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/crystallite.f90

4063 lines
218 KiB
Fortran
Raw Normal View History

doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
! $Id$
!--------------------------------------------------------------------------------------------------
!> @author Franz Roters, Max-Planck-Institut für Eisenforschung GmbH
!> @author Philip Eisenlohr, Max-Planck-Institut für Eisenforschung GmbH
!> @author Christoph Kords, Max-Planck-Institut für Eisenforschung GmbH
!> @brief crystallite state integration functions and reporting of results
!--------------------------------------------------------------------------------------------------
* now able to choose method for state integration (integrator and integratorStiffness in numerics.config); standard is "Fixed Point Iteration", which is basically the same as before; others available are "Explicit Euler", "AdaptiveEuler", "Classical Runge-Kutta" and "Runge-Kutta Cash-Karp" * now remembering stiffness similar to how we do it for Lp etc.; avoids undefined stiffness values for nonconverged stiffness calculation * non-local stuff: * changed non-local kinetics (Gilman2002) * enforce zero shearrate for overall carrrier density below relevant density * enforce zero density for those states that become negative and were below relevant density before * dislocation velocity is not limited by V^(1/3) / dt anymore
2010-10-01 17:48:49 +05:30
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
module crystallite
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
use prec, only: &
pReal, &
pInt
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
implicit none
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
private
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
character(len=64), dimension(:,:), allocatable, private :: &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_output !< name of each post result output
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
integer(pInt), public, protected :: &
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
crystallite_maxSizePostResults !< description not available
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
integer(pInt), dimension(:), allocatable, public, protected :: &
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
crystallite_sizePostResults !< description not available
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
integer(pInt), dimension(:,:), allocatable, private :: &
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
crystallite_sizePostResult !< description not available
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
real(pReal), dimension(:,:), allocatable, public :: &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
crystallite_temperature !< temperature (same on all components on one IP)
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
real(pReal), dimension(:,:,:), allocatable, public, protected :: &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
crystallite_heat !< heat source
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
real(pReal), dimension(:,:,:), allocatable, public :: &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
crystallite_dt !< requested time increment of each grain
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
real(pReal), dimension(:,:,:), allocatable, private :: &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_subdt, & !< substepped time increment of each grain
crystallite_subFrac, & !< already calculated fraction of increment
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
crystallite_subStep !< size of next integration step
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
real(pReal), dimension(:,:,:,:), allocatable, public :: &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_Tstar_v, & !< current 2nd Piola-Kirchhoff stress vector (end of converged time step)
crystallite_Tstar0_v, & !< 2nd Piola-Kirchhoff stress vector at start of FE inc
crystallite_partionedTstar0_v !< 2nd Piola-Kirchhoff stress vector at start of homog inc
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
real(pReal), dimension(:,:,:,:), allocatable, private :: &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_subTstar0_v, & !< 2nd Piola-Kirchhoff stress vector at start of crystallite inc
crystallite_orientation, & !< orientation as quaternion
crystallite_orientation0, & !< initial orientation as quaternion
crystallite_rotation !< grain rotation away from initial orientation as axis-angle (in degrees) in crystal reference frame
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
real(pReal), dimension(:,:,:,:,:), allocatable, public :: &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_Fp, & !< current plastic def grad (end of converged time step)
crystallite_Fp0, & !< plastic def grad at start of FE inc
crystallite_partionedFp0,& !< plastic def grad at start of homog inc
crystallite_F0, & !< def grad at start of FE inc
crystallite_partionedF, & !< def grad to be reached at end of homog inc
crystallite_partionedF0, & !< def grad at start of homog inc
crystallite_Lp, & !< current plastic velocitiy grad (end of converged time step)
crystallite_Lp0, & !< plastic velocitiy grad at start of FE inc
crystallite_partionedLp0,& !< plastic velocity grad at start of homog inc
crystallite_P !< 1st Piola-Kirchhoff stress per grain
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
real(pReal), dimension(:,:,:,:,:), allocatable, private :: &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_Fe, & !< current "elastic" def grad (end of converged time step)
crystallite_subFe0,& !< "elastic" def grad at start of crystallite inc
crystallite_invFp, & !< inverse of current plastic def grad (end of converged time step)
crystallite_subFp0,& !< plastic def grad at start of crystallite inc
crystallite_subF, & !< def grad to be reached at end of crystallite inc
crystallite_subF0, & !< def grad at start of crystallite inc
crystallite_subLp0,& !< plastic velocity grad at start of crystallite inc
crystallite_disorientation !< disorientation between two neighboring ips (only calculated for single grain IPs)
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
real(pReal), dimension(:,:,:,:,:,:,:), allocatable, public :: &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_dPdF, & !< current individual dPdF per grain (end of converged time step)
crystallite_dPdF0, & !< individual dPdF per grain at start of FE inc
crystallite_partioneddPdF0 !< individual dPdF per grain at start of homog inc
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
real(pReal), dimension(:,:,:,:,:,:,:), allocatable, private :: &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_fallbackdPdF !< dPdF fallback for non-converged grains (elastic prediction)
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
logical, dimension(:,:,:), allocatable, public :: &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_requested !< flag to request crystallite calculation
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
logical, dimension(:,:,:), allocatable, public, protected :: &
reorganized calculation modes for CPFEM, now having better readable and cleaner structure
2013-03-01 17:18:29 +05:30
crystallite_converged, & !< convergence flag
crystallite_localPlasticity !< indicates this grain to have purely local constitutive law
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
logical, dimension(:,:,:), allocatable, private :: &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_todo !< flag to indicate need for further computation
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
logical, dimension(:,:), allocatable, private :: &
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
crystallite_clearToWindForward, & !< description not available
crystallite_clearToCutback, & !< description not available
crystallite_syncSubFrac, & !< description not available
crystallite_syncSubFracCompleted, & !< description not available
crystallite_neighborEnforcedCutback !< description not available
doxygen comments
2013-02-22 04:38:36 +05:30
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
enum, bind(c)
enumerator :: undefined_ID, &
phase_ID, &
texture_ID, &
volume_ID, &
grainrotationx_ID, &
grainrotationy_ID, &
grainrotationz_ID, &
heat_ID, &
orientation_ID, &
grainrotation_ID, &
eulerangles_ID, &
defgrad_ID, &
fe_ID, &
fp_ID, &
lp_ID, &
e_ID, &
ee_ID, &
p_ID, &
s_ID, &
elasmatrix_ID, &
neighboringip_ID, &
neighboringelement_ID
end enum
integer(kind(undefined_ID)),dimension(:,:), allocatable, private :: &
crystallite_outputID !< ID of each post result output
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
public :: &
crystallite_init, &
crystallite_stressAndItsTangent, &
crystallite_orientations, &
crystallite_postResults
private :: &
crystallite_integrateStateFPI, &
crystallite_integrateStateEuler, &
crystallite_integrateStateAdaptiveEuler, &
crystallite_integrateStateRK4, &
crystallite_integrateStateRKCK45, &
crystallite_integrateStress, &
crystallite_stateJump
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
contains
major restructuring of code. homogenization as well as constitutive are now free to choose. the runtime got somewhat longer (25% on simple tests) compared to a hardcoded isostrain homogenization. this might be a point of further optimization at a later stage... please use homogenization_isostrain.f90 as starting point / example for future developments of homog-schemes. the homogenization scheme now can additionally output certain results. hence, the userdata structure at each integration point now looks like this: - sizeHomogPostResults - block of that size containing homogPostResults then for each grain: - sizeGrainPostResults - block of that size containing crystallitePostResults, which consist of: + phaseID + volFrac + Eulers (3) + any constitutive post results requested
2009-05-07 21:57:36 +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
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief allocates and initialize per grain variables
!--------------------------------------------------------------------------------------------------
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
subroutine crystallite_init(temperature)
doxygen comments
2013-02-22 04:38:36 +05:30
use, intrinsic :: iso_fortran_env ! to get compiler_version and compiler_options (at least for gfortran 4.6 at the moment)
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
use debug, only: &
debug_info, &
debug_reset, &
debug_level, &
debug_crystallite, &
debug_levelBasic
further improvements on implementing the plain mode
2013-04-16 22:37:27 +05:30
use numerics, only: &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
usePingPong
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
use math, only: &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
math_I3, &
math_EulerToR, &
math_inv33, &
math_transpose33, &
math_mul33xx33, &
math_mul33x33
use FEsolving, only: &
FEsolving_execElem, &
FEsolving_execIP
use mesh, only: &
mesh_element, &
mesh_NcpElems, &
mesh_maxNips, &
mesh_maxNipNeighbors
use IO, only: &
added possibility for multi-level inclusion of files in *.config and loadcase files. include subfiles by stating {path/to/include}
2013-06-27 00:49:00 +05:30
IO_read, &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
IO_timeStamp, &
IO_open_jobFile_stat, &
IO_open_file, &
IO_lc, &
IO_getTag, &
IO_isBlank, &
IO_stringPos, &
IO_stringValue, &
IO_write_jobFile, &
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
IO_error, &
IO_EOF
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
use material
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
use lattice, only: &
moved reading in of lattice type and elastic constants to lattice module removed structure type for hex, fcc, bcc, now defining slip/twin systems for each phase found in material.config constitutive modules will only be initialized if needed homogenizedC function is only needed for models incorporating twinning in a physical way (titanmod and dislotwin)
2014-03-09 02:20:31 +05:30
lattice_structure
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
use constitutive, only: &
constitutive_microstructure
reworked crystallite part to allow for flexible user output --> new "crystallite" part in config file --> new "crystallite" option for microstructures --> new output file "...job.outputCrystallite" to be used in conjunction with marc_addUserOutput for meaningful naming of User Defined Vars.
2010-02-25 23:09:11 +05:30
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
implicit none
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
real(pReal), intent(in) :: temperature
integer(pInt), parameter :: &
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
FILEUNIT = 200_pInt, &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
MAXNCHUNKS = 2_pInt
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
integer(pInt), dimension(1+2*MAXNCHUNKS) :: positions
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
integer(pInt) :: &
g, & !< grain number
i, & !< integration point number
e, & !< element number
gMax, & !< maximum number of grains
iMax, & !< maximum number of integration points
eMax, & !< maximum number of elements
nMax, & !< maximum number of ip neighbors
myNgrains, & !< number of grains in current IP
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
section = 0_pInt, &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
j, &
p, &
some small improvements, still something wrong, probably in crystallite.f90
2013-12-13 19:44:17 +05:30
output = 0_pInt, &
moved reading in of lattice type and elastic constants to lattice module removed structure type for hex, fcc, bcc, now defining slip/twin systems for each phase found in material.config constitutive modules will only be initialized if needed homogenizedC function is only needed for models incorporating twinning in a physical way (titanmod and dislotwin)
2014-03-09 02:20:31 +05:30
mySize
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
character(len=65536) :: &
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
tag = '', &
line= ''
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(/,a)') ' <<<+- crystallite init -+>>>'
write(6,'(a)') ' $Id$'
write(6,'(a15,a)') ' Current time: ',IO_timeStamp()
made code standard conform to Fortran 2008 (ignoring warning concerning comments beyond character 132). Basically, changing "x" format specifier to "Nx" ("1x") plus removing $ format specifier added compiler switches for gfortran and ifort to check for standard conformity old gnu compilers <4.4 are not longer supported because they don't provide the c binding for fftw
2012-02-01 00:48:55 +05:30
#include "compilation_info.f90"
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
gMax = homogenization_maxNgrains
iMax = mesh_maxNips
eMax = mesh_NcpElems
nMax = mesh_maxNipNeighbors
more work on the new state
2014-05-27 20:16:03 +05:30
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
allocate(crystallite_temperature(iMax,eMax), source=temperature)
allocate(crystallite_heat(gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_Tstar0_v(6,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_partionedTstar0_v(6,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_subTstar0_v(6,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_Tstar_v(6,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_P(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_F0(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_partionedF0(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_partionedF(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_subF0(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_subF(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_Fp0(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_partionedFp0(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_subFp0(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_Fp(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_invFp(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_Fe(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_subFe0(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_Lp0(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_partionedLp0(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_subLp0(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_Lp(3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_dPdF(3,3,3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_dPdF0(3,3,3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_partioneddPdF0(3,3,3,3,gMax,iMax,eMax),source=0.0_pReal)
allocate(crystallite_fallbackdPdF(3,3,3,3,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_dt(gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_subdt(gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_subFrac(gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_subStep(gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_orientation(4,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_orientation0(4,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_rotation(4,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_disorientation(4,nMax,gMax,iMax,eMax), source=0.0_pReal)
allocate(crystallite_localPlasticity(gMax,iMax,eMax), source=.true.)
allocate(crystallite_requested(gMax,iMax,eMax), source=.false.)
allocate(crystallite_todo(gMax,iMax,eMax), source=.false.)
allocate(crystallite_converged(gMax,iMax,eMax), source=.true.)
allocate(crystallite_clearToWindForward(iMax,eMax), source=.true.)
allocate(crystallite_syncSubFrac(iMax,eMax), source=.false.)
allocate(crystallite_syncSubFracCompleted(iMax,eMax), source=.false.)
allocate(crystallite_clearToCutback(iMax,eMax), source=.true.)
allocate(crystallite_neighborEnforcedCutback(iMax,eMax), source=.false.)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
allocate(crystallite_output(maxval(crystallite_Noutput), &
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
material_Ncrystallite)) ; crystallite_output = ''
allocate(crystallite_outputID(maxval(crystallite_Noutput), &
material_Ncrystallite), source=undefined_ID)
allocate(crystallite_sizePostResults(material_Ncrystallite),source=0_pInt)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
allocate(crystallite_sizePostResult(maxval(crystallite_Noutput), &
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
material_Ncrystallite), source=0_pInt)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
if (.not. IO_open_jobFile_stat(FILEUNIT,material_localFileExt)) & ! no local material configuration present...
call IO_open_file(FILEUNIT,material_configFile) ! ...open material.config file
rewind(FILEUNIT)
do while (trim(line) /= IO_EOF .and. IO_lc(IO_getTag(line,'<','>')) /= material_partCrystallite) ! wind forward to <crystallite>
line = IO_read(FILEUNIT)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
enddo
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
do while (trim(line) /= IO_EOF) ! read through sections of crystallite part
line = IO_read(FILEUNIT)
doxygen comments
2013-02-22 04:38:36 +05:30
if (IO_isBlank(line)) cycle ! skip empty lines
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
if (IO_getTag(line,'<','>') /= '') then ! stop at next part
line = IO_read(FILEUNIT, .true.) ! reset IO_read
exit
endif
doxygen comments
2013-02-22 04:38:36 +05:30
if (IO_getTag(line,'[',']') /= '') then ! next section
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
section = section + 1_pInt
doxygen comments
2013-02-22 04:38:36 +05:30
output = 0_pInt ! reset output counter
added missing lp output to crystallite
2013-12-20 16:43:12 +05:30
cycle ! skip to next line
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
endif
if (section > 0_pInt) then
positions = IO_stringPos(line,maxNchunks)
doxygen comments
2013-02-22 04:38:36 +05:30
tag = IO_lc(IO_stringValue(line,positions,1_pInt)) ! extract key
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
select case(tag)
case ('(output)')
output = output + 1_pInt
crystallite_output(output,section) = IO_lc(IO_stringValue(line,positions,2_pInt))
some small improvements, still something wrong, probably in crystallite.f90
2013-12-13 19:44:17 +05:30
select case(crystallite_output(output,section))
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('phase')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = phase_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('texture')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = texture_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('volume')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = volume_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('grainrotationx')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = grainrotationx_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('grainrotationy')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = grainrotationy_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('grainrotationz')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = grainrotationx_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('heat')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = heat_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('orientation')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = orientation_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('grainrotation')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = grainrotation_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('eulerangles')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = eulerangles_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('defgrad','f')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = defgrad_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('fe')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = fe_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('fp')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = fp_ID
added missing lp output to crystallite
2013-12-20 16:43:12 +05:30
case ('lp')
crystallite_outputID(output,section) = lp_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('e')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = e_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('ee')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = ee_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('p','firstpiola','1piola')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = p_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('s','tstar','secondpiola','2ndpiola')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = s_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('elasmatrix')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = elasmatrix_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('neighboringip')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = neighboringip_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case ('neighboringelement')
fixed small issues, ATTENTION: Still a bug introduced with rev. 2783 not fixed
2013-12-13 18:49:17 +05:30
crystallite_outputID(output,section) = neighboringelement_ID
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case default
call IO_error(105_pInt,ext_msg=IO_stringValue(line,positions,2_pInt)//' (Crystallite)')
end select
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
end select
endif
enddo
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
close(FILEUNIT)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
do i = 1_pInt,material_Ncrystallite
do j = 1_pInt,crystallite_Noutput(i)
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
select case(crystallite_outputID(j,i))
case(phase_ID,texture_ID,volume_ID,grainrotationx_ID,grainrotationy_ID,grainrotationz_ID,heat_ID)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
mySize = 1_pInt
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case(orientation_ID,grainrotation_ID) ! orientation as quaternion, or deviation from initial grain orientation in axis-angle form (angle in degrees)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
mySize = 4_pInt
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case(eulerangles_ID)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
mySize = 3_pInt
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case(defgrad_ID,fe_ID,fp_ID,lp_ID,e_ID,ee_ID,p_ID,s_ID)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
mySize = 9_pInt
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case(elasmatrix_ID)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
mySize = 36_pInt
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case(neighboringip_ID,neighboringelement_ID)
added crystallite output "neighboringelement" and "neighboringip" some polishing in crystallite_postResults
2013-05-08 17:32:30 +05:30
mySize = mesh_maxNipNeighbors
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
case default
mySize = 0_pInt
end select
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
outputFound: if (mySize > 0_pInt) then
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
crystallite_sizePostResult(j,i) = mySize
crystallite_sizePostResults(i) = crystallite_sizePostResults(i) + mySize
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
endif outputFound
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
enddo
enddo
crystallite_maxSizePostResults = 0_pInt
do j = 1_pInt,material_Nmicrostructure
if (microstructure_active(j)) &
crystallite_maxSizePostResults = max(crystallite_maxSizePostResults,&
crystallite_sizePostResults(microstructure_crystallite(j)))
enddo
Tstar_v is now restored at different stages of cutbacking; we need this because preguess of state relies on consistent Tstar_v
2009-06-16 14:33:30 +05:30
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
!--------------------------------------------------------------------------------------------------
reworked crystallite part to allow for flexible user output --> new "crystallite" part in config file --> new "crystallite" option for microstructures --> new output file "...job.outputCrystallite" to be used in conjunction with marc_addUserOutput for meaningful naming of User Defined Vars.
2010-02-25 23:09:11 +05:30
! write description file for crystallite output
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
call IO_write_jobFile(FILEUNIT,'outputCrystallite')
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
do p = 1_pInt,material_Ncrystallite
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
write(FILEUNIT,'(/,a,/)') '['//trim(crystallite_name(p))//']'
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
do e = 1_pInt,crystallite_Noutput(p)
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
write(FILEUNIT,'(a,i4)') trim(crystallite_output(e,p))//char(9),crystallite_sizePostResult(e,p)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
enddo
enddo
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
close(FILEUNIT)
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
!--------------------------------------------------------------------------------------------------
! initialize
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL DO PRIVATE(myNgrains)
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over all cp elements
myNgrains = homogenization_Ngrains(mesh_element(3,e)) ! look up homogenization-->grainCount
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), g = 1_pInt:myNgrains)
crystallite_Fp0(1:3,1:3,g,i,e) = math_EulerToR(material_EulerAngles(1:3,g,i,e)) ! plastic def gradient reflects init orientation
crystallite_F0(1:3,1:3,g,i,e) = math_I3
crystallite_localPlasticity(g,i,e) = phase_localPlasticity(material_phase(g,i,e))
crystallite_Fe(1:3,1:3,g,i,e) = math_transpose33(crystallite_Fp0(1:3,1:3,g,i,e))
crystallite_Fp(1:3,1:3,g,i,e) = crystallite_Fp0(1:3,1:3,g,i,e)
crystallite_requested(g,i,e) = .true.
endforall
enddo
!$OMP END PARALLEL DO
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
if(any(.not. crystallite_localPlasticity) .and. .not. usePingPong) call IO_error(601_pInt) ! exit if nonlocal but no ping-pong
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
crystallite_partionedFp0 = crystallite_Fp0
crystallite_partionedF0 = crystallite_F0
crystallite_partionedF = crystallite_F0
call crystallite_orientations()
crystallite_orientation0 = crystallite_orientation ! store initial orientations for calculation of grain rotations
homogenization_RGC.f90 >> just switching-off one of the debugging statement. crystallite.f90 >> optimizing the perturbation algorithm for local tangent.
2009-12-22 17:58:02 +05:30
* checked and corrected parallelization of code. compiles now successfully, but simply aborts computation with first parallel directive without any comment :-((( * also put a call to constitutive_microstructure at the start of each crystallite_integration subroutine like it was before. need that for nonlocal model in case of crystallite cutback
2010-11-03 22:52:48 +05:30
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
!$OMP PARALLEL DO PRIVATE(myNgrains)
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
do g = 1_pInt,myNgrains
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
call constitutive_microstructure(temperature, &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
crystallite_Fe(1:3,1:3,g,i,e), crystallite_Fp(1:3,1:3,g,i,e),g,i,e) ! update dependent state variables to be consistent with basic states
enddo
enddo
enddo
!$OMP END PARALLEL DO
fixed two severe bugs in new state, now a little bit faster (399 vs. 378 sec) and more memory efficicient (251312 vs. 236832 kbytes) for Phenopowerlaw 16x16x16 example (short load only)
2014-06-17 00:49:38 +05:30
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
call crystallite_stressAndItsTangent(.true.,.false.) ! request elastic answers
crystallite_fallbackdPdF = crystallite_dPdF ! use initial elastic stiffness as fallback
major restructuring of code. homogenization as well as constitutive are now free to choose. the runtime got somewhat longer (25% on simple tests) compared to a hardcoded isostrain homogenization. this might be a point of further optimization at a later stage... please use homogenization_isostrain.f90 as starting point / example for future developments of homog-schemes. the homogenization scheme now can additionally output certain results. hence, the userdata structure at each integration point now looks like this: - sizeHomogPostResults - block of that size containing homogPostResults then for each grain: - sizeGrainPostResults - block of that size containing crystallitePostResults, which consist of: + phaseID + volFrac + Eulers (3) + any constitutive post results requested
2009-05-07 21:57:36 +05:30
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
!--------------------------------------------------------------------------------------------------
! debug output
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
write(6,'(a35,1x,7(i8,1x))') 'crystallite_temperature: ', shape(crystallite_temperature)
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
write(6,'(a35,1x,7(i8,1x))') 'crystallite_heat: ', shape(crystallite_heat)
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a35,1x,7(i8,1x))') 'crystallite_Fe: ', shape(crystallite_Fe)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_Fp: ', shape(crystallite_Fp)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_Lp: ', shape(crystallite_Lp)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_F0: ', shape(crystallite_F0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_Fp0: ', shape(crystallite_Fp0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_Lp0: ', shape(crystallite_Lp0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_partionedF: ', shape(crystallite_partionedF)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_partionedF0: ', shape(crystallite_partionedF0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_partionedFp0: ', shape(crystallite_partionedFp0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_partionedLp0: ', shape(crystallite_partionedLp0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_subF: ', shape(crystallite_subF)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_subF0: ', shape(crystallite_subF0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_subFe0: ', shape(crystallite_subFe0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_subFp0: ', shape(crystallite_subFp0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_subLp0: ', shape(crystallite_subLp0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_P: ', shape(crystallite_P)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_Tstar_v: ', shape(crystallite_Tstar_v)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_Tstar0_v: ', shape(crystallite_Tstar0_v)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_partionedTstar0_v: ', shape(crystallite_partionedTstar0_v)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_subTstar0_v: ', shape(crystallite_subTstar0_v)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_dPdF: ', shape(crystallite_dPdF)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_dPdF0: ', shape(crystallite_dPdF0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_partioneddPdF0: ', shape(crystallite_partioneddPdF0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_fallbackdPdF: ', shape(crystallite_fallbackdPdF)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_orientation: ', shape(crystallite_orientation)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_orientation0: ', shape(crystallite_orientation0)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_rotation: ', shape(crystallite_rotation)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_disorientation: ', shape(crystallite_disorientation)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_dt: ', shape(crystallite_dt)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_subdt: ', shape(crystallite_subdt)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_subFrac: ', shape(crystallite_subFrac)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_subStep: ', shape(crystallite_subStep)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_localPlasticity: ', shape(crystallite_localPlasticity)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_requested: ', shape(crystallite_requested)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_todo: ', shape(crystallite_todo)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_converged: ', shape(crystallite_converged)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_sizePostResults: ', shape(crystallite_sizePostResults)
write(6,'(a35,1x,7(i8,1x))') 'crystallite_sizePostResult: ', shape(crystallite_sizePostResult)
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
write(6,'(/,a35,1x,i10)') 'Number of nonlocal grains: ',count(.not. crystallite_localPlasticity)
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
flush(6)
endif
* checked and corrected parallelization of code. compiles now successfully, but simply aborts computation with first parallel directive without any comment :-((( * also put a call to constitutive_microstructure at the start of each crystallite_integration subroutine like it was before. need that for nonlocal model in case of crystallite cutback
2010-11-03 22:52:48 +05:30
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4 new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing) updated preprocessing for documentation to handle precision correctly
2012-08-31 01:56:28 +05:30
call debug_info
call debug_reset
* checked and corrected parallelization of code. compiles now successfully, but simply aborts computation with first parallel directive without any comment :-((( * also put a call to constitutive_microstructure at the start of each crystallite_integration subroutine like it was before. need that for nonlocal model in case of crystallite cutback
2010-11-03 22:52:48 +05:30
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
end subroutine crystallite_init
major restructuring of code. homogenization as well as constitutive are now free to choose. the runtime got somewhat longer (25% on simple tests) compared to a hardcoded isostrain homogenization. this might be a point of further optimization at a later stage... please use homogenization_isostrain.f90 as starting point / example for future developments of homog-schemes. the homogenization scheme now can additionally output certain results. hence, the userdata structure at each integration point now looks like this: - sizeHomogPostResults - block of that size containing homogPostResults then for each grain: - sizeGrainPostResults - block of that size containing crystallitePostResults, which consist of: + phaseID + volFrac + Eulers (3) + any constitutive post results requested
2009-05-07 21:57:36 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief calculate stress (P) and tangent (dPdF) for crystallites
!--------------------------------------------------------------------------------------------------
simplified analytic jacobian calculation. removed Lpfrac, time_sensitive. introduced rate_sensitivity flag when calling crystallite_stressAndItsTangent that is currently set to .false. and is to be set according to which dPdF the FE solver is asking for
2012-03-14 19:26:50 +05:30
subroutine crystallite_stressAndItsTangent(updateJaco,rate_sensitivity)
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
use numerics, only: &
subStepMinCryst, &
subStepSizeCryst, &
stepIncreaseCryst, &
pert_Fg, &
pert_method, &
nCryst, &
numerics_integrator, &
numerics_integrationMode, &
numerics_timeSyncing, &
relevantStrain, &
analyticJaco
use debug, only: &
debug_level, &
debug_crystallite, &
debug_levelBasic, &
debug_levelExtensive, &
debug_levelSelective, &
debug_e, &
debug_i, &
debug_g, &
debug_CrystalliteLoopDistribution
use IO, only: &
IO_warning
use math, only: &
math_inv33, &
math_identity2nd, &
math_transpose33, &
math_mul33x33, &
math_mul66x6, &
math_Mandel6to33, &
math_Mandel33to6, &
math_I3, &
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
math_mul3333xx3333, &
math_mul33xx33
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
use FEsolving, only: &
FEsolving_execElem, &
FEsolving_execIP
use mesh, only: &
mesh_element, &
mesh_NcpElems, &
mesh_maxNips, &
mesh_ipNeighborhood, &
FE_NipNeighbors, &
FE_geomtype, &
FE_cellType
use material, only: &
homogenization_Ngrains, &
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#ifdef NEWSTATE
plasticState, &
#endif
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
homogenization_maxNgrains
use constitutive, only: &
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
constitutive_sizeState, &
constitutive_sizeDotState, &
constitutive_state, &
constitutive_state_backup, &
constitutive_subState0, &
constitutive_partionedState0, &
constitutive_dotState, &
constitutive_dotState_backup, &
more work on the new state
2014-05-27 20:16:03 +05:30
#else
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
mappingConstitutive, &
#endif
DAMASK (except nonlocal) now sees and interacts with FEM solver
2014-06-06 06:08:29 +05:30
constitutive_TandItsTangent, &
constitutive_localDamage, &
constitutive_gradientDamage
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
implicit none
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
logical, intent(in) :: &
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
updateJaco, & !< whether to update the Jacobian (stiffness) or not
rate_sensitivity !< rate sensitiv calculation
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
real(pReal) :: &
myPert, & ! perturbation with correct sign
formerSubStep, &
subFracIntermediate
real(pReal), dimension(3,3) :: &
invFp, & ! inverse of the plastic deformation gradient
Fe_guess, & ! guess for elastic deformation gradient
Tstar ! 2nd Piola-Kirchhoff stress tensor
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
real(pReal), dimension(3,3,3,3,homogenization_maxNgrains,mesh_maxNips,mesh_NcpElems) :: &
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
dPdF_perturbation1, &
dPdF_perturbation2
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
real(pReal), dimension(3,3,homogenization_maxNgrains,mesh_maxNips,mesh_NcpElems) :: &
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
F_backup, &
Fp_backup, &
InvFp_backup, &
Fe_backup, &
Lp_backup, &
P_backup
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
real(pReal), dimension(6,homogenization_maxNgrains,mesh_maxNips,mesh_NcpElems) :: &
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
Tstar_v_backup
integer(pInt) :: &
NiterationCrystallite, & ! number of iterations in crystallite loop
e, & ! element index
i, & ! integration point index
g, & ! grain index
k, &
l, &
n, startIP, endIP, &
neighboring_e, &
neighboring_i, &
o, &
p, &
perturbation , & ! loop counter for forward,backward perturbation mode
myNgrains
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
logical, dimension(homogenization_maxNgrains,mesh_maxNips,mesh_NcpElems) :: &
convergenceFlag_backup
! local variables used for calculating analytic Jacobian
real(pReal), dimension(3,3):: Fpinv_rate, &
FDot_inv, &
junk
real(pReal), dimension(3,3,3,3) :: dSdFe, &
dFedF, &
dFedFdot, &
dSdF, &
dSdFdot, &
dFp_invdFdot, &
junk2
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
real(pReal) :: counter
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
fixed selective debugging logic flaw in crystallite_stressAndItsTangent
2014-05-19 22:13:32 +05:30
if (iand(debug_level(debug_crystallite),debug_levelSelective) /= 0_pInt &
.and. FEsolving_execElem(1) <= debug_e &
.and. debug_e <= FEsolving_execElem(2)) then
write(6,'(/,a,i8,1x,i2,1x,i3)') '<< CRYST >> values at el ip g ', debug_e, debug_i, debug_g
more work on the new state
2014-05-27 20:16:03 +05:30
write(6,'(a,/,3(12x,3(f14.9,1x)/))') '<< CRYST >> F0 ', &
math_transpose33(crystallite_partionedF0(1:3,1:3,debug_g,debug_i,debug_e))
write(6,'(a,/,3(12x,3(f14.9,1x)/))') '<< CRYST >> Fp0', &
math_transpose33(crystallite_partionedFp0(1:3,1:3,debug_g,debug_i,debug_e))
write(6,'(a,/,3(12x,3(f14.9,1x)/))') '<< CRYST >> Lp0', &
math_transpose33(crystallite_partionedLp0(1:3,1:3,debug_g,debug_i,debug_e))
write(6,'(a,/,3(12x,3(f14.9,1x)/))') '<< CRYST >> F ', &
math_transpose33(crystallite_partionedF(1:3,1:3,debug_g,debug_i,debug_e))
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
endif
!--------------------------------------------------------------------------------------------------
! initialize to starting condition
crystallite_subStep = 0.0_pReal
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL DO PRIVATE(myNgrains)
elementLooping1: do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), &
g = 1_pInt:myNgrains, crystallite_requested(g,i,e))
#ifdef NEWSTATE
plasticState(mappingConstitutive(2,g,i,e))%subState0(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%partionedState0(:,mappingConstitutive(1,g,i,e))
#else
constitutive_subState0(g,i,e)%p = constitutive_partionedState0(g,i,e)%p ! ...microstructure
#endif
crystallite_subFp0(1:3,1:3,g,i,e) = crystallite_partionedFp0(1:3,1:3,g,i,e) ! ...plastic def grad
crystallite_subLp0(1:3,1:3,g,i,e) = crystallite_partionedLp0(1:3,1:3,g,i,e) ! ...plastic velocity grad
crystallite_dPdF0(1:3,1:3,1:3,1:3,g,i,e) = crystallite_partioneddPdF0(1:3,1:3,1:3,1:3,g,i,e) ! ...stiffness
crystallite_subF0(1:3,1:3,g,i,e) = crystallite_partionedF0(1:3,1:3,g,i,e) ! ...def grad
crystallite_subTstar0_v(1:6,g,i,e) = crystallite_partionedTstar0_v(1:6,g,i,e) !...2nd PK stress
crystallite_subFe0(1:3,1:3,g,i,e) = math_mul33x33(crystallite_subF0(1:3,1:3,g,i,e), &
math_inv33(crystallite_subFp0(1:3,1:3,g,i,e))) ! only needed later on for stiffness calculation
crystallite_subFrac(g,i,e) = 0.0_pReal
crystallite_subStep(g,i,e) = 1.0_pReal/subStepSizeCryst
crystallite_todo(g,i,e) = .true.
crystallite_converged(g,i,e) = .false. ! pretend failed step of twice the required size
endforall
enddo elementLooping1
!$OMP END PARALLEL DO
singleRun: if (FEsolving_execELem(1) == FEsolving_execElem(2) .and. &
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
FEsolving_execIP(1,FEsolving_execELem(1))==FEsolving_execIP(2,FEsolving_execELem(1))) then
startIP = FEsolving_execIP(1,FEsolving_execELem(1))
endIP = startIP
more work on the new state
2014-05-27 20:16:03 +05:30
else singleRun
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
startIP = 1_pInt
endIP = mesh_maxNips
more work on the new state
2014-05-27 20:16:03 +05:30
endif singleRun
homogenization_RGC.f90 >> just switching-off one of the debugging statement. crystallite.f90 >> optimizing the perturbation algorithm for local tangent.
2009-12-22 17:58:02 +05:30
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
NiterationCrystallite = 0_pInt
numerics_integrationMode = 1_pInt
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
cutbackLooping: do while (any(crystallite_todo(:,startIP:endIP,FEsolving_execELem(1):FEsolving_execElem(2))))
timeSyncing1: if (any(.not. crystallite_localPlasticity) .and. numerics_timeSyncing) then
fixed two severe bugs in new state, now a little bit faster (399 vs. 378 sec) and more memory efficicient (251312 vs. 236832 kbytes) for Phenopowerlaw 16x16x16 example (short load only)
2014-06-17 00:49:38 +05:30
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
! Time synchronization can only be used for nonlocal calculations, and only there it makes sense.
! The idea is that in nonlocal calculations often the vast amjority of the ips
! converges in one iteration whereas a small fraction of ips has to do a lot of cutbacks.
! Hence, we try to minimize the computational effort by just doing a lot of cutbacks
! in the vicinity of the "bad" ips and leave the easily converged volume more or less as it is.
! However, some synchronization of the time step has to be done at the border between "bad" ips
! and the ones that immediately converged.
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i6)') '<< CRYST >> crystallite iteration ',NiterationCrystallite
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
if (any(crystallite_syncSubFrac)) then
! Just did a time synchronization.
! If all synchrnizers converged, then do nothing else than winding them forward.
! If any of the cynchronizers did not converge, something went completely wrong
! and its not clear how to fix this, so all nonlocals become terminally ill.
if (any(crystallite_syncSubFrac .and. .not. crystallite_converged(1,:,:))) then
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) then
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
more work on the new state
2014-05-27 20:16:03 +05:30
if (crystallite_syncSubFrac(i,e) .and. .not. crystallite_converged(1,i,e)) &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i8,1x,i2)') '<< CRYST >> time synchronization: failed at el,ip ',e,i
enddo
enddo
endif
crystallite_syncSubFrac = .false.
where(.not. crystallite_localPlasticity)
crystallite_substep = 0.0_pReal
crystallite_todo = .false.
endwhere
else
!$OMP PARALLEL DO PRIVATE(myNgrains)
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
crystallite_clearToWindForward(i,e) = crystallite_localPlasticity(1,i,e) .or. crystallite_syncSubFrac(i,e)
crystallite_clearToCutback(i,e) = crystallite_localPlasticity(1,i,e)
enddo
enddo
!$OMP END PARALLEL DO
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i6)') '<< CRYST >> time synchronization: wind forward'
endif
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
elseif (any(crystallite_syncSubFracCompleted)) then
! Just completed a time synchronization.
! Make sure that the ips that synchronized their time step start non-converged
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
if (crystallite_syncSubFracCompleted(i,e)) crystallite_converged(1,i,e) = .false.
crystallite_syncSubFracCompleted(i,e) = .false.
crystallite_clearToWindForward(i,e) = crystallite_localPlasticity(1,i,e)
crystallite_clearToCutback(i,e) = crystallite_localPlasticity(1,i,e) .or. .not. crystallite_converged(1,i,e)
enddo
enddo
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i6)') '<< CRYST >> time synchronization: done, proceed with cutback'
else
! Normal calculation.
! If all converged and are at the end of the time increment, then just do a final wind forward.
! If all converged, but not all reached the end of the time increment, then we only wind
! those forward that are still on their way, all others have to wait.
! If some did not converge and all are still at the start of the time increment,
! then all non-convergers force their converged neighbors to also do a cutback.
! In case that some ips have already wound forward to an intermediate time (subfrac),
! then all those ips that converged in the first iteration, but now have a non-converged neighbor
! have to synchronize their time step to the same intermediate time. If such a synchronization
! takes place, all other ips have to wait and only the synchronizers do a cutback. In the next
! iteration those will do a wind forward while all others still wait.
!$OMP PARALLEL DO PRIVATE(myNgrains)
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
crystallite_clearToWindForward(i,e) = crystallite_localPlasticity(1,i,e)
crystallite_clearToCutback(i,e) = crystallite_localPlasticity(1,i,e)
enddo
enddo
!$OMP END PARALLEL DO
if (all(crystallite_localPlasticity .or. crystallite_converged)) then
if (all(crystallite_localPlasticity .or. crystallite_subStep + crystallite_subFrac >= 1.0_pReal)) then
crystallite_clearToWindForward = .true. ! final wind forward
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i6)') '<< CRYST >> final wind forward'
else
!$OMP PARALLEL DO PRIVATE(myNgrains)
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
crystallite_clearToWindForward(i,e) = crystallite_localPlasticity(1,i,e) .or. crystallite_subStep(1,i,e) < 1.0_pReal
enddo
enddo
!$OMP END PARALLEL DO
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i6)') '<< CRYST >> wind forward'
endif
else
subFracIntermediate = maxval(crystallite_subFrac, mask=.not.crystallite_localPlasticity)
if (subFracIntermediate == 0.0_pReal) then
crystallite_neighborEnforcedCutback = .false. ! look for ips that require a cutback because of a nonconverged neighbor
!$OMP PARALLEL
!$OMP DO PRIVATE(neighboring_e,neighboring_i)
do e = FEsolving_execElem(1),FEsolving_execElem(2)
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
if (.not. crystallite_localPlasticity(1,i,e) .and. crystallite_converged(1,i,e)) then
do n = 1_pInt,FE_NipNeighbors(FE_celltype(FE_geomtype(mesh_element(2,e))))
neighboring_e = mesh_ipNeighborhood(1,n,i,e)
neighboring_i = mesh_ipNeighborhood(2,n,i,e)
if (neighboring_e > 0_pInt .and. neighboring_i > 0_pInt) then
if (.not. crystallite_localPlasticity(1,neighboring_i,neighboring_e) &
.and. .not. crystallite_converged(1,neighboring_i,neighboring_e)) then
crystallite_neighborEnforcedCutback(i,e) = .true.
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
#ifndef _OPENMP
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) &
write(6,'(a12,i5,1x,i2,a,i5,1x,i2)') '<< CRYST >> ', neighboring_e,neighboring_i, &
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
' enforced cutback at ',e,i
#endif
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
exit
endif
endif
enddo
endif
enddo
enddo
!$OMP END DO
!$OMP DO PRIVATE(myNgrains)
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
if(crystallite_neighborEnforcedCutback(i,e)) crystallite_converged(1,i,e) = .false.
enddo
enddo
!$OMP END DO
!$OMP END PARALLEL
else
crystallite_syncSubFrac = .false. ! look for ips that have to do a time synchronization because of a nonconverged neighbor
!$OMP PARALLEL
!$OMP DO PRIVATE(neighboring_e,neighboring_i)
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2)
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
if (.not. crystallite_localPlasticity(1,i,e) .and. crystallite_subFrac(1,i,e) == 0.0_pReal) then
do n = 1_pInt,FE_NipNeighbors(FE_celltype(FE_geomtype(mesh_element(2,e))))
neighboring_e = mesh_ipNeighborhood(1,n,i,e)
neighboring_i = mesh_ipNeighborhood(2,n,i,e)
if (neighboring_e > 0_pInt .and. neighboring_i > 0_pInt) then
if (.not. crystallite_localPlasticity(1,neighboring_i,neighboring_e) &
.and. .not. crystallite_converged(1,neighboring_i,neighboring_e)) then
crystallite_syncSubFrac(i,e) = .true.
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
#ifndef _OPENMP
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) &
write(6,'(a12,i5,1x,i2,a,i5,1x,i2)') '<< CRYST >> ',neighboring_e,neighboring_i, &
' enforced time synchronization at ',e,i
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
#endif
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
exit
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
endif
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
endif
enddo
endif
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
enddo
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
enddo
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
!$OMP END DO
!$OMP DO PRIVATE(myNgrains)
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
if(crystallite_syncSubFrac(i,e)) crystallite_converged(1,i,e) = .false.
enddo
enddo
!$OMP END DO
!$OMP END PARALLEL
endif
where(.not. crystallite_localPlasticity .and. crystallite_subStep < 1.0_pReal) &
crystallite_converged = .false.
if (any(crystallite_syncSubFrac)) then ! have to do syncing now, so all wait except for the synchronizers which do a cutback
!$OMP PARALLEL DO PRIVATE(myNgrains)
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
crystallite_clearToWindForward(i,e) = crystallite_localPlasticity(1,i,e)
crystallite_clearToCutback(i,e) = crystallite_localPlasticity(1,i,e) .or. crystallite_syncSubFrac(i,e)
enddo
enddo
!$OMP END PARALLEL DO
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i6)') '<< CRYST >> time synchronization: cutback'
else
!$OMP PARALLEL DO PRIVATE(myNgrains)
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
if(.not. crystallite_converged(1,i,e)) crystallite_clearToCutback(i,e) = .true.
enddo
enddo
!$OMP END PARALLEL DO
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i6)') '<< CRYST >> cutback'
endif
endif
endif
! Make sure that all cutbackers start with the same substep
where(.not. crystallite_localPlasticity .and. .not. crystallite_converged) &
crystallite_subStep = minval(crystallite_subStep, mask=.not. crystallite_localPlasticity &
.and. .not. crystallite_converged)
! Those that do neither wind forward nor cutback are not to do
!$OMP PARALLEL DO PRIVATE(myNgrains)
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
elementLooping2: do e = FEsolving_execElem(1),FEsolving_execElem(2)
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
if(.not. crystallite_clearToWindForward(i,e) .and. .not. crystallite_clearToCutback(i,e)) &
crystallite_todo(1,i,e) = .false.
enddo
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
enddo elementLooping2
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
!$OMP END PARALLEL DO
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
endif timeSyncing1
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
!$OMP PARALLEL DO PRIVATE(myNgrains,formerSubStep)
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
elementLooping3: do e = FEsolving_execElem(1),FEsolving_execElem(2)
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
do g = 1,myNgrains
! --- wind forward ---
if (crystallite_converged(g,i,e) .and. crystallite_clearToWindForward(i,e)) then
formerSubStep = crystallite_subStep(g,i,e)
crystallite_subFrac(g,i,e) = crystallite_subFrac(g,i,e) + crystallite_subStep(g,i,e)
!$OMP FLUSH(crystallite_subFrac)
crystallite_subStep(g,i,e) = min(1.0_pReal - crystallite_subFrac(g,i,e), &
stepIncreaseCryst * crystallite_subStep(g,i,e))
!$OMP FLUSH(crystallite_subStep)
if (crystallite_subStep(g,i,e) > 0.0_pReal) then
crystallite_subF0(1:3,1:3,g,i,e) = crystallite_subF(1:3,1:3,g,i,e) ! ...def grad
!$OMP FLUSH(crystallite_subF0)
crystallite_subFp0(1:3,1:3,g,i,e) = crystallite_Fp(1:3,1:3,g,i,e) ! ...plastic def grad
crystallite_subFe0(1:3,1:3,g,i,e) = &
math_mul33x33(crystallite_subF(1:3,1:3,g,i,e), crystallite_invFp(1:3,1:3,g,i,e)) ! only needed later on for stiffness calculation
crystallite_subLp0(1:3,1:3,g,i,e) = crystallite_Lp(1:3,1:3,g,i,e) ! ...plastic velocity gradient
more work on the new state
2014-05-27 20:16:03 +05:30
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#ifdef NEWSTATE
more work on the new state
2014-05-27 20:16:03 +05:30
!if abbrevation, make c and p private in omp
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
plasticState(mappingConstitutive(2,g,i,e))%subState0(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%state(:,mappingConstitutive(1,g,i,e))
more work on the new state
2014-05-27 20:16:03 +05:30
#else
constitutive_subState0(g,i,e)%p = constitutive_state(g,i,e)%p ! ...microstructure
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#endif
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
crystallite_subTstar0_v(1:6,g,i,e) = crystallite_Tstar_v(1:6,g,i,e) ! ...2nd PK stress
if (crystallite_syncSubFrac(i,e)) then ! if we just did a synchronization of states, then we wind forward without any further time integration
crystallite_syncSubFracCompleted(i,e) = .true.
crystallite_syncSubFrac(i,e) = .false.
crystallite_todo(g,i,e) = .false.
else
crystallite_todo(g,i,e) = .true.
endif
!$OMP FLUSH(crystallite_todo)
* 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
#ifndef _OPENMP
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
if (iand(debug_level(debug_crystallite),debug_levelBasic) /= 0_pInt &
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g) &
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) &
write(6,'(a,f12.8,a,f12.8,a,i8,1x,i2,1x,i3,/)') '<< CRYST >> winding forward from ', &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
crystallite_subFrac(g,i,e)-formerSubStep,' to current crystallite_subfrac ', &
crystallite_subFrac(g,i,e),' in crystallite_stressAndItsTangent at el ip g ',e,i,g
* 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
#endif
last commit also solves problem with infinite loop in case of a cutback in homogenization corrected indentation of a comment
2013-08-02 13:29:55 +05:30
else ! this crystallite just converged for the entire timestep
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
crystallite_todo(g,i,e) = .false. ! so done here
!$OMP FLUSH(crystallite_todo)
remnoved some unused fuctions from use statements
2013-08-02 11:48:41 +05:30
if (iand(debug_level(debug_crystallite),debug_levelBasic) /= 0_pInt &
.and. formerSubStep > 0.0_pReal) then
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
!$OMP CRITICAL (distributionCrystallite)
debug_CrystalliteLoopDistribution(min(nCryst+1_pInt,NiterationCrystallite)) = &
debug_CrystalliteLoopDistribution(min(nCryst+1_pInt,NiterationCrystallite)) + 1_pInt
!$OMP END CRITICAL (distributionCrystallite)
endif
endif
! --- cutback ---
elseif (.not. crystallite_converged(g,i,e) .and. crystallite_clearToCutback(i,e)) then
if (crystallite_syncSubFrac(i,e)) then ! synchronize time
crystallite_subStep(g,i,e) = subFracIntermediate
else
crystallite_subStep(g,i,e) = subStepSizeCryst * crystallite_subStep(g,i,e) ! cut step in half and restore...
endif
!$OMP FLUSH(crystallite_subStep)
crystallite_Fp(1:3,1:3,g,i,e) = crystallite_subFp0(1:3,1:3,g,i,e) ! ...plastic def grad
!$OMP FLUSH(crystallite_Fp)
crystallite_invFp(1:3,1:3,g,i,e) = math_inv33(crystallite_Fp(1:3,1:3,g,i,e))
!$OMP FLUSH(crystallite_invFp)
crystallite_Lp(1:3,1:3,g,i,e) = crystallite_subLp0(1:3,1:3,g,i,e) ! ...plastic velocity grad
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
constitutive_state(g,i,e)%p = constitutive_subState0(g,i,e)%p ! ...microstructure
more work on the new state
2014-05-27 20:16:03 +05:30
#else
plasticState(mappingConstitutive(2,g,i,e))%state(:,mappingConstitutive(1,g,i,e)) = &
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
plasticState(mappingConstitutive(2,g,i,e))%subState0(:,mappingConstitutive(1,g,i,e))
#endif
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
crystallite_Tstar_v(1:6,g,i,e) = crystallite_subTstar0_v(1:6,g,i,e) ! ...2nd PK stress
more work on the new state
2014-05-27 20:16:03 +05:30
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
! cant restore dotState here, since not yet calculated in first cutback after initialization
crystallite_todo(g,i,e) = crystallite_subStep(g,i,e) > subStepMinCryst ! still on track or already done (beyond repair)
!$OMP FLUSH(crystallite_todo)
polishing
2012-11-22 15:28:36 +05:30
#ifndef _OPENMP
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
if(iand(debug_level(debug_crystallite),debug_levelBasic) /= 0_pInt &
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g) &
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
if (crystallite_todo(g,i,e)) then
write(6,'(a,f12.8,a,i8,1x,i2,1x,i3)') '<< CRYST >> cutback step in crystallite_stressAndItsTangent &
&with new crystallite_subStep: ',&
crystallite_subStep(g,i,e),' at el ip g ',e,i,g
else
write(6,'(a,i8,1x,i2,1x,i3)') '<< CRYST >> reached minimum step size &
&in crystallite_stressAndItsTangent at el ip g ',e,i,g
endif
write(6,*)
endif
polishing
2012-11-22 15:28:36 +05:30
#endif
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
endif
! --- prepare for integration ---
if (crystallite_todo(g,i,e) .and. (crystallite_clearToWindForward(i,e) .or. crystallite_clearToCutback(i,e))) then
crystallite_subF(1:3,1:3,g,i,e) = crystallite_subF0(1:3,1:3,g,i,e) &
+ crystallite_subStep(g,i,e) &
* (crystallite_partionedF(1:3,1:3,g,i,e) &
- crystallite_partionedF0(1:3,1:3,g,i,e))
!$OMP FLUSH(crystallite_subF)
crystallite_Fe(1:3,1:3,g,i,e) = math_mul33x33(crystallite_subF(1:3,1:3,g,i,e), crystallite_invFp(1:3,1:3,g,i,e))
crystallite_subdt(g,i,e) = crystallite_subStep(g,i,e) * crystallite_dt(g,i,e)
crystallite_converged(g,i,e) = .false. ! start out non-converged
endif
enddo ! grains
enddo ! IPs
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
enddo elementLooping3
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
!$OMP END PARALLEL DO
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
timeSyncing2: if(numerics_timeSyncing) then
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
if (any(.not. crystallite_localPlasticity .and. .not. crystallite_todo .and. .not. crystallite_converged &
.and. crystallite_subStep <= subStepMinCryst)) then ! no way of rescuing a nonlocal ip that violated the lower time step limit, ...
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) then
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
elementLooping4: do e = FEsolving_execElem(1),FEsolving_execElem(2)
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
do g = 1,myNgrains
if (.not. crystallite_localPlasticity(g,i,e) .and. .not. crystallite_todo(g,i,e) &
more work on the new state
2014-05-27 20:16:03 +05:30
.and. .not. crystallite_converged(g,i,e) .and. crystallite_subStep(g,i,e) <= subStepMinCryst) &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i8,1x,i2,1x,i3)') '<< CRYST >> nonlocal violated minimum subStep at el,ip,g ',e,i,g
enddo
enddo
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
enddo elementLooping4
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
endif
where(.not. crystallite_localPlasticity)
crystallite_todo = .false. ! ... so let all nonlocal ips die peacefully
crystallite_subStep = 0.0_pReal
endwhere
endif
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
endif timeSyncing2
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt) then
write(6,'(/,a,e12.5)') '<< CRYST >> min(subStep) ',minval(crystallite_subStep)
write(6,'(a,e12.5)') '<< CRYST >> max(subStep) ',maxval(crystallite_subStep)
write(6,'(a,e12.5)') '<< CRYST >> min(subFrac) ',minval(crystallite_subFrac)
write(6,'(a,e12.5,/)') '<< CRYST >> max(subFrac) ',maxval(crystallite_subFrac)
more work on the new state
2014-05-27 20:16:03 +05:30
flush(6)
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
endif
! --- integrate --- requires fully defined state array (basic + dependent state)
more work on the new state
2014-05-27 20:16:03 +05:30
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
if (any(crystallite_todo)) then
select case(numerics_integrator(numerics_integrationMode))
case(1_pInt)
call crystallite_integrateStateFPI()
case(2_pInt)
call crystallite_integrateStateEuler()
case(3_pInt)
call crystallite_integrateStateAdaptiveEuler()
case(4_pInt)
call crystallite_integrateStateRK4()
case(5_pInt)
call crystallite_integrateStateRKCK45()
end select
endif
more work on the new state
2014-05-27 20:16:03 +05:30
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
where(.not. crystallite_converged .and. crystallite_subStep > subStepMinCryst) & ! do not try non-converged & fully cutbacked any further
crystallite_todo = .true.
NiterationCrystallite = NiterationCrystallite + 1_pInt
enddo cutbackLooping
* checked and corrected parallelization of code. compiles now successfully, but simply aborts computation with first parallel directive without any comment :-((( * also put a call to constitutive_microstructure at the start of each crystallite_integration subroutine like it was before. need that for nonlocal model in case of crystallite cutback
2010-11-03 22:52:48 +05:30
! --+>> CHECK FOR NON-CONVERGED CRYSTALLITES <<+--
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
elementLooping5: do e = FEsolving_execElem(1),FEsolving_execElem(2)
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
do g = 1,myNgrains
if (.not. crystallite_converged(g,i,e)) then ! respond fully elastically (might be not required due to becoming terminally ill anyway)
more work on the new state
2014-05-27 20:16:03 +05:30
if(iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) &
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
write(6,'(a,i8,1x,i2,1x,i3,/)') '<< CRYST >> no convergence: respond fully elastic at el ip g ',e,i,g
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
invFp = math_inv33(crystallite_partionedFp0(1:3,1:3,g,i,e))
Fe_guess = math_mul33x33(crystallite_partionedF(1:3,1:3,g,i,e), invFp)
call constitutive_TandItsTangent(Tstar, junk2, Fe_guess,g,i,e)
crystallite_P(1:3,1:3,g,i,e) = math_mul33x33(Fe_guess,math_mul33x33(Tstar,transpose(invFp)))
endif
if(iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt &
fixed bug in crystallite_FPI: stateDamper always has to be defined for each grain added some OMP FLUSH statements were necessary replaced openmp do by forall construct where possible; this is much safer and perhaps even as fast for small loops
2012-11-07 21:13:29 +05:30
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g) &
.or. .not. iand(debug_level(debug_crystallite),debug_levelSelective) /= 0_pInt)) then
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
write(6,'(a,i8,1x,i2,1x,i3)') '<< CRYST >> central solution of cryst_StressAndTangent at el ip g ',e,i,g
write(6,'(/,a,/,3(12x,3(f12.4,1x)/))') '<< CRYST >> P / MPa', &
math_transpose33(crystallite_P(1:3,1:3,g,i,e))/1.0e6_pReal
write(6,'(a,/,3(12x,3(f14.9,1x)/))') '<< CRYST >> Fp', &
math_transpose33(crystallite_Fp(1:3,1:3,g,i,e))
write(6,'(a,/,3(12x,3(f14.9,1x)/),/)') '<< CRYST >> Lp', &
math_transpose33(crystallite_Lp(1:3,1:3,g,i,e))
more work on the new state
2014-05-27 20:16:03 +05:30
flush(6)
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
endif
enddo
enddo
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
enddo elementLooping5
* checked and corrected parallelization of code. compiles now successfully, but simply aborts computation with first parallel directive without any comment :-((( * also put a call to constitutive_microstructure at the start of each crystallite_integration subroutine like it was before. need that for nonlocal model in case of crystallite cutback
2010-11-03 22:52:48 +05:30
Tstar_v is now restored at different stages of cutbacking; we need this because preguess of state relies on consistent Tstar_v
2009-06-16 14:33:30 +05:30
* checked and corrected parallelization of code. compiles now successfully, but simply aborts computation with first parallel directive without any comment :-((( * also put a call to constitutive_microstructure at the start of each crystallite_integration subroutine like it was before. need that for nonlocal model in case of crystallite cutback
2010-11-03 22:52:48 +05:30
! --+>> STIFFNESS CALCULATION <<+--
Tstar_v is now restored at different stages of cutbacking; we need this because preguess of state relies on consistent Tstar_v
2009-06-16 14:33:30 +05:30
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
computeJacobian: if(updateJaco) then
jacobianMethod: if (analyticJaco) then
constitutive_nonlocal: - corrected flux term - multiplication is now aware of dislocation type - corrected change rate for "dipole size" dupper - corrected term for dipole dissociation by stress change - added transmissivity term in fluxes which accounts for misorientation between two neighboring grains (yet hardcoded transmissivity according to misorientation angle) - added more output variables constitutive: - 2 additional variables "previousDotState" and "previousDotState2", which are used to store the previous and second previous dotState (used in crystallite for acceleration/stabilization of state integration) - timer for dotState now measures the time for calls to constitutive_ collectState (used to reside in crystallite_updateState, which is not critical in terms of calculation time anymore) crystallite: - convergence check for nonlocal elments is now done at end of crystallite loop, not at the beginning; we simple set all elements to not converged if there is at least one nonlocal element that did not converge - need call to microstructure before first call to collect dotState for dependent states - stiffness calculation (jacobian): if there are nonlocal elements, we also have to consider changes in our neighborhood's states; so for every perturbed component in a single ip, we have to loop over all elements; since this is extremely time-consuming, we just perturb one component per cycle, starting with the one that changes the most during regular time step. - updateState gets a damping prefactor for our dotState that helps to improve convergence; prefactor is calculated according to change of dotState IO: - additional warning message for unknown crystal symmetry
2009-12-15 13:50:31 +05:30
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
! --- ANALYTIC JACOBIAN ---
!$OMP PARALLEL DO PRIVATE(dFedF,dSdF,dSdFe,myNgrains)
elementLooping6: do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
do g = 1_pInt,myNgrains
dFedF = 0.0_pReal
forall(p=1_pInt:3_pInt, o=1_pInt:3_pInt) &
fixed wrong indices in analytic jacobian calculation following christoph's changes. convergence is restored at least for the spectral solver
2014-01-22 15:46:55 +05:30
dFedF(o,p,o,1:3) = crystallite_invFp(1:3,p,g,i,e) ! dFe^T_ij/dF_kl = delta_jk * (Fp current^-1)_li
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_TandItsTangent(junk,dSdFe,crystallite_subFe0(1:3,1:3,g,i,e),g,i,e) ! call constitutive law to calculate 2nd Piola-Kirchhoff stress and its derivative
dSdF = math_mul3333xx3333(dSdFe,dFedF) ! dS/dF = dS/dFe * dFe/dF
forall(p=1_pInt:3_pInt, o=1_pInt:3_pInt) &
crystallite_dPdF(1:3,1:3,o,p,g,i,e) = math_mul33x33(math_mul33x33(dFedF(1:3,1:3,o,p),&
fixed bug in heat generation and analytic jacobian calculations
2013-12-18 15:49:48 +05:30
math_Mandel6to33(crystallite_Tstar_v(1:6,g,i,e))),math_transpose33(&
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
crystallite_invFp(1:3,1:3,g,i,e))) & ! dP/dF = dFe/dF * S * Fp^-T...
+ math_mul33x33(crystallite_subFe0(1:3,1:3,g,i,e),&
math_mul33x33(dSdF(1:3,1:3,o,p),math_transpose33(crystallite_invFp(1:3,1:3,g,i,e)))) ! + Fe * dS/dF * Fp^-T
enddo; enddo
enddo elementLooping6
!$OMP END PARALLEL DO
moved rate sensitivity block to where it belongs to, namely the analytic jacobian calculation
2014-01-22 21:04:10 +05:30
rateSensitivity: if (rate_sensitivity) then
!$OMP PARALLEL DO PRIVATE(dFedFdot,dSdFdot,dSdFe,Fpinv_rate,FDot_inv,counter,dFp_invdFdot,myNgrains)
elementLooping11: do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
do g = 1_pInt,myNgrains
Fpinv_rate = math_mul33x33(crystallite_invFp(1:3,1:3,g,i,e),crystallite_Lp(1:3,1:3,g,i,e)) ! dFp^-1 = dFp^-1/dt *dt... dFp may overshoot dF by small ammount as
FDot_inv = crystallite_subF(1:3,1:3,g,i,e) - crystallite_F0(1:3,1:3,g,i,e)
counter = 0.0_pReal
do p=1_pInt,3_pInt; do o=1_pInt,3_pInt
if (abs(FDot_inv(o,p)) < relevantStrain) then
FDot_inv(o,p) = 0.0_pReal
else
counter = counter + 1.0_pReal
FDot_inv(o,p) = crystallite_dt(g,i,e)/FDot_inv(o,p)
endif
enddo; enddo
if (counter > 0.0_pReal) FDot_inv = FDot_inv/counter
forall(p=1_pInt:3_pInt, o=1_pInt:3_pInt) &
dFp_invdFdot(o,p,1:3,1:3) = Fpinv_rate(o,p)*FDot_inv
forall(p=1_pInt:3_pInt, o=1_pInt:3_pInt) &
dFedFdot(1:3,1:3,o,p) = math_transpose33(math_mul33x33(crystallite_subF(1:3,1:3,g,i,e), &
dFp_invdFdot(1:3,1:3,o,p)))
call constitutive_TandItsTangent(junk,dSdFe,crystallite_subFe0(1:3,1:3,g,i,e),g,i,e) ! call constitutive law to calculate 2nd Piola-Kirchhoff stress and its derivative
dSdFdot = math_mul3333xx3333(dSdFe,dFedFdot)
forall(p=1_pInt:3_pInt, o=1_pInt:3_pInt) &
crystallite_dPdF(1:3,1:3,o,p,g,i,e) = crystallite_dPdF(1:3,1:3,o,p,g,i,e) - &
(math_mul33x33(math_mul33x33(dFedFdot(1:3,1:3,o,p), &
math_Mandel6to33(crystallite_Tstar_v(1:6,g,i,e))),math_transpose33( &
crystallite_invFp(1:3,1:3,g,i,e))) + & ! dP/dFdot = dFe/dFdot * S * Fp^-T...
math_mul33x33(math_mul33x33(crystallite_subFe0(1:3,1:3,g,i,e), &
math_Mandel6to33(crystallite_Tstar_v(1:6,g,i,e))),math_transpose33(dFp_invdFdot(1:3,1:3,o,p))) & ! + Fe * S * dFp^-T/dFdot...
+ math_mul33x33(crystallite_subFe0(1:3,1:3,g,i,e), &
math_mul33x33(dSdFdot(1:3,1:3,o,p),math_transpose33(crystallite_invFp(1:3,1:3,g,i,e))))) ! + Fe * dS/dFdot * Fp^-T
enddo; enddo;
enddo elementLooping11
!$OMP END PARALLEL DO
endif rateSensitivity
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
else jacobianMethod
! --- STANDARD (PERTURBATION METHOD) FOR JACOBIAN ---
numerics_integrationMode = 2_pInt
! --- BACKUP ---
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL DO PRIVATE(myNgrains)
elementLooping7: do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), g = 1:myNgrains)
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#ifdef NEWSTATE
more work on the new state
2014-05-27 20:16:03 +05:30
plasticState(mappingConstitutive(2,g,i,e))%state_backup(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%state(:,mappingConstitutive(1,g,i,e))
plasticState(mappingConstitutive(2,g,i,e))%dotState_backup(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e))
#else
constitutive_state_backup(g,i,e)%p(1:constitutive_sizeState(g,i,e)) = &
constitutive_state(g,i,e)%p(1:constitutive_sizeState(g,i,e)) ! remember unperturbed, converged state, ...
constitutive_dotState_backup(g,i,e)%p(1:constitutive_sizeDotState(g,i,e)) = &
constitutive_dotState(g,i,e)%p(1:constitutive_sizeDotState(g,i,e)) ! ... dotStates, ...
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#endif
more work on the new state
2014-05-27 20:16:03 +05:30
F_backup(1:3,1:3,g,i,e) = crystallite_subF(1:3,1:3,g,i,e) ! ... and kinematics
Fp_backup(1:3,1:3,g,i,e) = crystallite_Fp(1:3,1:3,g,i,e)
InvFp_backup(1:3,1:3,g,i,e) = crystallite_invFp(1:3,1:3,g,i,e)
Fe_backup(1:3,1:3,g,i,e) = crystallite_Fe(1:3,1:3,g,i,e)
Lp_backup(1:3,1:3,g,i,e) = crystallite_Lp(1:3,1:3,g,i,e)
Tstar_v_backup(1:6,g,i,e) = crystallite_Tstar_v(1:6,g,i,e)
P_backup(1:3,1:3,g,i,e) = crystallite_P(1:3,1:3,g,i,e)
convergenceFlag_backup(g,i,e) = crystallite_converged(g,i,e)
endforall
enddo elementLooping7
!$END PARALLEL DO
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
! --- CALCULATE STATE AND STRESS FOR PERTURBATION ---
dPdF_perturbation1 = crystallite_dPdF0 ! initialize stiffness with known good values from last increment
dPdF_perturbation2 = crystallite_dPdF0 ! initialize stiffness with known good values from last increment
more work on the new state
2014-05-27 20:16:03 +05:30
pertubationLoop: do perturbation = 1,2 ! forward and backward perturbation
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
if (iand(pert_method,perturbation) > 0_pInt) then ! mask for desired direction
myPert = -pert_Fg * (-1.0_pReal)**perturbation ! set perturbation step
do k = 1,3; do l = 1,3 ! ...alter individual components
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
write(6,'(a,2(1x,i1),1x,a,/)') '<< CRYST >> [[[[[[ Stiffness perturbation',k,l,']]]]]]'
! --- INITIALIZE UNPERTURBED STATE ---
select case(numerics_integrator(numerics_integrationMode))
more work on the new state
2014-05-27 20:16:03 +05:30
case(1_pInt)
!why not OMP? ! Fix-point method: restore to last converged state at end of subinc, since this is probably closest to perturbed state
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), g = 1:myNgrains)
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#ifdef NEWSTATE
plasticState(mappingConstitutive(2,g,i,e))%state(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%state_backup(:,mappingConstitutive(1,g,i,e))
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%dotState_backup(:,mappingConstitutive(1,g,i,e))
more work on the new state
2014-05-27 20:16:03 +05:30
#else
constitutive_state(g,i,e)%p(1:constitutive_sizeState(g,i,e)) = &
constitutive_state_backup(g,i,e)%p(1:constitutive_sizeState(g,i,e))
constitutive_dotState(g,i,e)%p(1:constitutive_sizeDotState(g,i,e)) = &
constitutive_dotState_backup(g,i,e)%p(1:constitutive_sizeDotState(g,i,e))
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#endif
perturbation calculation in crystallite_stressAndItsTangent: only backup and restore those chunks of crystallite_Fe, _Fp etc that are actually used in order to avoid pointless data copying in non ping pong mode
2014-01-21 22:05:12 +05:30
crystallite_Fp(1:3,1:3,g,i,e) = Fp_backup(1:3,1:3,g,i,e)
crystallite_invFp(1:3,1:3,g,i,e) = InvFp_backup(1:3,1:3,g,i,e)
crystallite_Fe(1:3,1:3,g,i,e) = Fe_backup(1:3,1:3,g,i,e)
crystallite_Lp(1:3,1:3,g,i,e) = Lp_backup(1:3,1:3,g,i,e)
crystallite_Tstar_v(1:6,g,i,e) = Tstar_v_backup(1:6,g,i,e)
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
endforall
enddo
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
case(2_pInt,3_pInt) ! explicit Euler methods: nothing to restore (except for F), since we are only doing a stress integration step
more work on the new state
2014-05-27 20:16:03 +05:30
case(4_pInt,5_pInt)
!why not OMP? ! explicit Runge-Kutta methods: restore to start of subinc, since we are doing a full integration of state and stress
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), g = 1:myNgrains)
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#ifdef NEWSTATE
plasticState(mappingConstitutive(2,g,i,e))%state(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%subState0(:,mappingConstitutive(1,g,i,e))
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) = &
more work on the new state
2014-05-27 20:16:03 +05:30
plasticState(mappingConstitutive(2,g,i,e))%dotState_backup(:,mappingConstitutive(1,g,i,e))
#else
constitutive_state(g,i,e)%p(1:constitutive_sizeState(g,i,e)) = &
constitutive_subState0(g,i,e)%p(1:constitutive_sizeState(g,i,e))
constitutive_dotState(g,i,e)%p(1:constitutive_sizeDotState(g,i,e)) = &
constitutive_dotState_backup(g,i,e)%p(1:constitutive_sizeDotState(g,i,e))
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#endif
perturbation calculation in crystallite_stressAndItsTangent: only backup and restore those chunks of crystallite_Fe, _Fp etc that are actually used in order to avoid pointless data copying in non ping pong mode
2014-01-21 22:05:12 +05:30
crystallite_Fp(1:3,1:3,g,i,e) = crystallite_subFp0(1:3,1:3,g,i,e)
crystallite_Fe(1:3,1:3,g,i,e) = crystallite_subFe0(1:3,1:3,g,i,e)
crystallite_Lp(1:3,1:3,g,i,e) = crystallite_subLp0(1:3,1:3,g,i,e)
crystallite_Tstar_v(1:6,g,i,e) = crystallite_subTstar0_v(1:6,g,i,e)
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
endforall
enddo
end select
! --- PERTURB EITHER FORWARD OR BACKWARD ---
more work on the new state
2014-05-27 20:16:03 +05:30
!why not OMP?
setting todo and converged flag only for those e,i,g that are to be executed (FE_elemExec etc) used to spoil convergence checking in non ping pong mode and lead to slow performance
2014-01-21 21:39:00 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
do g = 1,myNgrains
crystallite_subF(1:3,1:3,g,i,e) = F_backup(1:3,1:3,g,i,e)
crystallite_subF(k,l,g,i,e) = crystallite_subF(k,l,g,i,e) + myPert
crystallite_todo(g,i,e) = crystallite_requested(g,i,e) &
.and. convergenceFlag_backup(g,i,e)
more work on the new state
2014-05-27 20:16:03 +05:30
if (crystallite_todo(g,i,e)) crystallite_converged(g,i,e) = .false. ! start out non-converged
setting todo and converged flag only for those e,i,g that are to be executed (FE_elemExec etc) used to spoil convergence checking in non ping pong mode and lead to slow performance
2014-01-21 21:39:00 +05:30
enddo; enddo; enddo
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
more work on the new state
2014-05-27 20:16:03 +05:30
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
select case(numerics_integrator(numerics_integrationMode))
case(1_pInt)
call crystallite_integrateStateFPI()
case(2_pInt)
call crystallite_integrateStateEuler()
case(3_pInt)
call crystallite_integrateStateAdaptiveEuler()
case(4_pInt)
call crystallite_integrateStateRK4()
case(5_pInt)
call crystallite_integrateStateRKCK45()
end select
more work on the new state
2014-05-27 20:16:03 +05:30
!why not OMP?
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
elementLooping8: do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
select case(perturbation)
case(1_pInt)
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), g = 1:myNgrains, &
more work on the new state
2014-05-27 20:16:03 +05:30
crystallite_requested(g,i,e) .and. crystallite_converged(g,i,e)) & ! converged state warrants stiffness update
dPdF_perturbation1(1:3,1:3,k,l,g,i,e) = &
(crystallite_P(1:3,1:3,g,i,e) - P_backup(1:3,1:3,g,i,e)) / myPert ! tangent dP_ij/dFg_kl
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
case(2_pInt)
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), g = 1:myNgrains, &
more work on the new state
2014-05-27 20:16:03 +05:30
crystallite_requested(g,i,e) .and. crystallite_converged(g,i,e)) & ! converged state warrants stiffness update
dPdF_perturbation2(1:3,1:3,k,l,g,i,e) = &
(crystallite_P(1:3,1:3,g,i,e) - P_backup(1:3,1:3,g,i,e)) / myPert ! tangent dP_ij/dFg_kl
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
end select
enddo elementLooping8
more work on the new state
2014-05-27 20:16:03 +05:30
enddo; enddo ! k,l component perturbation loop
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
endif
more work on the new state
2014-05-27 20:16:03 +05:30
enddo pertubationLoop
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
! --- STIFFNESS ACCORDING TO PERTURBATION METHOD AND CONVERGENCE ---
elementLooping9: do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
select case(pert_method)
case(1_pInt)
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), g = 1:myNgrains, &
crystallite_requested(g,i,e) .and. convergenceFlag_backup(g,i,e)) & ! perturbation mode 1: central solution converged
crystallite_dPdF(1:3,1:3,1:3,1:3,g,i,e) = dPdF_perturbation1(1:3,1:3,1:3,1:3,g,i,e)
case(2_pInt)
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), g = 1:myNgrains, &
crystallite_requested(g,i,e) .and. convergenceFlag_backup(g,i,e)) & ! perturbation mode 2: central solution converged
crystallite_dPdF(1:3,1:3,1:3,1:3,g,i,e) = dPdF_perturbation2(1:3,1:3,1:3,1:3,g,i,e)
case(3_pInt)
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), g = 1:myNgrains, &
crystallite_requested(g,i,e) .and. convergenceFlag_backup(g,i,e)) & ! perturbation mode 3: central solution converged
crystallite_dPdF(1:3,1:3,1:3,1:3,g,i,e) = 0.5_pReal* ( dPdF_perturbation1(1:3,1:3,1:3,1:3,g,i,e) &
+ dPdF_perturbation2(1:3,1:3,1:3,1:3,g,i,e))
end select
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), g = 1:myNgrains, &
crystallite_requested(g,i,e) .and. .not. convergenceFlag_backup(g,i,e)) & ! for any pertubation mode: if central solution did not converge...
crystallite_dPdF(1:3,1:3,1:3,1:3,g,i,e) = crystallite_fallbackdPdF(1:3,1:3,1:3,1:3,g,i,e) ! ...use (elastic) fallback
enddo elementLooping9
! --- RESTORE ---
more work on the new state
2014-05-27 20:16:03 +05:30
!why not OMP?
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
elementLooping10: do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
forall (i = FEsolving_execIP(1,e):FEsolving_execIP(2,e), g = 1:myNgrains)
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#ifdef NEWSTATE
plasticState(mappingConstitutive(2,g,i,e))%state(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%state_backup(:,mappingConstitutive(1,g,i,e))
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%dotState_backup(:,mappingConstitutive(1,g,i,e))
more work on the new state
2014-05-27 20:16:03 +05:30
#else
constitutive_state(g,i,e)%p(1:constitutive_sizeState(g,i,e)) = &
constitutive_state_backup(g,i,e)%p(1:constitutive_sizeState(g,i,e))
constitutive_dotState(g,i,e)%p(1:constitutive_sizeDotState(g,i,e)) = &
constitutive_dotState_backup(g,i,e)%p(1:constitutive_sizeDotState(g,i,e))
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#endif
perturbation calculation in crystallite_stressAndItsTangent: only backup and restore those chunks of crystallite_Fe, _Fp etc that are actually used in order to avoid pointless data copying in non ping pong mode (missed some part in last commit)
2014-01-21 23:58:21 +05:30
crystallite_subF(1:3,1:3,g,i,e) = F_backup(1:3,1:3,g,i,e)
crystallite_Fp(1:3,1:3,g,i,e) = Fp_backup(1:3,1:3,g,i,e)
crystallite_invFp(1:3,1:3,g,i,e) = InvFp_backup(1:3,1:3,g,i,e)
crystallite_Fe(1:3,1:3,g,i,e) = Fe_backup(1:3,1:3,g,i,e)
crystallite_Lp(1:3,1:3,g,i,e) = Lp_backup(1:3,1:3,g,i,e)
crystallite_Tstar_v(1:6,g,i,e) = Tstar_v_backup(1:6,g,i,e)
crystallite_P(1:3,1:3,g,i,e) = P_backup(1:3,1:3,g,i,e)
crystallite_converged(g,i,e) = convergenceFlag_backup(g,i,e)
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
endforall
enddo elementLooping10
moved rate sensitivity block to where it belongs to, namely the analytic jacobian calculation
2014-01-22 21:04:10 +05:30
endif jacobianMethod
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
endif computeJacobian
more work on the new state
2014-05-27 20:16:03 +05:30
!why not OMP?
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
elementLooping12: do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
do g = 1,myNgrains
crystallite_heat(g,i,e) = 0.98_pReal* &
fixed bug in heat generation and analytic jacobian calculations
2013-12-18 15:49:48 +05:30
abs(math_mul33xx33(math_Mandel6to33(crystallite_Tstar_v(1:6,g,i,e)), &
crystallite_Lp(1:3,1:3,g,i,e)))
DAMASK (except nonlocal) now sees and interacts with FEM solver
2014-06-06 06:08:29 +05:30
constitutive_localDamage(g,i,e) = &
1.0_pReal* &
sum(math_Mandel6to33(crystallite_Tstar_v(1:6,g,i,e)/constitutive_gradientDamage(g,i,e))* &
(math_mul33x33(math_transpose33(crystallite_Fe(1:3,1:3,g,i,e)), &
crystallite_Fe(1:3,1:3,g,i,e))-math_I3))/4.0_pReal + &
0.0_pReal* &
sum(abs(math_mul33x33(math_transpose33(crystallite_Fp(1:3,1:3,g,i,e)), &
crystallite_Fp(1:3,1:3,g,i,e))-math_I3)/2.0_pReal)
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
enddo
enddo
enddo elementLooping12
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
end subroutine crystallite_stressAndItsTangent
major restructuring of code. homogenization as well as constitutive are now free to choose. the runtime got somewhat longer (25% on simple tests) compared to a hardcoded isostrain homogenization. this might be a point of further optimization at a later stage... please use homogenization_isostrain.f90 as starting point / example for future developments of homog-schemes. the homogenization scheme now can additionally output certain results. hence, the userdata structure at each integration point now looks like this: - sizeHomogPostResults - block of that size containing homogPostResults then for each grain: - sizeGrainPostResults - block of that size containing crystallitePostResults, which consist of: + phaseID + volFrac + Eulers (3) + any constitutive post results requested
2009-05-07 21:57:36 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
!> @brief integrate stress, state with 4th order explicit Runge Kutta method
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
subroutine crystallite_integrateStateRK4()
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
use numerics, only: &
numerics_integrationMode
use debug, only: &
debug_level, &
debug_crystallite, &
debug_levelBasic, &
debug_levelExtensive, &
debug_levelSelective, &
debug_e, &
debug_i, &
debug_g, &
debug_StateLoopDistribution
use FEsolving, only: &
FEsolving_execElem, &
FEsolving_execIP
use mesh, only: &
mesh_element, &
mesh_NcpElems, &
mesh_maxNips
use material, only: &
homogenization_Ngrains, &
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#ifdef NEWSTATE
plasticState, &
#endif
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
homogenization_maxNgrains
use constitutive, only: &
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_collectDotState, &
constitutive_microstructure, &
#ifndef NEWSTATE
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
constitutive_sizeDotState, &
constitutive_state, &
constitutive_subState0, &
constitutive_dotState, &
constitutive_RK4dotState, &
constitutive_deltaState, &
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_collectDeltaState
#else
mappingConstitutive
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
implicit none
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
real(pReal), dimension(4), parameter :: &
more work on the new state
2014-05-27 20:16:03 +05:30
TIMESTEPFRACTION = [0.5_pReal, 0.5_pReal, 1.0_pReal, 1.0_pReal] ! factor giving the fraction of the original timestep used for Runge Kutta Integration
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
real(pReal), dimension(4), parameter :: &
more work on the new state
2014-05-27 20:16:03 +05:30
WEIGHT = [1.0_pReal, 2.0_pReal, 2.0_pReal, 1.0_pReal] ! weight of slope used for Runge Kutta integration
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
more work on the new state
2014-05-27 20:16:03 +05:30
integer(pInt) :: e, & ! element index in element loop
doxygen comments
2013-02-22 04:38:36 +05:30
i, & ! integration point index in ip loop
g, & ! grain index in grain loop
more work on the new state
2014-05-27 20:16:03 +05:30
p, & ! phase loop
c, &
doxygen comments
2013-02-22 04:38:36 +05:30
n, &
mySizeDotState
integer(pInt), dimension(2) :: eIter ! bounds for element iteration
integer(pInt), dimension(2,mesh_NcpElems) :: iIter, & ! bounds for ip iteration
gIter ! bounds for grain iteration
fixed out of range bug introduced with singleRun changes in rev. 2356
2013-04-30 17:44:07 +05:30
logical :: singleRun ! flag indicating computation for single (g,i,e) triple
doxygen comments
2013-02-22 04:38:36 +05:30
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
eIter = FEsolving_execElem(1:2)
do e = eIter(1),eIter(2)
iIter(1:2,e) = FEsolving_execIP(1:2,e)
gIter(1:2,e) = [ 1_pInt,homogenization_Ngrains(mesh_element(3,e))]
enddo
fixed out of range bug introduced with singleRun changes in rev. 2356
2013-04-30 17:44:07 +05:30
singleRun = (eIter(1) == eIter(2) .and. iIter(1,eIter(1)) == iIter(2,eIter(2)))
doxygen comments
2013-02-22 04:38:36 +05:30
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#ifdef NEWSTATE
more work on the new state
2014-05-27 20:16:03 +05:30
!--------------------------------------------------------------------------------------------------
! initialize dotState
if (.not. singleRun) then
forall(p = 1_pInt:size(plasticState)) plasticState(p)%RK4dotState = 0.0_pReal
else
e = eIter(1)
i = iIter(1,e)
do g = iIter(1,e), iIter(2,e)
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
plasticState(mappingConstitutive(2,g,i,e))%RK4dotState(:,mappingConstitutive(1,g,i,e)) = 0.0_pReal
more work on the new state
2014-05-27 20:16:03 +05:30
enddo
endif
#endif
!--------------------------------------------------------------------------------------------------
! first Runge-Kutta step
!$OMP PARALLEL
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
#ifndef NEWSTATE
constitutive_RK4dotState(g,i,e)%p = 0.0_pReal ! initialize Runge-Kutta dotState
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
if (crystallite_todo(g,i,e)) then
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_collectDotState(crystallite_Tstar_v(1:6,g,i,e), crystallite_Fe, &
crystallite_Fp, crystallite_temperature(i,e), &
crystallite_subdt(g,i,e), crystallite_subFrac, g,i,e)
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP DO PRIVATE(p,c)
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
if ( any(constitutive_dotState(g,i,e)%p /= constitutive_dotState(g,i,e)%p)) then ! NaN occured in dotState
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
!$OMP CRITICAL (checkTodo)
more work on the new state
2014-05-27 20:16:03 +05:30
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
!$OMP END CRITICAL (checkTodo)
more work on the new state
2014-05-27 20:16:03 +05:30
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
endif
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#else
c = mappingConstitutive(1,g,i,e)
p = mappingConstitutive(2,g,i,e)
if ( any(plasticState(p)%dotState(:,c) /= plasticState(p)%dotState(:,c))) then ! NaN occured in dotState
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP CRITICAL (checkTodo)
more work on the new state
2014-05-27 20:16:03 +05:30
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP END CRITICAL (checkTodo)
more work on the new state
2014-05-27 20:16:03 +05:30
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
doxygen comments
2013-02-22 04:38:36 +05:30
endif
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP END PARALLEL
!--------------------------------------------------------------------------------------------------
! --- SECOND TO FOURTH RUNGE KUTTA STEP PLUS FINAL INTEGRATION ---
doxygen comments
2013-02-22 04:38:36 +05:30
do n = 1_pInt,4_pInt
! --- state update ---
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL
!$OMP DO PRIVATE(p,c)
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifdef NEWSTATE
p = mappingConstitutive(2,g,i,e)
c = mappingConstitutive(1,g,i,e)
#endif
first3steps: if (n < 4) then
#ifndef NEWSTATE
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_RK4dotState(g,i,e)%p = constitutive_RK4dotState(g,i,e)%p &
+ weight(n)*constitutive_dotState(g,i,e)%p
more work on the new state
2014-05-27 20:16:03 +05:30
#else
plasticState(p)%RK4dotState(:,c) = plasticState(p)%RK4dotState(:,c) &
+ weight(n)*plasticState(p)%dotState(:,c)
#endif
else first3steps
#ifndef NEWSTATE
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_dotState(g,i,e)%p = (constitutive_RK4dotState(g,i,e)%p &
+ weight(n)*constitutive_dotState(g,i,e)%p) / 6.0_pReal ! use weighted RKdotState for final integration
more work on the new state
2014-05-27 20:16:03 +05:30
#else
plasticState(p)%RK4dotState(:,c) = plasticState(p)%RK4dotState(:,c) &
+ weight(n)*plasticState(p)%dotState(:,c) / 6.0_pReal
#endif
endif first3steps
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP DO PRIVATE(mySizeDotState,p,c)
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
constitutive_state(g,i,e)%p(1:mySizeDotState) = constitutive_subState0(g,i,e)%p(1:mySizeDotState) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
+ constitutive_dotState(g,i,e)%p(1:mySizeDotState) &
* crystallite_subdt(g,i,e) * timeStepFraction(n)
more work on the new state
2014-05-27 20:16:03 +05:30
#else
p = mappingConstitutive(2,g,i,e)
c = mappingConstitutive(1,g,i,e)
plasticState(p)%State(:,c) = plasticState(p)%subState0(:,c) &
+ plasticState(p)%dotState (:,c) * crystallite_subdt(g,i,e) &
* timeStepFraction(n)
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
if (n == 4) then ! final integration step
new function stateJump, which takes care of immediate changes in the state (deltaState); this introduces discontinuities in the state evolution; therefore this is always and only once done after each integration step, so no evaluation of deltaState for intermediate steps of e.g. the Runge-Kutta integrator; otherwise integration becomes a pain without significant gain in accuracy deltaState now successfully tested for nonlocal model; was not correct for integrators 1,4,5 before
2012-06-06 20:41:30 +05:30
#ifndef _OPENMP
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt &
more work on the new state
2014-05-27 20:16:03 +05:30
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g) &
doxygen comments
2013-02-22 04:38:36 +05:30
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
write(6,'(a,i8,1x,i2,1x,i3,/)') '<< CRYST >> updateState at el ip g ',e,i,g
more work on the new state
2014-05-27 20:16:03 +05:30
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> dotState', constitutive_dotState(g,i,e)%p(1:mySizeDotState)
fixed some e12.6 string formats to avoid compile time warnings
2013-10-08 19:21:36 +05:30
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> new state', constitutive_state(g,i,e)%p(1:mySizeDotState)
more work on the new state
2014-05-27 20:16:03 +05:30
#else
write(6,'(a,i8,1x,i2,1x,i3,/)') '<< CRYST >> updateState at el ip g ',e,i,g
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> dotState', plasticState(p)%dotState(:,c)
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> new state', plasticState(p)%State(:,c)
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
new function stateJump, which takes care of immediate changes in the state (deltaState); this introduces discontinuities in the state evolution; therefore this is always and only once done after each integration step, so no evaluation of deltaState for intermediate steps of e.g. the Runge-Kutta integrator; otherwise integration becomes a pain without significant gain in accuracy deltaState now successfully tested for nonlocal model; was not correct for integrators 1,4,5 before
2012-06-06 20:41:30 +05:30
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- state jump ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
crystallite_todo(g,i,e) = crystallite_stateJump(g,i,e)
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e) .and. .not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- update dependent states ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_microstructure(crystallite_temperature(i,e), crystallite_Fe(1:3,1:3,g,i,e), &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_Fp(1:3,1:3,g,i,e), g, i, e) ! update dependent state variables to be consistent with basic states
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- stress integration ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
crystallite_todo(g,i,e) = crystallite_integrateStress(g,i,e,timeStepFraction(n)) ! fraction of original times step
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e) .and. .not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- dot state and RK dot state---
more work on the new state
2014-05-27 20:16:03 +05:30
first3steps2: if (n < 4) then
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_collectDotState(crystallite_Tstar_v(1:6,g,i,e), crystallite_Fe, &
crystallite_Fp, crystallite_temperature(i,e), &
timeStepFraction(n)*crystallite_subdt(g,i,e), & ! fraction of original timestep
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_subFrac, g,i,e)
endif
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP DO PRIVATE(p,c)
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
if ( any(constitutive_dotState(g,i,e)%p /= constitutive_dotState(g,i,e)%p)) then ! NaN occured in dotState
doxygen comments
2013-02-22 04:38:36 +05:30
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
endif
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#else
p = mappingConstitutive(2,g,i,e)
c = mappingConstitutive(1,g,i,e)
if ( any(plasticState(p)%dotState(:,c) /= plasticState(p)%dotState(:,c))) then
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
endif
endif
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
endif first3steps2
!$OMP END PARALLEL
doxygen comments
2013-02-22 04:38:36 +05:30
enddo
! --- SET CONVERGENCE FLAG ---
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
crystallite_converged(g,i,e) = .true. ! if still "to do" then converged per definitionem
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
!$OMP CRITICAL (distributionState)
debug_StateLoopDistribution(4,numerics_integrationMode) = &
debug_StateLoopDistribution(4,numerics_integrationMode) + 1_pInt
!$OMP END CRITICAL (distributionState)
endif
endif
enddo; enddo; enddo
! --- CHECK NONLOCAL CONVERGENCE ---
if (.not. singleRun) then ! if not requesting Integration of just a single IP
if (any(.not. crystallite_converged .and. .not. crystallite_localPlasticity)) then ! any non-local not yet converged (or broken)...
crystallite_converged = crystallite_converged .and. crystallite_localPlasticity ! ...restart all non-local as not converged
endif
endif
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
end subroutine crystallite_integrateStateRK4
* now able to choose method for state integration (integrator and integratorStiffness in numerics.config); standard is "Fixed Point Iteration", which is basically the same as before; others available are "Explicit Euler", "AdaptiveEuler", "Classical Runge-Kutta" and "Runge-Kutta Cash-Karp" * now remembering stiffness similar to how we do it for Lp etc.; avoids undefined stiffness values for nonconverged stiffness calculation * non-local stuff: * changed non-local kinetics (Gilman2002) * enforce zero shearrate for overall carrrier density below relevant density * enforce zero density for those states that become negative and were below relevant density before * dislocation velocity is not limited by V^(1/3) / dt anymore
2010-10-01 17:48:49 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
!> @brief integrate stress, state with 5th order Runge-Kutta Cash-Karp method with
doxygen comments
2013-02-22 04:38:36 +05:30
!> adaptive step size (use 5th order solution to advance = "local extrapolation")
!--------------------------------------------------------------------------------------------------
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
subroutine crystallite_integrateStateRKCK45()
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
use debug, only: &
debug_level, &
debug_crystallite, &
debug_levelBasic, &
debug_levelExtensive, &
debug_levelSelective, &
debug_e, &
debug_i, &
debug_g, &
debug_StateLoopDistribution
use numerics, only: &
rTol_crystalliteState, &
numerics_integrationMode
use FEsolving, only: &
FEsolving_execElem, &
FEsolving_execIP
use mesh, only: &
mesh_element, &
mesh_NcpElems, &
mesh_maxNips
use material, only: &
homogenization_Ngrains, &
more work on the new state
2014-05-27 20:16:03 +05:30
#ifdef NEWSTATE
plasticState, &
#endif
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
homogenization_maxNgrains
use constitutive, only: &
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_collectDotState, &
changes to make newstate aware of damage variables
2014-06-11 22:02:09 +05:30
constitutive_maxSizeDotState, &
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
constitutive_sizeDotState, &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_state, &
constitutive_aTolState, &
constitutive_subState0, &
constitutive_dotState, &
constitutive_RKCK45dotState, &
constitutive_deltaState, &
constitutive_collectDeltaState, &
more work on the new state
2014-05-27 20:16:03 +05:30
#else
mappingConstitutive, &
#endif
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_microstructure
doxygen comments
2013-02-22 04:38:36 +05:30
implicit none
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
real(pReal), dimension(5,5), parameter :: &
A = reshape([&
.2_pReal, .075_pReal, .3_pReal, -11.0_pReal/54.0_pReal, 1631.0_pReal/55296.0_pReal, &
.0_pReal, .225_pReal, -.9_pReal, 2.5_pReal, 175.0_pReal/512.0_pReal, &
.0_pReal, .0_pReal, 1.2_pReal, -70.0_pReal/27.0_pReal, 575.0_pReal/13824.0_pReal, &
.0_pReal, .0_pReal, .0_pReal, 35.0_pReal/27.0_pReal, 44275.0_pReal/110592.0_pReal, &
.0_pReal, .0_pReal, .0_pReal, .0_pReal, 253.0_pReal/4096.0_pReal], &
[5,5], order=[2,1]) !< coefficients in Butcher tableau (used for preliminary integration in stages 2 to 6)
real(pReal), dimension(6), parameter :: &
B = &
[37.0_pReal/378.0_pReal, .0_pReal, 250.0_pReal/621.0_pReal, &
125.0_pReal/594.0_pReal, .0_pReal, 512.0_pReal/1771.0_pReal], & !< coefficients in Butcher tableau (used for final integration and error estimate)
DB = B - &
more work on the new state
2014-05-27 20:16:03 +05:30
[2825.0_pReal/27648.0_pReal, .0_pReal, 18575.0_pReal/48384.0_pReal,&
13525.0_pReal/55296.0_pReal, 277.0_pReal/14336.0_pReal, 0.25_pReal] !< coefficients in Butcher tableau (used for final integration and error estimate)
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
real(pReal), dimension(5), parameter :: &
C = [0.2_pReal, 0.3_pReal, 0.6_pReal, 1.0_pReal, 0.875_pReal] !< coefficients in Butcher tableau (fractions of original time step in stages 2 to 6)
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
integer(pInt) :: &
e, & ! element index in element loop
i, & ! integration point index in ip loop
g, & ! grain index in grain loop
n, & ! stage index in integration stage loop
mySizeDotState, & ! size of dot State
s ! state index
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
integer(pInt), dimension(2) :: &
eIter ! bounds for element iteration
integer(pInt), dimension(2,mesh_NcpElems) :: &
iIter, & ! bounds for ip iteration
gIter ! bounds for grain iteration
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
real(pReal), dimension(constitutive_maxSizeDotState,homogenization_maxNgrains,mesh_maxNips,mesh_NcpElems) :: &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
stateResiduum, & ! residuum from evolution in micrstructure
relStateResiduum ! relative residuum from evolution in microstructure
logical :: &
singleRun ! flag indicating computation for single (g,i,e) triple
doxygen comments
2013-02-22 04:38:36 +05:30
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
eIter = FEsolving_execElem(1:2)
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt) &
write(6,'(a,1x,i1)') '<< CRYST >> RUNGE KUTTA STEP',1
! --- LOOP ITERATOR FOR ELEMENT, GRAIN, IP ---
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
do e = eIter(1),eIter(2)
iIter(1:2,e) = FEsolving_execIP(1:2,e)
gIter(1:2,e) = [ 1_pInt,homogenization_Ngrains(mesh_element(3,e))]
enddo
doxygen comments
2013-02-22 04:38:36 +05:30
fixed out of range bug introduced with singleRun changes in rev. 2356
2013-04-30 17:44:07 +05:30
singleRun = (eIter(1) == eIter(2) .and. iIter(1,eIter(1)) == iIter(2,eIter(2)))
doxygen comments
2013-02-22 04:38:36 +05:30
! --- FIRST RUNGE KUTTA STEP ---
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_collectDotState(crystallite_Tstar_v(1:6,g,i,e), crystallite_Fe, &
crystallite_Fp, crystallite_temperature(i,e), &
crystallite_subdt(g,i,e), crystallite_subFrac, g,i,e)
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
if ( any(constitutive_dotState(g,i,e)%p /= constitutive_dotState(g,i,e)%p)) then ! NaN occured in dotState
doxygen comments
2013-02-22 04:38:36 +05:30
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
endif
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#else
if ( any(plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) &
/= plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)))) then ! NaN occured in dotState
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
endif
endif
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
* Introduced preprocessor directives in order to suppress compilation of most write statements when using openmp. This tremendously improves efficiency of parallelization. * Also added some more openmp directives to increase percentage of parallelized code. * "implicit none" was missing in two subroutines of homogenization and constitutive.
2011-03-29 12:57:19 +05:30
enddo; enddo; enddo
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP ENDDO
!$OMP END PARALLEL
! --- SECOND TO SIXTH RUNGE KUTTA STEP ---
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
do n = 1_pInt,5_pInt
! --- state update ---
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL
!$OMP DO PRIVATE(mySizeDotState)
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
constitutive_RKCK45dotState(n,g,i,e)%p = constitutive_dotState(g,i,e)%p ! store Runge-Kutta dotState
more work on the new state
2014-05-27 20:16:03 +05:30
#else
mySizeDotState = plasticState(mappingConstitutive(2,g,i,e))%sizeDotState
plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(n,:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) ! store Runge-Kutta dotState
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
if (n == 1) then ! NEED TO DO THE ADDITION IN THIS LENGTHY WAY BECAUSE OF PARALLELIZATION (CAN'T USE A REDUCTION CLAUSE ON A POINTER OR USER DEFINED TYPE)
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
constitutive_dotState(g,i,e)%p = A(1,1) * constitutive_RKCK45dotState(1,g,i,e)%p
#else
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) = &
A(1,1) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(1,:,mappingConstitutive(1,g,i,e))
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
elseif (n == 2) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_dotState(g,i,e)%p = A(1,2) * constitutive_RKCK45dotState(1,g,i,e)%p &
+ A(2,2) * constitutive_RKCK45dotState(2,g,i,e)%p
more work on the new state
2014-05-27 20:16:03 +05:30
#else
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) = &
A(1,2) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(1,:,mappingConstitutive(1,g,i,e))&
+ A(2,2) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(2,:,mappingConstitutive(1,g,i,e))
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
elseif (n == 3) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_dotState(g,i,e)%p = A(1,3) * constitutive_RKCK45dotState(1,g,i,e)%p &
+ A(2,3) * constitutive_RKCK45dotState(2,g,i,e)%p &
+ A(3,3) * constitutive_RKCK45dotState(3,g,i,e)%p
more work on the new state
2014-05-27 20:16:03 +05:30
#else
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) = &
A(1,3) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(1,:,mappingConstitutive(1,g,i,e))&
+ A(2,3) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(2,:,mappingConstitutive(1,g,i,e))&
+ A(3,3) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(3,:,mappingConstitutive(1,g,i,e))
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
elseif (n == 4) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_dotState(g,i,e)%p = A(1,4) * constitutive_RKCK45dotState(1,g,i,e)%p &
+ A(2,4) * constitutive_RKCK45dotState(2,g,i,e)%p &
+ A(3,4) * constitutive_RKCK45dotState(3,g,i,e)%p &
+ A(4,4) * constitutive_RKCK45dotState(4,g,i,e)%p
more work on the new state
2014-05-27 20:16:03 +05:30
#else
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) = &
A(1,4) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(1,:,mappingConstitutive(1,g,i,e))&
+ A(2,4) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(2,:,mappingConstitutive(1,g,i,e))&
+ A(3,4) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(3,:,mappingConstitutive(1,g,i,e))&
+ A(4,4) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(4,:,mappingConstitutive(1,g,i,e))
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
elseif (n == 5) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_dotState(g,i,e)%p = A(1,5) * constitutive_RKCK45dotState(1,g,i,e)%p &
+ A(2,5) * constitutive_RKCK45dotState(2,g,i,e)%p &
+ A(3,5) * constitutive_RKCK45dotState(3,g,i,e)%p &
+ A(4,5) * constitutive_RKCK45dotState(4,g,i,e)%p &
+ A(5,5) * constitutive_RKCK45dotState(5,g,i,e)%p
more work on the new state
2014-05-27 20:16:03 +05:30
#else
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) = &
A(1,5) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(1,:,mappingConstitutive(1,g,i,e))&
+ A(2,5) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(2,:,mappingConstitutive(1,g,i,e))&
+ A(3,5) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(3,:,mappingConstitutive(1,g,i,e))&
+ A(4,5) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(4,:,mappingConstitutive(1,g,i,e))&
+ A(5,5) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(5,:,mappingConstitutive(1,g,i,e))
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP DO PRIVATE(mySizeDotState)
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
constitutive_state(g,i,e)%p(1:mySizeDotState) = constitutive_subState0(g,i,e)%p(1:mySizeDotState) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
+ constitutive_dotState(g,i,e)%p(1:mySizeDotState) &
* crystallite_subdt(g,i,e)
more work on the new state
2014-05-27 20:16:03 +05:30
#else
mySizeDotState = plasticState(mappingConstitutive(2,g,i,e))%sizeDotState
plasticState(mappingConstitutive(2,g,i,e))%state(1:mySizeDotState,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%subState0(1:mySizeDotState,mappingConstitutive(1,g,i,e))&
+ plasticState(mappingConstitutive(2,g,i,e))%dotState (1:mySizeDotState,mappingConstitutive(1,g,i,e))&
* crystallite_subdt(g,i,e)
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- state jump ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
crystallite_todo(g,i,e) = crystallite_stateJump(g,i,e)
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e) .and. .not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- update dependent states ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_microstructure(crystallite_temperature(i,e), crystallite_Fe(1:3,1:3,g,i,e), &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_Fp(1:3,1:3,g,i,e), g, i, e) ! update dependent state variables to be consistent with basic states
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- stress integration ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
crystallite_todo(g,i,e) = crystallite_integrateStress(g,i,e,c(n)) ! fraction of original time step
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e) .and. .not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- dot state and RK dot state---
* 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
#ifndef _OPENMP
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt) &
doxygen comments
2013-02-22 04:38:36 +05:30
write(6,'(a,1x,i1)') '<< CRYST >> Runge--Kutta step',n+1_pInt
* 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
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_collectDotState(crystallite_Tstar_v(1:6,g,i,e), crystallite_Fe, &
crystallite_Fp, crystallite_temperature(i,e), &
C(n)*crystallite_subdt(g,i,e), & ! fraction of original timestep
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_subFrac, g,i,e)
endif
enddo; enddo; enddo
!$OMP ENDDO
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
if ( any(constitutive_dotState(g,i,e)%p/=constitutive_dotState(g,i,e)%p)) then ! NaN occured in dotState
more work on the new state
2014-05-27 20:16:03 +05:30
#else
if ( any(plasticState(mappingConstitutive(2,g,i,e))%dotState (:,mappingConstitutive(1,g,i,e))/= &
plasticState(mappingConstitutive(2,g,i,e))%dotState (:,mappingConstitutive(1,g,i,e)))) then ! NaN occured in dotState
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
endif
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
!$OMP END PARALLEL
enddo
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
!--------------------------------------------------------------------------------------------------
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- STATE UPDATE WITH ERROR ESTIMATE FOR STATE ---
doxygen comments
2013-02-22 04:38:36 +05:30
relStateResiduum = 0.0_pReal
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL
!$OMP DO PRIVATE(mySizeDotState)
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
constitutive_RKCK45dotState(6,g,i,e)%p = constitutive_dotState(g,i,e)%p ! store Runge-Kutta dotState
more work on the new state
2014-05-27 20:16:03 +05:30
#else
mySizeDotState = plasticState(mappingConstitutive(2,g,i,e))%sizeDotState
plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(6,:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%dotState (:,mappingConstitutive(1,g,i,e)) ! store Runge-Kutta dotState
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP DO PRIVATE(mySizeDotState)
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- absolute residuum in state ---
doxygen comments
2013-02-22 04:38:36 +05:30
! NEED TO DO THE ADDITION IN THIS LENGTHY WAY BECAUSE OF PARALLELIZATION
! CAN'T USE A REDUCTION CLAUSE ON A POINTER OR USER DEFINED TYPE
stateResiduum(1:mySizeDotState,g,i,e) = &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
( DB(1) * constitutive_RKCK45dotState(1,g,i,e)%p(1:mySizeDotState) &
+ DB(2) * constitutive_RKCK45dotState(2,g,i,e)%p(1:mySizeDotState) &
+ DB(3) * constitutive_RKCK45dotState(3,g,i,e)%p(1:mySizeDotState) &
+ DB(4) * constitutive_RKCK45dotState(4,g,i,e)%p(1:mySizeDotState) &
+ DB(5) * constitutive_RKCK45dotState(5,g,i,e)%p(1:mySizeDotState) &
+ DB(6) * constitutive_RKCK45dotState(6,g,i,e)%p(1:mySizeDotState)) &
doxygen comments
2013-02-22 04:38:36 +05:30
* crystallite_subdt(g,i,e)
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- dot state ---
doxygen comments
2013-02-22 04:38:36 +05:30
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_dotState(g,i,e)%p = B(1) * constitutive_RKCK45dotState(1,g,i,e)%p &
+ B(2) * constitutive_RKCK45dotState(2,g,i,e)%p &
+ B(3) * constitutive_RKCK45dotState(3,g,i,e)%p &
+ B(4) * constitutive_RKCK45dotState(4,g,i,e)%p &
+ B(5) * constitutive_RKCK45dotState(5,g,i,e)%p &
+ B(6) * constitutive_RKCK45dotState(6,g,i,e)%p
more work on the new state
2014-05-27 20:16:03 +05:30
#else
mySizeDotState = plasticState(mappingConstitutive(2,g,i,e))%sizeDotState
! --- absolute residuum in state ---
! NEED TO DO THE ADDITION IN THIS LENGTHY WAY BECAUSE OF PARALLELIZATION
! CAN'T USE A REDUCTION CLAUSE ON A POINTER OR USER DEFINED TYPE
stateResiduum(1:mySizeDotState,g,i,e) = matmul(DB, &
plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(1:6,1:mySizeDotState,mappingConstitutive(1,g,i,e))) &
* crystallite_subdt(g,i,e)
! --- dot state ---
plasticState(mappingConstitutive(2,g,i,e))%dotState (:,mappingConstitutive(1,g,i,e)) &
= B(1) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(1,:,mappingConstitutive(1,g,i,e)) &
+ B(2) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(2,:,mappingConstitutive(1,g,i,e)) &
+ B(3) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(3,:,mappingConstitutive(1,g,i,e)) &
+ B(4) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(4,:,mappingConstitutive(1,g,i,e)) &
+ B(5) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(5,:,mappingConstitutive(1,g,i,e)) &
+ B(6) * plasticState(mappingConstitutive(2,g,i,e))%RKCK45dotState(6,:,mappingConstitutive(1,g,i,e))
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- state and update ---
doxygen comments
2013-02-22 04:38:36 +05:30
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP DO PRIVATE(mySizeDotState)
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
constitutive_state(g,i,e)%p(1:mySizeDotState) = constitutive_subState0(g,i,e)%p(1:mySizeDotState) &
more work on the new state
2014-05-27 20:16:03 +05:30
+ constitutive_dotState(g,i,e)%p(1:mySizeDotState) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
* crystallite_subdt(g,i,e)
more work on the new state
2014-05-27 20:16:03 +05:30
#else
mySizeDotState = plasticState(mappingConstitutive(2,g,i,e))%sizeDotState
plasticState(mappingConstitutive(2,g,i,e))%state(1:mySizeDotState,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%subState0(1:mySizeDotState,mappingConstitutive(1,g,i,e))&
+ plasticState(mappingConstitutive(2,g,i,e))%dotState (1:mySizeDotState,mappingConstitutive(1,g,i,e))&
* crystallite_subdt(g,i,e)
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- relative residui and state convergence ---
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP DO PRIVATE(mySizeDotState)
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
forall (s = 1_pInt:mySizeDotState, abs(constitutive_state(g,i,e)%p(s)) > 0.0_pReal) &
relStateResiduum(s,g,i,e) = stateResiduum(s,g,i,e) / constitutive_state(g,i,e)%p(s)
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
!$OMP FLUSH(relStateResiduum)
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_todo(g,i,e) = &
( all( abs(relStateResiduum(:,g,i,e)) < rTol_crystalliteState &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
.or. abs(stateResiduum(1:mySizeDotState,g,i,e)) < constitutive_aTolState(g,i,e)%p(1:mySizeDotState) ))
doxygen comments
2013-02-22 04:38:36 +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
#ifndef _OPENMP
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt&
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g)&
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
write(6,'(a,i8,1x,i3,1x,i3,/)') '<< CRYST >> updateState at el ip g ',e,i,g
write(6,'(a,/,(12x,12(f12.1,1x)),/)') '<< CRYST >> absolute residuum tolerance', &
doxygen comments
2013-02-22 04:38:36 +05:30
stateResiduum(1:mySizeDotState,g,i,e) / constitutive_aTolState(g,i,e)%p(1:mySizeDotState)
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
write(6,'(a,/,(12x,12(f12.1,1x)),/)') '<< CRYST >> relative residuum tolerance', &
doxygen comments
2013-02-22 04:38:36 +05:30
relStateResiduum(1:mySizeDotState,g,i,e) / rTol_crystalliteState
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> dotState', constitutive_dotState(g,i,e)%p(1:mySizeDotState)
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> new state', constitutive_state(g,i,e)%p(1:mySizeDotState)
doxygen comments
2013-02-22 04:38:36 +05:30
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#endif
#else
mySizeDotState = plasticState(mappingConstitutive(2,g,i,e))%sizeDotState
forall (s = 1_pInt:mySizeDotState, abs(plasticState(mappingConstitutive(2,g,i,e))% &
state(s,mappingConstitutive(1,g,i,e))) > 0.0_pReal) &
relStateResiduum(s,g,i,e) = stateResiduum(s,g,i,e) / plasticState(mappingConstitutive(2,g,i,e))%&
state(s,mappingConstitutive(1,g,i,e))
!$OMP FLUSH(relStateResiduum)
crystallite_todo(g,i,e) = &
( all( abs(relStateResiduum(1:mySizeDotState,g,i,e)) < rTol_crystalliteState &
.or. abs(stateResiduum(1:mySizeDotState,g,i,e)) < plasticState(mappingConstitutive(2,g,i,e))% &
aTolState(1:mySizeDotState) ))
* 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
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
Depending on which state integrator one uses for the stiffness calculation, the initial state has to be chosen accordingly: e.g. for FPI choose last converged state, but for explicit RK choose converged state from start of increment (in case of explicit euler no state integration at all, but only stress integration). For this purpose we also need to remember "Fe" which now follows the cutbacking procedure as it is used for "Fp".
2011-11-04 18:14:50 +05:30
enddo; enddo; enddo
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP ENDDO
! --- STATE JUMP ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
crystallite_todo(g,i,e) = crystallite_stateJump(g,i,e)
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e) .and. .not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
Depending on which state integrator one uses for the stiffness calculation, the initial state has to be chosen accordingly: e.g. for FPI choose last converged state, but for explicit RK choose converged state from start of increment (in case of explicit euler no state integration at all, but only stress integration). For this purpose we also need to remember "Fe" which now follows the cutbacking procedure as it is used for "Fp".
2011-11-04 18:14:50 +05:30
enddo; enddo; enddo
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP ENDDO
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
!--------------------------------------------------------------------------------------------------
! --- UPDATE DEPENDENT STATES IF RESIDUUM BELOW TOLERANCE ---
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_microstructure(crystallite_temperature(i,e), crystallite_Fe(1:3,1:3,g,i,e), &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_Fp(1:3,1:3,g,i,e), g, i, e) ! update dependent state variables to be consistent with basic states
endif
* now able to choose method for state integration (integrator and integratorStiffness in numerics.config); standard is "Fixed Point Iteration", which is basically the same as before; others available are "Explicit Euler", "AdaptiveEuler", "Classical Runge-Kutta" and "Runge-Kutta Cash-Karp" * now remembering stiffness similar to how we do it for Lp etc.; avoids undefined stiffness values for nonconverged stiffness calculation * non-local stuff: * changed non-local kinetics (Gilman2002) * enforce zero shearrate for overall carrrier density below relevant density * enforce zero density for those states that become negative and were below relevant density before * dislocation velocity is not limited by V^(1/3) / dt anymore
2010-10-01 17:48:49 +05:30
enddo; enddo; enddo
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP ENDDO
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
!--------------------------------------------------------------------------------------------------
! --- FINAL STRESS INTEGRATION STEP IF RESIDUUM BELOW TOLERANCE ---
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
crystallite_todo(g,i,e) = crystallite_integrateStress(g,i,e)
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e) .and. .not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
!--------------------------------------------------------------------------------------------------
! --- SET CONVERGENCE FLAG ---
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
changes to make newstate aware of damage variables
2014-06-11 22:02:09 +05:30
crystallite_converged(g,i,e) = .true. ! if still "to do" then converged per definition
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
!$OMP CRITICAL (distributionState)
debug_StateLoopDistribution(6,numerics_integrationMode) = &
debug_StateLoopDistribution(6,numerics_integrationMode) + 1_pInt
!$OMP END CRITICAL (distributionState)
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
!$OMP END PARALLEL
! --- nonlocal convergence check ---
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt) &
fixed two severe bugs in new state, now a little bit faster (399 vs. 378 sec) and more memory efficicient (251312 vs. 236832 kbytes) for Phenopowerlaw 16x16x16 example (short load only)
2014-06-17 00:49:38 +05:30
write(6,'(a,i8,a,i2,/)') '<< CRYST >> ', count(crystallite_converged(:,:,:)), ' grains converged' ! if not requesting Integration of just a single IP
more work on the new state
2014-05-27 20:16:03 +05:30
if ((.not. singleRun) .and. any(.not. crystallite_converged .and. .not. crystallite_localPlasticity)) & ! any non-local not yet converged (or broken)...
crystallite_converged = crystallite_converged .and. crystallite_localPlasticity ! ...restart all non-local as not converged
doxygen comments
2013-02-22 04:38:36 +05:30
end subroutine crystallite_integrateStateRKCK45
* now able to choose method for state integration (integrator and integratorStiffness in numerics.config); standard is "Fixed Point Iteration", which is basically the same as before; others available are "Explicit Euler", "AdaptiveEuler", "Classical Runge-Kutta" and "Runge-Kutta Cash-Karp" * now remembering stiffness similar to how we do it for Lp etc.; avoids undefined stiffness values for nonconverged stiffness calculation * non-local stuff: * changed non-local kinetics (Gilman2002) * enforce zero shearrate for overall carrrier density below relevant density * enforce zero density for those states that become negative and were below relevant density before * dislocation velocity is not limited by V^(1/3) / dt anymore
2010-10-01 17:48:49 +05:30
instantaneous jumps in the state by constitutive_deltaState are now incorporated for all state integrators. still they (should) not influence the result, since all constitutive laws simply return zero for the deltaState
2012-05-17 20:55:21 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
!> @brief integrate stress, state with 1st order Euler method with adaptive step size
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
subroutine crystallite_integrateStateAdaptiveEuler()
more work on the new state
2014-05-27 20:16:03 +05:30
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
use debug, only: &
debug_level, &
debug_crystallite, &
debug_levelBasic, &
debug_levelExtensive, &
debug_levelSelective, &
debug_e, &
debug_i, &
debug_g, &
debug_StateLoopDistribution
use numerics, only: &
rTol_crystalliteState, &
numerics_integrationMode
use FEsolving, only: &
FEsolving_execElem, &
FEsolving_execIP
use mesh, only: &
mesh_element, &
mesh_NcpElems, &
mesh_maxNips
use material, only: &
homogenization_Ngrains, &
more work on the new state
2014-05-27 20:16:03 +05:30
#ifdef NEWSTATE
plasticState, &
#endif
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
homogenization_maxNgrains
use constitutive, only: &
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_collectDotState, &
constitutive_microstructure, &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_maxSizeDotState, &
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_subState0, &
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_state, &
constitutive_sizeDotState, &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_dotState, &
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_aTolState
#else
mappingConstitutive
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
implicit none
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
integer(pInt) :: &
e, & ! element index in element loop
i, & ! integration point index in ip loop
g, & ! grain index in grain loop
mySizeDotState, & ! size of dot State
s ! state index
integer(pInt), dimension(2) :: &
eIter ! bounds for element iteration
integer(pInt), dimension(2,mesh_NcpElems) :: &
iIter, & ! bounds for ip iteration
gIter ! bounds for grain iteration
doxygen comments
2013-02-22 04:38:36 +05:30
real(pReal), dimension(constitutive_maxSizeDotState,homogenization_maxNgrains,mesh_maxNips,mesh_NcpElems) :: &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
stateResiduum, & ! residuum from evolution in micrstructure
relStateResiduum ! relative residuum from evolution in microstructure
more work on the new state
2014-05-27 20:16:03 +05:30
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
logical :: &
singleRun ! flag indicating computation for single (g,i,e) triple
doxygen comments
2013-02-22 04:38:36 +05:30
! --- LOOP ITERATOR FOR ELEMENT, GRAIN, IP ---
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
eIter = FEsolving_execElem(1:2)
do e = eIter(1),eIter(2)
iIter(1:2,e) = FEsolving_execIP(1:2,e)
gIter(1:2,e) = [ 1_pInt,homogenization_Ngrains(mesh_element(3,e))]
enddo
doxygen comments
2013-02-22 04:38:36 +05:30
fixed out of range bug introduced with singleRun changes in rev. 2356
2013-04-30 17:44:07 +05:30
singleRun = (eIter(1) == eIter(2) .and. iIter(1,eIter(1)) == iIter(2,eIter(2)))
doxygen comments
2013-02-22 04:38:36 +05:30
more work on the new state
2014-05-27 20:16:03 +05:30
stateResiduum = 0.0_pReal
relStateResiduum = 0.0_pReal
doxygen comments
2013-02-22 04:38:36 +05:30
more work on the new state
2014-05-27 20:16:03 +05:30
integrationMode: if (numerics_integrationMode == 1_pInt) then
!$OMP PARALLEL
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- DOT STATE (EULER INTEGRATION) ---
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_collectDotState(crystallite_Tstar_v(1:6,g,i,e), crystallite_Fe, &
crystallite_Fp, crystallite_temperature(i,e), &
crystallite_subdt(g,i,e), crystallite_subFrac, g,i,e)
doxygen comments
2013-02-22 04:38:36 +05:30
endif
instantaneous jumps in the state by constitutive_deltaState are now incorporated for all state integrators. still they (should) not influence the result, since all constitutive laws simply return zero for the deltaState
2012-05-17 20:55:21 +05:30
enddo; enddo; enddo
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP ENDDO
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
more work on the new state
2014-05-27 20:16:03 +05:30
if (crystallite_todo(g,i,e)) then
#ifndef NEWSTATE
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
if ( any(constitutive_dotState(g,i,e)%p /= constitutive_dotState(g,i,e)%p)) then ! NaN occured in dotState
doxygen comments
2013-02-22 04:38:36 +05:30
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
endif
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#else
if ( any(plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) &
/= plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e))))then ! NaN occured in dotState
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
endif
endif
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
more work on the new state
2014-05-27 20:16:03 +05:30
enddo; enddo; enddo
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP ENDDO
! --- STATE UPDATE (EULER INTEGRATION) ---
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP DO PRIVATE(mySizeDotState)
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
stateResiduum(1:mySizeDotState,g,i,e) = - 0.5_pReal * constitutive_dotState(g,i,e)%p &
* crystallite_subdt(g,i,e) ! contribution to absolute residuum in state
doxygen comments
2013-02-22 04:38:36 +05:30
constitutive_state(g,i,e)%p(1:mySizeDotState) = constitutive_state(g,i,e)%p(1:mySizeDotState) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
+ constitutive_dotState(g,i,e)%p(1:mySizeDotState) &
* crystallite_subdt(g,i,e)
more work on the new state
2014-05-27 20:16:03 +05:30
#else
mySizeDotState = plasticState(mappingConstitutive(2,g,i,e))%sizeDotState
stateResiduum(1:mySizeDotState,g,i,e) = - 0.5_pReal * plasticState(mappingConstitutive(2,g,i,e))% &
dotstate(1:mySizeDotState,mappingConstitutive(1,g,i,e)) * crystallite_subdt(g,i,e) ! contribution to absolute residuum in state
applied new State for nonlocal model and a trivial changes in crystallite
2014-06-14 02:23:17 +05:30
plasticState(mappingConstitutive(2,g,i,e))%state(1:mySizeDotState,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%state(1:mySizeDotState,mappingConstitutive(1,g,i,e)) &
+ plasticState(mappingConstitutive(2,g,i,e))%dotstate(1:mySizeDotState,mappingConstitutive(1,g,i,e)) &
more work on the new state
2014-05-27 20:16:03 +05:30
* crystallite_subdt(g,i,e)
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- STATE JUMP ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
crystallite_todo(g,i,e) = crystallite_stateJump(g,i,e)
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e) .and. .not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- UPDATE DEPENDENT STATES (EULER INTEGRATION) ---
!$OMP DO
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
if (crystallite_todo(g,i,e)) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_microstructure(crystallite_temperature(i,e), crystallite_Fe(1:3,1:3,g,i,e), &
crystallite_Fp(1:3,1:3,g,i,e), g, i, e) ! update dependent state variables to be consistent with basic states
doxygen comments
2013-02-22 04:38:36 +05:30
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP END PARALLEL
endif integrationMode
doxygen comments
2013-02-22 04:38:36 +05:30
! --- STRESS INTEGRATION (EULER INTEGRATION) ---
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL DO
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
crystallite_todo(g,i,e) = crystallite_integrateStress(g,i,e)
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e) .and. .not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
enddo; enddo; enddo
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP END PARALLEL DO
doxygen comments
2013-02-22 04:38:36 +05:30
if (numerics_integrationMode == 1_pInt) then
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- DOT STATE (HEUN METHOD) ---
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
if (crystallite_todo(g,i,e)) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_collectDotState(crystallite_Tstar_v(1:6,g,i,e), crystallite_Fe, &
crystallite_Fp, crystallite_temperature(i,e), &
crystallite_subdt(g,i,e), crystallite_subFrac, g,i,e)
doxygen comments
2013-02-22 04:38:36 +05:30
enddo; enddo; enddo
!$OMP ENDDO
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
if ( any(constitutive_dotState(g,i,e)%p /= constitutive_dotState(g,i,e)%p) ) then ! NaN occured in dotState
doxygen comments
2013-02-22 04:38:36 +05:30
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
endif
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#else
if ( any(plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) &
/= plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e))))then ! NaN occured in dotState
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
endif
endif
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- ERROR ESTIMATE FOR STATE (HEUN METHOD) ---
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP SINGLE
relStateResiduum = 0.0_pReal
!$OMP END SINGLE
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP DO PRIVATE(mySizeDotState)
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
doxygen comments
2013-02-22 04:38:36 +05:30
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
! --- contribution of heun step to absolute residui ---
stateResiduum(1:mySizeDotState,g,i,e) = stateResiduum(1:mySizeDotState,g,i,e) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
+ 0.5_pReal * constitutive_dotState(g,i,e)%p &
* crystallite_subdt(g,i,e) ! contribution to absolute residuum in state
more work on the new state
2014-05-27 20:16:03 +05:30
#else
mySizeDotState = plasticState(mappingConstitutive(2,g,i,e))%sizeDotState
! --- contribution of heun step to absolute residui ---
stateResiduum(1:mySizeDotState,g,i,e) = stateResiduum(1:mySizeDotState,g,i,e) &
+ 0.5_pReal * plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) &
* crystallite_subdt(g,i,e) ! contribution to absolute residuum in state
#endif
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
!$OMP FLUSH(stateResiduum)
doxygen comments
2013-02-22 04:38:36 +05:30
more work on the new state
2014-05-27 20:16:03 +05:30
! --- relative residui ---
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
forall (s = 1_pInt:mySizeDotState, abs(constitutive_state(g,i,e)%p(s)) > 0.0_pReal) &
more work on the new state
2014-05-27 20:16:03 +05:30
relStateResiduum(s,g,i,e) = stateResiduum(s,g,i,e) / constitutive_state(g,i,e)%p(s)
#else
forall (s = 1_pInt:mySizeDotState, abs(plasticState(mappingConstitutive(2,g,i,e))% &
dotState(s,mappingConstitutive(1,g,i,e))) > 0.0_pReal) &
relStateResiduum(s,g,i,e) = stateResiduum(s,g,i,e) / &
plasticState(mappingConstitutive(2,g,i,e))%dotState(s,mappingConstitutive(1,g,i,e))
#endif
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
!$OMP FLUSH(relStateResiduum)
doxygen comments
2013-02-22 04:38:36 +05:30
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef _OPENMP
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt &
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g)&
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
write(6,'(a,i8,1x,i2,1x,i3)') '<< CRYST >> updateState at el ip g ',e,i,g
write(6,*)
write(6,'(a,/,(12x,12(f12.1,1x)))') '<< CRYST >> absolute residuum tolerance', &
stateResiduum(1:mySizeDotState,g,i,e) / constitutive_aTolState(g,i,e)%p(1:mySizeDotState)
write(6,*)
write(6,'(a,/,(12x,12(f12.1,1x)))') '<< CRYST >> relative residuum tolerance', &
relStateResiduum(1:mySizeDotState,g,i,e) / rTol_crystalliteState
write(6,*)
write(6,'(a,/,(12x,12(e12.5,1x)))') '<< CRYST >> dotState', constitutive_dotState(g,i,e)%p(1:mySizeDotState) &
one line was beyond 132 characters
2013-02-22 11:08:02 +05:30
- 2.0_pReal * stateResiduum(1:mySizeDotState,g,i,e) / crystallite_subdt(g,i,e) ! calculate former dotstate from higher order solution and state residuum
doxygen comments
2013-02-22 04:38:36 +05:30
write(6,*)
write(6,'(a,/,(12x,12(e12.5,1x)))') '<< CRYST >> new state', constitutive_state(g,i,e)%p(1:mySizeDotState)
write(6,*)
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#else
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt &
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g)&
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
write(6,'(a,i8,1x,i2,1x,i3,/)') '<< CRYST >> updateState at el ip g ',e,i,g
write(6,'(a,/,(12x,12(f12.1,1x)),/)') '<< CRYST >> absolute residuum tolerance', &
stateResiduum(1:mySizeDotState,g,i,e) / &
plasticState(mappingConstitutive(2,g,i,e))%aTolState(1:mySizeDotState)
write(6,'(a,/,(12x,12(f12.1,1x)),/)') '<< CRYST >> relative residuum tolerance', &
relStateResiduum(1:mySizeDotState,g,i,e) / rTol_crystalliteState
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> dotState', plasticState(mappingConstitutive(2,g,i,e))%&
dotState(1:mySizeDotState,mappingConstitutive(1,g,i,e)) &
- 2.0_pReal * stateResiduum(1:mySizeDotState,g,i,e) / crystallite_subdt(g,i,e) ! calculate former dotstate from higher order solution and state residuum
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> new state', &
plasticState(mappingConstitutive(2,g,i,e))%State(1:mySizeDotState,mappingConstitutive(1,g,i,e))
endif
#endif
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
! --- converged ? ---
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
if ( all( abs(relStateResiduum(1:mySizeDotState,g,i,e)) < rTol_crystalliteState &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
.or. abs(stateResiduum(1:mySizeDotState,g,i,e)) < constitutive_aTolState(g,i,e)%p(1:mySizeDotState))) then
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_converged(g,i,e) = .true. ! ... converged per definitionem
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
!$OMP CRITICAL (distributionState)
debug_StateLoopDistribution(2,numerics_integrationMode) = &
debug_StateLoopDistribution(2,numerics_integrationMode) + 1_pInt
!$OMP END CRITICAL (distributionState)
endif
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#else
if ( all( abs(relStateResiduum(1:mySizeDotState,g,i,e)) < rTol_crystalliteState &
.or. abs(stateResiduum(1:mySizeDotState,g,i,e)) < &
plasticState(mappingConstitutive(2,g,i,e))%aTolState(1:mySizeDotState))) then
crystallite_converged(g,i,e) = .true. ! ... converged per definitionem
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
!$OMP CRITICAL (distributionState)
debug_StateLoopDistribution(2,numerics_integrationMode) = &
debug_StateLoopDistribution(2,numerics_integrationMode) + 1_pInt
!$OMP END CRITICAL (distributionState)
endif
endif
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP END PARALLEL
doxygen comments
2013-02-22 04:38:36 +05:30
elseif (numerics_integrationMode > 1) then ! stiffness calculation
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL DO
doxygen comments
2013-02-22 04:38:36 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
crystallite_converged(g,i,e) = .true. ! ... converged per definitionem
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
!$OMP CRITICAL (distributionState)
debug_StateLoopDistribution(2,numerics_integrationMode) = &
debug_StateLoopDistribution(2,numerics_integrationMode) + 1_pInt
!$OMP END CRITICAL (distributionState)
endif
endif
enddo; enddo; enddo
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP END PARALLEL DO
doxygen comments
2013-02-22 04:38:36 +05:30
endif
! --- NONLOCAL CONVERGENCE CHECK ---
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt) &
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i8,a,i2,/)') '<< CRYST >> ', count(crystallite_converged(:,:,:)), ' grains converged'
more work on the new state
2014-05-27 20:16:03 +05:30
if ((.not. singleRun) .and. any(.not. crystallite_converged .and. .not. crystallite_localPlasticity)) & ! any non-local not yet converged (or broken)...
crystallite_converged = crystallite_converged .and. crystallite_localPlasticity ! ...restart all non-local as not converged
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
end subroutine crystallite_integrateStateAdaptiveEuler
* now able to choose method for state integration (integrator and integratorStiffness in numerics.config); standard is "Fixed Point Iteration", which is basically the same as before; others available are "Explicit Euler", "AdaptiveEuler", "Classical Runge-Kutta" and "Runge-Kutta Cash-Karp" * now remembering stiffness similar to how we do it for Lp etc.; avoids undefined stiffness values for nonconverged stiffness calculation * non-local stuff: * changed non-local kinetics (Gilman2002) * enforce zero shearrate for overall carrrier density below relevant density * enforce zero density for those states that become negative and were below relevant density before * dislocation velocity is not limited by V^(1/3) / dt anymore
2010-10-01 17:48:49 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
!> @brief integrate stress, and state with 1st order explicit Euler method
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
subroutine crystallite_integrateStateEuler()
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
use debug, only: &
debug_level, &
debug_crystallite, &
debug_levelBasic, &
debug_levelExtensive, &
debug_levelSelective, &
debug_e, &
debug_i, &
debug_g, &
debug_StateLoopDistribution
use numerics, only: &
numerics_integrationMode, &
numerics_timeSyncing
use FEsolving, only: &
FEsolving_execElem, &
FEsolving_execIP
use mesh, only: &
mesh_element, &
mesh_NcpElems
use material, only: &
more work on the new state
2014-05-27 20:16:03 +05:30
#ifdef NEWSTATE
plasticState, &
#endif
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
homogenization_Ngrains
use constitutive, only: &
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_collectDotState, &
constitutive_microstructure, &
#ifndef NEWSTATE
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_subState0, &
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_state, &
constitutive_sizeDotState, &
constitutive_maxSizeDotState, &
fixed two severe bugs in new state, now a little bit faster (399 vs. 378 sec) and more memory efficicient (251312 vs. 236832 kbytes) for Phenopowerlaw 16x16x16 example (short load only)
2014-06-17 00:49:38 +05:30
constitutive_dotState
more work on the new state
2014-05-27 20:16:03 +05:30
#else
mappingConstitutive
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
implicit none
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
integer(pInt) :: &
e, & ! element index in element loop
i, & ! integration point index in ip loop
g, & ! grain index in grain loop
more work on the new state
2014-05-27 20:16:03 +05:30
mySizeDotState, & ! size of dot State
p
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
integer(pInt), dimension(2) :: &
eIter ! bounds for element iteration
integer(pInt), dimension(2,mesh_NcpElems) :: &
iIter, & ! bounds for ip iteration
gIter ! bounds for grain iteration
logical :: &
singleRun ! flag indicating computation for single (g,i,e) triple
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
eIter = FEsolving_execElem(1:2)
do e = eIter(1),eIter(2)
iIter(1:2,e) = FEsolving_execIP(1:2,e)
gIter(1:2,e) = [ 1_pInt,homogenization_Ngrains(mesh_element(3,e))]
enddo
doxygen comments
2013-02-22 04:38:36 +05:30
fixed out of range bug introduced with singleRun changes in rev. 2356
2013-04-30 17:44:07 +05:30
singleRun = (eIter(1) == eIter(2) .and. iIter(1,eIter(1)) == iIter(2,eIter(2)))
fixed two severe bugs in new state, now a little bit faster (399 vs. 378 sec) and more memory efficicient (251312 vs. 236832 kbytes) for Phenopowerlaw 16x16x16 example (short load only)
2014-06-17 00:49:38 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
if (numerics_integrationMode == 1_pInt) then
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL
doxygen comments
2013-02-22 04:38:36 +05:30
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- DOT STATE ---
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_collectDotState(crystallite_Tstar_v(1:6,g,i,e), crystallite_Fe, &
crystallite_Fp, crystallite_temperature(i,e), &
crystallite_subdt(g,i,e), crystallite_subFrac, g,i,e)
doxygen comments
2013-02-22 04:38:36 +05:30
enddo; enddo; enddo
!$OMP ENDDO
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
if ( any(constitutive_dotState(g,i,e)%p/=constitutive_dotState(g,i,e)%p) ) then ! NaN occured in dotState
if (.not. crystallite_localPlasticity(g,i,e) .and. .not. numerics_timeSyncing) then ! if broken non-local...
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
endif
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#else
if ( any(plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) &
/= plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e))))then ! NaN occured in dotState
if (.not. crystallite_localPlasticity(g,i,e) .and. .not. numerics_timeSyncing) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
else ! if broken local...
crystallite_todo(g,i,e) = .false. ! ... skip this one next time
endif
endif
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- UPDATE STATE ---
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO PRIVATE(mySizeDotState)
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
constitutive_state(g,i,e)%p(1:mySizeDotState) = constitutive_state(g,i,e)%p(1:mySizeDotState) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
+ constitutive_dotState(g,i,e)%p(1:mySizeDotState) &
more work on the new state
2014-05-27 20:16:03 +05:30
* crystallite_subdt(g,i,e)
#else
mySizeDotState = size(plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)))
plasticState(mappingConstitutive(2,g,i,e))%State(1:mySizeDotState,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%subState0(1:mySizeDotState,mappingConstitutive(1,g,i,e)) &
+ plasticState(mappingConstitutive(2,g,i,e))%dotState (1:mySizeDotState,mappingConstitutive(1,g,i,e)) &
* crystallite_subdt(g,i,e)
deleted space in front of preprocessor line
2012-11-23 01:34:33 +05:30
#endif
more work on the new state
2014-05-27 20:16:03 +05:30
!#ifndef _OPENMP
! if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt &
! .and. ((e == debug_e .and. i == debug_i .and. g == debug_g) &
! .or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
! write(6,'(a,i8,1x,i2,1x,i3,/)') '<< CRYST >> update state at el ip g ',e,i,g
! write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> dotState', constitutive_dotState(g,i,e)%p(1:mySizeDotState)
! write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> new state', constitutive_state(g,i,e)%p(1:mySizeDotState)
! endif
!#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- STATE JUMP ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) then
crystallite_todo(g,i,e) = crystallite_stateJump(g,i,e)
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e) .and. .not. crystallite_localPlasticity(g,i,e) & ! if broken non-local...
.and. .not. numerics_timeSyncing) then
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
! --- UPDATE DEPENDENT STATES ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_microstructure(crystallite_temperature(i,e), crystallite_Fe(1:3,1:3,g,i,e), &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_Fp(1:3,1:3,g,i,e), g, i, e) ! update dependent state variables to be consistent with basic states
order of integration now same for all state integrators: first calculate the dotState and then the deltaState
2012-11-22 18:34:19 +05:30
enddo; enddo; enddo
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP END PARALLEL
doxygen comments
2013-02-22 04:38:36 +05:30
endif
more work on the new state
2014-05-27 20:16:03 +05:30
!$OMP PARALLEL
doxygen comments
2013-02-22 04:38:36 +05:30
! --- STRESS INTEGRATION ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) then
crystallite_todo(g,i,e) = crystallite_integrateStress(g,i,e)
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e) .and. .not. crystallite_localPlasticity(g,i,e) & ! if broken non-local...
.and. .not. numerics_timeSyncing) then
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
Depending on which state integrator one uses for the stiffness calculation, the initial state has to be chosen accordingly: e.g. for FPI choose last converged state, but for explicit RK choose converged state from start of increment (in case of explicit euler no state integration at all, but only stress integration). For this purpose we also need to remember "Fe" which now follows the cutbacking procedure as it is used for "Fp".
2011-11-04 18:14:50 +05:30
enddo; enddo; enddo
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP ENDDO
! --- SET CONVERGENCE FLAG ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) then
crystallite_converged(g,i,e) = .true. ! if still "to do" then converged per definitionem
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
!$OMP CRITICAL (distributionState)
debug_StateLoopDistribution(1,numerics_integrationMode) = &
debug_StateLoopDistribution(1,numerics_integrationMode) + 1_pInt
!$OMP END CRITICAL (distributionState)
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
!$OMP END PARALLEL
! --- CHECK NON-LOCAL CONVERGENCE ---
if (.not. singleRun) then ! if not requesting Integration of just a single IP
if (any(.not. crystallite_converged .and. .not. crystallite_localPlasticity) & ! any non-local not yet converged (or broken)...
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
.and. .not. numerics_timeSyncing) &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_converged = crystallite_converged .and. crystallite_localPlasticity ! ...restart all non-local as not converged
endif
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
end subroutine crystallite_integrateStateEuler
* now able to choose method for state integration (integrator and integratorStiffness in numerics.config); standard is "Fixed Point Iteration", which is basically the same as before; others available are "Explicit Euler", "AdaptiveEuler", "Classical Runge-Kutta" and "Runge-Kutta Cash-Karp" * now remembering stiffness similar to how we do it for Lp etc.; avoids undefined stiffness values for nonconverged stiffness calculation * non-local stuff: * changed non-local kinetics (Gilman2002) * enforce zero shearrate for overall carrrier density below relevant density * enforce zero density for those states that become negative and were below relevant density before * dislocation velocity is not limited by V^(1/3) / dt anymore
2010-10-01 17:48:49 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
!> @brief integrate stress, state with adaptive 1st order explicit Euler method
doxygen comments
2013-02-22 04:38:36 +05:30
!> using Fixed Point Iteration to adapt the stepsize
!--------------------------------------------------------------------------------------------------
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
subroutine crystallite_integrateStateFPI()
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
use debug, only: &
debug_e, &
debug_i, &
debug_g, &
debug_level,&
debug_crystallite, &
debug_levelBasic, &
debug_levelExtensive, &
debug_levelSelective, &
debug_StateLoopDistribution
use numerics, only: &
nState, &
numerics_integrationMode, &
rTol_crystalliteState
use FEsolving, only: &
FEsolving_execElem, &
FEsolving_execIP
use mesh, only: &
mesh_element, &
mesh_NcpElems
use material, only: &
more work on the new state
2014-05-27 20:16:03 +05:30
#ifdef NEWSTATE
plasticState, &
#endif
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
homogenization_Ngrains
use constitutive, only: &
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_collectDotState, &
constitutive_microstructure, &
constitutive_maxSizeDotState, &
#ifndef NEWSTATE
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_subState0, &
constitutive_state, &
constitutive_sizeDotState, &
constitutive_dotState, &
constitutive_previousDotState, &
constitutive_previousDotState2, &
constitutive_aTolState
more work on the new state
2014-05-27 20:16:03 +05:30
#else
mappingConstitutive
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
implicit none
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
integer(pInt) :: &
NiterationState, & !< number of iterations in state loop
e, & !< element index in element loop
i, & !< integration point index in ip loop
g, & !< grain index in grain loop
more work on the new state
2014-05-27 20:16:03 +05:30
mySizeDotState, &
p ,&
ss
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
integer(pInt), dimension(2) :: &
eIter ! bounds for element iteration
integer(pInt), dimension(2,mesh_NcpElems) :: &
iIter, & ! bounds for ip iteration
gIter ! bounds for grain iteration
real(pReal) :: &
dot_prod12, &
dot_prod22, &
stateDamper ! damper for integration of state
more work on the new state
2014-05-27 20:16:03 +05:30
real(pReal), dimension(constitutive_maxSizeDotState) :: &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
stateResiduum, &
tempState
logical :: &
crystallite_integrateStateFPI: convergence check always ran about all e,i,g; now checking only those specified in FEsolving_execElem, FEsolving_execIP
2014-01-22 00:15:41 +05:30
singleRun, & ! flag indicating computation for single (g,i,e) triple
doneWithIntegration
doxygen comments
2013-02-22 04:38:36 +05:30
simplified logic for single run, now not checking for extra arguments but simply checking if start and end elem/IP are the same
2013-04-26 18:53:36 +05:30
eIter = FEsolving_execElem(1:2)
do e = eIter(1),eIter(2)
iIter(1:2,e) = FEsolving_execIP(1:2,e)
gIter(1:2,e) = [ 1_pInt,homogenization_Ngrains(mesh_element(3,e))]
enddo
fixed out of range bug introduced with singleRun changes in rev. 2356
2013-04-30 17:44:07 +05:30
singleRun = (eIter(1) == eIter(2) .and. iIter(1,eIter(1)) == iIter(2,eIter(2)))
more work on the new state
2014-05-27 20:16:03 +05:30
#ifdef NEWSTATE
!--------------------------------------------------------------------------------------------------
! initialize dotState
if (.not. singleRun) then
forall(p = 1_pInt:size(plasticState))
plasticState(p)%previousDotState = 0.0_pReal
plasticState(p)%previousDotState2 = 0.0_pReal
end forall
else
e = eIter(1)
i = iIter(1,e)
do g = iIter(1,e), iIter(2,e)
plasticState(mappingConstitutive(2,g,i,e))%previousDotState(:,mappingConstitutive(1,g,i,e)) = 0.0_pReal
plasticState(mappingConstitutive(2,g,i,e))%previousDotState2(:,mappingConstitutive(1,g,i,e)) = 0.0_pReal
enddo
endif
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
! --+>> PREGUESS FOR STATE <<+--
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
! --- DOT STATES ---
doxygen comments
2013-02-22 04:38:36 +05:30
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
!$OMP PARALLEL
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO
more work on the new state
2014-05-27 20:16:03 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
constitutive_previousDotState(g,i,e)%p = 0.0_pReal
constitutive_previousDotState2(g,i,e)%p = 0.0_pReal
more work on the new state
2014-05-27 20:16:03 +05:30
#endif
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
if (crystallite_todo(g,i,e)) then
call constitutive_collectDotState(crystallite_Tstar_v(1:6,g,i,e), crystallite_Fe, &
crystallite_Fp, crystallite_temperature(i,e), &
crystallite_subdt(g,i,e), crystallite_subFrac, g,i,e)
endif
doxygen comments
2013-02-22 04:38:36 +05:30
enddo; enddo; enddo
!$OMP ENDDO
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
if ( any(constitutive_dotState(g,i,e)%p/=constitutive_dotState(g,i,e)%p) ) then ! NaN occured in dotState
doxygen comments
2013-02-22 04:38:36 +05:30
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken is a non-local...
!$OMP CRITICAL (checkTodo)
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals done (and broken)
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP END CRITICAL (checkTodo)
else ! broken one was local...
crystallite_todo(g,i,e) = .false. ! ... done (and broken)
endif
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#else
if ( any(plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e))/= &
new state related changes
2014-06-11 17:57:41 +05:30
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)))) then ! NaN occured in dotState
more work on the new state
2014-05-27 20:16:03 +05:30
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken is a non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals done (and broken)
!$OMP END CRITICAL (checkTodo)
else ! broken one was local...
crystallite_todo(g,i,e) = .false. ! ... done (and broken)
endif
endif
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- UPDATE STATE ---
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO PRIVATE(mySizeDotState)
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
constitutive_state(g,i,e)%p(1:mySizeDotState) = constitutive_subState0(g,i,e)%p(1:mySizeDotState) &
found bug (array index out of bounds), Philip/Franz please check if correct (Gfortran was complaining)
2013-12-20 14:03:46 +05:30
+ constitutive_dotState(g,i,e)%p(1:mySizeDotState) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
* crystallite_subdt(g,i,e)
more work on the new state
2014-05-27 20:16:03 +05:30
#else
new state related changes
2014-06-11 17:57:41 +05:30
mySizeDotState = plasticState(mappingConstitutive(2,g,i,e))%sizeDotState
plasticState(mappingConstitutive(2,g,i,e))%state(1:mySizeDotState,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%subState0(1:mySizeDotState,mappingConstitutive(1,g,i,e)) &
+ plasticState(mappingConstitutive(2,g,i,e))%dotState(1:mySizeDotState,mappingConstitutive(1,g,i,e)) &
more work on the new state
2014-05-27 20:16:03 +05:30
* crystallite_subdt(g,i,e)
fixed two severe bugs in new state, now a little bit faster (399 vs. 378 sec) and more memory efficicient (251312 vs. 236832 kbytes) for Phenopowerlaw 16x16x16 example (short load only)
2014-06-17 00:49:38 +05:30
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP END PARALLEL
! --+>> STATE LOOP <<+--
NiterationState = 0_pInt
crystallite_integrateStateFPI: convergence check always ran about all e,i,g; now checking only those specified in FEsolving_execElem, FEsolving_execIP
2014-01-22 00:15:41 +05:30
doneWithIntegration = .false.
crystalliteLooping: do while (.not. doneWithIntegration .and. NiterationState < nState)
doxygen comments
2013-02-22 04:38:36 +05:30
NiterationState = NiterationState + 1_pInt
!$OMP PARALLEL
! --- UPDATE DEPENDENT STATES ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_microstructure(crystallite_temperature(i,e), crystallite_Fe(1:3,1:3,g,i,e), &
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_Fp(1:3,1:3,g,i,e), g, i, e) ! update dependent state variables to be consistent with basic states
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
constitutive_previousDotState2(g,i,e)%p = constitutive_previousDotState(g,i,e)%p ! remember previous dotState
constitutive_previousDotState(g,i,e)%p = constitutive_dotState(g,i,e)%p ! remember current dotState
more work on the new state
2014-05-27 20:16:03 +05:30
#else
plasticState(mappingConstitutive(2,g,i,e))%previousDotState2(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%previousDotState(:,mappingConstitutive(1,g,i,e))
plasticState(mappingConstitutive(2,g,i,e))%previousDotState(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e))
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
! --- STRESS INTEGRATION ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) then
crystallite_todo(g,i,e) = crystallite_integrateStress(g,i,e)
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e) .and. .not. crystallite_localPlasticity(g,i,e)) then ! broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ... then all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP SINGLE
!$OMP CRITICAL (write2out)
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt) &
doxygen comments
2013-02-22 04:38:36 +05:30
write(6,'(a,i8,a)') '<< CRYST >> ', count(crystallite_todo(:,:,:)),' grains todo after stress integration'
!$OMP END CRITICAL (write2out)
!$OMP END SINGLE
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- DOT STATE ---
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_collectDotState(crystallite_Tstar_v(1:6,g,i,e), crystallite_Fe, &
crystallite_Fp, crystallite_temperature(i,e), &
crystallite_subdt(g,i,e), crystallite_subFrac, g,i,e)
doxygen comments
2013-02-22 04:38:36 +05:30
enddo; enddo; enddo
!$OMP ENDDO
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
if ( any(constitutive_dotState(g,i,e)%p/=constitutive_dotState(g,i,e)%p) ) then ! NaN occured in dotState
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_todo(g,i,e) = .false. ! ... skip me next time
if (.not. crystallite_localPlasticity(g,i,e)) then ! if me is non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#else
if ( any(plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e))/= &
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e))) ) then ! NaN occured in dotState
crystallite_todo(g,i,e) = .false. ! ... skip me next time
if (.not. crystallite_localPlasticity(g,i,e)) then ! if me is non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
endif
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- UPDATE STATE ---
doxygen comments
2013-02-22 04:38:36 +05:30
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
!$OMP DO PRIVATE(dot_prod12,dot_prod22,statedamper,mySizeDotState,stateResiduum,tempState)
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
doxygen comments
2013-02-22 04:38:36 +05:30
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
! --- state damper ---
dot_prod12 = dot_product( constitutive_dotState(g,i,e)%p - constitutive_previousDotState(g,i,e)%p, &
constitutive_previousDotState(g,i,e)%p - constitutive_previousDotState2(g,i,e)%p )
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
dot_prod22 = dot_product( constitutive_previousDotState(g,i,e)%p - constitutive_previousDotState2(g,i,e)%p, &
constitutive_previousDotState(g,i,e)%p - constitutive_previousDotState2(g,i,e)%p )
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
if ( dot_prod22 > 0.0_pReal &
.and. ( dot_prod12 < 0.0_pReal &
.or. dot_product(constitutive_dotState(g,i,e)%p, constitutive_previousDotState(g,i,e)%p) < 0.0_pReal) ) then
statedamper = 0.75_pReal + 0.25_pReal * tanh(2.0_pReal + 4.0_pReal * dot_prod12 / dot_prod22)
else
statedamper = 1.0_pReal
endif
! --- get residui ---
mySizeDotState = constitutive_sizeDotState(g,i,e)
stateResiduum(1:mySizeDotState) = constitutive_state(g,i,e)%p(1:mySizeDotState) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
- constitutive_subState0(g,i,e)%p(1:mySizeDotState) &
found bug (array index out of bounds), Philip/Franz please check if correct (Gfortran was complaining)
2013-12-20 14:03:46 +05:30
- (constitutive_dotState(g,i,e)%p(1:mySizeDotState) * statedamper &
+ constitutive_previousDotState(g,i,e)%p(1:mySizeDotState) &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
* (1.0_pReal - statedamper)) * crystallite_subdt(g,i,e)
doxygen comments
2013-02-22 04:38:36 +05:30
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
! --- correct state with residuum ---
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
tempState(1:mySizeDotState) = constitutive_state(g,i,e)%p(1:mySizeDotState) &
- stateResiduum(1:mySizeDotState) ! need to copy to local variable, since we cant flush a pointer in openmp
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
more work on the new state
2014-05-27 20:16:03 +05:30
#else
dot_prod12 = dot_product( plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) &
- plasticState(mappingConstitutive(2,g,i,e))%previousDotState(:,mappingConstitutive(1,g,i,e)), &
fixed two severe bugs in new state, now a little bit faster (399 vs. 378 sec) and more memory efficicient (251312 vs. 236832 kbytes) for Phenopowerlaw 16x16x16 example (short load only)
2014-06-17 00:49:38 +05:30
plasticState(mappingConstitutive(2,g,i,e))%previousDotState(:,mappingConstitutive(1,g,i,e)) &
more work on the new state
2014-05-27 20:16:03 +05:30
- plasticState(mappingConstitutive(2,g,i,e))%previousDotState2(:,mappingConstitutive(1,g,i,e)))
fixed two severe bugs in new state, now a little bit faster (399 vs. 378 sec) and more memory efficicient (251312 vs. 236832 kbytes) for Phenopowerlaw 16x16x16 example (short load only)
2014-06-17 00:49:38 +05:30
dot_prod22 = dot_product( &
plasticState(mappingConstitutive(2,g,i,e))%previousDotState(:,mappingConstitutive(1,g,i,e)) &
more work on the new state
2014-05-27 20:16:03 +05:30
- plasticState(mappingConstitutive(2,g,i,e))%previousDotState2(:,mappingConstitutive(1,g,i,e)), &
plasticState(mappingConstitutive(2,g,i,e))%previousDotState(:,mappingConstitutive(1,g,i,e)) &
- plasticState(mappingConstitutive(2,g,i,e))%previousDotState2(:,mappingConstitutive(1,g,i,e)))
if ( dot_prod22 > 0.0_pReal &
.and. ( dot_prod12 < 0.0_pReal &
.or. dot_product(plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)), &
plasticState(mappingConstitutive(2,g,i,e))%previousDotState(:,mappingConstitutive(1,g,i,e))) &
< 0.0_pReal) ) then
statedamper = 0.75_pReal + 0.25_pReal * tanh(2.0_pReal + 4.0_pReal * dot_prod12 / dot_prod22)
else
statedamper = 1.0_pReal
endif
! --- get residui ---
new state related changes
2014-06-11 17:57:41 +05:30
mySizeDotState = plasticState(mappingConstitutive(2,g,i,e))%sizeDotState
stateResiduum(1:mySizeDotState) = plasticState(mappingConstitutive(2,g,i,e))% &
state(1:mySizeDotState,mappingConstitutive(1,g,i,e)) &
- plasticState(mappingConstitutive(2,g,i,e))% &
subState0(1:mySizeDotState,mappingConstitutive(1,g,i,e)) &
-(plasticState(mappingConstitutive(2,g,i,e))% &
dotState(1:mySizeDotState,mappingConstitutive(1,g,i,e)) * &
fixed two severe bugs in new state, now a little bit faster (399 vs. 378 sec) and more memory efficicient (251312 vs. 236832 kbytes) for Phenopowerlaw 16x16x16 example (short load only)
2014-06-17 00:49:38 +05:30
statedamper &
new state related changes
2014-06-11 17:57:41 +05:30
+ plasticState(mappingConstitutive(2,g,i,e))% &
previousDotState(1:mySizeDotState,mappingConstitutive(1,g,i,e)) &
more work on the new state
2014-05-27 20:16:03 +05:30
* (1.0_pReal - statedamper)) * crystallite_subdt(g,i,e)
! --- correct state with residuum ---
new state related changes
2014-06-11 17:57:41 +05:30
tempState(1:mySizeDotState) = plasticState(mappingConstitutive(2,g,i,e))% &
state(1:mySizeDotState,mappingConstitutive(1,g,i,e)) &
more work on the new state
2014-05-27 20:16:03 +05:30
- stateResiduum(1:mySizeDotState) ! need to copy to local variable, since we cant flush a pointer in openmp
#endif
instantaneous jumps in the state by constitutive_deltaState are now incorporated for all state integrators. still they (should) not influence the result, since all constitutive laws simply return zero for the deltaState
2012-05-17 20:55:21 +05:30
#ifndef _OPENMP
increased debug threshold for some outputs (decreased verbosity level at "basic")
2014-05-15 18:31:54 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt &
doxygen comments
2013-02-22 04:38:36 +05:30
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g) &
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i8,1x,i2,1x,i3,/)') '<< CRYST >> update state at el ip g ',e,i,g
write(6,'(a,f6.1,/)') '<< CRYST >> statedamper ',statedamper
fixed some e12.6 string formats to avoid compile time warnings
2013-10-08 19:21:36 +05:30
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> state residuum',stateResiduum(1:mySizeDotState)
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> new state',tempState(1:mySizeDotState)
doxygen comments
2013-02-22 04:38:36 +05:30
endif
instantaneous jumps in the state by constitutive_deltaState are now incorporated for all state integrators. still they (should) not influence the result, since all constitutive laws simply return zero for the deltaState
2012-05-17 20:55:21 +05:30
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
! --- store corrected dotState --- (cannot do this before state update, because not sure how to flush pointers in openmp)
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
constitutive_dotState(g,i,e)%p = constitutive_dotState(g,i,e)%p * statedamper &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
+ constitutive_previousDotState(g,i,e)%p &
* (1.0_pReal - statedamper)
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
! --- converged ? ---
if ( all( abs(stateResiduum(1:mySizeDotState)) < constitutive_aTolState(g,i,e)%p(1:mySizeDotState) &
.or. abs(stateResiduum(1:mySizeDotState)) < rTol_crystalliteState &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
* abs(tempState(1:mySizeDotState)) ) ) then
more work on the new state
2014-05-27 20:16:03 +05:30
crystallite_converged(g,i,e) = .true. ! ... converged per definitionem
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
!$OMP CRITICAL (distributionState)
debug_StateLoopDistribution(NiterationState,numerics_integrationMode) = &
debug_StateLoopDistribution(NiterationState,numerics_integrationMode) + 1_pInt
!$OMP END CRITICAL (distributionState)
endif
endif
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_state(g,i,e)%p(1:mySizeDotState) = tempState(1:mySizeDotState) ! copy local backup to global pointer
doxygen comments
2013-02-22 04:38:36 +05:30
more work on the new state
2014-05-27 20:16:03 +05:30
#else
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%dotState(:,mappingConstitutive(1,g,i,e)) * statedamper &
+ plasticState(mappingConstitutive(2,g,i,e))%previousDotState(:,mappingConstitutive(1,g,i,e)) &
* (1.0_pReal - statedamper)
! --- converged ? ---
if ( all( abs(stateResiduum(1:mySizeDotState)) < plasticState(mappingConstitutive(2,g,i,e))%aTolState &
(1:mySizeDotState).or. abs(stateResiduum(1:mySizeDotState)) < rTol_crystalliteState &
* abs(tempState(1:mySizeDotState)) ) ) then
crystallite_converged(g,i,e) = .true. ! ... converged per definition
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
!$OMP CRITICAL (distributionState)
debug_StateLoopDistribution(NiterationState,numerics_integrationMode) = &
debug_StateLoopDistribution(NiterationState,numerics_integrationMode) + 1_pInt
!$OMP END CRITICAL (distributionState)
endif
endif
fixed two severe bugs in new state, now a little bit faster (399 vs. 378 sec) and more memory efficicient (251312 vs. 236832 kbytes) for Phenopowerlaw 16x16x16 example (short load only)
2014-06-17 00:49:38 +05:30
plasticState(mappingConstitutive(2,g,i,e))%state(1:mySizeDotState,mappingConstitutive(1,g,i,e))= &
more work on the new state
2014-05-27 20:16:03 +05:30
tempState(1:mySizeDotState) ! copy local backup to global pointer
#endif
endif
doxygen comments
2013-02-22 04:38:36 +05:30
enddo; enddo; enddo
!$OMP ENDDO
more work on the new state
2014-05-27 20:16:03 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
! --- STATE JUMP ---
!$OMP DO
do e = eIter(1),eIter(2); do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
!$OMP FLUSH(crystallite_todo)
if (crystallite_todo(g,i,e) .and. crystallite_converged(g,i,e)) then ! converged and still alive...
more work on the new state
2014-05-27 20:16:03 +05:30
crystallite_todo(g,i,e) = crystallite_stateJump(g,i,e)
doxygen comments
2013-02-22 04:38:36 +05:30
!$OMP FLUSH(crystallite_todo)
if (.not. crystallite_todo(g,i,e)) then ! if state jump fails, then convergence is broken
more work on the new state
2014-05-27 20:16:03 +05:30
crystallite_converged(g,i,e) = .false.
doxygen comments
2013-02-22 04:38:36 +05:30
if (.not. crystallite_localPlasticity(g,i,e)) then ! if broken non-local...
!$OMP CRITICAL (checkTodo)
crystallite_todo = crystallite_todo .and. crystallite_localPlasticity ! ...all non-locals skipped
!$OMP END CRITICAL (checkTodo)
endif
endif
endif
enddo; enddo; enddo
!$OMP ENDDO
!$OMP END PARALLEL
more work on the new state
2014-05-27 20:16:03 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt) &
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
write(6,'(a,i8,a,i2,/)') '<< CRYST >> ', count(crystallite_converged(:,:,:)), &
doxygen comments
2013-02-22 04:38:36 +05:30
' grains converged after state integration no. ', NiterationState
! --- NON-LOCAL CONVERGENCE CHECK ---
if (.not. singleRun) then ! if not requesting Integration of just a single IP
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
if (any(.not. crystallite_converged .and. .not. crystallite_localPlasticity)) & ! any non-local not yet converged (or broken)...
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_converged = crystallite_converged .and. crystallite_localPlasticity ! ...restart all non-local as not converged
endif
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt) then
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
write(6,'(a,i8,a)') '<< CRYST >> ', count(crystallite_converged(:,:,:)), &
' grains converged after non-local check'
write(6,'(a,i8,a,i2,/)') '<< CRYST >> ', count(crystallite_todo(:,:,:)), &
' grains todo after state integration no. ', NiterationState
doxygen comments
2013-02-22 04:38:36 +05:30
endif
crystallite_integrateStateFPI: convergence check always ran about all e,i,g; now checking only those specified in FEsolving_execElem, FEsolving_execIP
2014-01-22 00:15:41 +05:30
! --- CHECK IF DONE WITH INTEGRATION ---
doneWithIntegration = .true.
elemLoop: do e = eIter(1),eIter(2)
do i = iIter(1,e),iIter(2,e); do g = gIter(1,e),gIter(2,e) ! iterate over elements, ips and grains
if (crystallite_todo(g,i,e) .and. .not. crystallite_converged(g,i,e)) then
doneWithIntegration = .false.
exit elemLoop
endif
enddo; enddo
enddo elemLoop
introduced plain mode (no ping pong) again and added test for MSC.MArc 2012 as prove that its working
2013-05-17 23:22:46 +05:30
enddo crystalliteLooping
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
end subroutine crystallite_integrateStateFPI
* now able to choose method for state integration (integrator and integratorStiffness in numerics.config); standard is "Fixed Point Iteration", which is basically the same as before; others available are "Explicit Euler", "AdaptiveEuler", "Classical Runge-Kutta" and "Runge-Kutta Cash-Karp" * now remembering stiffness similar to how we do it for Lp etc.; avoids undefined stiffness values for nonconverged stiffness calculation * non-local stuff: * changed non-local kinetics (Gilman2002) * enforce zero shearrate for overall carrrier density below relevant density * enforce zero density for those states that become negative and were below relevant density before * dislocation velocity is not limited by V^(1/3) / dt anymore
2010-10-01 17:48:49 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief calculates a jump in the state according to the current state and the current stress
removed delta state for J2 and phenopowerlaw when using new state
2014-06-17 12:24:49 +05:30
!> returns true, if state jump was successfull or not needed. false indicates NaN in delta state
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
logical function crystallite_stateJump(g,i,e)
use debug, only: &
debug_level, &
debug_crystallite, &
debug_levelExtensive, &
debug_levelSelective, &
debug_e, &
debug_i, &
debug_g
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
use FEsolving, only: &
FEsolving_execElem, &
FEsolving_execIP
use mesh, only: &
mesh_element, &
mesh_NcpElems
use material, only: &
more work on the new state
2014-05-27 20:16:03 +05:30
#ifdef NEWSTATE
plasticState, &
#endif
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
homogenization_Ngrains
use constitutive, only: &
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_collectDeltaState, &
removed delta state for J2 and phenopowerlaw when using new state
2014-06-17 12:24:49 +05:30
#ifndef NEWSTATE
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
constitutive_sizeDotState, &
constitutive_state, &
more work on the new state
2014-05-27 20:16:03 +05:30
constitutive_deltaState
#else
mappingConstitutive
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
implicit none
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
integer(pInt), intent(in):: &
e, & ! element index
i, & ! integration point index
g ! grain index
integer(pInt) :: &
mySizeDotState
doxygen comments
2013-02-22 04:38:36 +05:30
removed delta state for J2 and phenopowerlaw when using new state
2014-06-17 12:24:49 +05:30
if (constitutive_collectDeltaState(crystallite_Tstar_v(1:6,g,i,e), g,i,e)) then
more work on the new state
2014-05-27 20:16:03 +05:30
#ifdef NEWSTATE
removed delta state for J2 and phenopowerlaw when using new state
2014-06-17 12:24:49 +05:30
mySizeDotState = plasticState(mappingConstitutive(2,g,i,e))%sizeDotState
if( any(plasticState(mappingConstitutive(2,g,i,e))%deltaState(:,mappingConstitutive(1,g,i,e)) &
/= plasticState(mappingConstitutive(2,g,i,e))%deltaState(:,mappingConstitutive(1,g,i,e)))) then ! NaN occured in deltaState
crystallite_stateJump = .false.
return
endif
fixed warning message in constitutive
2014-06-17 20:54:44 +05:30
plasticState( mappingConstitutive(2,g,i,e))%state (1:mySizeDotState,mappingConstitutive(1,g,i,e)) = &
plasticState(mappingConstitutive(2,g,i,e))%state (1:mySizeDotState,mappingConstitutive(1,g,i,e)) + &
plasticState(mappingConstitutive(2,g,i,e))%deltaState(1:mySizeDotState,mappingConstitutive(1,g,i,e))
more work on the new state
2014-05-27 20:16:03 +05:30
#else
removed delta state for J2 and phenopowerlaw when using new state
2014-06-17 12:24:49 +05:30
mySizeDotState = constitutive_sizeDotState(g,i,e)
if (any(constitutive_deltaState(g,i,e)%p(1:mySizeDotState) &
/= constitutive_deltaState(g,i,e)%p(1:mySizeDotState))) then
crystallite_stateJump = .false.
return
endif
more work on the new state
2014-05-27 20:16:03 +05:30
removed delta state for J2 and phenopowerlaw when using new state
2014-06-17 12:24:49 +05:30
constitutive_state(g,i,e)%p(1:mySizeDotState) = constitutive_state(g,i,e)%p(1:mySizeDotState) &
doxygen comments
2013-02-22 04:38:36 +05:30
+ constitutive_deltaState(g,i,e)%p(1:mySizeDotState)
new function stateJump, which takes care of immediate changes in the state (deltaState); this introduces discontinuities in the state evolution; therefore this is always and only once done after each integration step, so no evaluation of deltaState for intermediate steps of e.g. the Runge-Kutta integrator; otherwise integration becomes a pain without significant gain in accuracy deltaState now successfully tested for nonlocal model; was not correct for integrators 1,4,5 before
2012-06-06 20:41:30 +05:30
#ifndef _OPENMP
removed delta state for J2 and phenopowerlaw when using new state
2014-06-17 12:24:49 +05:30
if (any(constitutive_deltaState(g,i,e)%p(1:mySizeDotState) /= 0.0_pReal) &
.and. iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt &
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g) &
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
write(6,'(a,i8,1x,i2,1x,i3, /)') '<< CRYST >> update state at el ip g ',e,i,g
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> deltaState', constitutive_deltaState(g,i,e)%p(1:mySizeDotState)
write(6,'(a,/,(12x,12(e12.5,1x)),/)') '<< CRYST >> new state', constitutive_state(g,i,e)%p(1:mySizeDotState)
endif
new function stateJump, which takes care of immediate changes in the state (deltaState); this introduces discontinuities in the state evolution; therefore this is always and only once done after each integration step, so no evaluation of deltaState for intermediate steps of e.g. the Runge-Kutta integrator; otherwise integration becomes a pain without significant gain in accuracy deltaState now successfully tested for nonlocal model; was not correct for integrators 1,4,5 before
2012-06-06 20:41:30 +05:30
#endif
removed delta state for J2 and phenopowerlaw when using new state
2014-06-17 12:24:49 +05:30
#endif
endif
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_stateJump = .true.
new function stateJump, which takes care of immediate changes in the state (deltaState); this introduces discontinuities in the state evolution; therefore this is always and only once done after each integration step, so no evaluation of deltaState for intermediate steps of e.g. the Runge-Kutta integrator; otherwise integration becomes a pain without significant gain in accuracy deltaState now successfully tested for nonlocal model; was not correct for integrators 1,4,5 before
2012-06-06 20:41:30 +05:30
end function crystallite_stateJump
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief calculation of stress (P) with time integration based on a residuum in Lp and
!> intermediate acceleration of the Newton-Raphson correction
!--------------------------------------------------------------------------------------------------
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
logical function crystallite_integrateStress(&
doxygen comments
2013-02-22 04:38:36 +05:30
g,& ! grain number
i,& ! integration point number
e,& ! element number
timeFraction &
)
use prec, only: pLongInt
use numerics, only: nStress, &
aTol_crystalliteStress, &
rTol_crystalliteStress, &
iJacoLpresiduum, &
numerics_integrationMode
use debug, only: debug_level, &
debug_crystallite, &
debug_levelBasic, &
debug_levelExtensive, &
debug_levelSelective, &
debug_e, &
debug_i, &
debug_g, &
debug_cumLpCalls, &
debug_cumLpTicks, &
debug_StressLoopDistribution
use constitutive, only: constitutive_LpAndItsTangent, &
remnoved some unused fuctions from use statements
2013-08-02 11:48:41 +05:30
constitutive_TandItsTangent
doxygen comments
2013-02-22 04:38:36 +05:30
use math, only: math_mul33x33, &
math_mul33xx33, &
math_mul66x6, &
math_mul99x99, &
math_transpose33, &
math_inv33, &
math_invert33, &
math_invert, &
math_det33, &
math_norm33, &
math_I3, &
math_identity2nd, &
math_Mandel66to3333, &
math_Mandel6to33, &
math_Mandel33to6, &
math_Plain3333to99, &
math_Plain33to9, &
math_Plain9to33
implicit none
integer(pInt), intent(in):: e, & ! element index
i, & ! integration point index
g ! grain index
real(pReal), optional, intent(in) :: timeFraction ! fraction of timestep
named some loops and made BUTCHER TABLEAUs parameters in crystallite_integrateStateRKCK45
2013-04-29 16:47:30 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!*** local variables ***!
real(pReal), dimension(3,3):: Fg_new, & ! deformation gradient at end of timestep
Fp_current, & ! plastic deformation gradient at start of timestep
Fp_new, & ! plastic deformation gradient at end of timestep
Fe_new, & ! elastic deformation gradient at end of timestep
invFp_new, & ! inverse of Fp_new
invFp_current, & ! inverse of Fp_current
Lpguess, & ! current guess for plastic velocity gradient
Lpguess_old, & ! known last good guess for plastic velocity gradient
Lp_constitutive, & ! plastic velocity gradient resulting from constitutive law
residuum, & ! current residuum of plastic velocity gradient
residuum_old, & ! last residuum of plastic velocity gradient
deltaLp, & ! direction of next guess
Tstar,& ! 2nd Piola-Kirchhoff Stress
A,&
B, &
Fe ! elastic deformation gradient
real(pReal), dimension(6):: Tstar_v ! 2nd Piola-Kirchhoff Stress in Mandel-Notation
real(pReal), dimension(9):: work ! needed for matrix inversion by LAPACK
integer(pInt), dimension(9) :: ipiv ! needed for matrix inversion by LAPACK
real(pReal), dimension(9,9) :: dLp_dT_constitutive, & ! partial derivative of plastic velocity gradient calculated by constitutive law
dT_dFe_constitutive, & ! partial derivative of 2nd Piola-Kirchhoff stress calculated by constitutive law
dFe_dLp, & ! partial derivative of elastic deformation gradient
dR_dLp, & ! partial derivative of residuum (Jacobian for NEwton-Raphson scheme)
dR_dLp2 ! working copy of dRdLp
real(pReal), dimension(3,3,3,3):: dT_dFe3333, & ! partial derivative of 2nd Piola-Kirchhoff stress
dFe_dLp3333 ! partial derivative of elastic deformation gradient
setting todo and converged flag only for those e,i,g that are to be executed (FE_elemExec etc) used to spoil convergence checking in non ping pong mode and lead to slow performance
2014-01-21 21:39:00 +05:30
real(pReal) det, & ! determinant
doxygen comments
2013-02-22 04:38:36 +05:30
steplength0, &
steplength, &
dt, & ! time increment
aTol
logical error ! flag indicating an error
integer(pInt) NiterationStress, & ! number of stress integrations
ierr, & ! error indicator for LAPACK
o, &
p, &
jacoCounter ! counter to check for Jacobian update
integer(pLongInt) tick, &
tock, &
tickrate, &
maxticks
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
made doxygen working for MSC.Marc again, small changes on the todo-statements for doxygen, they don't work on single lines of codes but only on module/variables
2013-06-11 22:05:04 +05:30
external :: &
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
#if(FLOAT==8)
made doxygen working for MSC.Marc again, small changes on the todo-statements for doxygen, they don't work on single lines of codes but only on module/variables
2013-06-11 22:05:04 +05:30
dgesv
#elif(FLOAT==4)
sgesv
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
!* be pessimistic
crystallite_integrateStress = .false.
* 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
#ifndef _OPENMP
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt &
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g) &
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
write(6,'(a,i8,1x,i2,1x,i3)') '<< CRYST >> integrateStress at el ip g ',e,i,g
endif
* 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
#endif
restructured stress integration. Double loop for stress integration (Lp and stress) with cutback step in IntegrateStress is replaced by single loop for Lp without cutback. loop counter and limits are accordingly renamed and implemented
2009-05-28 22:08:40 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!* only integrate over fraction of timestep?
if (present(timeFraction)) then
dt = crystallite_subdt(g,i,e) * timeFraction
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
Fg_new = crystallite_subF0(1:3,1:3,g,i,e) &
+ (crystallite_subF(1:3,1:3,g,i,e) - crystallite_subF0(1:3,1:3,g,i,e)) * timeFraction
doxygen comments
2013-02-22 04:38:36 +05:30
else
dt = crystallite_subdt(g,i,e)
Fg_new = crystallite_subF(1:3,1:3,g,i,e)
endif
!* feed local variables
Fp_current = crystallite_subFp0(1:3,1:3,g,i,e) ! "Fp_current" is only used as temp var here...
Lpguess_old = crystallite_Lp(1:3,1:3,g,i,e) ! consider present Lp good (i.e. worth remembering) ...
Lpguess = crystallite_Lp(1:3,1:3,g,i,e) ! ... and take it as first guess
!* inversion of Fp_current...
invFp_current = math_inv33(Fp_current)
if (all(invFp_current == 0.0_pReal)) then ! ... failed?
* 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
#ifndef _OPENMP
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
write(6,'(a,i8,1x,i2,1x,i3)') '<< CRYST >> integrateStress failed on inversion of Fp_current at el ip g ',e,i,g
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) > 0_pInt) &
write(6,'(/,a,/,3(12x,3(f12.7,1x)/))') '<< CRYST >> Fp_current',math_transpose33(Fp_current(1:3,1:3))
doxygen comments
2013-02-22 04:38:36 +05:30
endif
* 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
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
return
endif
A = math_mul33x33(Fg_new,invFp_current) ! intermediate tensor needed later to calculate dFe_dLp
restructured stress integration. Double loop for stress integration (Lp and stress) with cutback step in IntegrateStress is replaced by single loop for Lp without cutback. loop counter and limits are accordingly renamed and implemented
2009-05-28 22:08:40 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!* start LpLoop with normal step length
NiterationStress = 0_pInt
jacoCounter = 0_pInt
steplength0 = 1.0_pReal
steplength = steplength0
residuum_old = 0.0_pReal
LpLoop: do
NiterationStress = NiterationStress + 1_pInt
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
loopsExeced: if (NiterationStress > nStress) then
* 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
#ifndef _OPENMP
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) &
write(6,'(a,i3,a,i8,1x,i2,1x,i3,/)') '<< CRYST >> integrateStress reached loop limit',nStress,' at el ip g ',e,i,g
* 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
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
return
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
endif loopsExeced
doxygen comments
2013-02-22 04:38:36 +05:30
!* calculate (elastic) 2nd Piola--Kirchhoff stress tensor and its tangent from constitutive law
B = math_I3 - dt*Lpguess
Fe = math_mul33x33(A,B) ! current elastic deformation tensor
call constitutive_TandItsTangent(Tstar, dT_dFe3333, Fe, g,i,e) ! call constitutive law to calculate 2nd Piola-Kirchhoff stress and its derivative
Tstar_v = math_Mandel33to6(Tstar)
restructured stress integration. Double loop for stress integration (Lp and stress) with cutback step in IntegrateStress is replaced by single loop for Lp without cutback. loop counter and limits are accordingly renamed and implemented
2009-05-28 22:08:40 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!* calculate plastic velocity gradient and its tangent from constitutive law
* enclose wall time measurement for LpAndItsTangent and count of leapfrogbreaks by CRITICAL construct * better use SINGLE (having an implicit barrier at the end) instead of MASTER construct * deleted all explicit BARRIERs after do loops since parallel loop construct implies barrier at the end
2010-11-19 22:59:29 +05:30
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
doxygen comments
2013-02-22 04:38:36 +05:30
call system_clock(count=tick,count_rate=tickrate,count_max=maxticks)
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
endif
doxygen comments
2013-02-22 04:38:36 +05:30
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
call constitutive_LpAndItsTangent(Lp_constitutive, dLp_dT_constitutive, Tstar_v, &
crystallite_temperature(i,e), g, i, e)
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
call system_clock(count=tock,count_rate=tickrate,count_max=maxticks)
!$OMP CRITICAL (debugTimingLpTangent)
debug_cumLpCalls = debug_cumLpCalls + 1_pInt
debug_cumLpTicks = debug_cumLpTicks + tock-tick
!$OMP FLUSH (debug_cumLpTicks)
if (tock < tick) debug_cumLpTicks = debug_cumLpTicks + maxticks
!$OMP END CRITICAL (debugTimingLpTangent)
endif
* 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
#ifndef _OPENMP
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt &
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g) &
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
write(6,'(a,i3,/)') '<< CRYST >> iteration ', NiterationStress
doxygen comments
2013-02-22 04:38:36 +05:30
write(6,'(a,/,3(12x,3(e20.7,1x)/))') '<< CRYST >> Lp_constitutive', math_transpose33(Lp_constitutive)
write(6,'(a,/,3(12x,3(e20.7,1x)/))') '<< CRYST >> Lpguess', math_transpose33(Lpguess)
endif
* 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
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
!* update current residuum and check for convergence of loop
aTol = max(rTol_crystalliteStress * max(math_norm33(Lpguess),math_norm33(Lp_constitutive)), & ! absolute tolerance from largest acceptable relative error
aTol_crystalliteStress) ! minimum lower cutoff
residuum = Lpguess - Lp_constitutive
if (any(residuum /= residuum)) then ! NaN in residuum...
* 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
#ifndef _OPENMP
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) &
doxygen comments
2013-02-22 04:38:36 +05:30
write(6,'(a,i8,1x,i2,1x,i3,a,i3,a)') '<< CRYST >> integrateStress encountered NaN at el ip g ',e,i,g,&
' ; iteration ', NiterationStress,&
' >> returning..!'
* 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
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
return ! ...me = .false. to inform integrator about problem
elseif (math_norm33(residuum) < aTol) then ! converged if below absolute tolerance
exit LpLoop ! ...leave iteration loop
elseif (math_norm33(residuum) < math_norm33(residuum_old) .or. NiterationStress == 1_pInt ) then ! not converged, but improved norm of residuum (always proceed in first iteration)...
residuum_old = residuum ! ...remember old values and...
Lpguess_old = Lpguess
steplength = steplength0 ! ...proceed with normal step length (calculate new search direction)
else ! not converged and residuum not improved...
steplength = 0.5_pReal * steplength ! ...try with smaller step length in same direction
Lpguess = Lpguess_old + steplength * deltaLp
cycle LpLoop
endif
!* calculate Jacobian for correction term
if (mod(jacoCounter, iJacoLpresiduum) == 0_pInt) then
dFe_dLp3333 = 0.0_pReal
do o=1_pInt,3_pInt; do p=1_pInt,3_pInt
Fixed wrong indices in tangents dT_dFe and dFe_dLp, which however luckily did not have any effect in the perturbed stiffness since they were transposed such that the double contraction of both remained unchanged. In contrast, the analytical jacobian will probably be affected by this change! @Pratheek: Can you check with me how this can be fixed?
2014-01-22 14:08:13 +05:30
dFe_dLp3333(o,p,1:3,p) = A(o,1:3) ! dFe_dLp(i,j,k,l) = -dt * A(i,k) delta(l,j)
doxygen comments
2013-02-22 04:38:36 +05:30
enddo; enddo
dFe_dLp3333 = -dt * dFe_dLp3333
dFe_dLp = math_Plain3333to99(dFe_dLp3333)
dT_dFe_constitutive = math_Plain3333to99(dT_dFe3333)
dR_dLp = math_identity2nd(9_pInt) - &
math_mul99x99(dLp_dT_constitutive, math_mul99x99(dT_dFe_constitutive , dFe_dLp))
dR_dLp2 = dR_dLp ! will be overwritten in first call to LAPACK routine
work = math_plain33to9(residuum)
substituted hand written matrix inversion by LAPACK version with precision selection. also introduced check for inversion into DAMASK_spectral_Utilities.f90 for the stress BC calculation. This part is further improved by using 5% of the reference stiffness to avoid trouble in the fully plastic regime (where the stiffness is underestimated) Test for Marc 2010 is updated because the new inversion give slightly different results near 0 (order of e-13)
2012-08-28 22:29:45 +05:30
#if(FLOAT==8)
doxygen comments
2013-02-22 04:38:36 +05:30
call dgesv(9,1,dR_dLp2,9,ipiv,work,9,ierr) ! solve dR/dLp * delta Lp = -res for dR/dLp
substituted hand written matrix inversion by LAPACK version with precision selection. also introduced check for inversion into DAMASK_spectral_Utilities.f90 for the stress BC calculation. This part is further improved by using 5% of the reference stiffness to avoid trouble in the fully plastic regime (where the stiffness is underestimated) Test for Marc 2010 is updated because the new inversion give slightly different results near 0 (order of e-13)
2012-08-28 22:29:45 +05:30
#elif(FLOAT==4)
doxygen comments
2013-02-22 04:38:36 +05:30
call sgesv(9,1,dR_dLp2,9,ipiv,work,9,ierr) ! solve dR/dLp * delta Lp = -res for dR/dLp
substituted hand written matrix inversion by LAPACK version with precision selection. also introduced check for inversion into DAMASK_spectral_Utilities.f90 for the stress BC calculation. This part is further improved by using 5% of the reference stiffness to avoid trouble in the fully plastic regime (where the stiffness is underestimated) Test for Marc 2010 is updated because the new inversion give slightly different results near 0 (order of e-13)
2012-08-28 22:29:45 +05:30
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
if (ierr /= 0_pInt) then
* 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
#ifndef _OPENMP
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
write(6,'(a,i8,1x,i2,1x,i3,a,i3)') '<< CRYST >> integrateStress failed on dR/dLp inversion at el ip g ',e,i,g
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt &
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g)&
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
write(6,*)
write(6,'(a,/,9(12x,9(e15.3,1x)/))') '<< CRYST >> dR_dLp',transpose(dR_dLp)
write(6,'(a,/,9(12x,9(e15.3,1x)/))') '<< CRYST >> dFe_dLp',transpose(dFe_dLp)
write(6,'(a,/,9(12x,9(e15.3,1x)/))') '<< CRYST >> dT_dFe_constitutive',transpose(dT_dFe_constitutive)
write(6,'(a,/,9(12x,9(e15.3,1x)/))') '<< CRYST >> dLp_dT_constitutive',transpose(dLp_dT_constitutive)
write(6,'(a,/,3(12x,3(e20.7,1x)/))') '<< CRYST >> A',math_transpose33(A)
write(6,'(a,/,3(12x,3(e20.7,1x)/))') '<< CRYST >> B',math_transpose33(B)
write(6,'(a,/,3(12x,3(e20.7,1x)/))') '<< CRYST >> Lp_constitutive',math_transpose33(Lp_constitutive)
write(6,'(a,/,3(12x,3(e20.7,1x)/))') '<< CRYST >> Lpguess',math_transpose33(Lpguess)
endif
endif
Stress integration now uses Armijo rule to find an appropriate correction of Lp: decreases step in case that residuum does not improve significantly, accelerates as usual in case of good convergence. This turned out to improve convergence behavior.
2011-08-02 16:59:08 +05:30
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
return
endif
deltaLp = - math_plain9to33(work)
endif
jacoCounter = jacoCounter + 1_pInt ! increase counter for jaco update
Lpguess = Lpguess + steplength * deltaLp
enddo LpLoop
!* calculate new plastic and elastic deformation gradient
invFp_new = math_mul33x33(invFp_current,B)
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
invFp_new = invFp_new / math_det33(invFp_new)**(1.0_pReal/3.0_pReal) ! regularize by det
doxygen comments
2013-02-22 04:38:36 +05:30
call math_invert33(invFp_new,Fp_new,det,error)
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
if (error .or. any(Fp_new /= Fp_new)) then
* 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
#ifndef _OPENMP
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
write(6,'(a,i8,1x,i2,1x,i3,a,i3)') '<< CRYST >> integrateStress failed on invFp_new inversion at el ip g ',&
e,i,g, ' ; iteration ', NiterationStress
if (iand(debug_level(debug_crystallite), debug_levelExtensive) /= 0_pInt &
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g) &
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) &
write(6,'(/,a,/,3(12x,3(f12.7,1x)/))') '<< CRYST >> invFp_new',math_transpose33(invFp_new)
doxygen comments
2013-02-22 04:38:36 +05:30
endif
* 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
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
return
endif
Fe_new = math_mul33x33(Fg_new,invFp_new) ! calc resulting Fe
restructured stress integration. Double loop for stress integration (Lp and stress) with cutback step in IntegrateStress is replaced by single loop for Lp without cutback. loop counter and limits are accordingly renamed and implemented
2009-05-28 22:08:40 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
in crystallite_integrateStress: pass full 2nd PK stress to LpAndItsTangent instead of only the deviatoric part; otherwise constitutive laws that depend on the hydrostatic pressure do not work; constitutive laws that require a deviatoric stress need to calculate it internally (as is e.g. already done in the J2 model)
2014-01-16 16:06:40 +05:30
!* calculate 1st Piola-Kirchhoff stress
doxygen comments
2013-02-22 04:38:36 +05:30
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
crystallite_P(1:3,1:3,g,i,e) = math_mul33x33(Fe_new, math_mul33x33(math_Mandel6to33(Tstar_v), &
math_transpose33(invFp_new)))
doxygen comments
2013-02-22 04:38:36 +05:30
!* store local values in global variables
crystallite_Lp(1:3,1:3,g,i,e) = Lpguess
crystallite_Tstar_v(1:6,g,i,e) = Tstar_v
crystallite_Fp(1:3,1:3,g,i,e) = Fp_new
crystallite_Fe(1:3,1:3,g,i,e) = Fe_new
crystallite_invFp(1:3,1:3,g,i,e) = invFp_new
!* set return flag to true
crystallite_integrateStress = .true.
* 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
#ifndef _OPENMP
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite),debug_levelExtensive) /= 0_pInt &
.and. ((e == debug_e .and. i == debug_i .and. g == debug_g) &
.or. .not. iand(debug_level(debug_crystallite), debug_levelSelective) /= 0_pInt)) then
write(6,'(a,/,3(12x,3(f12.7,1x)/))') '<< CRYST >> P / MPa',math_transpose33(crystallite_P(1:3,1:3,g,i,e))/1.0e6_pReal
write(6,'(a,/,3(12x,3(f12.7,1x)/))') '<< CRYST >> Cauchy / MPa', &
math_mul33x33(crystallite_P(1:3,1:3,g,i,e), math_transpose33(Fg_new)) / 1.0e6_pReal / math_det33(Fg_new)
write(6,'(a,/,3(12x,3(f12.7,1x)/))') '<< CRYST >> Fe Lp Fe^-1', &
math_transpose33(math_mul33x33(Fe_new, math_mul33x33(crystallite_Lp(1:3,1:3,g,i,e), math_inv33(Fe_new)))) ! transpose to get correct print out order
write(6,'(a,/,3(12x,3(f12.7,1x)/))') '<< CRYST >> Fp',math_transpose33(crystallite_Fp(1:3,1:3,g,i,e))
endif
* 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
#endif
restructured stress integration. Double loop for stress integration (Lp and stress) with cutback step in IntegrateStress is replaced by single loop for Lp without cutback. loop counter and limits are accordingly renamed and implemented
2009-05-28 22:08:40 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
if (iand(debug_level(debug_crystallite), debug_levelBasic) /= 0_pInt) then
!$OMP CRITICAL (distributionStress)
debug_StressLoopDistribution(NiterationStress,numerics_integrationMode) = &
debug_StressLoopDistribution(NiterationStress,numerics_integrationMode) + 1_pInt
!$OMP END CRITICAL (distributionStress)
endif
end function crystallite_integrateStress
restructured stress integration. Double loop for stress integration (Lp and stress) with cutback step in IntegrateStress is replaced by single loop for Lp without cutback. loop counter and limits are accordingly renamed and implemented
2009-05-28 22:08:40 +05:30
major restructuring of code. homogenization as well as constitutive are now free to choose. the runtime got somewhat longer (25% on simple tests) compared to a hardcoded isostrain homogenization. this might be a point of further optimization at a later stage... please use homogenization_isostrain.f90 as starting point / example for future developments of homog-schemes. the homogenization scheme now can additionally output certain results. hence, the userdata structure at each integration point now looks like this: - sizeHomogPostResults - block of that size containing homogPostResults then for each grain: - sizeGrainPostResults - block of that size containing crystallitePostResults, which consist of: + phaseID + volFrac + Eulers (3) + any constitutive post results requested
2009-05-07 21:57:36 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief calculates orientations and disorientations (in case of single grain ips)
!--------------------------------------------------------------------------------------------------
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
subroutine crystallite_orientations
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
use math, only: &
math_pDecomposition, &
math_RtoQ, &
math_qConj
use FEsolving, only: &
FEsolving_execElem, &
FEsolving_execIP
use IO, only: &
IO_warning
use material, only: &
material_phase, &
homogenization_Ngrains, &
fixed bug introduced with lattice_structure change nonlocal, but with DEBUG=ON OPTIMIZATION=OFF there is an FPE. Division by zero? Marked in the code forgot to commit dislotwin last time, now seems to work
2014-03-12 05:25:40 +05:30
phase_localPlasticity
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
use mesh, only: &
mesh_element, &
mesh_ipNeighborhood, &
FE_NipNeighbors, &
FE_geomtype, &
FE_celltype
moved quaternion disorientation to lattice because it requires the knowledge of the lattice structure.
2014-02-28 18:58:27 +05:30
use lattice, only: &
moved reading in of lattice type and elastic constants to lattice module removed structure type for hex, fcc, bcc, now defining slip/twin systems for each phase found in material.config constitutive modules will only be initialized if needed homogenizedC function is only needed for models incorporating twinning in a physical way (titanmod and dislotwin)
2014-03-09 02:20:31 +05:30
lattice_qDisorientation, &
lattice_structure
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
use constitutive_nonlocal, only: &
constitutive_nonlocal_updateCompatibility
more work on the new state
2014-05-27 20:16:03 +05:30
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
implicit none
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
integer(pInt) &
e, & ! element index
i, & ! integration point index
g, & ! grain index
n, & ! neighbor index
neighboring_e, & ! element index of my neighbor
neighboring_i, & ! integration point index of my neighbor
myPhase, & ! phase
fixed bug introduced with lattice_structure change nonlocal, but with DEBUG=ON OPTIMIZATION=OFF there is an FPE. Division by zero? Marked in the code forgot to commit dislotwin last time, now seems to work
2014-03-12 05:25:40 +05:30
neighboringPhase
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
real(pReal), dimension(3,3) :: &
U, &
R
real(pReal), dimension(4) :: &
orientation
logical &
error
doxygen comments
2013-02-22 04:38:36 +05:30
! --- CALCULATE ORIENTATION AND LATTICE ROTATION ---
!$OMP PARALLEL DO PRIVATE(error,U,R,orientation)
do e = FEsolving_execElem(1),FEsolving_execElem(2)
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
do g = 1_pInt,homogenization_Ngrains(mesh_element(3,e))
added shortcuts for two do loops in the "timesyncing" procedure math_pDecomposition enclosed by omp critical statement; somehow, this subroutine is not thread safe, not clear what exactly the problem is
2013-02-27 16:02:37 +05:30
!$OMP CRITICAL (polarDecomp) ! somehow this subroutine is not threadsafe, so need critical statement here; not clear, what exactly the problem is
call math_pDecomposition(crystallite_Fe(1:3,1:3,g,i,e), U, R, error) ! polar decomposition of Fe
!$OMP END CRITICAL (polarDecomp)
doxygen comments
2013-02-22 04:38:36 +05:30
if (error) then
call IO_warning(650_pInt, e, i, g)
added shortcuts for two do loops in the "timesyncing" procedure math_pDecomposition enclosed by omp critical statement; somehow, this subroutine is not thread safe, not clear what exactly the problem is
2013-02-27 16:02:37 +05:30
orientation = [1.0_pReal, 0.0_pReal, 0.0_pReal, 0.0_pReal] ! fake orientation
doxygen comments
2013-02-22 04:38:36 +05:30
else
orientation = math_RtoQ(transpose(R))
endif
moved quaternion disorientation to lattice because it requires the knowledge of the lattice structure.
2014-02-28 18:58:27 +05:30
crystallite_rotation(1:4,g,i,e) = lattice_qDisorientation(crystallite_orientation0(1:4,g,i,e), & ! active rotation from ori0
moved reading in of lattice type and elastic constants to lattice module removed structure type for hex, fcc, bcc, now defining slip/twin systems for each phase found in material.config constitutive modules will only be initialized if needed homogenizedC function is only needed for models incorporating twinning in a physical way (titanmod and dislotwin)
2014-03-09 02:20:31 +05:30
orientation) ! to current orientation (with no symmetry)
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_orientation(1:4,g,i,e) = orientation
enddo
enddo
enddo
!$OMP END PARALLEL DO
! --- UPDATE SOME ADDITIONAL VARIABLES THAT ARE NEEDED FOR NONLOCAL MATERIAL ---
fixed bug introduced with lattice_structure change nonlocal, but with DEBUG=ON OPTIMIZATION=OFF there is an FPE. Division by zero? Marked in the code forgot to commit dislotwin last time, now seems to work
2014-03-12 05:25:40 +05:30
! --- we use crystallite_orientation from above, so need a separate loop
doxygen comments
2013-02-22 04:38:36 +05:30
fixed wrong omp statement
2014-03-12 20:59:14 +05:30
!$OMP PARALLEL DO PRIVATE(myPhase,neighboring_e,neighboring_i,neighboringPhase)
doxygen comments
2013-02-22 04:38:36 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2)
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
myPhase = material_phase(1,i,e) ! get my phase
more work on the new state
2014-05-27 20:16:03 +05:30
#ifndef NEWSTATE
doxygen comments
2013-02-22 04:38:36 +05:30
if (.not. phase_localPlasticity(myPhase)) then ! if nonlocal model
! --- calculate disorientation between me and my neighbor ---
neighborhood relations were screwed up for some elementtypes as of version 2318, might have caused segmentation faults with nonlocal model, now fixed
2013-04-22 19:05:35 +05:30
do n = 1_pInt,FE_NipNeighbors(FE_celltype(FE_geomtype(mesh_element(2,e)))) ! loop through my neighbors
doxygen comments
2013-02-22 04:38:36 +05:30
neighboring_e = mesh_ipNeighborhood(1,n,i,e)
neighboring_i = mesh_ipNeighborhood(2,n,i,e)
fixed wrong omp statement
2014-03-12 20:59:14 +05:30
if (neighboring_e > 0 .and. neighboring_i > 0) then ! if neighbor exists
doxygen comments
2013-02-22 04:38:36 +05:30
neighboringPhase = material_phase(1,neighboring_i,neighboring_e) ! get my neighbor's phase
if (.not. phase_localPlasticity(neighboringPhase)) then ! neighbor got also nonlocal plasticity
fixed bug introduced with lattice_structure change nonlocal, but with DEBUG=ON OPTIMIZATION=OFF there is an FPE. Division by zero? Marked in the code forgot to commit dislotwin last time, now seems to work
2014-03-12 05:25:40 +05:30
if (lattice_structure(myPhase) == lattice_structure(neighboringPhase)) then ! if my neighbor has same crystal structure like me
doxygen comments
2013-02-22 04:38:36 +05:30
crystallite_disorientation(:,n,1,i,e) = &
moved quaternion disorientation to lattice because it requires the knowledge of the lattice structure.
2014-02-28 18:58:27 +05:30
lattice_qDisorientation( crystallite_orientation(1:4,1,i,e), &
fixed bug introduced with lattice_structure change nonlocal, but with DEBUG=ON OPTIMIZATION=OFF there is an FPE. Division by zero? Marked in the code forgot to commit dislotwin last time, now seems to work
2014-03-12 05:25:40 +05:30
crystallite_orientation(1:4,1,neighboring_i,neighboring_e), &
lattice_structure(myPhase)) ! calculate disorientation for given symmetry
doxygen comments
2013-02-22 04:38:36 +05:30
else ! for neighbor with different phase
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
crystallite_disorientation(:,n,1,i,e) = [0.0_pReal, 1.0_pReal, 0.0_pReal, 0.0_pReal] ! 180 degree rotation about 100 axis
doxygen comments
2013-02-22 04:38:36 +05:30
endif
else ! for neighbor with local plasticity
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
crystallite_disorientation(:,n,1,i,e) = [-1.0_pReal, 0.0_pReal, 0.0_pReal, 0.0_pReal] ! homomorphic identity
doxygen comments
2013-02-22 04:38:36 +05:30
endif
else ! no existing neighbor
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
crystallite_disorientation(:,n,1,i,e) = [-1.0_pReal, 0.0_pReal, 0.0_pReal, 0.0_pReal] ! homomorphic identity
doxygen comments
2013-02-22 04:38:36 +05:30
endif
enddo
! --- calculate compatibility and transmissivity between me and my neighbor ---
call constitutive_nonlocal_updateCompatibility(crystallite_orientation,i,e)
endif
more work on the new state
2014-05-27 20:16:03 +05:30
#endif
doxygen comments
2013-02-22 04:38:36 +05:30
enddo
enddo
!$OMP END PARALLEL DO
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
end subroutine crystallite_orientations
new subroutine crystallite_orientations: calculates the orientation matrix and the euler angles and - in case of single grain ips - the misorientation for each neighboring ip. The calculation of the misorientation is now done once in crystallite init and at the end of every FE increment. This saves a lot of time compared to doing it in dotState for every crystallite subinc.
2009-12-18 21:16:33 +05:30
in crystallite_stressAndItsTangent we now update the state first (explicit integration) and then integrate the stress (implicit integration) also inside the stress integration, it is now possible to define the frequency of the Jacobian update oin the LpLoop through iJacoLpResiduum. The frequency of the Jacobian update for the stiffness in the crystallite loop is controlled by the parameter iJacoStiffness. also updated the corresponding stuctograms in the documentation
2009-06-09 16:35:29 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief return results of particular grain
!--------------------------------------------------------------------------------------------------
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
function crystallite_postResults(ipc, ip, el)
added crystallite output "neighboringelement" and "neighboringip" some polishing in crystallite_postResults
2013-05-08 17:32:30 +05:30
use math, only: &
math_qToEuler, &
grainrotation output now uses corrected rotation math
2013-06-07 15:05:00 +05:30
math_qToEulerAxisAngle, &
added crystallite output "neighboringelement" and "neighboringip" some polishing in crystallite_postResults
2013-05-08 17:32:30 +05:30
math_mul33x33, &
math_transpose33, &
math_det33, &
math_I3, &
inDeg, &
math_Mandel6to33, &
math_qMul, &
math_qConj
use mesh, only: &
mesh_element, &
mesh_ipVolume, &
mesh_ipCoordinates, &
mesh_maxNipNeighbors, &
mesh_ipNeighborhood, &
FE_NipNeighbors, &
FE_geomtype, &
FE_celltype
use material, only: &
microstructure_crystallite, &
crystallite_Noutput, &
material_phase, &
material_texture, &
homogenization_Ngrains
use constitutive, only: &
constitutive_sizePostResults, &
constitutive_postResults, &
constitutive_homogenizedC
in crystallite_stressAndItsTangent we now update the state first (explicit integration) and then integrate the stress (implicit integration) also inside the stress integration, it is now possible to define the frequency of the Jacobian update oin the LpLoop through iJacoLpResiduum. The frequency of the Jacobian update for the stiffness in the crystallite loop is controlled by the parameter iJacoStiffness. also updated the corresponding stuctograms in the documentation
2009-06-09 16:35:29 +05:30
major restructuring of code. homogenization as well as constitutive are now free to choose. the runtime got somewhat longer (25% on simple tests) compared to a hardcoded isostrain homogenization. this might be a point of further optimization at a later stage... please use homogenization_isostrain.f90 as starting point / example for future developments of homog-schemes. the homogenization scheme now can additionally output certain results. hence, the userdata structure at each integration point now looks like this: - sizeHomogPostResults - block of that size containing homogPostResults then for each grain: - sizeGrainPostResults - block of that size containing crystallitePostResults, which consist of: + phaseID + volFrac + Eulers (3) + any constitutive post results requested
2009-05-07 21:57:36 +05:30
implicit none
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
integer(pInt), intent(in):: &
el, & !< element index
ip, & !< integration point index
ipc !< grain index
major restructuring of code. homogenization as well as constitutive are now free to choose. the runtime got somewhat longer (25% on simple tests) compared to a hardcoded isostrain homogenization. this might be a point of further optimization at a later stage... please use homogenization_isostrain.f90 as starting point / example for future developments of homog-schemes. the homogenization scheme now can additionally output certain results. hence, the userdata structure at each integration point now looks like this: - sizeHomogPostResults - block of that size containing homogPostResults then for each grain: - sizeGrainPostResults - block of that size containing crystallitePostResults, which consist of: + phaseID + volFrac + Eulers (3) + any constitutive post results requested
2009-05-07 21:57:36 +05:30
added crystallite output "neighboringelement" and "neighboringip" some polishing in crystallite_postResults
2013-05-08 17:32:30 +05:30
real(pReal), dimension(1+crystallite_sizePostResults(microstructure_crystallite(mesh_element(4,el)))+ &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
1+constitutive_sizePostResults(ipc,ip,el)) :: &
crystallite_postResults
real(pReal), dimension(3,3) :: &
Ee
real(pReal), dimension(4) :: &
rotation
real(pReal) :: &
detF
integer(pInt) :: &
o, &
c, &
crystID, &
mySize, &
n
reworked crystallite part to allow for flexible user output --> new "crystallite" part in config file --> new "crystallite" option for microstructures --> new output file "...job.outputCrystallite" to be used in conjunction with marc_addUserOutput for meaningful naming of User Defined Vars.
2010-02-25 23:09:11 +05:30
added crystallite output "neighboringelement" and "neighboringip" some polishing in crystallite_postResults
2013-05-08 17:32:30 +05:30
crystID = microstructure_crystallite(mesh_element(4,el))
reworked crystallite part to allow for flexible user output --> new "crystallite" part in config file --> new "crystallite" option for microstructures --> new output file "...job.outputCrystallite" to be used in conjunction with marc_addUserOutput for meaningful naming of User Defined Vars.
2010-02-25 23:09:11 +05:30
crystallite_postResults = 0.0_pReal
c = 0_pInt
some lines in crystallite.f90 exceeded limit of 132 characters fixed bug in Lib_VTK_IO.f90 (informed author as well) added zip file containing source of Lib_VTK_IO.f90
2013-05-08 23:18:00 +05:30
crystallite_postResults(c+1) = real(crystallite_sizePostResults(crystID),pReal) ! size of results from cryst
openmp parallelization working again (at least for j2 and nonlocal constitutive model). In order to keep it like that, please follow these simple rules: DON'T use implicit array subscripts: example: real, dimension(3,3) :: A,B A(:,2) = B(:,1) <--- DON'T USE A(1:3,2) = B(1:3,1) <--- BETTER USE In many cases the use of explicit array subscripts is inevitable for parallelization. Additionally, it is an easy means to prevent memory leaks. Enclose all write statements with the following: !$OMP CRITICAL (write2out) <your write statement> !$OMP END CRITICAL (write2out) Whenever you change something in the code and are not sure if it affects parallelization and leads to nonconforming behavior, please ask me and/or Franz to check this.
2011-03-17 16:16:17 +05:30
c = c + 1_pInt
major restructuring of code. homogenization as well as constitutive are now free to choose. the runtime got somewhat longer (25% on simple tests) compared to a hardcoded isostrain homogenization. this might be a point of further optimization at a later stage... please use homogenization_isostrain.f90 as starting point / example for future developments of homog-schemes. the homogenization scheme now can additionally output certain results. hence, the userdata structure at each integration point now looks like this: - sizeHomogPostResults - block of that size containing homogPostResults then for each grain: - sizeGrainPostResults - block of that size containing crystallitePostResults, which consist of: + phaseID + volFrac + Eulers (3) + any constitutive post results requested
2009-05-07 21:57:36 +05:30
again: adding pInt, removing redundant use statments, chang in dble to real(,pReal)
2012-02-21 22:01:37 +05:30
do o = 1_pInt,crystallite_Noutput(crystID)
shape mismatch in crystallite, invalid names for variables(pos, size) in homogenization corrected. polishing of DAMASK_spectral_interface and makefile
2011-08-01 23:40:55 +05:30
mySize = 0_pInt
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
select case(crystallite_outputID(o,crystID))
case (phase_ID)
shape mismatch in crystallite, invalid names for variables(pos, size) in homogenization corrected. polishing of DAMASK_spectral_interface and makefile
2011-08-01 23:40:55 +05:30
mySize = 1_pInt
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
crystallite_postResults(c+1) = real(material_phase(ipc,ip,el),pReal) ! phaseID of grain
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (texture_ID)
shape mismatch in crystallite, invalid names for variables(pos, size) in homogenization corrected. polishing of DAMASK_spectral_interface and makefile
2011-08-01 23:40:55 +05:30
mySize = 1_pInt
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
crystallite_postResults(c+1) = real(material_texture(ipc,ip,el),pReal) ! textureID of grain
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (volume_ID)
shape mismatch in crystallite, invalid names for variables(pos, size) in homogenization corrected. polishing of DAMASK_spectral_interface and makefile
2011-08-01 23:40:55 +05:30
mySize = 1_pInt
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
detF = math_det33(crystallite_partionedF(1:3,1:3,ipc,ip,el)) ! V_current = det(F) * V_reference
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
crystallite_postResults(c+1) = detF * mesh_ipVolume(ip,el) &
/ homogenization_Ngrains(mesh_element(3,el)) ! grain volume (not fraction but absolute)
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (heat_ID)
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
mySize = 1_pInt
crystallite_postResults(c+1) = crystallite_heat(ipc,ip,el) ! heat production
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (orientation_ID)
shape mismatch in crystallite, invalid names for variables(pos, size) in homogenization corrected. polishing of DAMASK_spectral_interface and makefile
2011-08-01 23:40:55 +05:30
mySize = 4_pInt
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
crystallite_postResults(c+1:c+mySize) = crystallite_orientation(1:4,ipc,ip,el) ! grain orientation as quaternion
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (eulerangles_ID)
shape mismatch in crystallite, invalid names for variables(pos, size) in homogenization corrected. polishing of DAMASK_spectral_interface and makefile
2011-08-01 23:40:55 +05:30
mySize = 3_pInt
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
crystallite_postResults(c+1:c+mySize) = inDeg &
* math_qToEuler(crystallite_orientation(1:4,ipc,ip,el)) ! grain orientation as Euler angles in degree
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (grainrotation_ID)
shape mismatch in crystallite, invalid names for variables(pos, size) in homogenization corrected. polishing of DAMASK_spectral_interface and makefile
2011-08-01 23:40:55 +05:30
mySize = 4_pInt
grainrotation output now uses corrected rotation math
2013-06-07 15:05:00 +05:30
crystallite_postResults(c+1:c+mySize) = &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
math_qToEulerAxisAngle(crystallite_rotation(1:4,ipc,ip,el)) ! grain rotation away from initial orientation as axis-angle in sample reference coordinates
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
crystallite_postResults(c+4) = inDeg * crystallite_postResults(c+4) ! angle in degree
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (grainrotationx_ID)
new crystallite output "grainrotationX", "grainrotationY", "grainrotationZ" that gives the deviation of the initial orientation in degrees around the sample reference x/y/z axis, while "grainrotation" still gives the same as axis angle pair in crystal coordinates
2013-01-16 14:15:41 +05:30
mySize = 1_pInt
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
rotation = math_qToEulerAxisAngle(crystallite_rotation(1:4,ipc,ip,el)) ! grain rotation away from initial orientation as axis-angle in sample reference coordinates
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
crystallite_postResults(c+1) = inDeg * rotation(1) * rotation(4) ! angle in degree
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (grainrotationy_ID)
new crystallite output "grainrotationX", "grainrotationY", "grainrotationZ" that gives the deviation of the initial orientation in degrees around the sample reference x/y/z axis, while "grainrotation" still gives the same as axis angle pair in crystal coordinates
2013-01-16 14:15:41 +05:30
mySize = 1_pInt
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
rotation = math_qToEulerAxisAngle(crystallite_rotation(1:4,ipc,ip,el)) ! grain rotation away from initial orientation as axis-angle in sample reference coordinates
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
crystallite_postResults(c+1) = inDeg * rotation(2) * rotation(4) ! angle in degree
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (grainrotationz_ID)
new crystallite output "grainrotationX", "grainrotationY", "grainrotationZ" that gives the deviation of the initial orientation in degrees around the sample reference x/y/z axis, while "grainrotation" still gives the same as axis angle pair in crystal coordinates
2013-01-16 14:15:41 +05:30
mySize = 1_pInt
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
rotation = math_qToEulerAxisAngle(crystallite_rotation(1:4,ipc,ip,el)) ! grain rotation away from initial orientation as axis-angle in sample reference coordinates
introduced dummy heat calculation, available as an crystallite output temperature, avgP and avgF are available as homogenization output. move crystallite output of ipcoordinates to homogenization
2013-10-19 00:27:28 +05:30
crystallite_postResults(c+1) = inDeg * rotation(3) * rotation(4) ! angle in degree
new substructure with 'include' and 'config' directories renaming of mpie_... to DAMASK for main usersubroutines extension of element outputs from 5 to 8 digits (FFT!!!)
2011-05-11 22:08:45 +05:30
! remark: tensor output is of the form 11,12,13, 21,22,23, 31,32,33
! thus row index i is slow, while column index j is fast. reminder: "row is slow"
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (defgrad_ID)
added some output variables in crystallite: F, Fe, Fp, Ee, P, S
2010-05-18 13:27:13 +05:30
mySize = 9_pInt
some lines in crystallite.f90 exceeded limit of 132 characters fixed bug in Lib_VTK_IO.f90 (informed author as well) added zip file containing source of Lib_VTK_IO.f90
2013-05-08 23:18:00 +05:30
crystallite_postResults(c+1:c+mySize) = &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
reshape(math_transpose33(crystallite_partionedF(1:3,1:3,ipc,ip,el)),[mySize])
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (e_ID)
added total Green-Lagrange strain to crystallite outputs
2012-01-20 15:55:35 +05:30
mySize = 9_pInt
crystallite_postResults(c+1:c+mySize) = 0.5_pReal * reshape((math_mul33x33( &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
math_transpose33(crystallite_partionedF(1:3,1:3,ipc,ip,el)), &
crystallite_partionedF(1:3,1:3,ipc,ip,el)) - math_I3),[mySize])
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (fe_ID)
added some output variables in crystallite: F, Fe, Fp, Ee, P, S
2010-05-18 13:27:13 +05:30
mySize = 9_pInt
some lines in crystallite.f90 exceeded limit of 132 characters fixed bug in Lib_VTK_IO.f90 (informed author as well) added zip file containing source of Lib_VTK_IO.f90
2013-05-08 23:18:00 +05:30
crystallite_postResults(c+1:c+mySize) = &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
reshape(math_transpose33(crystallite_Fe(1:3,1:3,ipc,ip,el)),[mySize])
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (ee_ID)
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
Ee = 0.5_pReal *(math_mul33x33(math_transpose33(crystallite_Fe(1:3,1:3,ipc,ip,el)), &
crystallite_Fe(1:3,1:3,ipc,ip,el)) - math_I3)
added some output variables in crystallite: F, Fe, Fp, Ee, P, S
2010-05-18 13:27:13 +05:30
mySize = 9_pInt
polishing: removed variable names like 'unit' and 'data' that are keywords of fortran and ensured that integer and real precision matches independent of machine standard. removed cut_off parameter for damask_spectral removed outpot of derived divergence measures and added RMS output in brackets added comments and options to the makefile
2012-02-16 00:28:38 +05:30
crystallite_postResults(c+1:c+mySize) = reshape(Ee,[mySize])
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (fp_ID)
added some output variables in crystallite: F, Fe, Fp, Ee, P, S
2010-05-18 13:27:13 +05:30
mySize = 9_pInt
some lines in crystallite.f90 exceeded limit of 132 characters fixed bug in Lib_VTK_IO.f90 (informed author as well) added zip file containing source of Lib_VTK_IO.f90
2013-05-08 23:18:00 +05:30
crystallite_postResults(c+1:c+mySize) = &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
reshape(math_transpose33(crystallite_Fp(1:3,1:3,ipc,ip,el)),[mySize])
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (lp_ID)
added "Lp" as crystallite output
2011-02-25 13:45:26 +05:30
mySize = 9_pInt
some lines in crystallite.f90 exceeded limit of 132 characters fixed bug in Lib_VTK_IO.f90 (informed author as well) added zip file containing source of Lib_VTK_IO.f90
2013-05-08 23:18:00 +05:30
crystallite_postResults(c+1:c+mySize) = &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
reshape(math_transpose33(crystallite_Lp(1:3,1:3,ipc,ip,el)),[mySize])
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (p_ID)
added some output variables in crystallite: F, Fe, Fp, Ee, P, S
2010-05-18 13:27:13 +05:30
mySize = 9_pInt
some lines in crystallite.f90 exceeded limit of 132 characters fixed bug in Lib_VTK_IO.f90 (informed author as well) added zip file containing source of Lib_VTK_IO.f90
2013-05-08 23:18:00 +05:30
crystallite_postResults(c+1:c+mySize) = &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
reshape(math_transpose33(crystallite_P(1:3,1:3,ipc,ip,el)),[mySize])
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (s_ID)
added some output variables in crystallite: F, Fe, Fp, Ee, P, S
2010-05-18 13:27:13 +05:30
mySize = 9_pInt
some lines in crystallite.f90 exceeded limit of 132 characters fixed bug in Lib_VTK_IO.f90 (informed author as well) added zip file containing source of Lib_VTK_IO.f90
2013-05-08 23:18:00 +05:30
crystallite_postResults(c+1:c+mySize) = &
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
reshape(math_Mandel6to33(crystallite_Tstar_v(1:6,ipc,ip,el)),[mySize])
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case (elasmatrix_ID)
added elastic stiffness matrix as crystallite output
2012-03-01 18:36:09 +05:30
mySize = 36_pInt
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
crystallite_postResults(c+1:c+mySize) = reshape(constitutive_homogenizedC(ipc,ip,el),[mySize])
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case(neighboringelement_ID)
added crystallite output "neighboringelement" and "neighboringip" some polishing in crystallite_postResults
2013-05-08 17:32:30 +05:30
mySize = mesh_maxNipNeighbors
crystallite_postResults(c+1:c+mySize) = 0.0_pReal
forall (n = 1_pInt:FE_NipNeighbors(FE_celltype(FE_geomtype(mesh_element(2,el))))) &
crystallite_postResults(c+n) = real(mesh_ipNeighborhood(1,n,ip,el),pReal)
indroduced sourced allocation, enums where applicable (some parts still missing). fixed bug when having recursive file input
2013-12-12 22:39:59 +05:30
case(neighboringip_ID)
added crystallite output "neighboringelement" and "neighboringip" some polishing in crystallite_postResults
2013-05-08 17:32:30 +05:30
mySize = mesh_maxNipNeighbors
crystallite_postResults(c+1:c+mySize) = 0.0_pReal
forall (n = 1_pInt:FE_NipNeighbors(FE_celltype(FE_geomtype(mesh_element(2,el))))) &
crystallite_postResults(c+n) = real(mesh_ipNeighborhood(2,n,ip,el),pReal)
reworked crystallite part to allow for flexible user output --> new "crystallite" part in config file --> new "crystallite" option for microstructures --> new output file "...job.outputCrystallite" to be used in conjunction with marc_addUserOutput for meaningful naming of User Defined Vars.
2010-02-25 23:09:11 +05:30
end select
shape mismatch in crystallite, invalid names for variables(pos, size) in homogenization corrected. polishing of DAMASK_spectral_interface and makefile
2011-08-01 23:40:55 +05:30
c = c + mySize
reworked crystallite part to allow for flexible user output --> new "crystallite" part in config file --> new "crystallite" option for microstructures --> new output file "...job.outputCrystallite" to be used in conjunction with marc_addUserOutput for meaningful naming of User Defined Vars.
2010-02-25 23:09:11 +05:30
enddo
new substructure with 'include' and 'config' directories renaming of mpie_... to DAMASK for main usersubroutines extension of element outputs from 5 to 8 digits (FFT!!!)
2011-05-11 22:08:45 +05:30
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
crystallite_postResults(c+1) = real(constitutive_sizePostResults(ipc,ip,el),pReal) ! size of constitutive results
openmp parallelization working again (at least for j2 and nonlocal constitutive model). In order to keep it like that, please follow these simple rules: DON'T use implicit array subscripts: example: real, dimension(3,3) :: A,B A(:,2) = B(:,1) <--- DON'T USE A(1:3,2) = B(1:3,1) <--- BETTER USE In many cases the use of explicit array subscripts is inevitable for parallelization. Additionally, it is an easy means to prevent memory leaks. Enclose all write statements with the following: !$OMP CRITICAL (write2out) <your write statement> !$OMP END CRITICAL (write2out) Whenever you change something in the code and are not sure if it affects parallelization and leads to nonconforming behavior, please ask me and/or Franz to check this.
2011-03-17 16:16:17 +05:30
c = c + 1_pInt
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
if (constitutive_sizePostResults(ipc,ip,el) > 0_pInt) &
crystallite_postResults(c+1:c+constitutive_sizePostResults(ipc,ip,el)) = &
constitutive_postResults(crystallite_Tstar_v(1:6,ipc,ip,el), crystallite_Fe, &
in preguess of FPintegrator: resetting of previous dotstates can be done in same loop as collectdotstate polishing: indentation level and capitalization
2013-11-21 16:28:41 +05:30
crystallite_temperature(ip,el), ipc, ip, el)
removed temperature integration and corresponding data structures and debugging options temperature is stored in crystallite, but homogeneous on one IP (not an component (grain) quantity and an input value parsed in by the BVP solver. introduced heat, a component (grain) quantity which is homogenized before returned to the heat transfer solver. went ahead with removal of dummy functions in homogenization and constitutive, this time mainly reduced function signatures to reflect actually needed quantities.
2013-10-16 18:34:59 +05:30
c = c + constitutive_sizePostResults(ipc,ip,el)
did a lot of polishing: - removed unnecessary "return" before end of subroutine or function: - changed undetermined array length (:) to (1:3) To prevent problems with some code analysing tools: - "3D oneliner loops" (with ";) only for "do" and "enddo" at the same time - removed line continuation in OMP statements made the makefile more flexible, removed heap-arrays switch
2011-09-13 21:24:06 +05:30
added new, flexible debugging scheme. now all modules have their own debug specification. compiles and runs, I hope nothing is broken did a lot of polishing
2012-03-09 01:55:28 +05:30
end function crystallite_postResults
major restructuring of code. homogenization as well as constitutive are now free to choose. the runtime got somewhat longer (25% on simple tests) compared to a hardcoded isostrain homogenization. this might be a point of further optimization at a later stage... please use homogenization_isostrain.f90 as starting point / example for future developments of homog-schemes. the homogenization scheme now can additionally output certain results. hence, the userdata structure at each integration point now looks like this: - sizeHomogPostResults - block of that size containing homogPostResults then for each grain: - sizeGrainPostResults - block of that size containing crystallitePostResults, which consist of: + phaseID + volFrac + Eulers (3) + any constitutive post results requested
2009-05-07 21:57:36 +05:30
fixed small allocation flaws for new state
2014-05-12 18:30:37 +05:30
doxygen comments
2013-02-22 04:38:36 +05:30
end module crystallite