turbine modelling

[stage75]

Thjnk you Pisolini,

I use version 2.0.4 and i can't change it, since it is installed on a network and the IT department does not want to change it (the only stable version according to them!!) . So I have to work on this version.

I followed the steps given by you and guillaume.

I copied my three meshes (mesh turbine, fluid level mesh and the mesh of my pelvis).

Then i create my 2 directories (BASSIN and TURBINE):

code_saturne -s create -SIMULATION -c BASSIN -c TURBINE

This line also creates two icons of "SaturneGUI" in DATA of "BASSIN" and "DATA of "TURBINE".

Then i copied "usclim.f90", "usini1.f90", "usiniv.f90" and "ussatc.f90" in "SRC" of "TURBINE" and "SRC" of "BASSIN".

Since I want my turbine rotates with an angle of 360 ° until convergence, I modified the "usini1.90" present in "DATA" of "TURBINE", i marked "omegay=360.d0" (since my turbine rotates arround y direction).

So now I want to know is that it is correct that I did. And what are the steps to follow to compile the "usini1.f90" subroutine? I use the GUI mode by running the two SaturneGUI "and read my mesh Turbine by the first" SAturneGUI "and the other by the second mailages SaturneGUI"?!!!

I am beginning to Saturn code


please help me

[JamesMcNaughton]

Hi

The rotation vector should be in radians per second, dimensions 1/s so you should convert 360 deg to 2 pi, assuming you want 1 revolution per second.

To run the case you should go to the SCRIPTS folder and use the runcase script. To run you type ./runcase.

I've attached an example runcase that was for the genovas pump case in v2.0rc2 so hopefully that can help you with your set up. You'll need to set the runcase up to read your .xml files which goes in the PARAMETER argument in runcase.

(Had to change runcase to runcase.txt so the board would allow upload).

Hope that helps a bit!

James

[stage75]

Thank you James,

Is that we can not put the mesh form. "Med". why the form oh the meshes in [d1, d2] (runcase) are on ".des"?

when I changed my "runcase", i change "d" by (d1 and d2) I got an error message by running. / runcase :

d2 = domain(meshes = 'Mesh_BASSIN.med',
^
IndentationError: unindent does not match any outer indentation level

Since i put my two meshes (Mesh-DOMAINE.med and Mesh_TURBINE.med) in "MESH" !

For this simulation can i use SATURNE GUI or not?

[JamesMcNaughton]

The error sounds more like a white-space issue in python. Does it give you a line number with the problem. Make sure you are using spaces and not tabs and that all the lines are indented properly as in the example. If you can't see a problem could try posting it up here.

As far as I know you can use the GUI for the case set-up but not to run from (hence using runcase instead), I think you need to only set the line:

param = PARAMETERS,

to param = 'case.xml',

Cheers,

James

[stage75]

I forgot to add the line

Code: Select all

d2.valgrind = VALGRIND 

Now, I have another error :

Code: Select all

  File "./runcase", line 268
    if VALGRIND != None:
    ^
IndentationError: unexpected indent

?

[Yvan Fournier]

Hello,

The error is produced by the Python interpreter, and gives you all the information required to fix it.

Regards,

Yvan

[stage75]

Hi ,

I tried all methods but it does not work yet. So I summarize the steps what I've done lately:

Firstly i created my directories:

Code: Select all

code_saturne create -s SIMULATION -c BASSIN -c TURBINE

Then I pumy two meshes in "MESH" (Mesh_Turbine.med ans Mesh_Bassin.med)

Then I started Code_Saturne gui, and I read my two meshes and I check it, I went into the menu "periodic boundary' and I defined "rotation: omegay360 degré) and DX=0, DY=1, Dz=0) then i saved my "case.xml" then I started the calculation

Code: Select all

"./runcase"

in the terminal.

HOW CAN I COUPLE MY TWO MESHES (whats the name of subroutine in saturne 2.0.4 please)

I obtiened this error :

ingle processor Code_Saturne simulation

Code: Select all

                   Code_Saturne is running 
                      *********************** 
  Working directory (to be periodically cleaned) : 
     /home/mmeziane/tmp_Saturne/SIMULATION.DOMAINE.06191653

 Kernel version:           /serveurVega/logiciels/saturne/cs-2.0
 Preprocessor:             /serveurVega/logiciels/saturne/cs-2.0/bin

  ***************************************************************
   Compilation of user subroutines and linking of Code_Saturne
  ***************************************************************

  ********************************************
             Preparing calculation            
  ********************************************

  ********************************************
             Starting calculation
  ********************************************

Error running the preprocessor.
Check preprocessor log (listpre) for details.

  ********************************************
         Error in preprocessing stage.

Thank you for your response and proposition

[Yvan Fournier]

As I said in this post: http://code-saturne.org/forum/viewtopic ... 1036#p5821

with version 2.0, you must use the --new-runcase option when running code-saturne create. The directory hierarchy is also different (as you can se by comparing with the old example), so you can only define 1 case (-c option), with 2 domains (--nsat 2).

With the --new-runcase option, the GUI may not be used for the mesh selection, preprocessing and periodicity parts (you need to edit runcase).

In any case, bugs were corrected in version 2.2, and you will need less time learning to indent Python (Google python indent) than downgrading Code_Saturne versions, so I do ned recommend going back to Code_Saturne 2.0 for rotor/stator applications.

Regards,

Yvan

[stage75]

Hi,
thanks for all!

Maybe, we begin to converge towards a good solution. But there are still errors!

I followed all the steps you described. Finally I run "rancase". This is what I get:

Code: Select all

                      Code_Saturne is running
                      ***********************

 Working directory (to be periodically cleaned):
   /home/mmeziane/tmp_Saturne/SIMULATION2.CASE1.06201031

 Kernel version:  2.0.4
 Preprocessor:    /serveurVega/logiciels/saturne/cs-2.0/bin/cs_preprocess

 Code_Saturne domain 1 on 1 process(es).
 Code_Saturne domain 2 on 1 process(es).

 ****************************************
  Compiling user subroutines and linking
 ****************************************


 ****************************
  Preparing calculation data
 ****************************


 ***************************
  Preprocessing calculation
 ***************************


 **********************
  Starting calculation
 **********************

*** An error occurred in MPI_Intercomm_create
*** on communicator MPI COMMUNICATOR 3 SPLIT FROM 0
*** MPI_ERR_INTERN: internal error
 *** MPI_ERRORS_ARE_FATAL (your MPI job will now abort)
--------------------------------------------------------------------------
mpiexec has exited due to process rank 1 with PID 1927 on
node a7-lgc-c2-137 exiting without calling "finalize". This may
have caused other processes in the application to be
terminated by signals sent by mpiexec (as reported here).
--------------------------------------------------------------------------
Code_Saturne solver script exited with status 17.

Error running the calculation.

Check Code_Saturne log (listing) and error* files for details.


 ****************************
  Saving calculation results
 ****************************

 Error in calculation stage.

[stage75]

no idea of this error!!