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

You are not logged in.

#1 2018-07-11 20:40:57

Malleus
Member
Registered: 2017-09-02
Posts: 1

Fatigue calculation with nonlinear static analysis

Hello,

I am having troubles combining CALC_FATIGUE with STAT_NON_LINE. I am basically following your test case sslv135a, where a huge number of different fatigue options is exemplified. Amongst these I chose the first block in each i-loop, which gives me two MATAKE_MODI_AC simulations, one with linear MECA_STATIQUE, the other one with nonlinear STAT_NON_LINE. Then I changed my mesh to an elongated bar with a greater number of elements than in the original CUBE geometry of sslv135a.

The linear static analysis produces a fatigue result as expected. Both sides of the bar display a smooth variation of the variable NBRUPT1 between 2.62e4 and 5.85e6. So far so good.

But if I change the analysis to nonlinear,

resnonl = STAT_NON_LINE(
    CHAM_MATER=mater,
    COMPORTEMENT=_F(DEFORMATION='PETIT', RELATION='ELAS', TOUT='OUI'),
    EXCIT=(
        _F(CHARGE=bc, TYPE_CHARGE='FIXE_CSTE'),
        _F(CHARGE=last, FONC_MULT=last_zt)
    ),
    INCREMENT=_F(LIST_INST=zt_sch),
    MODELE=model,
    NEWTON=_F(MATRICE='ELASTIQUE', REAC_INCR=0)
)

resnonl = CALC_CHAMP(
    reuse=resnonl,
    CONTRAINTE=('SIGM_ELNO', 'SIGM_NOEU'),
    DEFORMATION=('EPSI_ELGA', 'EPSI_ELNO', 'EPSI_NOEU'),
    RESULTAT=resnonl
)

fatig = CALC_FATIGUE(
    CHAM_MATER=mater,
    CRITERE='MATAKE_MODI_AC',
    GROUP_MA=('balken', ),
    MAILLAGE=mesh,
    METHODE='CERCLE_EXACT',
    OPTION='DOMA_NOEUD',
    RESULTAT=resnonl,
    TYPE_CALCUL='FATIGUE_MULTI',
    TYPE_CHARGE='PERIODIQUE'
)

using the same loads, boundary conditions and list of time steps, I obtain the strange plot as in the attached png plot: One side of the bar shows the same, smooth, result as in the linear case (lower view), but the other side of the bar displays, for the same NBRUPT1, a zebra pattern, sort of (upper view). This happens only in the nonlinear case.

Can you please explain what is wrong with the nonlinear approach.

Thank you very much for your efforts


Attachments:
nonlinear.png, Size: 201 KiB, Downloads: 53

Offline

Board footer