Welcome to the forums. Please post in English or French.

You are not logged in.

#1 Re: Code_Aster usage » Using aster.putvectjev instead of DETRUIRE » 2011-10-07 17:08:41

BL

Richard,

I could never get this method to work.  And other priorities prevented me from looking more closely.

I still have problem with memory consumption, but I believe it has to do with the application of varied pressures.  The AFFE_CHAR_MECA that I use never fully releases all of the memory when I delete with DETRUIRE.  The command consumes some memory fixed memory for each command, 40kb or so.  It also seem to me some lesser amount of memory is used to remember each keyword within that command, which appear to be here my problem really lies.

However as I stated in http://www.code-aster.org/forum2/viewtopic.php?id=15050, this problem is fixed by restarting Aster.  It was much easier to write a script that relaunches than to try to actively handle the internal data.

#3 Code_Aster usage » [SOLVED] Large deformations » 2011-07-19 18:13:02

BL
Replies: 2

Hello

I am trying to model an object the will undergo large deformations and strain.  My current experience has been small deformations so I have been using shell elements either DKT or COQUE_3D as needed.  However from what I can translate from the documentation for large deformation I should be using SIMO_MIEHE and for that I need 3D elements.

Can someone please tell me if I have correctly interpreted the documentation?

#4 Code_Aster usage » What causes the error <F>_DECOMPRESSION? » 2011-06-21 23:09:37

BL
Replies: 1

In a lengthy chained calculation that I am running, my calculation starts and restarts fine for a time.  However eventually the restart ends with the error <F>_DECOMPRESSION.

What does this mean and how can I fix or prevent the problem?

Is it possible this actually an issue from the end of the previous calculation?

#5 Re: Code_Aster usage » impr_table to extraction nodes coordinates » 2011-06-17 17:33:29

BL

I agree with dbpatankar.  I do something similar for extracting my Stress Result into a python data structure that meets my needs.  I use the following code which extracts the Von Mises Stress and the Node it is associated with.  Take note of the val['NOEUD'][1:] in my example or the values[1][1:] used by dbpatankar.  This is removing the N from the node number.

Stress=POST_RELEVE_T(ACTION=_F(OPERATION='EXTRACTION',
        INTITULE='',
        RESULTAT=S[num],
        NOM_CHAM='EQUI_NOEU_SIGM',
        PRECISION=outputPrecision,
        INST=calcTime,
        GROUP_NO=(  postRelieveT ),
        NOM_CMP=('VMIS',),),);

StressT = Stress.EXTR_TABLE();
PyStress = {}
for val in StressT:
    PyStress[int(val['NOEUD'][1:])] = [ float(val['VMIS']) ]

#6 Re: Code_Aster usage » Rotation Modelling » 2011-04-15 15:07:29

BL

I am not sure but I think what you need is LIAISON_CYCL.

#7 Re: Code_Aster usage » [Solved] POURSUITE with PAR_LOT » 2011-04-15 15:01:25

BL

Hmm...seems I can import python directories before POURSUITE().  And that fixes everything.

#8 Re: Code_Aster usage » Using aster.putvectjev instead of DETRUIRE » 2011-04-07 17:20:22

BL

Richard, I don't have it working just yet.  But when I get it working or have problems I can't seem to solve I will post something.

#9 Re: Code_Aster usage » Using aster.putvectjev instead of DETRUIRE » 2011-04-06 17:10:16

BL

Thank you Thomas!

Yes, I was using an older version accidentally.   That seems to have been a large part of my problem.  It seems now all I need is to fix a few details.

#10 Code_Aster usage » Using aster.putvectjev instead of DETRUIRE » 2011-04-01 17:39:37

BL
Replies: 13

I am having a memory problem created by the use of a python loop.  As discussed in the thread:
http://www.code-aster.org/forum2/viewtopic.php?id=15050

I read this thread:
http://www.code-aster.org/forum2/viewtopic.php?id=13541

I am interested in doing the very thing this thread suggests.  However I fail in implementation. First I setup the pressure condition shown below and then try to access the method as described in the second link.

for some pressures:
    pressureList.append(  _F(MAILLE='M%s'%mesh,PRES=PressVal,) )

Pressure=AFFE_CHAR_MECA(MODELE=Mesh,
    FORCE_COQUE=(  pressureList  ),);

Previous = list(Pressure.CHME.PRESS.VALE.get());

By my understanding Previous should have accessed the internal structure of Aster which I can then modify with python and re-insert it with aster.putvecjev(). Instead I get an error:
   !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
   ! erreur non prevue et non traitee prevenir la maintenance fort.1 !
   ! Traceback (most recent call last):                              !
   !    File "./Python\Noyau\N_JDC.py", line 164, in exec_compile    !
   !     exec self.proc_compile in self.g_context                    !
   !    File "fort.1", line 583, in ?                                !
   !     Previous = list(Pressure.CHME.PRESS.VALE.get());            !
   !  AttributeError: 'char_meca' object has no attribute 'CHME'     !
   !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

But I believe it does as the command IMPR_CO(CHAINE='Pressure.CHME.PRESS.VALE',) works.

If anyone can explain this, or even tell me what documentation file I should be reading to learn more I would be very grateful.

#11 Re: Code_Aster usage » Problem/Question regarding DETRUIRE() » 2011-04-01 15:44:08

BL

Richard

I understood it the same way you describe.  The method as described might not work for you, but if you could do the same to the material as was suggested for the loading it might work.

Good Luck!

#12 Re: Code_Aster usage » Problem/Question regarding DETRUIRE() » 2011-04-01 14:51:29

BL

I am not to be able to read French that well, but Google is getting pretty good.

I haven't had a chance to try it out yet.  However the solution of modifying the loading data structure looks really promising.

#13 Re: Code_Aster usage » Problem/Question regarding DETRUIRE() » 2011-03-31 20:17:16

BL

I have noticed this too!  I couldn't find a good work around that does't require me to restart Aster.  Hopefully someone can answer this.

#14 Code_Aster usage » [Solved] POURSUITE with PAR_LOT » 2011-03-25 21:37:48

BL
Replies: 1

I am trying to run a study where I am using python for added control.  With the command DEBUT(PAR_LOT='NON',); everything runs well.  However when I try to re-launch the simulation with POURSUITE(PAR_LOT='NON',);.  It fails on initialization with this error message:

       FIN              &EXECUTE
       POURSUITE        &EXECUTE
JDC.py : ERREUR A L'INTERPRETATION DANS ACCAS - INTERRUPTION
>> JDC.py : DEBUT RAPPORT
CR phase d'initialisation
   !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
   ! Etape  POURSUITE ligne :  6 fichier :  fort.1  impossible d affecter un type au !
   ! resultat                                                                        !
   !    File "./Python\Cata\ops.py", line 150, in POURSUITE                          !
   !     self.jdc.restore_pickled_attrs(pickle_context)                              !
   !    File "./Python\Execution\E_JDC.py", line 307, in restore_pickled_attrs       !
   !     d = context.get('jdc_pickled_attributes', {})                               !
   !  AttributeError: 'NoneType' object has no attribute 'get'                       !
   !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
fin CR phase d'initialisation

This appear to be an issue reading the Base from the previous calculation but it appears to have finished reading all command from the original simulation.

What might the problem be?

#15 Re: Code_Aster usage » CA equivalent for RBE2 and RBE3 » 2011-03-25 20:46:33

BL

While I have no experience with Nastran, one of my previous coworkers claimed LIASON_SOLIDE was what Nastran called RBE2.

#16 Re: Code_Aster usage » Launch Aster on Windows with Perl script » 2011-03-18 16:24:41

BL

Sorry!  Yes you are right.  At the bottom I added.

python %aster%\ASTK\ASTK_SERV\bin\as_run %*

ENV.bat set up the enviroment properly, which I suppose I could do in my script but I didn't.  And then as_run launches Aster.

#17 Re: Code_Aster usage » Launch Aster on Windows with Perl script » 2011-03-17 17:45:45

BL

I have also used both of those sources.

for 9.1 downloaded from NECS website try Windows command line:
ENV.bat  Test.export

This requires a valid export file.  I create mine with a template I developed by examining the ones created by ASTK.

for 10.3 downloaded from PiMeca website try the Windows command line:
waster.bat  Test.comm  cwd=C:\CurrentWorkingDirectory

This one only requires a working command file, but without EFICAS that may require some work too.  I use a python script that scans my mesh file to create mine.


Good Luck!

#18 Re: Code_Aster usage » Launch Aster on Windows with Perl script » 2011-03-14 20:31:42

BL

I have a python script that launches Aster, so maybe I can give you a few ideas.  However it would be useful to know what version of Aster for Windows are you using and where you got it from.

#19 Re: Code_Aster usage » Help with CALC_NO » 2011-02-28 19:01:26

BL

Thomas DE SOZA wrote:

Just a word to say that the error encountered in STA10.3 was a bug that has already been fixed in the NEW10 branch.



Does this error exist in STA10.2?  For me (on Windows) it is much easier to return to a lower version than to update to the latest.

#20 Re: Code_Aster usage » Help with CALC_NO » 2011-02-18 18:52:26

BL

Well, to be honest this is only a simple test case.  I don't really care about the results.  Just that I can get it to work.  Usually I use this to get a static loading initial condition for use with DYNA_NON_LINE which doe in fact have a more useful time of calculation.

With your edit command file I get the attached .mess file.  Remember it is very possible that you have everything correct.  I am only beginning to test this installation it may be that the fault lies there.

#21 Re: Code_Aster usage » Help with CALC_NO » 2011-02-18 18:00:51

BL

jean pierre

I have reviewed your file.  Some of your comments are spot on, I realize I do some strange things in my command file.  That is because I have generate a script to create the command file from my mesh file.  Then I made some simplifications by hand.  This has led to some bizarre syntax I agree.

However your change to CALC_ELEM no ends my simulation in error!  As such I cannot even determine what you are trying to do in CALC_NO...my original question.

Additionally I also can not help but notice you have altered the fields of interest, the time scale involved.  How is this supposed to help?

#22 Re: Code_Aster usage » Help with CALC_NO » 2011-02-18 00:32:01

BL

Here is the mesh and command file.

#23 Re: Code_Aster usage » Help with CALC_NO » 2011-02-17 18:17:40

BL

Whoops!  I thought I included that crucial information.  Version 10.3

#24 Code_Aster usage » Help with CALC_NO » 2011-02-16 18:39:29

BL
Replies: 11

Hello.

I am trying to test the latest version of Aster posted to Windows.  However it does not include EFICAS.  This makes the changes in Syntax form version 9.1 difficult.  I have a test that worked in version 9.1 that I am trying to update.  However I am having problems with the CALC_NO command.

Originally in 9.1 it read:
Solver=CALC_NO(reuse =Solver,
               RESULTAT=Solver,
               OPTION=('SIGM_NOEU_COQU','EQUI_NOEU_SIGM','EPSI_NOEU_DEPL',),
               CHAM_MATER=AffectM,
               CARA_ELEM=Elements,);

Now I get the error:
!-------------------------------------------------------!
! <EXCEPTION> <DVP_1>                                   !
!                                                       !
! Erreur de programmation : condition non respectée.    !
!                                                       !
! Il y a probablement une erreur dans la programmation. !
! Veuillez contacter votre assistance technique.        !
!-------------------------------------------------------!

The documentation seems to suggest I need  something like:
Solver=CALC_NO(reuse =Solver,
               RESULTAT=Solver,
               OPTION=OPTION = 'FORC_NODA',
               CHAM_MATER=AffectM,
               CARA_ELEM=Elements,);

But I cannot determine where the fields I want to calculate should go, but I know they must go somewhere.  Can someone please help me?

#25 Re: Code_Aster usage » Error with FACE_IMPO [Solved] » 2011-02-09 22:51:10

BL

I found a way to force my version to behave properly by fixing a few edge condition and including a surface I really didn't want to model.  One of these days I'll have to update my version.