|
[Sponsors] |
Code for EARSM by Hellsten and RSM by Manceau, square duct, 2d turbulent channel |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
July 18, 2013, 08:39 |
Code for EARSM by Hellsten and RSM by Manceau, square duct, 2d turbulent channel
|
#1 |
Senior Member
Join Date: Apr 2009
Location: Karlsruhe, Germany
Posts: 104
Rep Power: 17 |
Hi everybody,
under the link http://www.istm.kit.edu/english/83.php you can download some codes and tutorials of turbulence models I implemented in OpenFOAM for different versions. They are implemented for incompressible flow (simpleFoam, pisoFoam,...). I validated them at serveral cases (2d channel, square duct, annular pipe flow, rod bundles). OpenFOAM 1.7.0 Code: nonlinear algebraic stress model ( EARSM ) by Hellsten and two validationcases: channel flow and square duct. Good agreement of the velocity profile at the 2d channel. The uv Reynolds stress is good (even for a annular pipe flow). (uu, vv, ww) are too low. Pic: EARSM_2dchannel_Reynolds_stresses.jpg OpenFOAM 2.1.0 Code: Reynolds stress model ( RSM ) by Manceau and the same two validation cases. Good agreement of velocity profiles, Reynolds stresses and secondary flows (see the pics) PICS: RSM: Reynolds stress profiles of a 2d channel RSM_2dchannel_Reynolds_stresses.png RSM: Square Duct RSM_ep_alpha.jpg RSM_square_duct1.jpg RSM_square_duct2.jpg Post-Processing yplus: To check the y+ values: post-processing tool to calculate them using low Reynolds models. Good luck Thomas |
|
October 16, 2013, 21:04 |
|
#2 |
New Member
Join Date: Jun 2013
Posts: 3
Rep Power: 13 |
Hello Thomas,
thank you very much for providing your files! I am interested in your square duct simulation using the Reynolds-stress model you implemented. Unfortunately, the link to the case setup (OF211_RSM_Manceau_quadchannel) doesn't work correctly (it's linked to the EARSM case). I wonder if you could fix it or upload your case folder here, since I have some problems with setting the right boundary conditions / fvSchemes. Thank you in advance! Alex |
|
October 23, 2013, 12:43 |
|
#3 |
Senior Member
Join Date: Apr 2009
Location: Karlsruhe, Germany
Posts: 104
Rep Power: 17 |
Hi,
I have uploaded it now. Try following link: https://dl.dropboxusercontent.com/u/...channel.tar.gz Regards Thomas |
|
October 29, 2013, 21:15 |
|
#4 |
New Member
Join Date: Jun 2013
Posts: 3
Rep Power: 13 |
Hello Thomas,
thank you very much for your reply. Could you also upload the 3D case of a square duct (OF211_RSM_Manceau_quadchannel) or do you use the same boundary conditions and fvSchemes as for the 2D channel? Furthermore, I have some questions concerning the 2D case you've just uploaded. What kind of boundary condition do you use for epsilon? Do you calculate the values for epsilon a priori and impose them as fixedValue? Is the initialization of the flow field important? If I get it right, there is no need for any lowReWallfunctions when using your code? Best wishes, Alex |
|
October 29, 2013, 23:10 |
|
#5 |
Senior Member
Join Date: Apr 2009
Location: Karlsruhe, Germany
Posts: 104
Rep Power: 17 |
Hi,
here is the correct link for the quad-channel. https://dl.dropboxusercontent.com/u/...channel.tar.gz It is a low Reynolds-number model with a elliptic factor, no damping functions are necessary... epsilon is calculated by a correction-function within the code. Its put to fixed value, but every time step the boundary values are updated. I made pre-simulations with the Launder-Sharma-model and used the results as starting conditions, because of the model is very sensitive. Of course, you have to correct the boundary conditions. You must read the papers of Manceau to understand the code. (an improved version ....) Best regards, Thomas |
|
October 31, 2013, 12:52 |
|
#6 |
New Member
Join Date: Jun 2013
Posts: 3
Rep Power: 13 |
Hi Thomas,
thank you once again. You helped me a lot! Alex |
|
February 17, 2014, 04:34 |
|
#7 |
New Member
Blaž Mikuž
Join Date: Sep 2011
Location: Ljubljana
Posts: 29
Rep Power: 15 |
Hi Thomas,
I've tried your quadchannel case and it works fine if I run it on a single core. Since I'd like to test it also on a bigger case I'll need to run it in parallel. The problem is that the case cannot be decomposed and it returns me the following error: HTML Code:
/*---------------------------------------------------------------------------*\ | ========= | | | \\ / F ield | OpenFOAM: The Open Source CFD Toolbox | | \\ / O peration | Version: 2.1.1 | | \\ / A nd | Web: www.OpenFOAM.org | | \\/ M anipulation | | \*---------------------------------------------------------------------------*/ Build : 2.1.1-221db2718bbb Exec : decomposePar Date : Feb 17 2014 Time : 09:29:22 Host : "krn" PID : 31017 Case : /home/r4/mikuz/MATIS-H/MATiS-bmk/steady/Mine_Fine_mesh/Baumann/quadchannel_paral nProcs : 1 sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE). fileModificationChecking : Monitoring run-time modified files using timeStampMaster allowSystemOperations : Disallowing user-supplied system call operations // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * // Create time --> FOAM Warning : From function dlOpen(const fileName&, const bool) in file POSIX.C at line 1175 dlopen error : /home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so: undefined symbol: _ZN4Foam14incompressible8RASModel11printCoeffsEv --> FOAM Warning : From function dlLibraryTable::open(const fileName&, const bool) in file db/dynamicLibrary/dlLibraryTable/dlLibraryTable.C at line 96 could not load "/home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so" Create mesh Calculating distribution of cells Selecting decompositionMethod simple Finished decomposition in 0.29 s Calculating original mesh data Distributing cells to processors Distributing faces to processors Distributing points to processors Constructing processor meshes --> FOAM Warning : From function dlOpen(const fileName&, const bool) in file POSIX.C at line 1175 dlopen error : /home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so: undefined symbol: _ZN4Foam14incompressible8RASModel11printCoeffsEv --> FOAM Warning : From function dlLibraryTable::open(const fileName&, const bool) in file db/dynamicLibrary/dlLibraryTable/dlLibraryTable.C at line 96 could not load "/home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so" Processor 0 Number of cells = 63000 Number of faces shared with processor 1 = 3718 Number of processor patches = 1 Number of processor faces = 3718 Number of boundary faces = 7800 --> FOAM Warning : From function dlOpen(const fileName&, const bool) in file POSIX.C at line 1175 dlopen error : /home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so: undefined symbol: _ZN4Foam14incompressible8RASModel11printCoeffsEv --> FOAM Warning : From function dlLibraryTable::open(const fileName&, const bool) in file db/dynamicLibrary/dlLibraryTable/dlLibraryTable.C at line 96 could not load "/home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so" Processor 1 Number of cells = 63000 Number of faces shared with processor 0 = 3718 Number of faces shared with processor 2 = 3600 Number of processor patches = 2 Number of processor faces = 7318 Number of boundary faces = 4200 --> FOAM Warning : From function dlOpen(const fileName&, const bool) in file POSIX.C at line 1175 dlopen error : /home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so: undefined symbol: _ZN4Foam14incompressible8RASModel11printCoeffsEv --> FOAM Warning : From function dlLibraryTable::open(const fileName&, const bool) in file db/dynamicLibrary/dlLibraryTable/dlLibraryTable.C at line 96 could not load "/home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so" Processor 2 Number of cells = 63000 Number of faces shared with processor 1 = 3600 Number of faces shared with processor 3 = 3718 Number of processor patches = 2 Number of processor faces = 7318 Number of boundary faces = 4200 --> FOAM Warning : From function dlOpen(const fileName&, const bool) in file POSIX.C at line 1175 dlopen error : /home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so: undefined symbol: _ZN4Foam14incompressible8RASModel11printCoeffsEv --> FOAM Warning : From function dlLibraryTable::open(const fileName&, const bool) in file db/dynamicLibrary/dlLibraryTable/dlLibraryTable.C at line 96 could not load "/home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so" Processor 3 Number of cells = 63000 Number of faces shared with processor 2 = 3718 Number of processor patches = 1 Number of processor faces = 3718 Number of boundary faces = 7800 Number of processor faces = 11036 Max number of cells = 63000 (0% above average 63000) Max number of processor patches = 2 (33.3333% above average 1.5) Max number of faces between processors = 7318 (32.6205% above average 5518) Time = 0 Processor 0: field transfer --> FOAM Warning : From function dlOpen(const fileName&, const bool) in file POSIX.C at line 1175 dlopen error : /home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so: undefined symbol: _ZN4Foam14incompressible8RASModel11printCoeffsEv --> FOAM Warning : From function dlLibraryTable::open(const fileName&, const bool) in file db/dynamicLibrary/dlLibraryTable/dlLibraryTable.C at line 96 could not load "/home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so" Processor 1: field transfer --> FOAM Warning : From function dlOpen(const fileName&, const bool) in file POSIX.C at line 1175 dlopen error : /home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so: undefined symbol: _ZN4Foam14incompressible8RASModel11printCoeffsEv --> FOAM Warning : From function dlLibraryTable::open(const fileName&, const bool) in file db/dynamicLibrary/dlLibraryTable/dlLibraryTable.C at line 96 could not load "/home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so" Processor 2: field transfer --> FOAM Warning : From function dlOpen(const fileName&, const bool) in file POSIX.C at line 1175 dlopen error : /home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so: undefined symbol: _ZN4Foam14incompressible8RASModel11printCoeffsEv --> FOAM Warning : From function dlLibraryTable::open(const fileName&, const bool) in file db/dynamicLibrary/dlLibraryTable/dlLibraryTable.C at line 96 could not load "/home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so" Processor 3: field transfer --> FOAM Warning : From function dlOpen(const fileName&, const bool) in file POSIX.C at line 1175 dlopen error : /home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so: undefined symbol: _ZN4Foam14incompressible8RASModel11printCoeffsEv --> FOAM Warning : From function dlLibraryTable::open(const fileName&, const bool) in file db/dynamicLibrary/dlLibraryTable/dlLibraryTable.C at line 96 could not load "/home/r4/mikuz/OpenFOAM/mikuz-2.1.1/platforms/linux64GccDPOpt/lib/libRSMEBMenhanced.so" End. Thanks! Blaz |
|
March 3, 2014, 15:52 |
|
#8 |
New Member
Mohammadreza Abyanaki
Join Date: Jan 2011
Posts: 5
Rep Power: 15 |
Dear Thomas,
Would you please write down exact reference to works of Manceau and Hellsten? I am not sure about title of the papers. Regards, Mohammadreza |
|
August 4, 2014, 16:23 |
Expressing the production term 'G'
|
#9 |
Member
Manan
Join Date: Oct 2013
Location: Göteborg
Posts: 37
Rep Power: 13 |
Hi Thomas
Thanks for the code! Last edited by MaLa; August 4, 2014 at 18:28. |
|
November 27, 2014, 16:31 |
|
#10 |
New Member
Join Date: Feb 2013
Posts: 21
Rep Power: 13 |
Thanks for sharing your code and the results.
Has anybody managed to compile the EARSM code for newer openfoam versions? I have been struggling with it for some time without any success. |
|
January 18, 2015, 16:12 |
|
#11 |
Senior Member
Join Date: Apr 2009
Location: Karlsruhe, Germany
Posts: 104
Rep Power: 17 |
Here are some infos; bit late
the EARSM from Hellsten is the version from his Dissertation: new two equation turbulence model for aerodynamics applications. The RSM by Manceau should be the one from the paper: AN IMPROVED VERSION OF THE ELLIPTIC BLENDING MODEL APPLICATION TO NON-ROTATING AND ROTATING CHANNEL FLOWS |
|
April 10, 2015, 09:56 |
|
#12 |
Member
Ardalan
Join Date: Jul 2012
Location: Atlanta, USA
Posts: 77
Rep Power: 14 |
I'm working with Remi Manceau now, your code is helpful a lot
A note: The current version of the code is not robust. You need to make it stable. Split the anisotoropy tensor to deviatic and diagonal and use Sp in deviatic part! check the implementation of Spezial-Sarkar-Gatski RSM as well. This is my REq: tmp<fvSymmTensorMatrix> REqn ( fvm::ddt(R_) + fvm::div(phi_, R_) //+ fvm::SuSp(-fvc::div(phi_), R_) - fvm::Sp(fvc::div(phi_), R_) - fvm::laplacian(nu(), R_) - fvm::laplacian(CMu/sigmak*TEBM*R_,R_) //anisotrope SSG-Modellierung + fvm::Sp(anisotrope, R_) == P - fvm::Sp(epsilonterm1, R_) +isotrope -epsilonterm2 +phistar ); //////////////////// If you have problem with it contact me! Ardalan Javadi Last edited by Ardali; April 23, 2015 at 13:43. |
|
July 10, 2020, 10:40 |
code for the RSM by Manceau und the EARSM by Hellsten
|
#14 |
Senior Member
Join Date: Apr 2009
Location: Karlsruhe, Germany
Posts: 104
Rep Power: 17 |
Hallo, the former links don't work anymore.
Here is the code for the RSM by Manceau und the EARSM by Hellsten. Its the OF 1.7 and the OpenFOAM 2.11 Version. |
|
October 16, 2020, 05:11 |
|
#15 | |
New Member
Join Date: Sep 2019
Posts: 1
Rep Power: 0 |
Quote:
thanks for the codes that you shared. Now I am trying to converting your RSTM Elliptic Blending model in OpenFoam 6: I did only minor syntax changes to your version and then I compiled it successfully. I tried then to reproduce your results for the channel flow at Re_tau=395, but I got very different results with poor accuracy. Could you please share again also the channel test case that you used for the validation? Thank you very much in advance Matilde |
||
|
|