223031012
/
DAMASK_EICMD
mirror of https://github.com/achalhp/DAMASK_EICMD.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
DAMASK_EICMD/code/DAMASK_abaqus_exp.f

300 lines
14 KiB
FortranFixed
Raw Normal View History

now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
!--------------------------------------------------------------------------------------------------
! $Id$
!--------------------------------------------------------------------------------------------------
!> @author Philip Eisenlohr, Max-Planck-Institut f<>r Eisenforschung GmbH
!> @author Franz Roters, Max-Planck-Institut f<>r Eisenforschung GmbH
!> @author Koen Janssens, Paul Scherrer Institut
!> @author Arun Prakash, Fraunhofer IWM
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
!> @brief interfaces DAMASK with Abaqus/Explicit
!> @details put the included file abaqus_v6.env in either your home or model directory,
!> it is a minimum Abaqus environment file containing all changes necessary to use the
!> DAMASK subroutine (see Abaqus documentation for more information on the use of abaqus_v6.env)
now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
!--------------------------------------------------------------------------------------------------
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
#ifndef INT
#define INT 4
#endif
#ifndef FLOAT
#define FLOAT 8
#endif
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
#define Abaqus
changed fileopen routines in IO.f90 to use solver dependant functions defined in the interface routines DID NOT test ABAQUS versions yet, not shure whether (V)GETOUTDIR returns path with or without terminating slash
2010-05-10 20:32:59 +05:30
merged precision info into the file prec.f90 and removed prec_single.f90 precision information is now parsed as a makro, defined by the compiler, eg. -DFLOAT=8 for 8 byte long floats or -- in case no values is given -- for the FEM wrapper by the interface files. Please don't forget to run setup_code.py !!!!!!!!!!! abaqus_v6.env is adjusted to specify the standard precision (float=8, int=4) and got some extra compilation options, some deprecated warning supressors are removed Optional Marc modification will follow that allow automatic selection of suitable precision
2012-08-28 21:38:17 +05:30
#include "prec.f90"
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
module DAMASK_interface
changed fileopen routines in IO.f90 to use solver dependant functions defined in the interface routines DID NOT test ABAQUS versions yet, not shure whether (V)GETOUTDIR returns path with or without terminating slash
2010-05-10 20:32:59 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
implicit none
now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
character(len=4), dimension(2), parameter :: INPUTFILEEXTENSION = ['.pes','.inp']
character(len=4), parameter :: LOGFILEEXTENSION = '.log'
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
contains
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief just reporting
!--------------------------------------------------------------------------------------------------
renamed interface from "mpie_interface" to "DAMASK_interface"
2011-05-11 22:31:03 +05:30
subroutine DAMASK_interface_init()
removed time information from interface routines, should work now
2013-02-26 18:19:07 +05:30
now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
write(6,'(/,a)') ' <<<+- DAMASK_abaqus init -+>>>'
write(6,'(a)') ' $Id$'
reporting current time for most module inits
2013-02-25 22:04:59 +05:30
#include "compilation_info.f90"
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
end subroutine DAMASK_interface_init
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
!--------------------------------------------------------------------------------------------------
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
!> @brief using Abaqus/Explicit function to get working directory name
now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
!--------------------------------------------------------------------------------------------------
character(1024) function getSolverWorkingDirectoryName()
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
changed fileopen routines in IO.f90 to use solver dependant functions defined in the interface routines DID NOT test ABAQUS versions yet, not shure whether (V)GETOUTDIR returns path with or without terminating slash
2010-05-10 20:32:59 +05:30
implicit none
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
integer :: lenOutDir
changed fileopen routines in IO.f90 to use solver dependant functions defined in the interface routines DID NOT test ABAQUS versions yet, not shure whether (V)GETOUTDIR returns path with or without terminating slash
2010-05-10 20:32:59 +05:30
improved version of general I/O functions renamed module cpfem_interface to mpie_interface
2010-05-11 12:27:15 +05:30
getSolverWorkingDirectoryName=''
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
call vgetOutDir(getSolverWorkingDirectoryName, lenOutDir)
abaqus routines VGETOUTDIR and GETOUTDIR do not return the trailing slash, which we assume to be there, so it has to be appended
2011-03-03 15:59:59 +05:30
getSolverWorkingDirectoryName=trim(getSolverWorkingDirectoryName)//'/'
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
end function getSolverWorkingDirectoryName
extended IO to cope with different name for solverJob and Model polishing, added error codes added FFTW library files
2011-02-21 20:07:38 +05:30
now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
!--------------------------------------------------------------------------------------------------
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
!> @brief using Abaqus/Explicit function to get solver job name
now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
!--------------------------------------------------------------------------------------------------
character(1024) function getSolverJobName()
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
changed fileopen routines in IO.f90 to use solver dependant functions defined in the interface routines DID NOT test ABAQUS versions yet, not shure whether (V)GETOUTDIR returns path with or without terminating slash
2010-05-10 20:32:59 +05:30
implicit none
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
integer :: lenJobName
changed fileopen routines in IO.f90 to use solver dependant functions defined in the interface routines DID NOT test ABAQUS versions yet, not shure whether (V)GETOUTDIR returns path with or without terminating slash
2010-05-10 20:32:59 +05:30
improved version of general I/O functions renamed module cpfem_interface to mpie_interface
2010-05-11 12:27:15 +05:30
getSolverJobName=''
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
call vGetJobName(getSolverJobName, lenJobName)
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
end function getSolverJobName
changed fileopen routines in IO.f90 to use solver dependant functions defined in the interface routines DID NOT test ABAQUS versions yet, not shure whether (V)GETOUTDIR returns path with or without terminating slash
2010-05-10 20:32:59 +05:30
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
end module DAMASK_interface
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
* Introduced preprocessor directives in order to suppress compilation of most write statements when using openmp. This tremendously improves efficiency of parallelization. * Also added some more openmp directives to increase percentage of parallelized code. * "implicit none" was missing in two subroutines of homogenization and constitutive.
2011-03-29 12:57:19 +05:30
#include "IO.f90"
introduced output of initial geometry (vtk file) to mesh
2013-04-18 22:10:49 +05:30
#include "libs.f90"
* Introduced preprocessor directives in order to suppress compilation of most write statements when using openmp. This tremendously improves efficiency of parallelization. * Also added some more openmp directives to increase percentage of parallelized code. * "implicit none" was missing in two subroutines of homogenization and constitutive.
2011-03-29 12:57:19 +05:30
#include "numerics.f90"
#include "debug.f90"
#include "math.f90"
#include "FEsolving.f90"
#include "mesh.f90"
#include "material.f90"
#include "lattice.f90"
added constitutive none to abaqus files
2012-08-25 00:51:39 +05:30
#include "constitutive_none.f90"
* Introduced preprocessor directives in order to suppress compilation of most write statements when using openmp. This tremendously improves efficiency of parallelization. * Also added some more openmp directives to increase percentage of parallelized code. * "implicit none" was missing in two subroutines of homogenization and constitutive.
2011-03-29 12:57:19 +05:30
#include "constitutive_j2.f90"
#include "constitutive_phenopowerlaw.f90"
#include "constitutive_titanmod.f90"
#include "constitutive_dislotwin.f90"
#include "constitutive_nonlocal.f90"
#include "constitutive.f90"
#include "crystallite.f90"
introduced homogenization_none to substitute isostrain with ngrains 1. cleaned up homogenization isostrain (has no state)
2014-03-14 04:50:50 +05:30
#include "homogenization_none.f90"
* Introduced preprocessor directives in order to suppress compilation of most write statements when using openmp. This tremendously improves efficiency of parallelization. * Also added some more openmp directives to increase percentage of parallelized code. * "implicit none" was missing in two subroutines of homogenization and constitutive.
2011-03-29 12:57:19 +05:30
#include "homogenization_isostrain.f90"
#include "homogenization_RGC.f90"
#include "homogenization.f90"
#include "CPFEM.f90"
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
subroutine vumat(nBlock, nDir, nshr, nStateV, nFieldV, nProps, lAnneal, &
stepTime, totalTime, dt, cmName, coordMp, charLength, &
props, density, strainInc, relSpinInc, &
tempOld, stretchOld, defgradOld, fieldOld, &
stressOld, stateOld, enerInternOld, enerInelasOld, &
tempNew, stretchNew, defgradNew, fieldNew, &
stressNew, stateNew, enerInternNew, enerInelasNew)
use prec, only: &
pReal, &
pInt
Abaqus explicit now with correct interface, seems to work now
2013-03-21 22:54:58 +05:30
use FEsolving, only: &
cycleCounter, &
theTime, &
outdatedByNewInc, &
outdatedFFN1, &
terminallyIll, &
symmetricSolver
use math, only: &
invnrmMandel
use debug, only: &
debug_info, &
debug_reset, &
debug_levelBasic, &
debug_level, &
debug_abaqus
use mesh, only: &
replaced numerics_unitlength with mesh_unitlength
2013-05-17 22:22:19 +05:30
mesh_unitlength, &
Abaqus explicit now with correct interface, seems to work now
2013-03-21 22:54:58 +05:30
mesh_FEasCP, &
mesh_ipCoordinates
reorganized calculation modes for CPFEM, now having better readable and cleaner structure
2013-03-01 17:18:29 +05:30
use CPFEM, only: &
CPFEM_general, &
CPFEM_init_done, &
CPFEM_initAll, &
CPFEM_CALCRESULTS, &
parallelExecution flag now passed as input variable by CPFEM_general; flag is set on the solver level (DAMASK_abaqus_exp.f, DAMASK_marc.f90, etc.)
2013-08-02 16:50:11 +05:30
CPFEM_AGERESULTS
Abaqus explicit now with correct interface, seems to work now
2013-03-21 22:54:58 +05:30
use homogenization, only: &
materialpoint_sizeResults, &
materialpoint_results
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
Abaqus explicit now with correct interface, seems to work now
2013-03-21 22:54:58 +05:30
implicit none
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
integer(pInt), intent(in) :: &
nDir, & !< number of direct components in a symmetric tensor
nshr, & !< number of indirect components in a symmetric tensor
nStateV, & !< number of user-defined state variables that are associated with this material type
nFieldV, & !< number of user-defined external field variables
nprops, & !< user-specified number of user-defined material properties
lAnneal !< indicating whether the routine is being called during an annealing process
integer(pInt), dimension(*), intent(in) :: &
nBlock !< 1: No of Materialpoints in this call, 2: No of Materialpoint (IP)
corrected comment explaining structure of nBlock
2013-03-26 00:06:21 +05:30
!< 3: No of layer, 4: No of secPoint, 5+: element numbers
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
character(len=80), intent(in) :: &
cmname !< uses-specified material name, left justified
real(pReal), dimension(nprops), intent(in) :: &
props !< user-supplied material properties
real(pReal), intent(in) :: &
stepTime, & !< value of time since the step began
totalTime, & !< value of total time
dt !< time increment size
real(pReal), dimension(nblock(1)), intent(in) :: &
density, & !< current density at material points in the midstep configuration
charLength, & !< characteristic element length
enerInternOld, &
enerInelasOld, &
tempOld, & !< temperature
tempNew
real(pReal), dimension(nblock(1),*), intent(in) :: &
coordMp !< material point coordinates
real(pReal), dimension(nblock(1),ndir+nshr), intent(in) :: &
strainInc, & !< strain increment tensor at each material point
stretchOld, & !< stretch tensor U at each material point
stretchNew, & !< stretch tensor U at each material point
stressOld !< stress tensor at each material point
real(pReal), dimension(nblock(1),nshr), intent(in) :: &
relSpinInc !< incremental relative rotation vector
real(pReal), dimension(nblock(1),nstatev), intent(in) :: &
stateOld
real(pReal), dimension(nblock(1),nfieldv), intent(in) :: &
fieldOld, & !< user-defined field variables
fieldNew !< user-defined field variables
real(pReal), dimension(nblock(1),ndir+2*nshr), intent(in) :: &
defgradOld, &
defgradNew
real(pReal), dimension(nblock(1)), intent(out) :: &
enerInternNew, & !< internal energy per unit mass at each material point at the end of the increment
enerInelasNew !< dissipated inelastic energy per unit mass at each material point at the end of the increment
real(pReal), dimension(nblock(1),ndir+nshr), intent(out) :: &
stressNew !< stress tensor at each material point at the end of the increment
real(pReal), dimension(nblock(1),nstatev), intent(out) :: &
stateNew !< state variables at each material point at the end of the increment
solved argument mismatch
2011-09-13 19:45:17 +05:30
real(pReal), dimension(3) :: coordinates
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
real(pReal), dimension(3,3) :: defgrd0,defgrd1
real(pReal), dimension(6) :: stress
real(pReal), dimension(6,6) :: ddsdde
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
real(pReal) :: temp, timeInc
integer(pInt) :: computationMode, n, i, cp_en
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
parallelExecution flag now passed as input variable by CPFEM_general; flag is set on the solver level (DAMASK_abaqus_exp.f, DAMASK_marc.f90, etc.)
2013-08-02 16:50:11 +05:30
computationMode = CPFEM_CALCRESULTS ! always calculate
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
do n = 1,nblock(1) ! loop over vector of IPs
Abaqus interfaces dapted to newest version of the rest of the code does not yet check for restart, have to check out first how restarts work in Abaqus explicit code not checked as we do not have a license
2011-03-07 18:19:27 +05:30
temp = tempOld(n)
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
if ( .not. CPFEM_init_done ) then
saved some memory not copying data within one function
2013-03-26 00:32:12 +05:30
call CPFEM_initAll(temp,nBlock(4_pInt+n),nBlock(2))
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
outdatedByNewInc = .false.
changed variable name "debug_what" to "debug_level"
2012-07-05 15:24:50 +05:30
if (iand(debug_level(debug_abaqus),debug_levelBasic) /= 0) then
enclosed output statements in an "if (verboseDebugger)"
2010-03-25 13:57:42 +05:30
!$OMP CRITICAL (write2out)
saved some memory not copying data within one function
2013-03-26 00:32:12 +05:30
write(6,'(i8,1x,i2,1x,a)') nBlock(4_pInt+n),nBlock(2),'first call special case..!'; flush(6)
enclosed output statements in an "if (verboseDebugger)"
2010-03-25 13:57:42 +05:30
!$OMP END CRITICAL (write2out)
endif
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
else if (theTime < totalTime) then ! reached convergence
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
outdatedByNewInc = .true.
changed variable name "debug_what" to "debug_level"
2012-07-05 15:24:50 +05:30
if (iand(debug_level(debug_abaqus),debug_levelBasic) /= 0) then
enclosed output statements in an "if (verboseDebugger)"
2010-03-25 13:57:42 +05:30
!$OMP CRITICAL (write2out)
saved some memory not copying data within one function
2013-03-26 00:32:12 +05:30
write (6,'(i8,1x,i2,1x,a)') nBlock(4_pInt+n),nBlock(2),'lastIncConverged + outdated'; flush(6)
enclosed output statements in an "if (verboseDebugger)"
2010-03-25 13:57:42 +05:30
!$OMP END CRITICAL (write2out)
endif
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
endif
outdatedFFN1 = .false.
terminallyIll = .false.
Abaqus explicit now with correct interface, seems to work now
2013-03-21 22:54:58 +05:30
cycleCounter = 1_pInt
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
if ( outdatedByNewInc ) then
outdatedByNewInc = .false.
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
call debug_info() ! first after new inc reports debugging
call debug_reset() ! resets debugging
computationMode = ior(computationMode, CPFEM_AGERESULTS) ! age results
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
endif
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
theTime = totalTime ! record current starting time
changed variable name "debug_what" to "debug_level"
2012-07-05 15:24:50 +05:30
if (iand(debug_level(debug_abaqus),debug_levelBasic) /= 0) then
enclosed output statements in an "if (verboseDebugger)"
2010-03-25 13:57:42 +05:30
!$OMP CRITICAL (write2out)
saved some memory not copying data within one function
2013-03-26 00:32:12 +05:30
write(6,'(a,i8,i2,a)') '(',nBlock(4_pInt+n),nBlock(2),')'; flush(6)
reorganized calculation modes for CPFEM, now having better readable and cleaner structure
2013-03-01 17:18:29 +05:30
write(6,'(a,l1)') 'Aging Results: ', iand(computationMode, CPFEM_AGERESULTS) /= 0_pInt
enclosed output statements in an "if (verboseDebugger)"
2010-03-25 13:57:42 +05:30
!$OMP END CRITICAL (write2out)
endif
defgrd0 = 0.0_pReal
defgrd1 = 0.0_pReal
timeInc = dt
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
! ABAQUS explicit: deformation gradient as vector 11, 22, 33, 12, 23, 31, 21, 32, 13
! ABAQUS explicit: deformation gradient as vector 11, 22, 33, 12, 21
forall (i=1:ndir)
defgrd0(i,i) = defgradOld(n,i)
defgrd1(i,i) = defgradNew(n,i)
end forall
if (nshr == 1) then
defgrd0(1,2) = defgradOld(n,4)
defgrd1(1,2) = defgradNew(n,4)
defgrd0(2,1) = defgradOld(n,5)
defgrd1(2,1) = defgradNew(n,5)
else
defgrd0(1,2) = defgradOld(n,4)
defgrd1(1,2) = defgradNew(n,4)
defgrd0(1,3) = defgradOld(n,9)
defgrd1(1,3) = defgradNew(n,9)
defgrd0(2,1) = defgradOld(n,7)
defgrd1(2,1) = defgradNew(n,7)
defgrd0(2,3) = defgradOld(n,5)
defgrd1(2,3) = defgradNew(n,5)
defgrd0(3,1) = defgradOld(n,6)
defgrd1(3,1) = defgradNew(n,6)
defgrd0(3,2) = defgradOld(n,8)
defgrd1(3,2) = defgradNew(n,8)
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
endif
saved some memory not copying data within one function
2013-03-26 00:32:12 +05:30
cp_en = mesh_FEasCP('elem',nBlock(4_pInt+n))
replaced numerics_unitlength with mesh_unitlength
2013-05-17 22:22:19 +05:30
mesh_ipCoordinates(1:3,n,cp_en) = mesh_unitlength * coordMp(n,1:3)
removed coordinates from call to CPFEM_general and introduced direct storage of coordinates to mesh_ipCoordinates
2012-11-14 20:08:10 +05:30
parallelExecution flag now passed as input variable by CPFEM_general; flag is set on the solver level (DAMASK_abaqus_exp.f, DAMASK_marc.f90, etc.)
2013-08-02 16:50:11 +05:30
call CPFEM_general(computationMode,.false.,defgrd0,defgrd1,temp,timeInc,cp_en,nBlock(2),stress,ddsdde)
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
! Mandel: 11, 22, 33, SQRT(2)*12, SQRT(2)*23, SQRT(2)*13
! straight: 11, 22, 33, 12, 23, 13
! ABAQUS implicit: 11, 22, 33, 12, 13, 23
! ABAQUS explicit: 11, 22, 33, 12, 23, 13
! ABAQUS explicit: 11, 22, 33, 12
stressNew(n,1:ndir+nshr) = stress(1:ndir+nshr)*invnrmMandel(1:ndir+nshr)
saved some memory not copying data within one function
2013-03-26 00:32:12 +05:30
stateNew(n,:) = materialpoint_results(1:min(nstatev,materialpoint_sizeResults),&
nBlock(2),mesh_FEasCP('elem', nBlock(4_pInt+n)))
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
enddo
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
end subroutine vumat
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief calls the exit function of Abaqus/Explicit
!--------------------------------------------------------------------------------------------------
now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
subroutine quit(mpie_error)
merged two-stepped subroutine call into one call, added comments and cleaned up. working for small example (with and without openMP), don't know if the results make any sense
2013-03-25 23:15:58 +05:30
use prec, only: &
pInt
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
renaming of abaqus to _std and adding of VUMAT interface for explicit.
2010-02-18 15:53:42 +05:30
implicit none
integer(pInt) mpie_error
now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
moved the the variable cutBack from FEsolving.f90 (where it was not used at all) to the interface routines to make responsibility of cut back control clear PLEASE RUN setup_code.py IF YOU DO MARC SIMULATIONS, THE INTERFACE HAS CHANGED
2012-10-02 18:16:58 +05:30
flush(6)
now using correct exit routine for Abaqus/Explicit
2013-03-25 19:22:22 +05:30
call xplb_exit
end subroutine quit