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

1759 lines
102 KiB
Fortran
Raw Normal View History

added version information to all files do NOT edit text like this: $Id: constitutive_phenopowerlaw.f90 406 2009-08-31 14:13:10Z MPIE\f.roters $
2009-08-31 20:39:15 +05:30
!* $Id$
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
!***************************************
!* Module: CRYSTALLITE *
!***************************************
!* contains: *
!* - _init *
!* - materialpoint_stressAndItsTangent *
!* - _partitionDeformation *
!* - _updateState *
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
!* - _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
!* - _postResults *
!***************************************
MODULE crystallite
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
use prec, only: pReal, pInt
implicit none
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
!
! ****************************************************************
! *** General variables for the crystallite calculation ***
! ****************************************************************
Major Update: all modules are now correctly submitted
2009-07-01 16:25:31 +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
integer(pInt) crystallite_maxSizePostResults
integer(pInt), dimension(:), allocatable :: crystallite_sizePostResults
integer(pInt), dimension(:,:), allocatable :: crystallite_sizePostResult
character(len=64), dimension(:,:), allocatable :: crystallite_output ! name of each post result output
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
integer(pInt), dimension (:,:,:), allocatable :: &
crystallite_symmetryID ! crystallographic symmetry 1=cubic 2=hexagonal, needed in all orientation calcs
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
small readability improvement
2010-02-18 20:36:08 +05:30
real(pReal), dimension (:,:,:), allocatable :: &
crystallite_dt, & ! requested time increment of each grain
crystallite_subdt, & ! substepped time increment of each grain
crystallite_subFrac, & ! already calculated fraction of increment
crystallite_subStep, & ! size of next integration step
crystallite_Temperature, & ! Temp of each grain
crystallite_partionedTemperature0, & ! Temp of each grain at start of homog inc
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
crystallite_subTemperature0, & ! Temp of each grain at start of crystallite inc
crystallite_statedamper ! damping for state update
small readability improvement
2010-02-18 20:36:08 +05:30
real(pReal), dimension (:,:,:,:), allocatable :: &
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
crystallite_subTstar0_v, & ! 2nd Piola-Kirchhoff stress vector at start of crystallite inc
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
crystallite_orientation, & ! orientation as quaternion
crystallite_orientation0, & ! initial orientation as quaternion
crystallite_rotation ! grain rotation away from initial orientation as axis-angle (in degrees)
small readability improvement
2010-02-18 20:36:08 +05:30
real(pReal), dimension (:,:,:,:,:), allocatable :: &
crystallite_Fe, & ! current "elastic" def grad (end of converged time step)
crystallite_Fp, & ! current plastic def grad (end of converged time step)
crystallite_invFp, & ! inverse of 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_subFp0,& ! plastic def grad at start of crystallite 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_subF, & ! def grad to be reached at end of crystallite inc
crystallite_subF0, & ! def grad at start of crystallite 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_subLp0,& ! plastic velocity grad at start of crystallite inc
crystallite_P, & ! 1st Piola-Kirchhoff stress per grain
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
crystallite_disorientation ! disorientation between two neighboring ips (only calculated for single grain IPs)
small readability improvement
2010-02-18 20:36:08 +05:30
real(pReal), dimension (:,:,:,:,:,:,:), allocatable :: &
crystallite_dPdF, & ! individual dPdF per grain
crystallite_fallbackdPdF ! dPdF fallback for non-converged grains (elastic prediction)
logical, dimension (:,:,:), allocatable :: &
crystallite_localConstitution, & ! indicates this grain to have purely local constitutive law
crystallite_requested, & ! flag to request crystallite calculation
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
crystallite_todo, & ! flag to indicate ongoing calculation
small readability improvement
2010-02-18 20:36:08 +05:30
crystallite_converged, & ! convergence flag
crystallite_stateConverged, & ! flag indicating convergence of state
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
crystallite_temperatureConverged ! flag indicating convergence of temperature
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
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
!********************************************************************
! allocate and initialize per grain variables
!********************************************************************
Major Update: all modules are now correctly submitted
2009-07-01 16:25:31 +05:30
subroutine crystallite_init(Temperature)
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
!*** variables and functions from other modules ***!
use prec, only: pInt, &
pReal
use debug, only: debug_info, &
debug_reset
use math, only: math_I3, &
restoring the crystallite.f90, IO.f90 and material.f90 to the version 575. the version 576 of these files contain some uncleared debugging statements.
2010-05-26 22:04:44 +05:30
math_EulerToR
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
use FEsolving, only: FEsolving_execElem, &
FEsolving_execIP
use mesh, only: mesh_element, &
mesh_NcpElems, &
mesh_maxNips, &
mesh_maxNipNeighbors
use IO
use material
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
use lattice, only: lattice_symmetryTypes
use constitutive_phenopowerlaw, only: constitutive_phenopowerlaw_label, &
constitutive_phenopowerlaw_structure
Updated corresponding files for titanmod constitutive
2010-09-13 14:59:03 +05:30
use constitutive_titanmod, only: constitutive_titanmod_label, &
constitutive_titanmod_structure
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
use constitutive_dislotwin, only: constitutive_dislotwin_label, &
constitutive_dislotwin_structure
use constitutive_nonlocal, only: constitutive_nonlocal_label, &
constitutive_nonlocal_structure
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
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
implicit none
integer(pInt), parameter :: file = 200
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
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
!*** input variables ***!
real(pReal) Temperature
!*** output variables ***!
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +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
!*** local variables ***!
integer(pInt), parameter :: maxNchunks = 2
integer(pInt), dimension(1+2*maxNchunks) :: positions
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
myCrystallite ! crystallite of current elem
integer(pInt) section, j,p, output, mySize
character(len=64) tag
character(len=1024) line
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
integer(pInt) myStructure, & ! lattice structure
myPhase
rearranged output statements in crystallite init for better readability in output file
2010-08-26 15:22:40 +05:30
write(6,*)
write(6,*) '<<<+- crystallite init -+>>>'
write(6,*) '$Id$'
write(6,*)
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
gMax = homogenization_maxNgrains
iMax = mesh_maxNips
eMax = mesh_NcpElems
nMax = mesh_maxNipNeighbors
allocate(crystallite_Temperature(gMax,iMax,eMax)); crystallite_Temperature = Temperature
allocate(crystallite_P(3,3,gMax,iMax,eMax)); crystallite_P = 0.0_pReal
allocate(crystallite_Fe(3,3,gMax,iMax,eMax)); crystallite_Fe = 0.0_pReal
allocate(crystallite_Fp(3,3,gMax,iMax,eMax)); crystallite_Fp = 0.0_pReal
allocate(crystallite_invFp(3,3,gMax,iMax,eMax)); crystallite_invFp = 0.0_pReal
allocate(crystallite_Lp(3,3,gMax,iMax,eMax)); crystallite_Lp = 0.0_pReal
allocate(crystallite_Tstar_v(6,gMax,iMax,eMax)); crystallite_Tstar_v = 0.0_pReal
allocate(crystallite_F0(3,3,gMax,iMax,eMax)); crystallite_F0 = 0.0_pReal
allocate(crystallite_Fp0(3,3,gMax,iMax,eMax)); crystallite_Fp0 = 0.0_pReal
allocate(crystallite_Lp0(3,3,gMax,iMax,eMax)); crystallite_Lp0 = 0.0_pReal
allocate(crystallite_Tstar0_v(6,gMax,iMax,eMax)); crystallite_Tstar0_v = 0.0_pReal
allocate(crystallite_partionedTemperature0(gMax,iMax,eMax)); crystallite_partionedTemperature0 = 0.0_pReal
allocate(crystallite_partionedF(3,3,gMax,iMax,eMax)); crystallite_partionedF = 0.0_pReal
allocate(crystallite_partionedF0(3,3,gMax,iMax,eMax)); crystallite_partionedF0 = 0.0_pReal
allocate(crystallite_partionedFp0(3,3,gMax,iMax,eMax)); crystallite_partionedFp0 = 0.0_pReal
allocate(crystallite_partionedLp0(3,3,gMax,iMax,eMax)); crystallite_partionedLp0 = 0.0_pReal
allocate(crystallite_partionedTstar0_v(6,gMax,iMax,eMax)); crystallite_partionedTstar0_v = 0.0_pReal
allocate(crystallite_subTemperature0(gMax,iMax,eMax)); crystallite_subTemperature0 = 0.0_pReal
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
allocate(crystallite_symmetryID(gMax,iMax,eMax)); crystallite_symmetryID = 0.0_pReal !NEW
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
allocate(crystallite_subF(3,3,gMax,iMax,eMax)); crystallite_subF = 0.0_pReal
allocate(crystallite_subF0(3,3,gMax,iMax,eMax)); crystallite_subF0 = 0.0_pReal
allocate(crystallite_subFp0(3,3,gMax,iMax,eMax)); crystallite_subFp0 = 0.0_pReal
allocate(crystallite_subLp0(3,3,gMax,iMax,eMax)); crystallite_subLp0 = 0.0_pReal
changed internal representation of orientation and misorientation from euler angles to quaternions (this should also fix some bugs in the math_misorientation subroutine). includes a couple of new functions in math.f90 and some changes in crystallite.f90. beware that crystallite output "orientation" now by default returns the orientation as quaternion. if you want euler angles instead, you have to add "eulerangles" as a crystallite output in your material.config file (see material.config template). for input of orientations in the texture block of the material.config you still have to specify the rotation in terms of euler angles, quaternions are not yet supported for input.
2010-03-18 17:53:17 +05:30
allocate(crystallite_orientation(4,gMax,iMax,eMax)); crystallite_orientation = 0.0_pReal
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
allocate(crystallite_orientation0(4,gMax,iMax,eMax)); crystallite_orientation = 0.0_pReal
allocate(crystallite_rotation(4,gMax,iMax,eMax)); crystallite_rotation = 0.0_pReal
small typo
2010-04-28 23:15:30 +05:30
allocate(crystallite_disorientation(4,nMax,gMax,iMax,eMax)); crystallite_disorientation = 0.0_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
allocate(crystallite_subTstar0_v(6,gMax,iMax,eMax)); crystallite_subTstar0_v = 0.0_pReal
allocate(crystallite_dPdF(3,3,3,3,gMax,iMax,eMax)); crystallite_dPdF = 0.0_pReal
allocate(crystallite_fallbackdPdF(3,3,3,3,gMax,iMax,eMax)); crystallite_fallbackdPdF = 0.0_pReal
allocate(crystallite_dt(gMax,iMax,eMax)); crystallite_dt = 0.0_pReal
allocate(crystallite_subdt(gMax,iMax,eMax)); crystallite_subdt = 0.0_pReal
allocate(crystallite_subFrac(gMax,iMax,eMax)); crystallite_subFrac = 0.0_pReal
allocate(crystallite_subStep(gMax,iMax,eMax)); crystallite_subStep = 0.0_pReal
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
allocate(crystallite_statedamper(gMax,iMax,eMax)); crystallite_statedamper = 1.0_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
allocate(crystallite_localConstitution(gMax,iMax,eMax)); crystallite_localConstitution = .true.
allocate(crystallite_requested(gMax,iMax,eMax)); crystallite_requested = .false.
allocate(crystallite_converged(gMax,iMax,eMax)); crystallite_converged = .true.
allocate(crystallite_stateConverged(gMax,iMax,eMax)); crystallite_stateConverged = .false.
allocate(crystallite_temperatureConverged(gMax,iMax,eMax)); crystallite_temperatureConverged = .false.
allocate(crystallite_todo(gMax,iMax,eMax)); crystallite_todo = .true.
allocate(crystallite_output(maxval(crystallite_Noutput), &
material_Ncrystallite)) ; crystallite_output = ''
allocate(crystallite_sizePostResults(material_Ncrystallite)) ; crystallite_sizePostResults = 0_pInt
allocate(crystallite_sizePostResult(maxval(crystallite_Noutput), &
material_Ncrystallite)) ; crystallite_sizePostResult = 0_pInt
if(.not. IO_open_file(file,material_configFile)) call IO_error (100) ! corrupt config file
line = ''
section = 0
do while (IO_lc(IO_getTag(line,'<','>')) /= material_partCrystallite) ! wind forward to <crystallite>
read(file,'(a1024)',END=100) line
enddo
do ! read thru sections of phase part
read(file,'(a1024)',END=100) line
if (IO_isBlank(line)) cycle ! skip empty lines
if (IO_getTag(line,'<','>') /= '') exit ! stop at next part
if (IO_getTag(line,'[',']') /= '') then ! next section
section = section + 1
output = 0 ! reset output counter
endif
if (section > 0) then
positions = IO_stringPos(line,maxNchunks)
tag = IO_lc(IO_stringValue(line,positions,1)) ! extract key
select case(tag)
case ('(output)')
output = output + 1
crystallite_output(output,section) = IO_lc(IO_stringValue(line,positions,2))
end select
endif
enddo
100 close(file)
do i = 1,material_Ncrystallite ! sanity checks
enddo
do i = 1,material_Ncrystallite
do j = 1,crystallite_Noutput(i)
select case(crystallite_output(j,i))
case('phase')
mySize = 1
case('volume')
mySize = 1
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
case('orientation') ! orientation as quaternion
changed internal representation of orientation and misorientation from euler angles to quaternions (this should also fix some bugs in the math_misorientation subroutine). includes a couple of new functions in math.f90 and some changes in crystallite.f90. beware that crystallite output "orientation" now by default returns the orientation as quaternion. if you want euler angles instead, you have to add "eulerangles" as a crystallite output in your material.config file (see material.config template). for input of orientations in the texture block of the material.config you still have to specify the rotation in terms of euler angles, quaternions are not yet supported for input.
2010-03-18 17:53:17 +05:30
mySize = 4
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
case('eulerangles') ! Bunge Euler angles
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
mySize = 3
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
case('grainrotation') ! Deviation from initial grain orientation in axis-angle form (angle in degrees)
mySize = 4
added some output variables in crystallite: F, Fe, Fp, Ee, P, S
2010-05-18 13:27:13 +05:30
case('defgrad','f','fe','fp','ee','p','firstpiola','1stpiola','s','tstar','secondpiola','2ndpiola')
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
mySize = 9
case default
mySize = 0
end select
if (mySize > 0_pInt) then ! any meaningful output found
crystallite_sizePostResult(j,i) = mySize
crystallite_sizePostResults(i) = crystallite_sizePostResults(i) + mySize
endif
enddo
enddo
crystallite_maxSizePostResults = maxval(crystallite_sizePostResults)
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
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
if(.not. IO_open_jobFile(file,'outputCrystallite')) call IO_error (50) ! problems in writing file
do p = 1,material_Ncrystallite
write(file,*)
write(file,'(a)') '['//trim(crystallite_name(p))//']'
write(file,*)
do e = 1,crystallite_Noutput(p)
write(file,'(a,i4)') trim(crystallite_output(e,p))//char(9),crystallite_sizePostResult(e,p)
enddo
enddo
close(file)
!$OMP PARALLEL DO
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over all cp elements
myNgrains = homogenization_Ngrains(mesh_element(3,e)) ! look up homogenization-->grainCount
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element
do g = 1,myNgrains
crystallite_partionedTemperature0(g,i,e) = Temperature ! isothermal assumption
crystallite_Fp0(:,:,g,i,e) = math_EulerToR(material_EulerAngles(:,g,i,e)) ! plastic def gradient reflects init orientation
crystallite_Fe(:,:,g,i,e) = transpose(crystallite_Fp0(:,:,g,i,e))
crystallite_F0(:,:,g,i,e) = math_I3
crystallite_partionedFp0(:,:,g,i,e) = crystallite_Fp0(:,:,g,i,e)
crystallite_partionedF0(:,:,g,i,e) = crystallite_F0(:,:,g,i,e)
crystallite_partionedF(:,:,g,i,e) = crystallite_F0(:,:,g,i,e)
crystallite_requested(g,i,e) = .true.
crystallite_localConstitution(g,i,e) = phase_localConstitution(material_phase(g,i,e))
enddo
enddo
enddo
!$OMPEND PARALLEL DO
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
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
! Initialize crystallite_symmetryID(g,i,e)
!$OMP PARALLEL DO
do e = FEsolving_execElem(1),FEsolving_execElem(2)
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
do g = 1,homogenization_Ngrains(mesh_element(3,e))
myPhase = material_phase(g,i,e)
select case (phase_constitution(myPhase))
case (constitutive_phenopowerlaw_label)
myStructure = constitutive_phenopowerlaw_structure(phase_constitutionInstance(myPhase))
Updated corresponding files for titanmod constitutive
2010-09-13 14:59:03 +05:30
case (constitutive_titanmod_label)
myStructure = constitutive_titanmod_structure(phase_constitutionInstance(myPhase))
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
case (constitutive_dislotwin_label)
myStructure = constitutive_dislotwin_structure(phase_constitutionInstance(myPhase))
case (constitutive_nonlocal_label)
myStructure = constitutive_nonlocal_structure(phase_constitutionInstance(myPhase))
case default
myStructure = -1_pInt ! does this happen for j2 material?
end select
if (myStructure>0_pInt) then
crystallite_symmetryID(g,i,e)=lattice_symmetryTypes(myStructure) ! structure = 1(fcc) or 2(bcc) => 1; 3(hex)=>2
endif
enddo
enddo
enddo
!$OMPEND PARALLEL DO
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
call crystallite_orientations()
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
crystallite_orientation0=crystallite_orientation ! Store initial orientations for calculation of grain rotations
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
call crystallite_stressAndItsTangent(.true.) ! 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
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
! *** Output to MARC output file ***
!$OMP CRITICAL (write2out)
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_Temperature: ', shape(crystallite_Temperature)
write(6,'(a35,x,7(i5,x))') 'crystallite_Fe: ', shape(crystallite_Fe)
write(6,'(a35,x,7(i5,x))') 'crystallite_Fp: ', shape(crystallite_Fp)
write(6,'(a35,x,7(i5,x))') 'crystallite_Lp: ', shape(crystallite_Lp)
write(6,'(a35,x,7(i5,x))') 'crystallite_F0: ', shape(crystallite_F0)
write(6,'(a35,x,7(i5,x))') 'crystallite_Fp0: ', shape(crystallite_Fp0)
write(6,'(a35,x,7(i5,x))') 'crystallite_Lp0: ', shape(crystallite_Lp0)
write(6,'(a35,x,7(i5,x))') 'crystallite_partionedF: ', shape(crystallite_partionedF)
write(6,'(a35,x,7(i5,x))') 'crystallite_partionedTemp0: ', shape(crystallite_partionedTemperature0)
write(6,'(a35,x,7(i5,x))') 'crystallite_partionedF0: ', shape(crystallite_partionedF0)
write(6,'(a35,x,7(i5,x))') 'crystallite_partionedFp0: ', shape(crystallite_partionedFp0)
write(6,'(a35,x,7(i5,x))') 'crystallite_partionedLp0: ', shape(crystallite_partionedLp0)
write(6,'(a35,x,7(i5,x))') 'crystallite_subF: ', shape(crystallite_subF)
write(6,'(a35,x,7(i5,x))') 'crystallite_subTemperature0: ', shape(crystallite_subTemperature0)
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_symmetryID: ', shape(crystallite_symmetryID) !NEW
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_subF0: ', shape(crystallite_subF0)
write(6,'(a35,x,7(i5,x))') 'crystallite_subFp0: ', shape(crystallite_subFp0)
write(6,'(a35,x,7(i5,x))') 'crystallite_subLp0: ', shape(crystallite_subLp0)
write(6,'(a35,x,7(i5,x))') 'crystallite_P: ', shape(crystallite_P)
write(6,'(a35,x,7(i5,x))') 'crystallite_Tstar_v: ', shape(crystallite_Tstar_v)
write(6,'(a35,x,7(i5,x))') 'crystallite_Tstar0_v: ', shape(crystallite_Tstar0_v)
write(6,'(a35,x,7(i5,x))') 'crystallite_partionedTstar0_v: ', shape(crystallite_partionedTstar0_v)
write(6,'(a35,x,7(i5,x))') 'crystallite_subTstar0_v: ', shape(crystallite_subTstar0_v)
write(6,'(a35,x,7(i5,x))') 'crystallite_dPdF: ', shape(crystallite_dPdF)
write(6,'(a35,x,7(i5,x))') 'crystallite_fallbackdPdF: ', shape(crystallite_fallbackdPdF)
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_orientation: ', shape(crystallite_orientation)
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_orientation0: ', shape(crystallite_orientation0)
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_rotation: ', shape(crystallite_rotation)
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_disorientation: ', shape(crystallite_disorientation)
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_dt: ', shape(crystallite_dt)
write(6,'(a35,x,7(i5,x))') 'crystallite_subdt: ', shape(crystallite_subdt)
write(6,'(a35,x,7(i5,x))') 'crystallite_subFrac: ', shape(crystallite_subFrac)
write(6,'(a35,x,7(i5,x))') 'crystallite_subStep: ', shape(crystallite_subStep)
homogenization_RGC.f90 >>> adding some lines, mostly for debugging purpose. no critical change. constitutive_phenopowerlaw.f90 >>> adding new parameter: constitutive_phenopowerlaw_w0_slip, i.e., the hardening rate exponent. homogenization.f90 >>> most important change is to add an if-else statement (line 379-380) to switch crystallite_requeted = .false. for already converged material point iteration (el/ip). the rest of the changes are cosmetics and debugging stuffs. crystallite.f90 >>> similar to homogenization.f90, the most important change is to add additional if-else statement (line 574) in the jacobian (perturbation) loop. now the jacobian calculation will only be performed when crystallite_requested = .true.. the rest is only cosmetic.
2009-08-27 17:40:06 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_localConstitution: ', shape(crystallite_localConstitution)
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_requested: ', shape(crystallite_requested)
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_todo: ', shape(crystallite_todo)
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_converged: ', shape(crystallite_converged)
write(6,'(a35,x,7(i5,x))') 'crystallite_stateConverged: ', shape(crystallite_stateConverged)
write(6,'(a35,x,7(i5,x))') 'crystallite_temperatureConverged: ', shape(crystallite_temperatureConverged)
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
write(6,'(a35,x,7(i5,x))') 'crystallite_sizePostResults: ', shape(crystallite_sizePostResults)
write(6,'(a35,x,7(i5,x))') 'crystallite_sizePostResult: ', shape(crystallite_sizePostResult)
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
write(6,*)
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
write(6,*) 'Number of nonlocal grains: ',count(.not. crystallite_localConstitution)
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
call flush(6)
!$OMPEND CRITICAL (write2out)
call debug_info()
call debug_reset()
return
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
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
endsubroutine
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
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
!********************************************************************
! calculate stress (P) and tangent (dPdF) for crystallites
!********************************************************************
subroutine crystallite_stressAndItsTangent(updateJaco)
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
!*** variables and functions from other modules ***!
use prec, only: pInt, &
pReal
minimum substepping size in homogenization and crystallite is now controlled by two independent parameters subStepMinHomog and subStepMinCryst respectively. deleted flush statements where they are not needed, since they seriously slow down computation.
2009-10-26 22:13:43 +05:30
use numerics, only: subStepMinCryst, &
introduced a flexibility in cut-backing scheme in homogenization.f90 and in crystallite.f90: (1) subStepSizeHomog and subStepSizeCryst := size of substep when cut-back is applied (initially was hard-coded). (2) stepIncreaseHomog and stepIncreaseCryst := step increase when calculation for substep converge (was also hardcoded). introduced a possibility to choose different finite difference scheme, i.e., forward-, backward- and central-difference, for computing grain numerical tangent. note that central-difference scheme will slow down the computation significantly. please use it only if necessary. parameters to set these new features have been included in numerics.f90 and numerics.config, whereas corresponding error messages have been introduced in the IO.f90
2009-11-10 19:06:27 +05:30
subStepSizeCryst, &
stepIncreaseCryst, &
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
pert_Fg, &
introduced a flexibility in cut-backing scheme in homogenization.f90 and in crystallite.f90: (1) subStepSizeHomog and subStepSizeCryst := size of substep when cut-back is applied (initially was hard-coded). (2) stepIncreaseHomog and stepIncreaseCryst := step increase when calculation for substep converge (was also hardcoded). introduced a possibility to choose different finite difference scheme, i.e., forward-, backward- and central-difference, for computing grain numerical tangent. note that central-difference scheme will slow down the computation significantly. please use it only if necessary. parameters to set these new features have been included in numerics.f90 and numerics.config, whereas corresponding error messages have been introduced in the IO.f90
2009-11-10 19:06:27 +05:30
pert_method, &
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
nState, &
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
nCryst, &
iJacoStiffness
iJacoStiffness belongs to numerics, not debug sorry for not having checked that before
2010-03-17 22:30:54 +05:30
use debug, only: debugger, &
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
selectiveDebugger, &
verboseDebugger, &
debug_e, &
debug_i, &
debug_g, &
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
debug_CrystalliteLoopDistribution, &
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
debug_CrystalliteStateLoopDistribution, &
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
debug_StiffnessStateLoopDistribution
use IO, only: IO_warning
use math, only: math_inv3x3, &
math_mul33x33, &
math_mul66x6, &
math_Mandel6to33, &
math_Mandel33to6, &
homogenization_RGC.f90 >>> adding some lines, mostly for debugging purpose. no critical change. constitutive_phenopowerlaw.f90 >>> adding new parameter: constitutive_phenopowerlaw_w0_slip, i.e., the hardening rate exponent. homogenization.f90 >>> most important change is to add an if-else statement (line 379-380) to switch crystallite_requeted = .false. for already converged material point iteration (el/ip). the rest of the changes are cosmetics and debugging stuffs. crystallite.f90 >>> similar to homogenization.f90, the most important change is to add additional if-else statement (line 574) in the jacobian (perturbation) loop. now the jacobian calculation will only be performed when crystallite_requested = .true.. the rest is only cosmetic.
2009-08-27 17:40:06 +05:30
math_I3, &
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
math_Plain3333to99
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
use FEsolving, only: FEsolving_execElem, &
FEsolving_execIP, &
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
theInc, &
cycleCounter
use mesh, only: mesh_element, &
mesh_NcpElems, &
mesh_maxNips
use material, only: homogenization_Ngrains, &
homogenization_maxNgrains
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
use constitutive, only: constitutive_maxSizeState, &
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
constitutive_maxSizeDotState, &
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
constitutive_sizeState, &
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
constitutive_sizeDotState, &
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
constitutive_state, &
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
constitutive_state_backup, &
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
constitutive_subState0, &
constitutive_partionedState0, &
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
constitutive_homogenizedC, &
constitutive_dotState, &
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
constitutive_dotState_backup, &
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
constitutive_previousDotState, &
constitutive_previousDotState2, &
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
constitutive_collectDotState, &
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
constitutive_dotTemperature, &
constitutive_microstructure
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
implicit none
!*** input variables ***!
logical, intent(in) :: updateJaco ! flag indicating wehther we want to update the Jacobian (stiffness) or not
!*** output variables ***!
!*** local variables ***!
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
real(pReal) myTemperature, & ! local copy of the temperature
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
myPert ! perturbation with correct sign
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
real(pReal), dimension(3,3) :: invFp, & ! inverse of the plastic deformation gradient
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
Fe_guess, & ! guess for elastic deformation gradient
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
Tstar ! 2nd Piola-Kirchhoff stress tensor
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
integer(pInt) NiterationCrystallite, & ! number of iterations in crystallite loop
NiterationState ! number of iterations in state loop
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
integer(pInt) e, ee, & ! element index
i, ii, & ! integration point index
g, gg, & ! grain index
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
k, &
l, &
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
perturbation , & ! loop counter for forward,backward perturbation mode
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
comp, &
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
myNgrains, &
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
mySizeState, &
mySizeDotState
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
integer(pInt), dimension(2,9,homogenization_maxNgrains,mesh_maxNips,mesh_NcpElems) :: &
kl
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
logical onTrack, & ! flag indicating whether we are still on track
temperatureConverged, & ! flag indicating if temperature converged
stateConverged, & ! flag indicating if state converged
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
converged ! flag indicating if iteration converged
introduced a flexibility in cut-backing scheme in homogenization.f90 and in crystallite.f90: (1) subStepSizeHomog and subStepSizeCryst := size of substep when cut-back is applied (initially was hard-coded). (2) stepIncreaseHomog and stepIncreaseCryst := step increase when calculation for substep converge (was also hardcoded). introduced a possibility to choose different finite difference scheme, i.e., forward-, backward- and central-difference, for computing grain numerical tangent. note that central-difference scheme will slow down the computation significantly. please use it only if necessary. parameters to set these new features have been included in numerics.f90 and numerics.config, whereas corresponding error messages have been introduced in the IO.f90
2009-11-10 19:06:27 +05:30
real(pReal), dimension(9,9) :: dPdF99
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
real(pReal), dimension(3,3,3,3,2) :: dPdF_perturbation
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
real(pReal) dot_prod12, &
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
dot_prod22, &
formerSubStep
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
real(pReal), dimension(3,3,homogenization_maxNgrains,mesh_maxNips,mesh_NcpElems) :: &
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
F_backup, &
Fp_backup, &
InvFp_backup, &
Fe_backup, &
Lp_backup, &
P_backup
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
real(pReal), dimension(6,homogenization_maxNgrains,mesh_maxNips,mesh_NcpElems) :: &
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
Tstar_v_backup
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
real(pReal), dimension(homogenization_maxNgrains,mesh_maxNips,mesh_NcpElems) :: &
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
Temperature_backup
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
logical, dimension(homogenization_maxNgrains,mesh_maxNips,mesh_NcpElems) :: &
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
ConvergenceFlag_backup
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
logical, dimension(3,3) :: mask
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
logical forceLocalStiffnessCalculation ! flag indicating that stiffness calculation is always done locally
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
forceLocalStiffnessCalculation = .true.
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
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
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
! ------ initialize to starting condition ------
introduced a flexibility in cut-backing scheme in homogenization.f90 and in crystallite.f90: (1) subStepSizeHomog and subStepSizeCryst := size of substep when cut-back is applied (initially was hard-coded). (2) stepIncreaseHomog and stepIncreaseCryst := step increase when calculation for substep converge (was also hardcoded). introduced a possibility to choose different finite difference scheme, i.e., forward-, backward- and central-difference, for computing grain numerical tangent. note that central-difference scheme will slow down the computation significantly. please use it only if necessary. parameters to set these new features have been included in numerics.f90 and numerics.config, whereas corresponding error messages have been introduced in the IO.f90
2009-11-10 19:06:27 +05:30
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
!$OMP CRITICAL (write2out)
! write (6,*)
! write (6,*) 'Crystallite request from Materialpoint'
! write (6,'(a,/,(f12.7,x))') 'crystallite_partionedTemp0 of 1 1 1' ,crystallite_partionedTemperature0(1,1,1)
! write (6,'(a,/,3(3(f12.7,x)/))') 'crystallite_partionedF0 of 1 1 1' ,crystallite_partionedF0(1:3,:,1,1,1)
! write (6,'(a,/,3(3(f12.7,x)/))') 'crystallite_partionedFp0 of 1 1 1' ,crystallite_partionedFp0(1:3,:,1,1,1)
! write (6,'(a,/,3(3(f12.7,x)/))') 'crystallite_partionedF of 1 1 1' ,crystallite_partionedF(1:3,:,1,1,1)
! write (6,'(a,/,3(3(f12.7,x)/))') 'crystallite_partionedLp0 of 1 1 1' ,crystallite_partionedLp0(1:3,:,1,1,1)
!$OMPEND CRITICAL (write2out)
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
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
crystallite_subStep = 0.0_pReal
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
!$OMP PARALLEL DO
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
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
myNgrains = homogenization_Ngrains(mesh_element(3,e))
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
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
do g = 1,myNgrains
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
if (crystallite_requested(g,i,e)) then ! initialize restoration point of ...
crystallite_subTemperature0(g,i,e) = crystallite_partionedTemperature0(g,i,e) ! ...temperature
constitutive_subState0(g,i,e)%p = constitutive_partionedState0(g,i,e)%p ! ...microstructure
crystallite_subFp0(:,:,g,i,e) = crystallite_partionedFp0(:,:,g,i,e) ! ...plastic def grad
crystallite_subLp0(:,:,g,i,e) = crystallite_partionedLp0(:,:,g,i,e) ! ...plastic velocity grad
crystallite_subF0(:,:,g,i,e) = crystallite_partionedF0(:,:,g,i,e) ! ...def grad
crystallite_subTstar0_v(:,g,i,e) = crystallite_partionedTstar0_v(:,g,i,e) !...2nd PK stress
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
crystallite_subFrac(g,i,e) = 0.0_pReal
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
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
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
endif
enddo
enddo
enddo
!$OMPEND PARALLEL DO
again corrections. but this version must work.
2010-03-24 21:53:21 +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
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
! --+>> crystallite loop <<+--
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
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
NiterationCrystallite = 0_pInt
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
minimum substepping size in homogenization and crystallite is now controlled by two independent parameters subStepMinHomog and subStepMinCryst respectively. deleted flush statements where they are not needed, since they seriously slow down computation.
2009-10-26 22:13:43 +05:30
do while (any(crystallite_subStep(:,:,FEsolving_execELem(1):FEsolving_execElem(2)) > subStepMinCryst)) ! cutback loop for crystallites
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
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
!$OMP PARALLEL DO
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
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
myNgrains = homogenization_Ngrains(mesh_element(3,e))
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
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
do g = 1,myNgrains
constitutive_nonlocal: - reworked contribution of immobile dislocation density for rate equations - flux is now calculated on the basis of interpolated velocities and densities at the interface; both incoming and outgoing fluxes are considered, so every material point only changes his own dotState - dislocation velocity is now globally defined and calculated by subroutine constitutive_nonlocal_kinetics; the subroutine is called inside _LpAndItsTangent as well as _microstructure; therefore, microstructure now needs Tstar_v as additional input; in the future one should perhaps create a subroutine constitutive_kinetics that calls constitutive_nonlocal_kinetics separately, to clearly distinguish between microstructural and kinetic variables - better use flux density vector as output variable instead of scalar flux values for each interface - added output variables internal and external resolved stress crystallite: - added flag to force local stiffness calculation in case of nonlocal model - misorientation angle is explicitly set to zero when no neighbor can be found debug: - added flag "selectiveDebugger" that is used when debugging statements should only affect a specific element, ip and grain; these are specified with the new variables debug_e, debug_i and debug_g - debugger can now be used in its original sense
2010-02-17 18:51:36 +05:30
selectiveDebugger = (e == debug_e .and. i == debug_i .and. g == debug_g)
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
if (crystallite_converged(g,i,e)) then
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
if (debugger .and. selectiveDebugger) then
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
!$OMP CRITICAL (write2out)
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
write(6,'(a21,f10.8,a32,f10.8,a35)') 'winding forward from ', &
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
crystallite_subFrac(g,i,e),' to current crystallite_subfrac ', &
crystallite_subFrac(g,i,e)+crystallite_subStep(g,i,e),' in crystallite_stressAndItsTangent'
write(6,*)
!$OMPEND CRITICAL (write2out)
endif
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
crystallite_subFrac(g,i,e) = crystallite_subFrac(g,i,e) + crystallite_subStep(g,i,e)
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
formerSubStep = crystallite_subStep(g,i,e)
crystallite_subStep(g,i,e) = min(1.0_pReal-crystallite_subFrac(g,i,e), stepIncreaseCryst*crystallite_subStep(g,i,e))
minimum substepping size in homogenization and crystallite is now controlled by two independent parameters subStepMinHomog and subStepMinCryst respectively. deleted flush statements where they are not needed, since they seriously slow down computation.
2009-10-26 22:13:43 +05:30
if (crystallite_subStep(g,i,e) > subStepMinCryst) then
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
crystallite_subTemperature0(g,i,e) = crystallite_Temperature(g,i,e) ! wind forward...
crystallite_subF0(:,:,g,i,e) = crystallite_subF(:,:,g,i,e) ! ...def grad
crystallite_subFp0(:,:,g,i,e) = crystallite_Fp(:,:,g,i,e) ! ...plastic def grad
crystallite_subLp0(:,:,g,i,e) = crystallite_Lp(:,:,g,i,e) ! ...plastic velocity gradient
constitutive_subState0(g,i,e)%p = constitutive_state(g,i,e)%p ! ...microstructure
crystallite_subTstar0_v(:,g,i,e) = crystallite_Tstar_v(:,g,i,e) ! ...2nd PK stress
elseif (formerSubStep > subStepMinCryst) then ! this crystallite just converged
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
!$OMP CRITICAL (distributionCrystallite)
debug_CrystalliteLoopDistribution(min(nCryst+1,NiterationCrystallite)) = &
debug_CrystalliteLoopDistribution(min(nCryst+1,NiterationCrystallite)) + 1
!$OMPEND CRITICAL (distributionCrystallite)
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
endif
else
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
crystallite_subStep(g,i,e) = subStepSizeCryst*crystallite_subStep(g,i,e) ! cut step in half and restore...
crystallite_Temperature(g,i,e) = crystallite_subTemperature0(g,i,e) ! ...temperature
crystallite_Fp(:,:,g,i,e) = crystallite_subFp0(:,:,g,i,e) ! ...plastic def grad
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
crystallite_invFp(:,:,g,i,e) = math_inv3x3(crystallite_Fp(:,:,g,i,e))
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
crystallite_Lp(:,:,g,i,e) = crystallite_subLp0(:,:,g,i,e) ! ...plastic velocity grad
constitutive_state(g,i,e)%p = constitutive_subState0(g,i,e)%p ! ...microstructure
crystallite_Tstar_v(:,g,i,e) = crystallite_subTstar0_v(:,g,i,e) ! ...2nd PK stress
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
if (debugger .and. selectiveDebugger) then
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
!$OMP CRITICAL (write2out)
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
write(6,'(a78,f10.8)') 'cutback step in crystallite_stressAndItsTangent with new crystallite_subStep: ',&
crystallite_subStep(g,i,e)
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
write(6,*)
!$OMPEND CRITICAL (write2out)
endif
endif
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
crystallite_todo(g,i,e) = crystallite_subStep(g,i,e) > subStepMinCryst ! still on track or already done (beyond repair)
if (crystallite_todo(g,i,e)) then ! specify task (according to substep)
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
crystallite_subF(:,:,g,i,e) = crystallite_subF0(:,:,g,i,e) + &
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
crystallite_subStep(g,i,e) * &
(crystallite_partionedF(:,:,g,i,e) - crystallite_partionedF0(:,:,g,i,e))
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
crystallite_Fe(:,:,g,i,e) = math_mul33x33(crystallite_subF(:,:,g,i,e),crystallite_invFp(:,:,g,i,e))
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
crystallite_subdt(g,i,e) = crystallite_subStep(g,i,e) * crystallite_dt(g,i,e)
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
crystallite_converged(g,i,e) = .false. ! start out non-converged
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
endif
enddo
enddo
enddo
!$OMPEND PARALLEL DO
! --+>> preguess for state <<+--
!
! incrementing by crystallite_subdt
! based on constitutive_subState0
! results in constitutive_state
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
! first loop for collection of state evolution based on old state
! second loop for updating to new state
adapted crystallite_stressAndItsTangent to do a pre-guess for the state before the actual state loop (with order stress integration, state update)
2009-06-10 20:38:33 +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
!$OMP PARALLEL DO
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
myNgrains = homogenization_Ngrains(mesh_element(3,e))
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
do g = 1,myNgrains
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
selectiveDebugger = (e == debug_e .and. i == debug_i .and. g == debug_g)
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
if (crystallite_todo(g,i,e)) then ! all undone crystallites
constitutive_nonlocal: - reworked contribution of immobile dislocation density for rate equations - flux is now calculated on the basis of interpolated velocities and densities at the interface; both incoming and outgoing fluxes are considered, so every material point only changes his own dotState - dislocation velocity is now globally defined and calculated by subroutine constitutive_nonlocal_kinetics; the subroutine is called inside _LpAndItsTangent as well as _microstructure; therefore, microstructure now needs Tstar_v as additional input; in the future one should perhaps create a subroutine constitutive_kinetics that calls constitutive_nonlocal_kinetics separately, to clearly distinguish between microstructural and kinetic variables - better use flux density vector as output variable instead of scalar flux values for each interface - added output variables internal and external resolved stress crystallite: - added flag to force local stiffness calculation in case of nonlocal model - misorientation angle is explicitly set to zero when no neighbor can be found debug: - added flag "selectiveDebugger" that is used when debugging statements should only affect a specific element, ip and grain; these are specified with the new variables debug_e, debug_i and debug_g - debugger can now be used in its original sense
2010-02-17 18:51:36 +05:30
call constitutive_microstructure(crystallite_Temperature(g,i,e), crystallite_Tstar_v(:,g,i,e), crystallite_Fe, &
when calculating dislocation stress at grain boundaries, densities are now extrapolated similarly to like it was already done at free surfaces
2010-06-07 21:31:37 +05:30
crystallite_Fp, crystallite_disorientation(:,:,g,i,e), g, i, e) ! update dependent state variables to be consistent with basic states
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
constitutive_previousDotState2(g,i,e)%p = 0.0_pReal
constitutive_previousDotState(g,i,e)%p = 0.0_pReal
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
constitutive_dotState(g,i,e)%p = 0.0_pReal ! zero out dotStates
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
endif
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
enddo; enddo; enddo
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
!$OMPEND PARALLEL DO
!$OMP PARALLEL DO
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
myNgrains = homogenization_Ngrains(mesh_element(3,e))
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
do g = 1,myNgrains
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
selectiveDebugger = (e == debug_e .and. i == debug_i .and. g == debug_g)
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
if (crystallite_todo(g,i,e)) then ! all undone crystallites
constitutive_nonlocal: - dipole dislocations with evolution crystallite.f90: - collect state uses subdt and subTstar0_v - in nonlocal modus: set all crystallites to broken if one is not on track anymore after either stress integration or state update - constitutive_microstructure is now called inside state update and not in integrate_stress anymore material.config: - new parameter for nonlocal constitution CPFEM.f90: - age Tstar after increment was finished
2009-08-28 19:20:47 +05:30
call constitutive_collectDotState(crystallite_Tstar_v(:,g,i,e), crystallite_subTstar0_v(:,g,i,e), &
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
crystallite_Fe, crystallite_Fp, crystallite_Temperature(g,i,e), &
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
crystallite_disorientation(:,:,g,i,e), crystallite_subdt(g,i,e), g, i, e)
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
endif
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
enddo; enddo; enddo
!$OMPEND PARALLEL DO
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
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
crystallite_statedamper = 1.0_pReal
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
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
!$OMP PARALLEL DO
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
myNgrains = homogenization_Ngrains(mesh_element(3,e))
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
do g = 1,myNgrains
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
selectiveDebugger = (e == debug_e .and. i == debug_i .and. g == debug_g)
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
if (crystallite_todo(g,i,e)) then ! all undone crystallites
crystallite_stateConverged(g,i,e) = crystallite_updateState(g,i,e) ! update state
crystallite_temperatureConverged(g,i,e) = crystallite_updateTemperature(g,i,e) ! update temperature
if ( .not. crystallite_localConstitution(g,i,e) &
.and. .not. crystallite_todo(g,i,e)) & ! if broken non-local...
crystallite_todo = crystallite_todo .and. crystallite_localConstitution ! ...all non-locals skipped
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
endif
enddo; enddo; 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
!$OMPEND PARALLEL DO
adapted crystallite_stressAndItsTangent to do a pre-guess for the state before the actual state loop (with order stress integration, state update)
2009-06-10 20:38:33 +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
! --+>> state loop <<+--
adapted crystallite_stressAndItsTangent to do a pre-guess for the state before the actual state loop (with order stress integration, state update)
2009-06-10 20:38:33 +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
NiterationState = 0_pInt
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
constitutive_nonlocal: - dipole dislocations with evolution crystallite.f90: - collect state uses subdt and subTstar0_v - in nonlocal modus: set all crystallites to broken if one is not on track anymore after either stress integration or state update - constitutive_microstructure is now called inside state update and not in integrate_stress anymore material.config: - new parameter for nonlocal constitution CPFEM.f90: - age Tstar after increment was finished
2009-08-28 19:20:47 +05:30
do while ( any(crystallite_todo(:,:,FEsolving_execELem(1):FEsolving_execElem(2))) &
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
.and. NiterationState < nState) ! convergence loop for crystallite
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
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
NiterationState = NiterationState + 1_pInt
! --+>> stress integration <<+--
!
! incrementing by crystallite_subdt
! based on crystallite_subF0,.._subFp0,.._subLp0
! constitutive_state is internally interpolated with .._subState0
! to account for substepping within _integrateStress
! results in crystallite_Fp,.._Lp
!$OMP PARALLEL DO
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
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
myNgrains = homogenization_Ngrains(mesh_element(3,e))
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
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
do g = 1,myNgrains
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
selectiveDebugger = (e == debug_e .and. i == debug_i .and. g == debug_g)
- corrected an if statement in the state loop - nonlocal stiffness calculation: we perturb all material points at the same time, so instead of N^2 loops we just need N - set "forceLocalStiffnessCalculation" to false as standard
2010-03-04 22:57:39 +05:30
if (crystallite_todo(g,i,e)) then ! all undone crystallites
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
crystallite_todo(g,i,e) = crystallite_integrateStress(1_pInt,g,i,e) ! mode: central solution
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
if ( .not. crystallite_localConstitution(g,i,e) &
.and. .not. crystallite_todo(g,i,e)) & ! if broken non-local...
crystallite_todo = crystallite_todo .and. crystallite_localConstitution ! ...all non-locals skipped
- corrected an if statement in the state loop - nonlocal stiffness calculation: we perturb all material points at the same time, so instead of N^2 loops we just need N - set "forceLocalStiffnessCalculation" to false as standard
2010-03-04 22:57:39 +05:30
endif
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
enddo; enddo; 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
!$OMPEND PARALLEL DO
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
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
if (verboseDebugger) 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
!$OMP CRITICAL (write2out)
write(6,*) count(crystallite_todo(:,:,:)),'grains todo after stress integration'
!$OMPEND CRITICAL (write2out)
endif
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
! --+>> state integration <<+--
!
! incrementing by crystallite_subdt
! based on constitutive_subState0
! results in constitutive_state
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
! first loop for collection of state evolution based on old state
! second loop for updating to new state
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
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
!$OMP PARALLEL DO
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
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
myNgrains = homogenization_Ngrains(mesh_element(3,e))
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
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
do g = 1,myNgrains
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
if (crystallite_todo(g,i,e)) then ! all undone crystallites
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
constitutive_previousDotState2(g,i,e)%p = constitutive_previousDotState(g,i,e)%p
constitutive_previousDotState(g,i,e)%p = constitutive_dotState(g,i,e)%p
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
constitutive_dotState(g,i,e)%p = 0.0_pReal ! zero out dotState
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
endif
enddo; enddo; enddo
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
!$OMPEND PARALLEL DO
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
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
crystallite_statedamper = 1.0_pReal
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
!$OMP PARALLEL DO
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
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
myNgrains = homogenization_Ngrains(mesh_element(3,e))
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
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
do g = 1,myNgrains
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
selectiveDebugger = (e == debug_e .and. i == debug_i .and. g == debug_g)
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
if (crystallite_todo(g,i,e)) then ! all undone crystallites
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
call constitutive_collectDotState(crystallite_Tstar_v(:,g,i,e), crystallite_subTstar0_v(:,g,i,e), &
crystallite_Fe, crystallite_Fp, crystallite_Temperature(g,i,e), &
corrected syntax errors (long lines, line continuation by \) and logical mistake in mpie_cpfem_marc / abq_std which went unnoticed in ifort. Using SunStudio f90 surfaced those...
2010-08-04 05:17:00 +05:30
crystallite_disorientation(:,:,g,i,e), crystallite_subdt(g,i,e), g, i, e)
calculation of state damper was erroneous for use with nonlocal and other constitution at the same time
2010-08-26 18:24:31 +05:30
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 )
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 )
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
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) ) &
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
crystallite_statedamper(g,i,e) = 0.75_pReal + 0.25_pReal * tanh(2.0_pReal + 4.0_pReal * dot_prod12 / dot_prod22)
endif
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
enddo; enddo; enddo
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
!$OMPEND PARALLEL DO
homogenization_RGC.f90 >> just switching-off one of the debugging statement. crystallite.f90 >> optimizing the perturbation algorithm for local tangent.
2009-12-22 17:58:02 +05:30
Tstar_v is now restored at different stages of cutbacking; we need this because preguess of state relies on consistent Tstar_v
2009-06-16 14:33:30 +05:30
!$OMP PARALLEL DO
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
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
myNgrains = homogenization_Ngrains(mesh_element(3,e))
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
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
do g = 1,myNgrains
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
selectiveDebugger = (e == debug_e .and. i == debug_i .and. g == debug_g)
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
if (crystallite_todo(g,i,e)) then ! all undone crystallites
crystallite_stateConverged(g,i,e) = crystallite_updateState(g,i,e) ! update state
crystallite_temperatureConverged(g,i,e) = crystallite_updateTemperature(g,i,e) ! update temperature
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
crystallite_converged(g,i,e) = crystallite_stateConverged(g,i,e) .and. crystallite_temperatureConverged(g,i,e)
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
if ( .not. crystallite_localConstitution(g,i,e) &
.and. .not. crystallite_todo(g,i,e)) & ! if updateState signals broken non-local...
crystallite_todo = crystallite_todo .and. crystallite_localConstitution ! ...all non-locals skipped
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
if (crystallite_converged(g,i,e)) then
!$OMP CRITICAL (distributionState)
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
debug_CrystalliteStateLoopDistribution(NiterationState) = &
debug_CrystalliteStateLoopDistribution(NiterationState) + 1
Tstar_v is now restored at different stages of cutbacking; we need this because preguess of state relies on consistent Tstar_v
2009-06-16 14:33:30 +05:30
!$OMPEND CRITICAL (distributionState)
endif
endif
enddo
enddo
enddo
!$OMPEND PARALLEL DO
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
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
if (verboseDebugger) 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
!$OMP CRITICAL (write2out)
write(6,*) count(crystallite_converged(:,:,:)),'grains converged after state integration no.', NiterationState
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
write(6,*)
constitutive_nonlocal: - reworked contribution of immobile dislocation density for rate equations - flux is now calculated on the basis of interpolated velocities and densities at the interface; both incoming and outgoing fluxes are considered, so every material point only changes his own dotState - dislocation velocity is now globally defined and calculated by subroutine constitutive_nonlocal_kinetics; the subroutine is called inside _LpAndItsTangent as well as _microstructure; therefore, microstructure now needs Tstar_v as additional input; in the future one should perhaps create a subroutine constitutive_kinetics that calls constitutive_nonlocal_kinetics separately, to clearly distinguish between microstructural and kinetic variables - better use flux density vector as output variable instead of scalar flux values for each interface - added output variables internal and external resolved stress crystallite: - added flag to force local stiffness calculation in case of nonlocal model - misorientation angle is explicitly set to zero when no neighbor can be found debug: - added flag "selectiveDebugger" that is used when debugging statements should only affect a specific element, ip and grain; these are specified with the new variables debug_e, debug_i and debug_g - debugger can now be used in its original sense
2010-02-17 18:51:36 +05:30
! write(6,'(8(L,x))') crystallite_converged(:,:,:)
! do e = FEsolving_execElem(1),FEsolving_execElem(2)
! if (any(.not. crystallite_converged(:,:,e))) &
! write(6,'(i4,8(x,L))') e, crystallite_converged(:,:,e)
! enddo
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
!$OMPEND CRITICAL (write2out)
endif
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
if (any(.not. crystallite_converged .and. .not. crystallite_localConstitution)) & ! any non-local not yet converged (or broken)...
crystallite_converged = crystallite_converged .and. crystallite_localConstitution ! ...restart all non-local as not converged
- corrected an if statement in the state loop - nonlocal stiffness calculation: we perturb all material points at the same time, so instead of N^2 loops we just need N - set "forceLocalStiffnessCalculation" to false as standard
2010-03-04 22:57:39 +05:30
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
crystallite_todo = crystallite_todo .and. .not. crystallite_converged ! skip all converged
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
if (verboseDebugger) then
constitutive_nonlocal: - read in activation energy for dislocation glide from material.config - changed naming of dDipMin/Max to dLower/dUpper - added new outputs: rho_dot, rho_dot_dip, rho_dot_gen, rho_dot_sgl2dip, rho_dot_dip2sgl, rho_dot_ann_ath, rho_dot_ann_the, rho_dot_flux, d_upper_edge, d_upper_screw, d_upper_dot_edge, d_upper_dot_screw - poisson's ratio is now calculated from elastic constants - microstrucutre has state as first argument, since this is our output variable - periodic boundary conditions are taken into account for fluxes and internal stresses. for the moment, flag has to be set in constitutive_nonlocal. - corrected calculation for dipole formation by glide - added terms for dipole formation/annihilation by stress decrease/increase constitutive: - passing of arguments is adapted for constitutive_nonlocal model crystallite: - in stiffness calculation: call to collect_dotState used wrong arguments - crystallite_postResults uses own Tstar_v and temperature, no need for passing them from materialpoint_postResults homogenization: - crystallite_postResults uses own Tstar_v and temperature, no need for passing them from materialpoint_postResults IO: - changed error message 229 material.config: - changed example for nonlocal constitution according to constitutive_nonlocal all: - added some flush statements
2009-10-20 20:06:03 +05:30
!$OMP CRITICAL (write2out)
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
write(6,*) count(crystallite_converged(:,:,:)),'grains converged after non-local check'
write(6,*) count(crystallite_todo(:,:,:)),'grains todo after state integration no.', NiterationState
constitutive_nonlocal: - read in activation energy for dislocation glide from material.config - changed naming of dDipMin/Max to dLower/dUpper - added new outputs: rho_dot, rho_dot_dip, rho_dot_gen, rho_dot_sgl2dip, rho_dot_dip2sgl, rho_dot_ann_ath, rho_dot_ann_the, rho_dot_flux, d_upper_edge, d_upper_screw, d_upper_dot_edge, d_upper_dot_screw - poisson's ratio is now calculated from elastic constants - microstrucutre has state as first argument, since this is our output variable - periodic boundary conditions are taken into account for fluxes and internal stresses. for the moment, flag has to be set in constitutive_nonlocal. - corrected calculation for dipole formation by glide - added terms for dipole formation/annihilation by stress decrease/increase constitutive: - passing of arguments is adapted for constitutive_nonlocal model crystallite: - in stiffness calculation: call to collect_dotState used wrong arguments - crystallite_postResults uses own Tstar_v and temperature, no need for passing them from materialpoint_postResults homogenization: - crystallite_postResults uses own Tstar_v and temperature, no need for passing them from materialpoint_postResults IO: - changed error message 229 material.config: - changed example for nonlocal constitution according to constitutive_nonlocal all: - added some flush statements
2009-10-20 20:06:03 +05:30
write(6,*)
!$OMPEND CRITICAL (write2out)
endif
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
enddo ! crystallite convergence loop
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
NiterationCrystallite = NiterationCrystallite + 1
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
enddo ! cutback loop
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +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
! ------ check for non-converged crystallites ------
!$OMP PARALLEL DO
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
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
myNgrains = homogenization_Ngrains(mesh_element(3,e))
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
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
do g = 1,myNgrains
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
if (.not. crystallite_converged(g,i,e)) then ! respond fully elastically (might be not required due to becoming terminally ill anyway)
homogenization_RGC.f90 >>> adding some lines, mostly for debugging purpose. no critical change. constitutive_phenopowerlaw.f90 >>> adding new parameter: constitutive_phenopowerlaw_w0_slip, i.e., the hardening rate exponent. homogenization.f90 >>> most important change is to add an if-else statement (line 379-380) to switch crystallite_requeted = .false. for already converged material point iteration (el/ip). the rest of the changes are cosmetics and debugging stuffs. crystallite.f90 >>> similar to homogenization.f90, the most important change is to add additional if-else statement (line 574) in the jacobian (perturbation) loop. now the jacobian calculation will only be performed when crystallite_requested = .true.. the rest is only cosmetic.
2009-08-27 17:40:06 +05:30
! call IO_warning(600,e,i,g)
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
invFp = math_inv3x3(crystallite_partionedFp0(:,:,g,i,e))
Fe_guess = math_mul33x33(crystallite_partionedF(:,:,g,i,e),invFp)
detabbed file for better readability
2010-02-19 19:14:38 +05:30
Tstar = math_Mandel6to33( math_mul66x6( 0.5_pReal*constitutive_homogenizedC(g,i,e), &
math_Mandel33to6( math_mul33x33(transpose(Fe_guess),Fe_guess) - math_I3 ) ) )
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
crystallite_P(:,:,g,i,e) = math_mul33x33(Fe_guess,math_mul33x33(Tstar,transpose(invFp)))
endif
enddo
enddo
enddo
!$OMPEND PARALLEL DO
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
! --+>> stiffness calculation <<+--
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
if(updateJaco) then ! Jacobian required
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
crystallite_statedamper = 1.0_pReal
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
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
myNgrains = homogenization_Ngrains(mesh_element(3,e))
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e) ! iterate over IPs of this element to be processed
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
do g = 1,myNgrains
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
mySizeState = constitutive_sizeState(g,i,e) ! number of state variables for this grain
mySizeDotState = constitutive_sizeDotState(g,i,e) ! number of dotStates for this grain
constitutive_state_backup(g,i,e)%p(1:mySizeState) = &
constitutive_state(g,i,e)%p(1:mySizeState) ! remember unperturbed, converged state, ...
constitutive_dotState_backup(g,i,e)%p(1:mySizeDotState) = &
constitutive_dotState(g,i,e)%p(1:mySizeDotState) ! ... dotStates, ...
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
enddo; enddo; enddo
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
Temperature_backup = crystallite_Temperature ! ... Temperature, ...
F_backup = crystallite_subF ! ... and kinematics
Fp_backup = crystallite_Fp
InvFp_backup = crystallite_invFp
Fe_backup = crystallite_Fe
Lp_backup = crystallite_Lp
Tstar_v_backup = crystallite_Tstar_v
P_backup = crystallite_P
ConvergenceFlag_backup = crystallite_converged
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
nonlocal stiffness calculation produced segmentation fault for cycle number > 16. Corrected calculation of perturbation indices k and l.
2010-03-10 15:23:41 +05:30
if (all(crystallite_localConstitution) .or. theInc < 1 .or. forceLocalStiffnessCalculation) then ! all grains have local constitution, so local convergence of perturbed grain is sufficient
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
!$OMP PARALLEL DO
do e = FEsolving_execElem(1),FEsolving_execElem(2) ! iterate over elements to be processed
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
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
selectiveDebugger = (e == debug_e .and. i == debug_i .and. g == debug_g)
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
if (crystallite_requested(g,i,e)) then ! first check whether is requested at all!
if (crystallite_converged(g,i,e)) then ! grain converged in above iteration
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
if (debugger .and. selectiveDebugger) 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
!$OMP CRITICAL (write2out)
write (6,*) '#############'
write (6,*) 'central solution of cryst_StressAndTangent'
write (6,*) '#############'
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
write (6,'(a8,3(x,i4),/,3(3(f12.4,x)/))') ' P of', g, i, e, P_backup(1:3,:,g,i,e)/1e6
write (6,'(a8,3(x,i4),/,3(3(f14.9,x)/))') ' Fp of', g, i, e, Fp_backup(1:3,:,g,i,e)
write (6,'(a8,3(x,i4),/,3(3(f14.9,x)/))') ' Lp of', g, i, e, Lp_backup(1:3,:,g,i,e)
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
!$OMPEND CRITICAL (write2out)
endif
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
do perturbation = 1,2 ! forward and backward perturbation
if (iand(pert_method,perturbation) > 0) 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
crystallite_subF(k,l,g,i,e) = crystallite_subF(k,l,g,i,e) + myPert ! perturb either forward or backward
if (debugger .and. selectiveDebugger) 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
!$OMP CRITICAL (write2out)
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
write (6,'(i1,x,i1)') k,l
write (6,'(a8,3(x,i4),/,3(3(f14.9,x)/))') 'pertF of', g, i, e, crystallite_subF(1:3,:,g,i,e)
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
!$OMPEND CRITICAL (write2out)
endif
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
onTrack = .true.
converged = .false.
NiterationState = 0_pInt
do while(.not. converged .and. onTrack .and. NiterationState < nState) ! keep cycling until done (potentially non-converged)
NiterationState = NiterationState + 1_pInt
onTrack = crystallite_integrateStress(2_pInt,g,i,e) ! stress of perturbed situation (overwrites _P,_Tstar_v,_Fp,_Lp,_Fe) with mode:2
if (onTrack) then
constitutive_dotState(g,i,e)%p = 0.0_pReal
call constitutive_collectDotState(crystallite_Tstar_v(:,g,i,e), crystallite_subTstar0_v(:,g,i,e), &
crystallite_Fe, crystallite_Fp, crystallite_Temperature(g,i,e), &
crystallite_disorientation(:,:,g,i,e), crystallite_subdt(g,i,e), &
g,i,e)
stateConverged = crystallite_updateState(g,i,e) ! update state
temperatureConverged = crystallite_updateTemperature(g,i,e) ! update temperature
converged = stateConverged .and. temperatureConverged
endif
if (debugger .and. selectiveDebugger) then
!$OMP CRITICAL (write2out)
write (6,*) '-------------'
write (6,'(a,x,l,x,l)') 'ontrack + converged:',onTrack,converged
write (6,'(a12,3(x,i4),/,3(3(f12.4,x)/))') 'pertP/MPa of', g, i, e, crystallite_P(1:3,:,g,i,e)/1e6
write (6,'(a12,3(x,i4),/,3(3(f12.4,x)/))') 'DP/MPa of', g, i, e, &
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
(crystallite_P(1:3,:,g,i,e)-P_backup(1:3,:,g,i,e))/1e6
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
!$OMPEND CRITICAL (write2out)
endif
enddo
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
if (converged) & ! converged state warrants stiffness update
dPdF_perturbation(:,:,k,l,perturbation) = (crystallite_P(:,:,g,i,e) - P_backup(:,:,g,i,e))/myPert ! tangent dP_ij/dFg_kl
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
mySizeState = constitutive_sizeState(g,i,e) ! number of state variables for this grain
mySizeDotState = constitutive_sizeDotState(g,i,e) ! number of dotStates for this grain
constitutive_state(g,i,e)%p(1:mySizeState) = constitutive_state_backup(g,i,e)%p(1:mySizeState)
constitutive_dotState(g,i,e)%p(1:mySizeDotState) = constitutive_dotState_backup(g,i,e)%p(1:mySizeDotState)
crystallite_Temperature(g,i,e) = Temperature_backup(g,i,e)
crystallite_subF(:,:,g,i,e) = F_backup(:,:,g,i,e)
crystallite_Fp(:,:,g,i,e) = Fp_backup(:,:,g,i,e)
crystallite_invFp(:,:,g,i,e) = InvFp_backup(:,:,g,i,e)
crystallite_Fe(:,:,g,i,e) = Fe_backup(:,:,g,i,e)
crystallite_Lp(:,:,g,i,e) = Lp_backup(:,:,g,i,e)
crystallite_Tstar_v(:,g,i,e) = Tstar_v_backup(:,g,i,e)
crystallite_P(:,:,g,i,e) = P_backup(:,:,g,i,e)
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
!$OMP CRITICAL (out)
debug_StiffnessStateLoopDistribution(NiterationState) = &
debug_StiffnessstateLoopDistribution(NiterationState) + 1
!$OMPEND CRITICAL (out)
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
enddo; enddo
endif
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
enddo ! perturbation direction
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
select case(pert_method)
case (1)
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
crystallite_dPdF(:,:,:,:,g,i,e) = dPdF_perturbation(:,:,:,:,1)
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
case (2)
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
crystallite_dPdF(:,:,:,:,g,i,e) = dPdF_perturbation(:,:,:,:,2)
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
case (3)
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
crystallite_dPdF(:,:,:,:,g,i,e) = 0.5_pReal*(dPdF_perturbation(:,:,:,:,1)+dPdF_perturbation(:,:,:,:,2))
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
end select
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
else ! grain did not converge
crystallite_dPdF(:,:,:,:,g,i,e) = crystallite_fallbackdPdF(:,:,:,:,g,i,e) ! use (elastic) fallback
endif ! grain convergence
endif ! grain request
enddo ! grain loop
enddo ! ip loop
enddo ! element loop
!$OMPEND PARALLEL DO
again corrections. but this version must work.
2010-03-24 21:53:21 +05:30
elseif (any(.not. crystallite_localConstitution)) then ! if any nonlocal grain present, we have to do a full loop over all grains after each perturbance
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
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
do k = 1,3
do l = 1,3
crystallite_subF(k,l,:,:,:) = crystallite_subF(k,l,:,:,:) + pert_Fg ! perturb single component
NiterationState = 0_pInt
crystallite_todo = .true.
do while ( any(crystallite_todo(:,:,FEsolving_execELem(1):FEsolving_execElem(2))) .and. NiterationState < nState)
NiterationState = NiterationState + 1_pInt
!$OMP PARALLEL DO
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
selectiveDebugger = (e == debug_e .and. i == debug_i .and. g == debug_g)
if (crystallite_todo(g,i,e)) then
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
crystallite_todo(g,i,e) = crystallite_integrateStress(2_pInt,g,i,e) ! stress integration for stiffness (mode:2)
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
if ( .not. crystallite_localConstitution(g,i,e) &
corrected syntax errors (long lines, line continuation by \) and logical mistake in mpie_cpfem_marc / abq_std which went unnoticed in ifort. Using SunStudio f90 surfaced those...
2010-08-04 05:17:00 +05:30
.and. .not. crystallite_todo(g,i,e)) & ! if broken non-local...
crystallite_todo = crystallite_todo .and. crystallite_localConstitution ! ...all non-locals skipped
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
endif
enddo; enddo; enddo
!$OMPEND PARALLEL DO
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
if (crystallite_todo(g,i,e)) &
corrected syntax errors (long lines, line continuation by \) and logical mistake in mpie_cpfem_marc / abq_std which went unnoticed in ifort. Using SunStudio f90 surfaced those...
2010-08-04 05:17:00 +05:30
constitutive_dotState(g,i,e)%p = 0.0_pReal ! zero out dotState
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
enddo; enddo; enddo
- corrected an if statement in the state loop - nonlocal stiffness calculation: we perturb all material points at the same time, so instead of N^2 loops we just need N - set "forceLocalStiffnessCalculation" to false as standard
2010-03-04 22:57:39 +05:30
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
crystallite_statedamper = 1.0_pReal
!$OMP PARALLEL DO
do e = FEsolving_execElem(1),FEsolving_execElem(2)
myNgrains = homogenization_Ngrains(mesh_element(3,e))
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
do g = 1,myNgrains
selectiveDebugger = (e == debug_e .and. i == debug_i .and. g == debug_g)
if (crystallite_todo(g,i,e)) then
call constitutive_collectDotState(crystallite_Tstar_v(:,g,i,e), crystallite_subTstar0_v(:,g,i,e), &
crystallite_Fe, crystallite_Fp, crystallite_Temperature(g,i,e), &
crystallite_disorientation(:,:,g,i,e), crystallite_subdt(g,i,e), &
g,i,e) ! collect dot state
calculation of state damper was erroneous for use with nonlocal and other constitution at the same time
2010-08-26 18:24:31 +05:30
dot_prod12 = dot_product( constitutive_dotState(g,i,e)%p - constitutive_previousDotState(g,i,e)%p, &
constitutive_previousDotState(g,i,e)%p - constitutive_previousDotState2(g,i,e)%p )
dot_prod22 = dot_product( constitutive_previousDotState(g,i,e)%p - constitutive_previousDotState2(g,i,e)%p, &
constitutive_previousDotState(g,i,e)%p - constitutive_previousDotState2(g,i,e)%p )
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
if ( dot_prod22 > 0.0_pReal &
.and. ( dot_prod12 < 0.0_pReal &
.or. dot_product(constitutive_dotState(g,i,e)%p, &
constitutive_previousDotState(g,i,e)%p) < 0.0_pReal) ) &
crystallite_statedamper(g,i,e) = 0.75_pReal &
+ 0.25_pReal * tanh(2.0_pReal + 4.0_pReal * dot_prod12 / dot_prod22)
endif
enddo; enddo; enddo
!$OMPEND PARALLEL DO
!$OMP PARALLEL DO
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
selectiveDebugger = (e == debug_e .and. i == debug_i .and. g == debug_g)
if (crystallite_todo(g,i,e)) then
corrected syntax errors (long lines, line continuation by \) and logical mistake in mpie_cpfem_marc / abq_std which went unnoticed in ifort. Using SunStudio f90 surfaced those...
2010-08-04 05:17:00 +05:30
crystallite_stateConverged(g,i,e) = crystallite_updateState(g,i,e) ! update state
crystallite_temperatureConverged(g,i,e) = crystallite_updateTemperature(g,i,e) ! update temperature
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
crystallite_converged(g,i,e) = crystallite_stateConverged(g,i,e) &
.and. crystallite_temperatureConverged(g,i,e)
if ( .not. crystallite_localConstitution(g,i,e) &
corrected syntax errors (long lines, line continuation by \) and logical mistake in mpie_cpfem_marc / abq_std which went unnoticed in ifort. Using SunStudio f90 surfaced those...
2010-08-04 05:17:00 +05:30
.and. .not. crystallite_todo(g,i,e)) & ! if updateState signals broken non-local...
crystallite_todo = crystallite_todo .and. crystallite_localConstitution ! ...all non-locals skipped
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
endif
enddo; enddo; enddo
!$OMPEND PARALLEL DO
corrected syntax errors (long lines, line continuation by \) and logical mistake in mpie_cpfem_marc / abq_std which went unnoticed in ifort. Using SunStudio f90 surfaced those...
2010-08-04 05:17:00 +05:30
if (any(.not. crystallite_converged .and. .not. crystallite_localConstitution)) & ! any non-local not yet converged?
crystallite_converged = crystallite_converged .and. crystallite_localConstitution ! all non-local not converged
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
corrected syntax errors (long lines, line continuation by \) and logical mistake in mpie_cpfem_marc / abq_std which went unnoticed in ifort. Using SunStudio f90 surfaced those...
2010-08-04 05:17:00 +05:30
crystallite_todo = crystallite_todo .and. .not. crystallite_converged ! skip all converged
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
enddo
- corrected an if statement in the state loop - nonlocal stiffness calculation: we perturb all material points at the same time, so instead of N^2 loops we just need N - set "forceLocalStiffnessCalculation" to false as standard
2010-03-04 22:57:39 +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
corrected syntax errors (long lines, line continuation by \) and logical mistake in mpie_cpfem_marc / abq_std which went unnoticed in ifort. Using SunStudio f90 surfaced those...
2010-08-04 05:17:00 +05:30
if (crystallite_converged(g,i,e)) then ! if stiffness calculation converged...
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
crystallite_dPdF(:,:,k,l,g,i,e) = (crystallite_P(:,:,g,i,e) - P_backup(:,:,g,i,e))/pert_Fg ! ... use tangent dP_ij/dFg_kl
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
!$OMP CRITICAL (out)
debug_StiffnessStateLoopDistribution(NiterationState) = &
debug_StiffnessstateLoopDistribution(NiterationState) + 1
!$OMPEND CRITICAL (out)
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
elseif (.not. ConvergenceFlag_backup(g,i,e)) then ! if crystallite didn<64>t converge before...
corrected syntax errors (long lines, line continuation by \) and logical mistake in mpie_cpfem_marc / abq_std which went unnoticed in ifort. Using SunStudio f90 surfaced those...
2010-08-04 05:17:00 +05:30
crystallite_dPdF(:,:,:,:,g,i,e) = crystallite_fallbackdPdF(:,:,:,:,g,i,e) ! ... use (elastic) fallback
- corrected an if statement in the state loop - nonlocal stiffness calculation: we perturb all material points at the same time, so instead of N^2 loops we just need N - set "forceLocalStiffnessCalculation" to false as standard
2010-03-04 22:57:39 +05:30
endif
enddo; enddo; enddo
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
- corrected an if statement in the state loop - nonlocal stiffness calculation: we perturb all material points at the same time, so instead of N^2 loops we just need N - set "forceLocalStiffnessCalculation" to false as standard
2010-03-04 22:57:39 +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
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
mySizeState = constitutive_sizeState(g,i,e)
mySizeDotState = constitutive_sizeDotState(g,i,e)
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
constitutive_state(g,i,e)%p(1:mySizeState) = constitutive_state_backup(g,i,e)%p(1:mySizeState)
constitutive_dotState(g,i,e)%p(1:mySizeDotState) = constitutive_dotState_backup(g,i,e)%p(1:mySizeDotState)
- corrected an if statement in the state loop - nonlocal stiffness calculation: we perturb all material points at the same time, so instead of N^2 loops we just need N - set "forceLocalStiffnessCalculation" to false as standard
2010-03-04 22:57:39 +05:30
enddo; enddo; enddo
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
crystallite_Temperature = Temperature_backup
crystallite_subF = F_backup
crystallite_Fp = Fp_backup
crystallite_invFp = InvFp_backup
crystallite_Fe = Fe_backup
crystallite_Lp = Lp_backup
crystallite_Tstar_v = Tstar_v_backup
crystallite_P = P_backup
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
leaner memory requirement to store states and their rates when calculating the crystallite stiffness
2010-09-13 14:43:25 +05:30
crystallite_converged = ConvergenceFlag_backup
- corrected an if statement in the state loop - nonlocal stiffness calculation: we perturb all material points at the same time, so instead of N^2 loops we just need N - set "forceLocalStiffnessCalculation" to false as standard
2010-03-04 22:57:39 +05:30
nonlocal stiffness calculation: rather perturb all components at once (and optionally decrease the frequency of the Jacobian update with the iJaco parameter) than perturbing only a single component per cycle
2010-06-17 12:02:56 +05:30
enddo;enddo ! k,l loop
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
endif
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
homogenization_RGC.f90 >>> adding some lines, mostly for debugging purpose. no critical change. constitutive_phenopowerlaw.f90 >>> adding new parameter: constitutive_phenopowerlaw_w0_slip, i.e., the hardening rate exponent. homogenization.f90 >>> most important change is to add an if-else statement (line 379-380) to switch crystallite_requeted = .false. for already converged material point iteration (el/ip). the rest of the changes are cosmetics and debugging stuffs. crystallite.f90 >>> similar to homogenization.f90, the most important change is to add additional if-else statement (line 574) in the jacobian (perturbation) loop. now the jacobian calculation will only be performed when crystallite_requested = .true.. the rest is only cosmetic.
2009-08-27 17:40:06 +05:30
endif ! jacobian calculation
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
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
endsubroutine
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
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
!********************************************************************
! update the internal state of the constitutive law
! and tell whether state has converged
!********************************************************************
function crystallite_updateState(&
g,& ! grain number
i,& ! integration point number
e & ! element number
)
!*** variables and functions from other modules ***!
use prec, only: pReal, &
pInt, &
pLongInt
use numerics, only: rTol_crystalliteState
use constitutive, only: constitutive_dotState, &
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
constitutive_previousDotState, &
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
constitutive_sizeDotState, &
constitutive_subState0, &
constitutive_nonlocal: - dipole dislocations with evolution crystallite.f90: - collect state uses subdt and subTstar0_v - in nonlocal modus: set all crystallites to broken if one is not on track anymore after either stress integration or state update - constitutive_microstructure is now called inside state update and not in integrate_stress anymore material.config: - new parameter for nonlocal constitution CPFEM.f90: - age Tstar after increment was finished
2009-08-28 19:20:47 +05:30
constitutive_state, &
convergence of state in crystallite is now tested as follows: (state < relevant state) or (residuum < relative tolerance * state) since the relevant value for the state variables depend on their nature and can vary by large scales (e.g. volume fraction: 1e-10, dislocation density: 1e5) it is not possible to set a unique value. instead the constitutive law has to decide what is relevant. therefore, all constitutive laws now read in parameters from the material.config that determine the values for relevantState [@luc: in dislobased law relevant State is for the moment generally set to 1e-200, so no additional parameters necessary in material.config. if you also want this feature, we can still implement it, no big deal] - added sanity checks in constitutive_nonlocal.f90 - corrected coordinate transformation for backstress calculation in constitutive_nonlocal.f90 - corrected equations for evolution of dipole dislocation densities (athermal annihilation and formation by glide)
2009-09-18 21:07:14 +05:30
constitutive_relevantState, &
constitutive_nonlocal: - dipole dislocations with evolution crystallite.f90: - collect state uses subdt and subTstar0_v - in nonlocal modus: set all crystallites to broken if one is not on track anymore after either stress integration or state update - constitutive_microstructure is now called inside state update and not in integrate_stress anymore material.config: - new parameter for nonlocal constitution CPFEM.f90: - age Tstar after increment was finished
2009-08-28 19:20:47 +05:30
constitutive_microstructure
constitutive_nonlocal: - reworked contribution of immobile dislocation density for rate equations - flux is now calculated on the basis of interpolated velocities and densities at the interface; both incoming and outgoing fluxes are considered, so every material point only changes his own dotState - dislocation velocity is now globally defined and calculated by subroutine constitutive_nonlocal_kinetics; the subroutine is called inside _LpAndItsTangent as well as _microstructure; therefore, microstructure now needs Tstar_v as additional input; in the future one should perhaps create a subroutine constitutive_kinetics that calls constitutive_nonlocal_kinetics separately, to clearly distinguish between microstructural and kinetic variables - better use flux density vector as output variable instead of scalar flux values for each interface - added output variables internal and external resolved stress crystallite: - added flag to force local stiffness calculation in case of nonlocal model - misorientation angle is explicitly set to zero when no neighbor can be found debug: - added flag "selectiveDebugger" that is used when debugging statements should only affect a specific element, ip and grain; these are specified with the new variables debug_e, debug_i and debug_g - debugger can now be used in its original sense
2010-02-17 18:51:36 +05:30
use debug, only: debugger, &
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
selectiveDebugger, &
verboseDebugger
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
use FEsolving, only: cycleCounter, theInc
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
!*** input variables ***!
integer(pInt), intent(in):: e, & ! element index
i, & ! integration point index
g ! grain index
!*** output variables ***!
logical crystallite_updateState ! flag indicating if integration suceeded
!*** local variables ***!
real(pReal), dimension(constitutive_sizeDotState(g,i,e)) :: residuum ! residuum from evolution of microstructure
integer(pInt) mySize
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
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
mySize = constitutive_sizeDotState(g,i,e)
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
! correct my dotState
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
constitutive_dotState(g,i,e)%p(1:mySize) = constitutive_dotState(g,i,e)%p(1:mySize) * crystallite_statedamper(g,i,e) &
+ constitutive_previousDotState(g,i,e)%p(1:mySize) * (1.0_pReal-crystallite_statedamper(g,i,e))
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
! calculate the residuum
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
residuum = constitutive_state(g,i,e)%p(1:mySize) - constitutive_subState0(g,i,e)%p(1:mySize) &
- constitutive_dotState(g,i,e)%p(1:mySize) * crystallite_subdt(g,i,e)
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
only need 2 logicals in crystallite_stressAndItsTangent to reflect all possible integration states, so crystallite_onTrack was obsolete and replaced by crystallite_todo; broken state update now directly produces a cutback
2010-02-23 15:16:39 +05:30
if (any(residuum/=residuum)) then ! if NaN occured then return without changing the state...
crystallite_updateState = .false. ! ...indicate state update failed
crystallite_todo(g,i,e) = .false. ! ...no need to calculate any further
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
if (verboseDebugger) then
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
!$OMP CRITICAL (write2out)
write(6,*) '::: updateState encountered NaN',g,i,e
!$OMPEND CRITICAL (write2out)
endif
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
return
endif
! update the microstructure
constitutive_state(g,i,e)%p(1:mySize) = constitutive_state(g,i,e)%p(1:mySize) - residuum
constitutive_nonlocal: - reworked contribution of immobile dislocation density for rate equations - flux is now calculated on the basis of interpolated velocities and densities at the interface; both incoming and outgoing fluxes are considered, so every material point only changes his own dotState - dislocation velocity is now globally defined and calculated by subroutine constitutive_nonlocal_kinetics; the subroutine is called inside _LpAndItsTangent as well as _microstructure; therefore, microstructure now needs Tstar_v as additional input; in the future one should perhaps create a subroutine constitutive_kinetics that calls constitutive_nonlocal_kinetics separately, to clearly distinguish between microstructural and kinetic variables - better use flux density vector as output variable instead of scalar flux values for each interface - added output variables internal and external resolved stress crystallite: - added flag to force local stiffness calculation in case of nonlocal model - misorientation angle is explicitly set to zero when no neighbor can be found debug: - added flag "selectiveDebugger" that is used when debugging statements should only affect a specific element, ip and grain; these are specified with the new variables debug_e, debug_i and debug_g - debugger can now be used in its original sense
2010-02-17 18:51:36 +05:30
call constitutive_microstructure(crystallite_Temperature(g,i,e), crystallite_Tstar_v(:,g,i,e), crystallite_Fe, crystallite_Fp, &
when calculating dislocation stress at grain boundaries, densities are now extrapolated similarly to like it was already done at free surfaces
2010-06-07 21:31:37 +05:30
crystallite_disorientation(:,:,g,i,e), g, i, e)
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
constitutive_nonlocal: - dipole dislocations with evolution crystallite.f90: - collect state uses subdt and subTstar0_v - in nonlocal modus: set all crystallites to broken if one is not on track anymore after either stress integration or state update - constitutive_microstructure is now called inside state update and not in integrate_stress anymore material.config: - new parameter for nonlocal constitution CPFEM.f90: - age Tstar after increment was finished
2009-08-28 19:20:47 +05:30
! setting flag to true if state is below relative tolerance, otherwise set it to false
convergence of state in crystallite is now tested as follows: (state < relevant state) or (residuum < relative tolerance * state) since the relevant value for the state variables depend on their nature and can vary by large scales (e.g. volume fraction: 1e-10, dislocation density: 1e5) it is not possible to set a unique value. instead the constitutive law has to decide what is relevant. therefore, all constitutive laws now read in parameters from the material.config that determine the values for relevantState [@luc: in dislobased law relevant State is for the moment generally set to 1e-200, so no additional parameters necessary in material.config. if you also want this feature, we can still implement it, no big deal] - added sanity checks in constitutive_nonlocal.f90 - corrected coordinate transformation for backstress calculation in constitutive_nonlocal.f90 - corrected equations for evolution of dipole dislocation densities (athermal annihilation and formation by glide)
2009-09-18 21:07:14 +05:30
crystallite_updateState = all( constitutive_state(g,i,e)%p(1:mySize) < constitutive_relevantState(g,i,e)%p(1:mySize) &
nonlocal stiffness calculation produced segmentation fault for cycle number > 16. Corrected calculation of perturbation indices k and l.
2010-03-10 15:23:41 +05:30
.or. abs(residuum) < rTol_crystalliteState*abs(constitutive_state(g,i,e)%p(1:mySize)) )
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
if (verboseDebugger .and. selectiveDebugger) then
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
!$OMP CRITICAL (write2out)
if (crystallite_updateState) then
write(6,*) '::: updateState converged',g,i,e
else
write(6,*) '::: updateState did not converge',g,i,e
endif
constitutive_nonlocal: - dipole dislocations with evolution crystallite.f90: - collect state uses subdt and subTstar0_v - in nonlocal modus: set all crystallites to broken if one is not on track anymore after either stress integration or state update - constitutive_microstructure is now called inside state update and not in integrate_stress anymore material.config: - new parameter for nonlocal constitution CPFEM.f90: - age Tstar after increment was finished
2009-08-28 19:20:47 +05:30
write(6,*)
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
write(6,'(a,f6.1)') 'crystallite_statedamper',crystallite_statedamper(g,i,e)
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
write(6,*)
write(6,'(a,/,12(e12.5,x))') 'dotState',constitutive_dotState(g,i,e)%p(1:mySize)
write(6,*)
write(6,'(a,/,12(e12.5,x))') 'new state',constitutive_state(g,i,e)%p(1:mySize)
constitutive_nonlocal: - dipole dislocations with evolution crystallite.f90: - collect state uses subdt and subTstar0_v - in nonlocal modus: set all crystallites to broken if one is not on track anymore after either stress integration or state update - constitutive_microstructure is now called inside state update and not in integrate_stress anymore material.config: - new parameter for nonlocal constitution CPFEM.f90: - age Tstar after increment was finished
2009-08-28 19:20:47 +05:30
write(6,*)
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
write(6,'(a,/,12(f12.1,x))') 'resid tolerance',abs(residuum/rTol_crystalliteState/constitutive_state(g,i,e)%p(1:mySize))
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
write(6,*)
!$OMPEND CRITICAL (write2out)
List of changes/modifications: * IO.f90 :: Adding error messages for RGC homogenization * crystallite.f90 :: Modifying convergent criteria in crystalline_updateState and crystalline_updateTemperature * material.f90 :: Adding IO_lc in homogenization_type(Name) * debug.f90 :: Adding debbugging statement and counter for material point loop * homogenization.f90 :: Adding homogenization_RGC blocks * homogenization_isostrain.f90 :: Modifying argument of homogenization_isostrain_stateInit * mpie_cpfem_marc.f90 :: Adding homogenization_RGC include * numerics.f90 :: Adding numerical parameters for RGC scheme * math.f90 :: Changing function name: math_permut to math_civita
2009-07-31 17:32:20 +05:30
endif
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
return
endfunction
!********************************************************************
! update the temperature of the grain
! and tell whether it has converged
!********************************************************************
function crystallite_updateTemperature(&
g,& ! grain number
i,& ! integration point number
e & ! element number
)
!*** variables and functions from other modules ***!
use prec, only: pReal, &
pInt, &
pLongInt
use numerics, only: rTol_crystalliteTemperature
use constitutive, only: constitutive_dotTemperature
use debug, only: debugger, &
debug_cumDotTemperatureCalls, &
debug_cumDotTemperatureTicks
!*** input variables ***!
integer(pInt), intent(in):: e, & ! element index
i, & ! integration point index
g ! grain index
!*** output variables ***!
logical crystallite_updateTemperature ! flag indicating if integration suceeded
!*** local variables ***!
real(pReal) residuum ! residuum from evolution of temperature
integer(pLongInt) tick, &
tock, &
tickrate, &
maxticks
List of changes/modifications: * IO.f90 :: Adding error messages for RGC homogenization * crystallite.f90 :: Modifying convergent criteria in crystalline_updateState and crystalline_updateTemperature * material.f90 :: Adding IO_lc in homogenization_type(Name) * debug.f90 :: Adding debbugging statement and counter for material point loop * homogenization.f90 :: Adding homogenization_RGC blocks * homogenization_isostrain.f90 :: Modifying argument of homogenization_isostrain_stateInit * mpie_cpfem_marc.f90 :: Adding homogenization_RGC include * numerics.f90 :: Adding numerical parameters for RGC scheme * math.f90 :: Changing function name: math_permut to math_civita
2009-07-31 17:32:20 +05:30
! calculate the residuum
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
call system_clock(count=tick,count_rate=tickrate,count_max=maxticks)
residuum = crystallite_Temperature(g,i,e) - crystallite_subTemperature0(g,i,e) - &
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
crystallite_subdt(g,i,e) * &
constitutive_dotTemperature(crystallite_Tstar_v(:,g,i,e),crystallite_Temperature(g,i,e),g,i,e)
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
call system_clock(count=tock,count_rate=tickrate,count_max=maxticks)
debug_cumDotTemperatureCalls = debug_cumDotTemperatureCalls + 1_pInt
debug_cumDotTemperatureTicks = debug_cumDotTemperatureTicks + tock-tick
if (tock < tick) debug_cumDotTemperatureTicks = debug_cumDotTemperatureTicks + maxticks
! if NaN occured then return without changing the state
if (residuum/=residuum) then
crystallite_updateTemperature = .false. ! indicate update failed
!$OMP CRITICAL (write2out)
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
write(6,*) '::: updateTemperature encountered NaN',g,i,e
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
!$OMPEND CRITICAL (write2out)
return
endif
! update the microstructure
crystallite_Temperature(g,i,e) = crystallite_Temperature(g,i,e) - residuum
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
! setting flag to true if residuum is below relative tolerance (or zero Kelvin), otherwise set it to false
List of changes/modifications: * IO.f90 :: Adding error messages for RGC homogenization * crystallite.f90 :: Modifying convergent criteria in crystalline_updateState and crystalline_updateTemperature * material.f90 :: Adding IO_lc in homogenization_type(Name) * debug.f90 :: Adding debbugging statement and counter for material point loop * homogenization.f90 :: Adding homogenization_RGC blocks * homogenization_isostrain.f90 :: Modifying argument of homogenization_isostrain_stateInit * mpie_cpfem_marc.f90 :: Adding homogenization_RGC include * numerics.f90 :: Adding numerical parameters for RGC scheme * math.f90 :: Changing function name: math_permut to math_civita
2009-07-31 17:32:20 +05:30
crystallite_updateTemperature = crystallite_Temperature(g,i,e) == 0.0_pReal .or. &
abs(residuum) < rTol_crystalliteTemperature*crystallite_Temperature(g,i,e)
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
return
endfunction
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
!***********************************************************************
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
!*** calculation of stress (P) with time integration ***
!*** based on a residuum in Lp and intermediate ***
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
!*** acceleration of the Newton-Raphson correction ***
!***********************************************************************
function crystallite_integrateStress(&
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
mode, & ! 1: central solution, 2: stiffness (by perturbation)
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
g,& ! grain number
i,& ! integration point number
e) ! element number
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +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
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
!*** variables and functions from other modules ***!
use prec, only: pReal, &
pInt, &
added a new module called numerics.f90 which reads in all numerical "parameters" from the file numerics.config (also being added). From now on this file has to be located in the working directory of the FEM-model and has to contain all necessary parameters.
2009-06-15 18:41:21 +05:30
pLongInt
use numerics, only: nStress, &
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
aTol_crystalliteStress, &
rTol_crystalliteStress, &
added a new module called numerics.f90 which reads in all numerical "parameters" from the file numerics.config (also being added). From now on this file has to be located in the working directory of the FEM-model and has to contain all necessary parameters.
2009-06-15 18:41:21 +05:30
iJacoLpresiduum, &
relevantStrain
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
use debug, only: debugger, &
constitutive_nonlocal: - reworked contribution of immobile dislocation density for rate equations - flux is now calculated on the basis of interpolated velocities and densities at the interface; both incoming and outgoing fluxes are considered, so every material point only changes his own dotState - dislocation velocity is now globally defined and calculated by subroutine constitutive_nonlocal_kinetics; the subroutine is called inside _LpAndItsTangent as well as _microstructure; therefore, microstructure now needs Tstar_v as additional input; in the future one should perhaps create a subroutine constitutive_kinetics that calls constitutive_nonlocal_kinetics separately, to clearly distinguish between microstructural and kinetic variables - better use flux density vector as output variable instead of scalar flux values for each interface - added output variables internal and external resolved stress crystallite: - added flag to force local stiffness calculation in case of nonlocal model - misorientation angle is explicitly set to zero when no neighbor can be found debug: - added flag "selectiveDebugger" that is used when debugging statements should only affect a specific element, ip and grain; these are specified with the new variables debug_e, debug_i and debug_g - debugger can now be used in its original sense
2010-02-17 18:51:36 +05:30
selectiveDebugger, &
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
verboseDebugger, &
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
debug_cumLpCalls, &
debug_cumLpTicks, &
debug_StressLoopDistribution
constitutive_nonlocal: - dipole dislocations with evolution crystallite.f90: - collect state uses subdt and subTstar0_v - in nonlocal modus: set all crystallites to broken if one is not on track anymore after either stress integration or state update - constitutive_microstructure is now called inside state update and not in integrate_stress anymore material.config: - new parameter for nonlocal constitution CPFEM.f90: - age Tstar after increment was finished
2009-08-28 19:20:47 +05:30
use constitutive, only: constitutive_homogenizedC, &
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
constitutive_LpAndItsTangent
use math, only: math_mul33x33, &
math_mul66x6, &
math_mul99x99, &
math_inv3x3, &
math_invert3x3, &
sorry, forgot to declare all variables, promise to be more careful to always commit a running version of the subroutine ;-)
2009-06-02 15:25:45 +05:30
math_invert, &
math_det3x3, &
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
math_I3, &
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
math_identity2nd, &
math_Mandel66to3333, &
sorry, forgot to declare all variables, promise to be more careful to always commit a running version of the subroutine ;-)
2009-06-02 15:25:45 +05:30
math_Mandel6to33, &
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
math_mandel33to6
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
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
!*** input variables ***!
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
integer(pInt), intent(in):: mode, & ! 1 or 2
e, & ! element index
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
i, & ! integration point index
g ! 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
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
!*** output variables ***!
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
logical crystallite_integrateStress ! flag indicating if integration suceeded
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
!*** local variables ***!
real(pReal), dimension(3,3):: Fg_new, & ! deformation gradient at end of timestep
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
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
A, &
B, &
BT, &
AB, &
BTA
real(pReal), dimension(6):: Tstar_v ! 2nd Piola-Kirchhoff Stress in Mandel-Notation
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
real(pReal), dimension(9,9):: dLpdT_constitutive, & ! partial derivative of plastic velocity gradient calculated by constitutive law
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
dTdLp, & ! partial derivative of 2nd Piola-Kirchhoff stress
dRdLp, & ! partial derivative of residuum (Jacobian for NEwton-Raphson scheme)
invdRdLp ! inverse of dRdLp
real(pReal), dimension(3,3,3,3):: C ! 4th rank elasticity tensor
real(pReal), dimension(6,6):: C_66 ! simplified 2nd rank elasticity tensor
real(pReal) p_hydro, & ! volumetric part of 2nd Piola-Kirchhoff Stress
det, & ! determinant
leapfrog, & ! acceleration factor for Newton-Raphson scheme
maxleap ! maximum acceleration factor
logical error ! flag indicating an error
integer(pInt) NiterationStress, & ! number of stress integrations
dummy, &
h, &
j, &
k, &
l, &
m, &
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
n, &
jacoCounter ! counter to check for Jacobian update
added NAssi-Schneidermann-Diagrams to the documentation for better understanding. Also reordered the declaration of variables in the corresponding function in crystallite.f90
2009-06-02 15:07:38 +05:30
integer(pLongInt) tick, &
tock, &
tickrate, &
maxticks
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
! be pessimistic
crystallite_integrateStress = .false.
! feed local variables
Fg_new = crystallite_subF(:,:,g,i,e)
Fp_current = crystallite_subFp0(:,:,g,i,e)
Tstar_v = crystallite_Tstar_v(:,g,i,e)
Lpguess_old = crystallite_Lp(:,:,g,i,e) ! consider present Lp good (i.e. worth remembering) ...
Lpguess = crystallite_Lp(:,:,g,i,e) ! ... and take it as first guess
constitutive_nonlocal: - dipole dislocations with evolution crystallite.f90: - collect state uses subdt and subTstar0_v - in nonlocal modus: set all crystallites to broken if one is not on track anymore after either stress integration or state update - constitutive_microstructure is now called inside state update and not in integrate_stress anymore material.config: - new parameter for nonlocal constitution CPFEM.f90: - age Tstar after increment was finished
2009-08-28 19:20:47 +05:30
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
! inversion of Fp_current...
invFp_current = math_inv3x3(Fp_current)
if (all(invFp_current == 0.0_pReal)) then ! ... failed?
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
if (verboseDebugger .and. selectiveDebugger) then
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
!$OMP CRITICAL (write2out)
write(6,*) '::: integrateStress failed on invFp_current inversion',g,i,e
write(6,*)
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
write(6,'(a11,i3,x,i2,x,i5,/,3(3(f12.7,x)/))') 'invFp_new at ',g,i,e,invFp_new
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
!$OMPEND CRITICAL (write2out)
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
return
endif
A = math_mul33x33(transpose(invFp_current), math_mul33x33(transpose(Fg_new),math_mul33x33(Fg_new,invFp_current)))
! get elasticity tensor
C_66 = constitutive_homogenizedC(g,i,e)
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
! if (debugger) write(6,'(a,/,6(6(f10.4,x)/))') 'elasticity',C_66(1:6,:)/1e9
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
C = math_Mandel66to3333(C_66)
! start LpLoop with no acceleration
NiterationStress = 0_pInt
leapfrog = 1.0_pReal
maxleap = 1024.0_pReal
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
jacoCounter = 0_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
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
LpLoop: do
! increase loop counter
NiterationStress = NiterationStress + 1
! too many loops required ?
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
if (NiterationStress > nStress) then
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
if (verboseDebugger) then
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
!$OMP CRITICAL (write2out)
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
write(6,*) '::: integrateStress reached loop limit at ',g,i,e
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
write(6,*)
!$OMPEND CRITICAL (write2out)
endif
return
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
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
B = math_I3 - crystallite_subdt(g,i,e)*Lpguess
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
BT = transpose(B)
AB = math_mul33x33(A,B)
BTA = math_mul33x33(BT,A)
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
! calculate 2nd Piola-Kirchhoff stress tensor
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
Tstar_v = 0.5_pReal*math_mul66x6(C_66,math_mandel33to6(math_mul33x33(BT,AB)-math_I3))
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
p_hydro = sum(Tstar_v(1:3))/3.0_pReal
forall(n=1:3) Tstar_v(n) = Tstar_v(n) - p_hydro ! get deviatoric stress tensor
! calculate plastic velocity gradient and its tangent according to constitutive law
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
call system_clock(count=tick,count_rate=tickrate,count_max=maxticks)
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
call constitutive_LpAndItsTangent(Lp_constitutive, dLpdT_constitutive, Tstar_v, crystallite_Temperature(g,i,e), g, i, e)
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
call system_clock(count=tock,count_rate=tickrate,count_max=maxticks)
debug_cumLpCalls = debug_cumLpCalls + 1_pInt
debug_cumLpTicks = debug_cumLpTicks + tock-tick
if (tock < tick) debug_cumLpTicks = debug_cumLpTicks + maxticks
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
if (verboseDebugger .and. selectiveDebugger) then
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
!$OMP CRITICAL (write2out)
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
write(6,'(a,i3,x,i2,x,i5,x,a,x,i3)') '::: integrateStress at ' ,g,i,e, ' ; iteration ', NiterationStress
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
write(6,*)
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
write(6,'(a,/,3(3(e20.7,x)/))') 'Lp_constitutive', Lp_constitutive
write(6,'(a,/,3(3(e20.7,x)/))') 'Lpguess', Lpguess
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
!$OMPEND CRITICAL (write2out)
endif
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
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
! update current residuum
residuum = Lpguess - Lp_constitutive
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
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
! Check for convergence of loop
if (.not.(any(residuum/=residuum)) .and. & ! exclude any NaN in residuum
( maxval(abs(residuum)) < aTol_crystalliteStress .or. & ! below absolute tolerance .or.
( any(abs(crystallite_subdt(g,i,e)*Lpguess) > relevantStrain) .and. & ! worth checking? .and.
maxval(abs(residuum/Lpguess), abs(crystallite_subdt(g,i,e)*Lpguess) > relevantStrain) < rTol_crystalliteStress & ! below relative tolerance
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
) &
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
) &
) &
exit LpLoop
! NaN occured at regular speed?
if (any(residuum/=residuum) .and. leapfrog == 1.0) then
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
if (verboseDebugger) then
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
!$OMP CRITICAL (write2out)
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
write(6,'(a,i3,x,i2,x,i5,x,a,x,i3)') '::: integrateStress encountered NaN at ',g,i,e,' ; iteration ', NiterationStress
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
!$OMPEND CRITICAL (write2out)
endif
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
return
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
! something went wrong at accelerated speed?
elseif (leapfrog > 1.0_pReal .and. & ! at fast pace .and.
( sum(residuum*residuum) > sum(residuum_old*residuum_old) .or. & ! worse residuum .or.
sum(residuum*residuum_old) < 0.0_pReal .or. & ! residuum changed sign (overshoot) .or.
any(residuum/=residuum) & ! NaN occured
) &
) then
maxleap = 0.5_pReal * leapfrog ! limit next acceleration
leapfrog = 1.0_pReal ! grinding halt
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
jacoCounter = 0_pInt ! reset counter for Jacobian update (we want to do an update next time!)
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
! restore old residuum and Lp
Lpguess = Lpguess_old
residuum = residuum_old
! residuum got better
else
! calculate Jacobian for correction term
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
if (mod(jacoCounter, iJacoLpresiduum) == 0_pInt) then
dTdLp = 0.0_pReal
exchanged forall (warning causers) to nested do loops deleted (currently unused) symmetrization of H tensor in CPFEM.f90 --> Denny pls advise?
2010-08-20 03:05:38 +05:30
do h=1,3; do j=1,3; do k=1,3; do l=1,3; do m=1,3
! forall (h=1:3,j=1:3,k=1:3,l=1:3,m=1:3) &
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
dTdLp(3*(h-1)+j,3*(k-1)+l) = dTdLp(3*(h-1)+j,3*(k-1)+l) + C(h,j,l,m)*AB(k,m)+C(h,j,m,l)*BTA(m,k)
exchanged forall (warning causers) to nested do loops deleted (currently unused) symmetrization of H tensor in CPFEM.f90 --> Denny pls advise?
2010-08-20 03:05:38 +05:30
enddo; enddo; enddo; enddo; enddo
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
dTdLp = -0.5_pReal*crystallite_subdt(g,i,e)*dTdLp
constitutive_nonlocal - take orientation gradients into account when calculating dislocation stress and dislocation fluxes - hard coded value for nu - changed kinetics (parameter G0 is currently defined as a parameter, needs to be read from material.config) - added some output statements constitutive: - some functions and subroutines needed additional input variables for passing to constitutive_nonlocal crystallite: - some functions and subroutines needed additional input variables for passing to constitutive - call microstructure with current temperature, Fp, Fe, not "sub0" values - show number of IPs, that are "onTrack" instead of those not "onTrack" - calculate Fe at beginning of substep, since we need it for state preguess
2009-10-07 21:01:52 +05:30
dRdLp = math_identity2nd(9) - math_mul99x99(dLpdT_constitutive,dTdLp)
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
invdRdLp = 0.0_pReal
call math_invert(9,dRdLp,invdRdLp,dummy,error) ! invert dR/dLp --> dLp/dR
if (error) then
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
if (verboseDebugger .and. selectiveDebugger) then
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
!$OMP CRITICAL (write2out)
corrected syntax errors (long lines, line continuation by \) and logical mistake in mpie_cpfem_marc / abq_std which went unnoticed in ifort. Using SunStudio f90 surfaced those...
2010-08-04 05:17:00 +05:30
write(6,'(a,i3,x,i2,x,i5,x,a,x,i3)') '::: integrateStress failed on dR/dLp inversion at ',g,i,e, &
' ; iteration ', NiterationStress
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
write(6,*)
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
write(6,'(a,/,9(9(e15.3,x)/))') 'dRdLp',dRdLp
write(6,'(a,/,9(9(e15.3,x)/))') 'dLpdT_constitutive',dLpdT_constitutive
write(6,'(a,/,3(3(e20.7,x)/))') 'Lp_constitutive',Lp_constitutive
write(6,'(a,/,3(3(e20.7,x)/))') 'Lpguess',Lpguess
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
!$OMPEND CRITICAL (write2out)
endif
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
return
endif
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
endif
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
jacoCounter = jacoCounter + 1_pInt ! increase counter for jaco update
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
! remember current residuum and Lpguess
residuum_old = residuum
Lpguess_old = Lpguess
! accelerate?
if (NiterationStress > 1 .and. leapfrog < maxleap) leapfrog = 2.0_pReal * leapfrog
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
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
! leapfrog to updated Lp
exchanged forall (warning causers) to nested do loops deleted (currently unused) symmetrization of H tensor in CPFEM.f90 --> Denny pls advise?
2010-08-20 03:05:38 +05:30
do k=1,3; do l=1,3; do m=1,3; do n=1,3
! forall (k=1:3,l=1:3,m=1:3,n=1:3) &
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
Lpguess(k,l) = Lpguess(k,l) - leapfrog*invdRdLp(3*(k-1)+l,3*(m-1)+n)*residuum(m,n)
exchanged forall (warning causers) to nested do loops deleted (currently unused) symmetrization of H tensor in CPFEM.f90 --> Denny pls advise?
2010-08-20 03:05:38 +05:30
enddo; enddo; enddo; enddo
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
enddo LpLoop
! calculate new plastic and elastic deformation gradient
invFp_new = math_mul33x33(invFp_current,B)
invFp_new = invFp_new/math_det3x3(invFp_new)**(1.0_pReal/3.0_pReal) ! regularize by det
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
call math_invert3x3(invFp_new,Fp_new,det,error)
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
if (error) then
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
if (verboseDebugger .and. selectiveDebugger) then
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
!$OMP CRITICAL (write2out)
corrected syntax errors (long lines, line continuation by \) and logical mistake in mpie_cpfem_marc / abq_std which went unnoticed in ifort. Using SunStudio f90 surfaced those...
2010-08-04 05:17:00 +05:30
write(6,'(a,i3,x,i2,x,i5,x,a,x,i3)') '::: integrateStress failed on invFp_new inversion at ',g,i,e, &
' ; iteration ', NiterationStress
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
write(6,*)
write(6,'(a11,3(i3,x),/,3(3(f12.7,x)/))') 'invFp_new at ',g,i,e,invFp_new
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
!$OMPEND CRITICAL (write2out)
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
return
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
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
Fe_new = math_mul33x33(Fg_new,invFp_new) ! calc resulting Fe
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
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
! add volumetric component to 2nd Piola-Kirchhoff stress
forall (n=1:3) Tstar_v(n) = Tstar_v(n) + p_hydro
! calculate 1st Piola-Kirchhoff stress
crystallite_P(:,:,g,i,e) = math_mul33x33(Fe_new,math_mul33x33(math_Mandel6to33(Tstar_v),transpose(invFp_new)))
! store local values in global variables
crystallite_Lp(:,:,g,i,e) = Lpguess
crystallite_Tstar_v(:,g,i,e) = Tstar_v
crystallite_Fp(:,:,g,i,e) = Fp_new
crystallite_Fe(:,:,g,i,e) = Fe_new
now with first draft of nonlocal constitutive law debugging memory leak closed debugging counters corrected center of gravity stored in mesh state updated is now split into a collecting loop and an execution updateState and updateTemperature fill sequentially separate logicals and evaluate afterwards to converged added 3x3 transposition function, norm for 3x1 matrix and 33x3 matrix multiplication in math non-converged crystallite triggers materialpoint cutback (used to respond elastically) non-converged materialpoint raises terminal illness which in turn renders whole FE increment useless by means of odd stress/stiffness and thus waits for FE cutback
2009-08-11 22:01:57 +05:30
crystallite_invFp(:,:,g,i,e) = invFp_new
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
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
! set return flag to true
crystallite_integrateStress = .true.
added flag verboseDebugger to debug in order to have more control about debugging statements set this to true if you want extended debugging info in the output file
2010-03-19 19:44:08 +05:30
if (verboseDebugger .and. selectiveDebugger) then
quite some changes: # non-greedy memory allocation # generation of outputConstitutive to allow for script-based T16 extraction # exchange of phenomenological by more general phenopowerlaw # lattice is based on slip and twin families which can be treated as individual entities (switched on/off, separate hardening, etc.) # nicer debugging output # changed some error/warning codes # plus potentially some minor additional brushes here and there
2009-07-22 21:37:19 +05:30
!$OMP CRITICAL (write2out)
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
write(6,'(a,i3,x,i2,x,i5,x,a,x,i3)') '::: integrateStress converged at ',g,i,e,' ; iteration ', NiterationStress
In crystallite_stressAndItsTangent: state is now correctly collected during perturbation method
2009-08-13 15:34:14 +05:30
write(6,*)
write(6,'(a,/,3(3(f12.7,x)/))') 'P / MPa',crystallite_P(:,:,g,i,e)/1e6
write(6,'(a,/,3(3(f12.7,x)/))') 'Lp',crystallite_Lp(:,:,g,i,e)
constitutive_nonlocal.f90 - completed postResults output function - connecting vector of neighboring material points is mapped to intermediate configuration of my neighbor crystallite.f90 - zero out dotState only when crystallite is non-finished - set nonfinished flag to false if crystallite is not on Track after state update - in updateState: set onTrack flag to false if encounter NaN - removed some old debugging outputs and added others homogenization.f90 - in debugging mode now telling when a cutback happens
2009-08-24 13:46:01 +05:30
write(6,'(a,/,3(3(f12.7,x)/))') 'Fp',crystallite_Fp(:,:,g,i,e)
corrected two OMP CRITICAL terminations
2010-04-06 12:17:15 +05:30
!$OMP END CRITICAL (write2out)
in crystallite_stressAndItsTangent we now update the state first (explicit integration) and then integrate the stress (implicit integration) also inside the stress integration, it is now possible to define the frequency of the Jacobian update oin the LpLoop through iJacoLpResiduum. The frequency of the Jacobian update for the stiffness in the crystallite loop is controlled by the parameter iJacoStiffness. also updated the corresponding stuctograms in the documentation
2009-06-09 16:35:29 +05:30
endif
major restructuring of code. homogenization as well as constitutive are now free to choose. the runtime got somewhat longer (25% on simple tests) compared to a hardcoded isostrain homogenization. this might be a point of further optimization at a later stage... please use homogenization_isostrain.f90 as starting point / example for future developments of homog-schemes. the homogenization scheme now can additionally output certain results. hence, the userdata structure at each integration point now looks like this: - sizeHomogPostResults - block of that size containing homogPostResults then for each grain: - sizeGrainPostResults - block of that size containing crystallitePostResults, which consist of: + phaseID + volFrac + Eulers (3) + any constitutive post results requested
2009-05-07 21:57:36 +05:30
adapted crystallite_stressAndItsTangent to do a pre-guess for the state before the actual state loop (with order stress integration, state update)
2009-06-10 20:38:33 +05:30
!$OMP CRITICAL (distributionStress)
1) terminallyIll was reset before FE did cutback --> useless extra calculations of same problem over and over... 2) local stiffness calculation is now standard for non-local grains 3) stressLoopDistribution discriminates between (a) central solution and (b) stiffness perturbation 4) debugger is switched on as standard... (but verboseDebugger not!)
2010-09-06 21:36:41 +05:30
debug_StressLoopDistribution(NiterationStress,mode) = debug_StressLoopDistribution(NiterationStress,mode) + 1
adapted crystallite_stressAndItsTangent to do a pre-guess for the state before the actual state loop (with order stress integration, state update)
2009-06-10 20:38:33 +05:30
!$OMPEND CRITICAL (distributionStress)
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
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
return
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
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
endfunction
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
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
!********************************************************************
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
! calculates orientations and disorientations (in case of single grain ips)
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
!********************************************************************
subroutine crystallite_orientations()
!*** variables and functions from other modules ***!
detabbed file for better readability
2010-02-19 19:14:38 +05:30
use prec, only: pInt, &
pReal
use math, only: math_pDecomposition, &
changed internal representation of orientation and misorientation from euler angles to quaternions (this should also fix some bugs in the math_misorientation subroutine). includes a couple of new functions in math.f90 and some changes in crystallite.f90. beware that crystallite output "orientation" now by default returns the orientation as quaternion. if you want euler angles instead, you have to add "eulerangles" as a crystallite output in your material.config file (see material.config template). for input of orientations in the texture block of the material.config you still have to specify the rotation in terms of euler angles, quaternions are not yet supported for input.
2010-03-18 17:53:17 +05:30
math_RtoQuaternion, &
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
math_QuaternionDisorientation, &
Like this Euler angles AND grain rotation should be OK.
2010-05-11 20:36:21 +05:30
inDeg, &
math_qConj
detabbed file for better readability
2010-02-19 19:14:38 +05:30
use FEsolving, only: FEsolving_execElem, &
FEsolving_execIP
use IO, only: IO_warning
use material, only: material_phase, &
homogenization_Ngrains, &
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
phase_constitution
detabbed file for better readability
2010-02-19 19:14:38 +05:30
use mesh, only: mesh_element, &
mesh_ipNeighborhood, &
FE_NipNeighbors
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
use debug, only: debugger, &
debug_e, debug_i, debug_g, &
verboseDebugger, &
selectiveDebugger
use constitutive_nonlocal, only: constitutive_nonlocal_label
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
implicit none
!*** input variables ***!
!*** output variables ***!
!*** local variables ***!
detabbed file for better readability
2010-02-19 19:14:38 +05:30
integer(pInt) e, & ! element index
i, & ! integration point index
g, & ! grain index
n, & ! neighbor index
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
myPhase, & ! phase
detabbed file for better readability
2010-02-19 19:14:38 +05:30
neighboring_e, & ! element index of my neighbor
neighboring_i, & ! integration point index of my neighbor
neighboringPhase, & ! phase of my neighbor
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
neighboringStructure ! lattice structure of my neighbor
changed internal representation of orientation and misorientation from euler angles to quaternions (this should also fix some bugs in the math_misorientation subroutine). includes a couple of new functions in math.f90 and some changes in crystallite.f90. beware that crystallite output "orientation" now by default returns the orientation as quaternion. if you want euler angles instead, you have to add "eulerangles" as a crystallite output in your material.config file (see material.config template). for input of orientations in the texture block of the material.config you still have to specify the rotation in terms of euler angles, quaternions are not yet supported for input.
2010-03-18 17:53:17 +05:30
real(pReal), dimension(3,3) :: U, R
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
logical error
!$OMP PARALLEL DO
detabbed file for better readability
2010-02-19 19:14:38 +05:30
do e = FEsolving_execElem(1),FEsolving_execElem(2)
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
do g = 1,homogenization_Ngrains(mesh_element(3,e))
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
detabbed file for better readability
2010-02-19 19:14:38 +05:30
! calculate orientation in terms of rotation matrix and euler angles
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
call math_pDecomposition(crystallite_Fe(:,:,g,i,e), U, R, error) ! polar decomposition of Fe
detabbed file for better readability
2010-02-19 19:14:38 +05:30
if (error) then
call IO_warning(650, e, i, g)
changed internal representation of orientation and misorientation from euler angles to quaternions (this should also fix some bugs in the math_misorientation subroutine). includes a couple of new functions in math.f90 and some changes in crystallite.f90. beware that crystallite output "orientation" now by default returns the orientation as quaternion. if you want euler angles instead, you have to add "eulerangles" as a crystallite output in your material.config file (see material.config template). for input of orientations in the texture block of the material.config you still have to specify the rotation in terms of euler angles, quaternions are not yet supported for input.
2010-03-18 17:53:17 +05:30
crystallite_orientation(:,g,i,e) = (/1.0_pReal, 0.0_pReal, 0.0_pReal, 0.0_pReal/) ! fake orientation
detabbed file for better readability
2010-02-19 19:14:38 +05:30
else
Like this Euler angles AND grain rotation should be OK.
2010-05-11 20:36:21 +05:30
crystallite_orientation(:,g,i,e) = math_RtoQuaternion(transpose(R))
detabbed file for better readability
2010-02-19 19:14:38 +05:30
endif
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
crystallite_rotation(:,g,i,e) = &
Like this Euler angles AND grain rotation should be OK.
2010-05-11 20:36:21 +05:30
math_QuaternionDisorientation( math_qConj(crystallite_orientation(:,g,i,e)), & ! calculate grainrotation
math_qConj(crystallite_orientation0(:,g,i,e)), &
crystallite: - grainrotation calculation now is done with symmetryID 0, i.e. without symmetry reduction since we want the absolute misorientation. - While math has everything in radians, post results eulerangles and axisangle are given in degrees. And: grain rotation seems OK after the previous changes in math module.
2010-05-07 17:31:46 +05:30
0_pInt ) ! we don't want symmetry here
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
enddo
enddo
enddo
!$OMPEND PARALLEL DO
!$OMP PARALLEL DO
! Another loop for nonlocal material which uses the orientations from the first one.
do e = FEsolving_execElem(1),FEsolving_execElem(2)
do i = FEsolving_execIP(1,e),FEsolving_execIP(2,e)
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
selectiveDebugger = (e == debug_e .and. i == debug_i)
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
myPhase = material_phase(1,i,e) ! get my crystal structure
if (phase_constitution(myPhase) == constitutive_nonlocal_label) then ! if nonlocal model
New array crystallite_symmetryID(i,g,e) is now filled during initialization run with 1 for bcc and fcc phase and 2 for hexagonal phase. The values are needed for misorientation calculations to apply the correct symmetry operators for cubic and hexagonal phases.
2010-04-19 15:33:34 +05:30
do n = 1,FE_NipNeighbors(mesh_element(2,e)) ! loop through my neighbors
detabbed file for better readability
2010-02-19 19:14:38 +05:30
neighboring_e = mesh_ipNeighborhood(1,n,i,e)
neighboring_i = mesh_ipNeighborhood(2,n,i,e)
if ((neighboring_e > 0) .and. (neighboring_i > 0)) then ! if neighbor exists
neighboringPhase = material_phase(1,neighboring_i,neighboring_e) ! get my neighbor's crystal structure
if (myPhase == neighboringPhase) then ! if my neighbor has same phase like me
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
crystallite_disorientation(:,n,1,i,e) = &
math_QuaternionDisorientation( crystallite_orientation(:,1,i,e), &
crystallite_orientation(:,1,neighboring_i,neighboring_e), &
crystallite_symmetryID(1,i,e)) ! calculate disorientation
detabbed file for better readability
2010-02-19 19:14:38 +05:30
else ! for neighbor with different phase
return 180° disorientation about 100 for neighbors with different phases
2010-04-29 17:58:51 +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
detabbed file for better readability
2010-02-19 19:14:38 +05:30
endif
else ! no existing neighbor
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
crystallite_disorientation(:,n,1,i,e) = (/-1.0_pReal, 0.0_pReal, 0.0_pReal, 0.0_pReal/) ! homomorphic identity
detabbed file for better readability
2010-02-19 19:14:38 +05:30
endif
statedamper has to be local (specific for each e,i,g); with a global damping we may produce spurious convergence
2010-04-29 13:11:29 +05:30
if (verboseDebugger .and. selectiveDebugger) then
!$OMP CRITICAL (write2out)
write(6,'(a27,i2,a3,4(f12.5,x))') 'disorientation to neighbor ',n,' : ',crystallite_disorientation(:,n,1,i,e)
!$OMP END CRITICAL (write2out)
endif
detabbed file for better readability
2010-02-19 19:14:38 +05:30
enddo
endif
enddo
enddo
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
!$OMPEND PARALLEL DO
endsubroutine
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
!********************************************************************
! return results of particular grain
!********************************************************************
function crystallite_postResults(&
dt,& ! time increment
g,& ! grain number
i,& ! integration point number
e & ! element number
)
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
!*** variables and functions from other modules ***!
use prec, only: pInt, &
pReal
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
use math, only: math_QuaternionToEuler, &
crystallite: - grainrotation calculation now is done with symmetryID 0, i.e. without symmetry reduction since we want the absolute misorientation. - While math has everything in radians, post results eulerangles and axisangle are given in degrees. And: grain rotation seems OK after the previous changes in math module.
2010-05-07 17:31:46 +05:30
math_QuaternionToAxisAngle, &
added some output variables in crystallite: F, Fe, Fp, Ee, P, S
2010-05-18 13:27:13 +05:30
math_mul33x33, &
math_I3, &
inDeg, &
math_Mandel6to33
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
use mesh, only: mesh_element
use material, only: microstructure_crystallite, &
crystallite_Noutput, &
material_phase, &
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
material_volume
use constitutive, only: constitutive_sizePostResults, &
constitutive_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
implicit none
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
!*** input variables ***!
integer(pInt), intent(in):: e, & ! element index
i, & ! integration point index
g ! grain index
constitutive_nonlocal: - read in activation energy for dislocation glide from material.config - changed naming of dDipMin/Max to dLower/dUpper - added new outputs: rho_dot, rho_dot_dip, rho_dot_gen, rho_dot_sgl2dip, rho_dot_dip2sgl, rho_dot_ann_ath, rho_dot_ann_the, rho_dot_flux, d_upper_edge, d_upper_screw, d_upper_dot_edge, d_upper_dot_screw - poisson's ratio is now calculated from elastic constants - microstrucutre has state as first argument, since this is our output variable - periodic boundary conditions are taken into account for fluxes and internal stresses. for the moment, flag has to be set in constitutive_nonlocal. - corrected calculation for dipole formation by glide - added terms for dipole formation/annihilation by stress decrease/increase constitutive: - passing of arguments is adapted for constitutive_nonlocal model crystallite: - in stiffness calculation: call to collect_dotState used wrong arguments - crystallite_postResults uses own Tstar_v and temperature, no need for passing them from materialpoint_postResults homogenization: - crystallite_postResults uses own Tstar_v and temperature, no need for passing them from materialpoint_postResults IO: - changed error message 229 material.config: - changed example for nonlocal constitution according to constitutive_nonlocal all: - added some flush statements
2009-10-20 20:06:03 +05:30
real(pReal), intent(in):: dt ! time increment
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
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
!*** output variables ***!
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
real(pReal), dimension(1+crystallite_sizePostResults(microstructure_crystallite(mesh_element(4,e)))+ &
1+constitutive_sizePostResults(g,i,e)) :: 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
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
!*** local variables ***!
sorry, in rev 568 some variables were not declared in crystallite_postResults
2010-05-18 18:06:09 +05:30
real(pReal), dimension(3,3) :: Ee
integer(pInt) k,l,o,c,crystID,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
logical error
crystID = microstructure_crystallite(mesh_element(4,e))
crystallite_postResults = 0.0_pReal
c = 0_pInt
crystallite_postResults(c+1) = crystallite_sizePostResults(crystID); c = c+1_pInt ! size of results from cryst
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
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
do o = 1,crystallite_Noutput(crystID)
select case(crystallite_output(o,crystID))
case ('phase')
crystallite_postResults(c+1) = material_phase(g,i,e) ! phaseID of grain
c = c + 1_pInt
case ('volume')
crystallite_postResults(c+1) = material_volume(g,i,e) ! grain volume (not fraction but absolute, right?)
c = c + 1_pInt
case ('orientation')
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
crystallite_postResults(c+1:c+4) = &
crystallite: - grainrotation calculation now is done with symmetryID 0, i.e. without symmetry reduction since we want the absolute misorientation. - While math has everything in radians, post results eulerangles and axisangle are given in degrees. And: grain rotation seems OK after the previous changes in math module.
2010-05-07 17:31:46 +05:30
crystallite_orientation(:,g,i,e) ! grain orientation as quaternion
changed internal representation of orientation and misorientation from euler angles to quaternions (this should also fix some bugs in the math_misorientation subroutine). includes a couple of new functions in math.f90 and some changes in crystallite.f90. beware that crystallite output "orientation" now by default returns the orientation as quaternion. if you want euler angles instead, you have to add "eulerangles" as a crystallite output in your material.config file (see material.config template). for input of orientations in the texture block of the material.config you still have to specify the rotation in terms of euler angles, quaternions are not yet supported for input.
2010-03-18 17:53:17 +05:30
c = c + 4_pInt
case ('eulerangles')
crystallite: - grainrotation calculation now is done with symmetryID 0, i.e. without symmetry reduction since we want the absolute misorientation. - While math has everything in radians, post results eulerangles and axisangle are given in degrees. And: grain rotation seems OK after the previous changes in math module.
2010-05-07 17:31:46 +05:30
crystallite_postResults(c+1:c+3) = inDeg * &
math_QuaternionToEuler(crystallite_orientation(:,g,i,e)) ! grain orientation as Euler angles in degree
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
c = c + 3_pInt
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
case ('grainrotation')
crystallite_postResults(c+1:c+4) = &
crystallite: - grainrotation calculation now is done with symmetryID 0, i.e. without symmetry reduction since we want the absolute misorientation. - While math has everything in radians, post results eulerangles and axisangle are given in degrees. And: grain rotation seems OK after the previous changes in math module.
2010-05-07 17:31:46 +05:30
math_QuaternionToAxisAngle(crystallite_rotation(1:4,g,i,e)) ! grain rotation away from initial orientation as axis-angle
crystallite_postResults(c+4) = inDeg * crystallite_postResults(c+4) ! angle in degree
New output can be requested from crystallite: (output) grainrotation it gives the deviation from the initial grain orientation in axis-angle representation with the angle in degrees.
2010-04-12 16:44:36 +05:30
c = c + 4_pInt
added some output variables in crystallite: F, Fe, Fp, Ee, P, S
2010-05-18 13:27:13 +05:30
case ('defgrad','f')
mySize = 9_pInt
crystallite_postResults(c+1:c+1+mySize) = reshape(crystallite_partionedF(:,:,g,i,e),(/mySize/))
c = c + mySize
case ('fe')
mySize = 9_pInt
crystallite_postResults(c+1:c+1+mySize) = reshape(crystallite_Fe(:,:,g,i,e),(/mySize/))
c = c + mySize
case ('ee')
Ee = 0.5_pReal * (math_mul33x33(transpose(crystallite_Fe(:,:,g,i,e)), crystallite_Fe(:,:,g,i,e)) - math_I3)
mySize = 9_pInt
crystallite_postResults(c+1:c+1+mySize) = reshape(Ee(:,:),(/mySize/))
c = c + mySize
case ('fp')
mySize = 9_pInt
crystallite_postResults(c+1:c+1+mySize) = reshape(crystallite_Fp(:,:,g,i,e),(/mySize/))
c = c + mySize
case ('p','firstpiola','1stpiola')
mySize = 9_pInt
crystallite_postResults(c+1:c+1+mySize) = reshape(crystallite_P(:,:,g,i,e),(/mySize/))
c = c + mySize
case ('s','tstar','secondpiola','2ndpiola')
mySize = 9_pInt
crystallite_postResults(c+1:c+1+mySize) = reshape(math_Mandel6to33(crystallite_Tstar_v(:,g,i,e)),(/mySize/))
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
end select
enddo
sorry, back to the previous version. grain deformation output is part of crystallite.f90, no longer done by homogenization. material.config reflects the changes.
2009-10-22 22:29:24 +05:30
crystallite_postResults(c+1) = constitutive_sizePostResults(g,i,e); c = c+1_pInt ! size of constitutive results
crystallite_postResults(c+1:c+constitutive_sizePostResults(g,i,e)) = &
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
constitutive_postResults(crystallite_Tstar_v(:,g,i,e), crystallite_subTstar0_v(:,g,i,e), crystallite_Fe, crystallite_Fp, &
corrected (?) disorientation calc and introduced some new assisting functions
2010-04-28 22:49:58 +05:30
crystallite_Temperature(g,i,e), crystallite_disorientation(:,:,g,i,e), dt, &
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
crystallite_subdt(g,i,e), g, i, e)
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
c = c + constitutive_sizePostResults(g,i,e)
constitutive_nonlocal: - read in activation energy for dislocation glide from material.config - changed naming of dDipMin/Max to dLower/dUpper - added new outputs: rho_dot, rho_dot_dip, rho_dot_gen, rho_dot_sgl2dip, rho_dot_dip2sgl, rho_dot_ann_ath, rho_dot_ann_the, rho_dot_flux, d_upper_edge, d_upper_screw, d_upper_dot_edge, d_upper_dot_screw - poisson's ratio is now calculated from elastic constants - microstrucutre has state as first argument, since this is our output variable - periodic boundary conditions are taken into account for fluxes and internal stresses. for the moment, flag has to be set in constitutive_nonlocal. - corrected calculation for dipole formation by glide - added terms for dipole formation/annihilation by stress decrease/increase constitutive: - passing of arguments is adapted for constitutive_nonlocal model crystallite: - in stiffness calculation: call to collect_dotState used wrong arguments - crystallite_postResults uses own Tstar_v and temperature, no need for passing them from materialpoint_postResults homogenization: - crystallite_postResults uses own Tstar_v and temperature, no need for passing them from materialpoint_postResults IO: - changed error message 229 material.config: - changed example for nonlocal constitution according to constitutive_nonlocal all: - added some flush statements
2009-10-20 20:06:03 +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
return
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
endfunction
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
END MODULE
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
!##############################################################