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

You are not logged in.

#1 2015-08-10 15:21:45

stephaneberger
Member
From: Strasbourg (France)
Registered: 2012-10-15
Posts: 70

Incompatibility between SIMO_MIEHE and ETAT_INIT

Hi,

I'm doing a simulation taking into account an initial stress (due to heat treatment) and doing an elastoplastic calculation using chaboche modele in STAT_NON_LINE.
The initial stress is introduced using ETAT_INIT in COMP_INCR.
I have seen that, if DEFORMATION in set to PETIT or GROT_GDEP, the initial stress is correctly taken into account whereas if I use SIMO_MIEHE the initial stress is loosen at the first iteration.
I have tried to find out some documentation on SIMO_MIEHE to see if this behaviour is the right one but nothing found up to now.

Could someone tell me if it's a bug or if it's due to the formulation of SIMO_MIEHE ?

Thanks for your Help

Stephane

------------------------------------
SALOME-MECA 2015.2

Offline

#2 2015-08-10 18:54:04

pierre_j
Member
Registered: 2010-01-19
Posts: 773

Re: Incompatibility between SIMO_MIEHE and ETAT_INIT

Hi,

I am no expert, and a Code_Aster developper may better answer your question than me.

However, as a workaround, did you also try with GDEF_LOG?

GDEF_LOG offers equivalent finite strain management capability than SIMO_MIEHE, being in addition able to handle anisotropic yield criterion, on the opposite to SIMO_MIEHE if my understanding is correct.

Best regards,

Pierre

Offline

#3 2015-08-16 14:27:31

Thomas DE SOZA
Guru
From: EDF
Registered: 2007-11-23
Posts: 3,066

Re: Incompatibility between SIMO_MIEHE and ETAT_INIT

Hi,

If you only input stresses with ETAT_INIT/SIGM then this may be normal. SIMO_MIEHE handles large strains and store information additional information in history (internal) variables.
Still if the prestressing keeps the structure in an elastic state with small strains then this should nonetheless work. Otherwise you'll need to input ETAT_INIT/VARI or generate the prestressing with a previous non-linear calculation and input the whole state with ETAT_INIT/EVOL_NOLI.

TdS

Offline

#4 2015-08-27 15:11:52

stephaneberger
Member
From: Strasbourg (France)
Registered: 2012-10-15
Posts: 70

Re: Incompatibility between SIMO_MIEHE and ETAT_INIT

HI,

Sorry for the late reply.

@Thomas,
Actually I have made several tests. I have tried to initialise the internal variable and I have made some simulations staying in the elastic domain. Each time, I have the same behaviour with SIMO_MIEHE.

@Pierre,
I will try with GDEF_LOG. In any cas I will have to use this because at the end I want to use a chaboche behaviour law wich cannot be used with SIMO_MIEHE.

Thanks again

Stephane

Offline

#5 2015-10-22 00:44:52

Thomas DE SOZA
Guru
From: EDF
Registered: 2007-11-23
Posts: 3,066

Re: Incompatibility between SIMO_MIEHE and ETAT_INIT

stephaneberger wrote:

@Pierre,
I will try with GDEF_LOG. In any cas I will have to use this because at the end I want to use a chaboche behaviour law wich cannot be used with SIMO_MIEHE.

Note that GDEF_LOG uses a similar approach to SIMO_MIEHE to store initial state.
There's a not about this in the STAT_NON_LINE manual (U4.51.03) :

En grandes déformations, l'utilisateur souhaitant utiliser le formalisme GDEF_LOG avec un champ de contrainte initial (ETAT_INIT) se reportera au cas test SSNP159B. En effet, pour imposer un champ de contrainte initial, l'utilisateur doit donner en entrée le tenseur de contrainte défini dans l'espace logarithmique T (et non celui de Cauchy σ). Les composantes de ce dernier étant stockées en tant que variables internes, il faut utiliser les opérandes VARI et DEPL décrites ci- dessous (ces champs peuvent par exemple être obtenus par la commande CREA_CHAMP [U4.72.04]).

In a nutshell, you may refer to testcase SSNP159B for an example on the way to initialize a large strain calculation with this kinematic hypotheses.

TdS

Offline

#6 2015-10-22 10:53:15

stephaneberger
Member
From: Strasbourg (France)
Registered: 2012-10-15
Posts: 70

Re: Incompatibility between SIMO_MIEHE and ETAT_INIT

Hi Thomas,

Thank you for your answer.
Effectively you are right. I have never used the operandes VARI and DEPL to initialise in Large strain.

I will try as soon as I can.

Thanks again.

Stephane

Offline