Non Converge on Drucker Prager Simulation

[Galesky]

Hello there,

I'm studying the displacement on an earth dam caused by the water on one of the sides.

The parameters i used were as follow :

USING SFGRAD

Spoiler

/prep7

K,1,0,0
K,2,122.5,30
K,3,134.5,30
K,4,257,0

L,1,2
L,2,3
L,3,4
L,1,4

FLST,2,4,4
FITEM,2,1
FITEM,2,2
FITEM,2,3
FITEM,2,4
AL,P51X



ET,1,PLANE82 !PLANE 182 E 183 NÃO PODEM SER USADO PARA DRUCKER-PRAGER

! DEFINIDO AS CARACTERISTICAS DO MATERIAL QUE FORMA A BARRAGEM
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,DENS,1,,1.68e3
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,EX,1,,16412.18e3
MPDATA,PRXY,1,,.49
TB,DP,1,,,
TBMODIF,1,1,20e3
TBMODIF,1,2,10.78
TBMODIF,1,3,0


!CRIANDO A MALHA
ESIZE,2,0,
MSHKEY,0
CM,_Y,AREA
ASEL, , , , 1
CM,_Y1,AREA
CHKMSH,'AREA'
CMSEL,S,_Y
!*
!*
ACLEAR,_Y1
AMESH,_Y1

!CONTRINGINDO A BASE EM Y E O LIMITE À DIREITA EM X e Y

FLST,2,1,3,ORDE,1
FITEM,2,4
!*
DK,P51X, , , ,0,ALL, , , , , ,
FLST,2,1,4,ORDE,1
FITEM,2,4
!*
/GO
DL,P51X, ,UY,



!APLICANDO O GRADIENTE DE PRESSÃO
SFGRAD,PRES,0,Y,0,-9810 ! PRESSAO INICIANDO EM 0,CADA METRO EM Y REDUZ 9810
nsel,s,NODE,,3,129 !SELECIONAR OS NÓS DO CONTATO (NSEL) ALTERAR SE ALTERA MALHA
SF,ALL,PRES,294300 !todo os nós , pressão inicial

/SOLU
!PARÂMETROS DA SOLUÇÃO
NROPT,UNSYM
NLGEOM,ON
ANTYPE,0
NSUB,500,700,400
TIME,1

Using constant pressure along the slope :

Spoiler

/prep7

K,1,0,0
K,2,122.5,30
K,3,134.5,30
K,4,257,0

L,1,2
L,2,3
L,3,4
L,1,4

FLST,2,4,4
FITEM,2,1
FITEM,2,2
FITEM,2,3
FITEM,2,4
AL,P51X



ET,1,PLANE82 !PLANE 182 E 183 NÃO PODEM SER USADO PARA DRUCKER-PRAGER

! DEFINIDO AS CARACTERISTICAS DO MATERIAL QUE FORMA A BARRAGEM
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,DENS,1,,16.8e3
MPTEMP,,,,,,,,
MPTEMP,1,0
MPDATA,EX,1,,16412.18e3
MPDATA,PRXY,1,,.49
TB,DP,1,,,
TBMODIF,1,1,20e3
TBMODIF,1,2,10.78
TBMODIF,1,3,0


!CRIANDO A MALHA
ESIZE,2,0,
MSHKEY,0
CM,_Y,AREA
ASEL, , , , 1
CM,_Y1,AREA
CHKMSH,'AREA'
CMSEL,S,_Y
!*
!*
ACLEAR,_Y1
AMESH,_Y1

!CONTRINGINDO A BASE EM Y E O LIMITE À DIREITA EM X e Y

FLST,2,1,3,ORDE,1
FITEM,2,4
!*
DK,P51X, , , ,0,ALL, , , , , ,

FLST,2,1,4,ORDE,1
FITEM,2,4
!*
/GO
DL,P51X, ,ALL,


!APLICANDO PRESSÃO CONSTANTE de 140kN/m
FLST,2,1,4,ORDE,1
FITEM,2,1
/GO
!*
SFL,P51X,PRES,140e3,


/SOLU
!PARÂMETROS DA SOLUÇÃO
NROPT,UNSYM
NLGEOM,ON
ANTYPE,0
NSUB,500,700,400
time,1

I noticed that it does converge if the pressure is not that intense (until around 110kN/m)
But, using a small value will not satisfy my goal.

I hope you guys can help me.

 

[rickfischer51]

If I remember correctly Drucker-Prager is elastic, perfectly plastic. Loading with a pressure will result in structural instability, as indicated by a failure to converge, when the load gets to a certain limiting value. Ramp up the load slowly and look at the last converged substep to see if you are near the yield point.

Rick Fischer
Principal Engineer
Argonne National Laboratory