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

You are not logged in.

#1 2015-07-22 09:00:00

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

[SOLVED] How to take into account Initial Stress

Hi everybody,

I would want to simulate a mechanical loading to a structure taking into account the fact that the material has residual stress after heat treatment.
To make this calculation, I have understand that I have to create the fields of residual stress and take into account this field in my STAT_NON_LINE in ETAT_INIT / SIGM.

Up to now, my calculation converges. When I look at the results, I see that at STEP 0, my model has the residual stress. But after the first iteration of the external load, the residual stress disappear and only the external load is considered.

(I join my comm file)

I have read the documentation on how-to implement this initial stress (U2.01.09 & U4.72.04) and I have still a lot of point that I don't understand.
- In my case, in ETAT_INIT, I only consider SIGM. Do I have to consider VARI and/or DEPL ?
- Is it necessary that the initial stress field is in equilibrium ? (In the attached .comm file, the initial stress in in equilibrium, but it's just an example. Sometime, in reality, we don't know all the components of this stress).
- Do I have to make a pre-solve, based on a "bidon" calculation to initialize some variables and/or displacements ?

Thanks for your help

Stephane

Last edited by stephaneberger (2015-08-01 18:52:00)


Attachments:
initial_stress.comm, Size: 3.61 KiB, Downloads: 331

Offline

#2 2015-07-22 12:34:38

katoulos
Member
Registered: 2011-06-10
Posts: 72

Re: [SOLVED] How to take into account Initial Stress

I would try AFFE_CHAR_MECA, keyword PRE_EPSI

best regards

mimis

Offline

#3 2015-07-23 07:28:25

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

Re: [SOLVED] How to take into account Initial Stress

Hi,

Thanks for your answer. I have tried to use PRE_EPSI/PRE_SIGM but as it explain in the U4.44.01, these operand are different compared to ETAT_INIT.
Actually, when I use PRE_SIGM, there is no pre-stress at the forst step of my non-linear calculation.

Since my post, I have made more tests. I have applied displacement and stress as initial conditions (ETAT_INIT / DEPL and ETAT_INIT / SIGM). It appears that the initial condition are well taken into account. Apparently there is no need of internal variable initialisation.
Now, I understand the situation as follow : when we use ETAT_INIT / SIGM, the stress field needs to be at equilibrium because otherwise the mechanical equilibrium is recalculated at the first iteration.

I'm still going on my understanding.

Thanks again

Stephane

Offline

#4 2015-08-01 18:54:23

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

Re: [SOLVED] How to take into account Initial Stress

HI,

I've just realised that the initial stress is well taken into account as long as the COMPORTEMENT in STAT_NON_LINE is equal to ELAS.
Now, I have to see if the initial stress can be introduced with a strain hardenning behaviourbut It will be the object of an other post.

Thanks to all.

stephane

Offline