abaqus restart problem

[sunlinfeng]

Hi all,
I've been facing a very strange problem now when I'm using abaqus to finish RVE multiscaling simulation: so far, what I've done is successfully linking fortran and abaqus to run the whole calculation. At first, I met a small problem--- overwrite the old files(y/n)?---which has been solved by adding a "abaqus_v6.env" file whose content is as follows in the abaqus working direction:
ask_delete=OFF
But another bigger problem has come. When abaqus running the RVE's initial step which has the identity deformation gradient loaded, everything goes fine. However, when the next "real" loading step comes to increment 2, the whole running process has been abort due to the errors. the macro input model's message file says:
***NOTE: THE SOLUTION APPEARS TO BE DIVERGING. CONVERGENCE IS JUDGED UNLIKELY.
***ERROR: TOO MANY ATTEMPTS MADE FOR THIS INCREMENT
Because when RVE's analysis comes to increment n (n>1), the RVE's restart job will begin. So I check the RVE's restart job's message file, it says:
***ERROR: USER SUBROUTINE DISP IS MISSING
This is so strange! Because you can see the RVE's initial job's message file, everything is good:
THE ANALYSIS HAS BEEN COMPLETED
So why there is a sentence "***ERROR: USER SUBROUTINE DISP IS MISSING" in the restart message file? where is the problem coming from? Is there anything wrong with my input file? So now I'll list the original RVE's input file and RVE's restart input file.

Original RVE's input file(name is R_0010.inp, used in initial step and 1st increment of "STEP: apply")
*Heading
** Job name: R_0010 Model name: Model-1
** Generated by: Abaqus/CAE 6.13-1
*Preprint, echo=NO, model=NO, history=NO, contact=NO
** ----------------------------------------------------------------
**
** PART INSTANCE: part-c-1
**
*Node
1, 0., 0., 0.200000003
*******************************************
*Element, type=C3D4
1, 733, 734, 735, 736
****************************************
*Nset, nset=part-c-1__PickedSet2, generate
1, 2189, 1
*Elset, elset=part-c-1__PickedSet2, generate
1, 11009, 1
** Section: Section-1
*Solid Section, elset=part-c-1__PickedSet2, material=Material-1
,
** ----------------------------------------------------------------
**
** PART INSTANCE: Part-merge-1
**
*System
*Nset, nset="Face XN", generate
547, 550, 1
***********
*Nset, nset="corner_nodes"
4, 7, 3, 1
***********
** Constraint: 2-1 edge node 0 on u
*Equation
3
"Node2 II_0", 1, 1.
"Node1 I_0", 1, -1.
"corner_node_2", 1, -1.
**********
**
** MATERIALS
**
*Material, name=Material-1
*Elastic
155000., 0.27
**
** BOUNDARY CONDITIONS
**
** Name: Translation stop corner_node_1 Type: Displacement/Rotation
*Boundary
"corner_node_1", 1, 1
"corner_node_1", 2, 2
"corner_node_1", 3, 3
** ----------------------------------------------------------------
**
** STEP: apply
**
*Step, name=apply, nlgeom=YES
*Static,direct
1, 1.,
*FILE FORMAT, ZERO INCREMENT
**
** BOUNDARY CONDITIONS
**
** Name: BC-5 Type: Displacement/Rotation
*Boundary, user
"corner_node_5", 1, 3
"corner_node_4", 1, 3
"corner_node_2", 1, 3
**
** OUTPUT REQUESTS
**
*EL PRINT, FREQ=0
*NODE PRINT, FREQ=0
*Restart, write, frequency=1
*EL FILE
S,
*NODE FILE, NSET="corner_nodes"
RF,COORD,U
*End Step
** ----------------------------------------------------------------
**
** STEP: SubstructureGeneration
**
*Step, name=SubstructureGeneration, nlgeom=YES
*Substructure Generate,Type=Z1,Library=R_0010,Overwrite
**
*Retained Nodal Dofs,Sorted=YES
"corner_nodes", 1,3
**
*Boundary,OP=NEW
**
*Substructure Matrix Output,Stiffness=YES,
Output File=USER DEFINED,File Name=KM
*End Step

Restart RVE's input file(name is R_re_0010.inp, used in Nth increment(N>1) of "STEP: apply")

*Heading
*Restart,read
** STEP: apply
**
*Step, name=apply, nlgeom=YES
*Static,direct
1, 1.,
*FILE FORMAT, ZERO INCREMENT
**
** BOUNDARY CONDITIONS
**
** Name: BC-5 Type: Displacement/Rotation
*Boundary, user
"corner_node_5", 1, 3
"corner_node_4", 1, 3
"corner_node_2", 1, 3
**
** OUTPUT REQUESTS
**
*EL PRINT, FREQ=0
*NODE PRINT, FREQ=0
*Restart, write, frequency=1
*EL FILE
S,
*NODE FILE, NSET="corner_nodes"
RF,COORD,U
*End Step
** ----------------------------------------------------------------
**
** STEP: SubstructureGeneration
**
*Step, name=SubstructureGeneration, nlgeom=YES
*Substructure Generate,Type=Z1,Library=R_re_0010,Overwrite
**
*Retained Nodal Dofs,Sorted=YES
"corner_nodes", 1,3
**
*Boundary,OP=NEW
**
*Substructure Matrix Output,Stiffness=YES,
Output File=USER DEFINED,File Name=KM
*End Step

So, is there anyone can help me get rid of this problem? It's a little urgent to me, many thanks!

 

[Mustaine3]

So in the first analysis you have no convergence and in the restart analysis the subroutine is not found, right?

Check the command that you use for the restart analysis.

It's impossible to say with these informations what's the problem in the first analysis. But the direct static method with that incrementation is usually not good.

 

[sunlinfeng]

Hi Mustaine3,
The attachment is the message file of the macro analysis, and the sta file of macro analysis's content is as follows:
Abaqus/Standard 6.13-1 DATE 23-Mar-2015 TIME 00:11:50
SUMMARY OF JOB INFORMATION:
STEP INC ATT SEVERE EQUIL TOTAL TOTAL STEP INC OF DOF IF
DISCON ITERS ITERS TIME/ TIME/LPF TIME/LPF MONITOR RIKS
ITERS FREQ
1 1 1 0 2 2 1.00e-005 1.00e-005 1.000e-005
1 2 1U 0 4 4 1.00e-005 1.00e-005 1.000e-005
1 2 2U 0 4 4 1.00e-005 1.00e-005 2.500e-006
1 2 3U 0 4 4 1.00e-005 1.00e-005 6.250e-007
1 2 4U 0 4 4 1.00e-005 1.00e-005 1.563e-007
1 2 5U 0 4 4 1.00e-005 1.00e-005 3.906e-008

THE ANALYSIS HAS NOT BEEN COMPLETED

the sta file of RVE's initial and 1st increment analysis's content is as follows:
Abaqus/Standard 6.13-1 DATE 23-Mar-2015 TIME 00:08:11
SUMMARY OF JOB INFORMATION:
STEP INC ATT SEVERE EQUIL TOTAL TOTAL STEP INC OF DOF IF
DISCON ITERS ITERS TIME/ TIME/LPF TIME/LPF MONITOR RIKS
ITERS FREQ
1 1 1 0 1 1 1.00 1.00 1.000
2 1 1 0 0 0 1.00 1.00e-036 1.000e-036

THE ANALYSIS HAS COMPLETED SUCCESSFULLY

when we start the "restart computation" by using the R_re_0010.inp, the errors come out. So I'm not sure if there is something wrong in my input file. Because I know that subroutines need to be specified again when submitting a restart job since they are not stored as part of restart information. Besides, I wrote a batch file to be called in subroutine UMAT in order to run the restart work. the code is as follows:

---------------------------------
IF (KINC.EQ.1) THEN
open(800,
* file='C:\Temp\Execute_1.bat',
* STATUS='REPLACE')
Write(800,*)'@echo off'
Write(800,*)'echo This is a batch file'
Write(800,*)'cd C:\temp && copy R_0010.inp R_',nnnn,'.inp'
Write(800,*)'copy C:\Temp\R_0010.inp C:\Temp\R_',nnnn,'.inp'
close (800)
C
open(900,
* file='C:\Temp\Execute.bat',
* STATUS='REPLACE')
Write(900,*)'@echo off'
Write(900,*)'echo This is a batch file'
Write(900,*)'cd C:\temp && abaqus job=R_',nnnn,
* ' input=R_',nnnn,'.inp user=userSub cpus=8 int'
C Write(900,*)'abaqus job=R_',nnnn,'.inp user=user_test int'
close (900)
C
C
result=SYSTEM('cd C:\temp && Execute_1.bat int')
result=SYSTEM('cd C:\temp && Execute.bat int')
write(*,*) 'result=',result
----------------------------------------------
this part runs good, now here is for the restart work

------------------------------------------------
ELSEIF(KINC.GT.1) THEN
WRITE(*,*) 'Running RVE analysis job ', nnnn
WRITE(*,*) 'Increment:', KINC
WRITE(*,*) 'Element number:', NOEL
WRITE(*,*) 'Integration point number:', NPT
open(600,
* file='C:\Temp\Exe_1.bat',
* STATUS='REPLACE')
Write(600,*)'@echo off'
Write(600,*)'echo This is a batch file'
Write(600,*)'cd C:\temp && copy R_re_0010.inp R_re_',nnnn,
* '.inp'
close(600)
C
open(500,
* file='C:\Temp\Exe.bat',
* STATUS='REPLACE')
Write(500,*)'@echo off'
Write(500,*)'echo This is a batch file'
Write(500,*)'cd C:\temp && abaqus job=R_re_',nnnn,
* ' oldjob=R_',nnnn, ' input=R_re_',nnnn,
* '.inp user=userSub cpus=8 int'
close (500)
C
result=SYSTEM('cd C:\temp && Exe_1.bat int')
result=SYSTEM('cd C:\temp && Exe.bat int')
----------------------------------------------------------
You can see I have already set the specific subroutine, but why the error "USER SUBROUTINE DISP IS MISSING" still comes out? the "userSub" subroutine has already included the DISP subroutine.

 

[sunlinfeng]

for your convenience, here is the msg file for R_0011

 

[sunlinfeng]

and also the msg file for R_re_0012