Noobie question about DEFI_GRUOP

Hi all,

I have a very noobie question. I have created a mesh in Salomé and exported to .med. In the mesh some groups were defined, but I have read some threads which say that eventually it might bring problems, although in most cases it shouldn't be an issue.

Therefore, and as I'm facing problems to have a model running, I thought in importing just the groups that actually are needed in the simulation. As far as I have understood, I need to do something like:

MAIL=DEFI_GROUP(reuse =MAIL,MAILLAGE=MAIL,
CREA_GROUP_MA=(_F(NOM='soil_1'),
_F(NOM='soil_2'),
_F(NOM='soil_3'),),
CREA_GROUP_NO=(_F(GROUP_MA='soil_1'),
_F(GROUP_MA='soil_2'),
_F(GROUP_MA='soil_3'),),
);

But it says

Etape : DEFI_GROUP ligne : 52 fichier : u'fort.1'
Mot-clé facteur multiple : CREA_GROUP_MA
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! L'occurrence numéro 1 du mot-clé facteur : CREA_GROUP_MA n'est pas valide !
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
CREA_GROUP_MA
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! Règle(s) non respectée(s) : !
! - Il faut un et un seul mot-clé parmi : ('TOUT', 'GROUP_MA', 'MAILLE', !
! 'INTERSEC', 'UNION', 'DIFFE', 'OPTION') !
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Fin CREA_GROUP_MA

I used the following to remediate it:

CREA_GROUP_MA=(_F(NOM='soil_1',TOUT='OUI'),
_F(NOM='soil_2',TOUT='OUI'),
_F(NOM='soil_3',TOUT='OUI'),


It doesn't show any error, but I still wonder what I'm doing with that " TOUT='OUI'" option. I did it because I think I'm saying YES to every property of each group, but I'm not sure. Because it's still showing the following alarm message:


LES MAILLES M842 ET M817 SONT DOUBLES (MEME NOEUDS EN SUPPORT) TYPE DES MAILLES:SEG3

<MODELISA4_8> !
! !
! -> Phase de vérification du maillage : présence de mailles doubles (ou triples, ...) !
! Les mailles multiples sont des mailles de noms différents qui ont la même connectivité !
! (elles s'appuient sur les mêmes noeuds). !
! !
! -> Risque & Conseil : !
! Le risque est de modéliser 2 fois (ou plus) l'espace. On peut par exemple avoir !
! un modèle 2 fois trop lourd ou 2 fois trop rigide. !
! Remarque : les mailles concernées sont imprimées dans le fichier "message". !
! Sur ce maillage, il est imprudent d'affecter des quantités avec le mot clé TOUT='OUI'.


and finally a closing error:

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
! <S> Exception utilisateur levee mais pas interceptee. !
! Les bases sont fermees. !
! Type de l'exception : error !
! !
! Le groupe de mailles 'soil_1' existe déjà. !
! !
! Conseil : !
! Si vous souhaitez utiliser un nom de groupe existant, il suffit !
! de le détruire avec DEFI_GROUP / DETR_GROUP_MA. !
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Does this mean that there was a problem of double definition of the groups? How can it be solved?

Thx and cheers, Ferney

Hoi Kees,

Thx for your answer. Actually, now I'm following what you said, just using the groups defined in Salomé. The original post arose from some problems I had at the beginning, because my study wasn't running properly and then in the mess file there were some alert messages about repeated elements SEG6, so I thought that the 'additional' groups, i.e. those defined in salomé, but not used in CA, were causing some problems, and therefore, I wanted to define just the needed meshed. However, I solved it by applying /repair/remove extra edges, since then I don't have the alarm message and I can simply define the input as:

MAIL=DEFI_GROUP(reuse =MAIL,MAILLAGE=MAIL,
CREA_GROUP_MA=_F(NOM='TOUT',TOUT='OUI',),
CREA_GROUP_NO=(_F(TOUT_GROUP_MA='OUI',),)
);

Once again, thank you for your help,

Ferney