Code_aster (meca_statique vs stat_non_line)

Project Support
1

After using SIMSCALE i thought to know code aster better to exploit simscale and know whats going on in there. So, I took a very simple linear analysis problem. Where a plate is fixed at both ends and a force is applied on the middle. So it produced succesful result in MECA_STATIQUE. So i tried it in STAT_NON_LINE because I have many project which demands static non linear analysis. But doing so, I found that results are wrong. The bending is negligible. Is there something I dont know about STAT_NON_LINE.

2

Hi @swa!

Please feel free to share your project with us so that I can have a look at it. In the meantime our Code_Aster expert @rszoeke might give you some more information on the capabilities of the platform.

Best,

Jousef

3

Hi @swa,
generally MECA_STATIQUE and STAT_NON_LINE produce the same result for a linear case (small strains, small rotations, etc.) if the setup is comparable and they use the same mesh.
I would need to see both input files to be able to tell you where the problem lies.

Best,
Richard

4

Hi @rszoeke. Here are the files. I am providing .comm files for MECA_STATIQUE and STAT_NON_LINE. I hope you can find the mistake. I am also providing the contour plots of the result for both STATIQUE and STAT NON LINEAR. There were many warning in STAT_NON_LINE test case so I am also providing the log file. Here is the link to those files : https://drive.google.com/file/d/1gAoLFAE4CxXaaEYFZh_lb2awloqNNR7j/view?usp=sharing
FIRST ONE IS MECA_STATIQUE AND THE SECOND IS STAT_NON_LINE

STAT_LINE
STAT_LINE1036×628 23 KB
STAT_NON_LINE
STAT_NON_LINE1036×628 12.1 KB

5

Hey @rszoeke.I gave you permission. I forgot modify the sharing settings

6

Hi @swa,
your files look ok.

I think the problem is just that you are visualizing time step 0, which in a nonlinear analysis shows only the initial conditions, which are always zero if you did not specify any other. This is evident from your screenshot as 1e-41 is essentially zero.

Btw. you are computing 10 time steps from 0 to 1 in 0.1 increments, but have a static load, so you repeat 9 times the same caluclation (from 0.2 to 1.0), which is unnecessary. You could also just use a time stepping with an increment of 1 and only compute one step.

Best,
Richard

1 Like
7

Thanks Richard, @rszoeke
Do you have any thoughts on the warning I had in the STAT_NON_LINE test case log. It is showing many warning.Another thing I forgot to mention earlier.Whenever I use CALC_CHAMP to extract result from CONSTRAINTE( SIGM_ELNO,SIPM_ELNO,SIPO_ELNO) the log is showing that it’s unable to calculate those result (this is common in both case STAT_NON_LINE and MECA_STATIQUE). In output settings i use NOM_CHAM to print those outputs. Does this have any reason.
PS: Sorry for asking the lame question. I dont know how I got stuck without checking. Man, I need to get my thoughts straight

8

Hi,
the warning just states that it was unable to print the fields of intern variables (as there is only a linear-elastic material which does not have any intern variables). You can ignore this warning, or just remove the IMPR_NOM_VARI=‘OUI’.

If I am not wrong *_ELNO fields can only exist for structural elements like shells and beams, which you don’t have in your model, so calculating them fails.

Best,
Richard