#1 2021-12-21 13:44:36

VonPire

Member

Registered: 2021-01-14

Posts: 33

Restart of unconverged Case & Post-processing of last converged step

Hello everyone !

I'm using Salome-Meca 2021 and Ubuntu 20.04 LTS.

I have tried to get working the restart of the unconverged RunCase and also postprocessing of the last converged step of unconverged RunCase with separate RunCase with no succes. I think that this problem is due to the Salome-Meca 2021 AsterStudy module. I have read from the other topics that there shouldn't be this kind of problem when using original Code_Aster "standalone" or ASTK.

However the restart of the RunCase and postprocessing of the results in separate RunCase works if the original simulation RunCase has converged and calculated fully to the end (Green icon indicating completed RunCase).

This is the procedure that works when Simulation RunCase is converged and calculated fully to the end:

General structure of the Simulation RunCase:

DEBUT()
...
...
resnonl = STAT_NON_LINE(
                 ARCHIVAGE=_F(
                 LIST_INST=listr),
                 INCREMENT=_F(
                 LIST_INST=listr),
                 ...,
                 ...,)

FIN()

General structure of the post-processing RunCase:

POURSUITE()

resnonl = CALC_CHAMP(...,
                                            ...,
                                           RESULTAT=resnonl,
                                           reuse=resnonl)
IMPR_RESU(
  FORMAT='MED',
  RESU=_F(
    RESULTAT=resnonl
  ),
  UNITE=80
)

FIN()

General structure of the restart RunCase:

POURSUITE()

resnonl = STAT_NON_LINE(
                 ARCHIVAGE=_F(
                 LIST_INST=storing
                 ),
                 ETAT_INIT=_F(
                 EVOL_NOLI=resnonl,
                 NUME_ORDRE=16
                 ),
                 INCREMENT=_F(
                 LIST_INST=listr
                 ),
                 RESULTAT=resnonl,
                 )
FIN()

This Procedure doesn't work anymore for restarting or postprocessing if Simulation RunCase is stopped/unconverged (RunCase is shown as red) for example due to Newton iteration limit in 80%/100% of time and we are requesting restart time (with NUME_ORDRE) before uncorverged timestep. For some reason the blue reuse button cannot activated anymore if the Simulation Case isn't converged fully even there is the ARCHIVAGE on (in the STAT_NON_LINE).

Does anyone have solution to this problem ?

Best regards,
VonPire