014
Location: Trieste (Italy) - Florence (Italy)
Posts: 14
Rep Power: 12 |
I will soon release the part doing the POD decomposition. For the part regarding the Galerkin projection, it will take some time.
Giovanni |
|
June 20, 2017, 09:45 |
|
#18 |
New Member
M. Salman Siddiqui
Join Date: Mar 2015
Posts: 16
Rep Power: 11 |
Great, just let us know here on the group once you release it
|
|
October 30, 2017, 07:21 |
|
#19 |
New Member
M. Salman Siddiqui
Join Date: Mar 2015
Posts: 16
Rep Power: 11 |
Dear Giovann,
Any updates on the release of your tool? Bye the way, i am Trond Kvamsdal PhD student at NTNU regards, salman |
|
October 30, 2017, 10:26 |
|
#20 |
New Member
Giovanni Stabile
Join Date: Sep 2014
Location: Trieste (Italy) - Florence (Italy)
Posts: 14
Rep Power: 12 |
Dear Salman,
I'm working right now to the first release of the code. I suppose in few weeks (two maximum three) we will come out with the first release. I'll keep you informed. Take my regards to Trond Kvamsdal. Giovanni |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
"PROPER ORTHOGONAL DECOMPOSITION" | dpshaka | Main CFD Forum | 7 | June 12, 2019 10:42 |
Proper Orthogonal Decomposition (POD) in Commercial Codes? | asajdak | Main CFD Forum | 1 | August 11, 2013 08:53 |
Proper orthogonal decomposition of turbulent flow with k-e | Blob | Main CFD Forum | 0 | June 5, 2009 10:32 |
Snapshot Proper Orthogonal Decomposition (POD) | Mehmet SEN | Main CFD Forum | 7 | May 19, 2006 13:38 |
On Proper Orthogonal Decomposition | R.D.Prabhu | Main CFD Forum | 0 | March 15, 2000 18:31 |