675 lines
34 KiB
Fortran
675 lines
34 KiB
Fortran
!--------------------------------------------------------------------------------------------------
|
|
!* $Id$
|
|
!--------------------------------------------------------------------------------------------------
|
|
!> @author Pratheek Shanthraj, Max-Planck-Institut für Eisenforschung GmbH
|
|
!> @author Martin Diehl, Max-Planck-Institut für Eisenforschung GmbH
|
|
!> @author Philip Eisenlohr, Max-Planck-Institut für Eisenforschung GmbH
|
|
!> @brief Utilities used by the different spectral solver variants
|
|
!--------------------------------------------------------------------------------------------------
|
|
module DAMASK_spectral_Utilities
|
|
use prec, only: &
|
|
pReal, &
|
|
pInt
|
|
use mesh, only : &
|
|
res, &
|
|
res1_red, &
|
|
geomdim, &
|
|
mesh_NcpElems, &
|
|
wgt
|
|
|
|
use math
|
|
|
|
use IO, only: &
|
|
IO_error
|
|
|
|
implicit none
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! variables storing information for spectral method and FFTW
|
|
type(C_PTR), private :: plan_forward, plan_backward ! plans for fftw
|
|
real(pReal), private, dimension(:,:,:,:,:,:,:), allocatable :: gamma_hat ! gamma operator (field) for spectral method
|
|
real(pReal), private, dimension(:,:,:,:), allocatable :: xi ! wave vector field for divergence and for gamma operator
|
|
complex(pReal),private, dimension(:,:,:,:,:), pointer :: field_fourier
|
|
real(pReal), private, dimension(3,3,3,3) :: C_ref
|
|
|
|
real(pReal), public, dimension(:,:,:,:,:), pointer :: field_real
|
|
logical, public :: cutBack =.false.
|
|
!--------------------------------------------------------------------------------------------------
|
|
! debug fftw
|
|
type(C_PTR), private :: plan_scalarField_forth, plan_scalarField_back
|
|
complex(pReal),private, dimension(:,:,:), pointer :: scalarField_real
|
|
complex(pReal),private, dimension(:,:,:), pointer :: scalarField_fourier
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! debug divergence
|
|
type(C_PTR), private :: plan_divergence
|
|
real(pReal), private, dimension(:,:,:,:), pointer :: divergence_real
|
|
complex(pReal), private, dimension(:,:,:,:), pointer :: divergence_fourier
|
|
real(pReal), private, dimension(:,:,:,:), allocatable :: divergence_post
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
!variables controlling debugging
|
|
logical,public :: debugGeneral, debugDivergence, debugRestart, debugFFTW
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! derived types
|
|
type tSolutionState
|
|
logical :: converged = .true.
|
|
logical :: regrid = .false.
|
|
logical :: termIll = .false.
|
|
end type tSolutionState
|
|
|
|
type tBoundaryCondition
|
|
real(pReal), dimension(3,3) :: values = 0.0_pReal
|
|
real(pReal), dimension(3,3) :: maskFloat = 0.0_pReal
|
|
logical, dimension(3,3) :: maskLogical = .false.
|
|
character(len=64) :: myType = 'None'
|
|
end type tBoundaryCondition
|
|
|
|
contains
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
!> @brief allocates all neccessary fields, sets debug flags, create plans for fftw
|
|
!> @details Sets the debug levels for general, divergence, restart and fftw from the biwise coding
|
|
!> provided by the debug module to logicals.
|
|
!> Allocates all fields used by FFTW and create the corresponding plans depending on the debug
|
|
!> level chosen.
|
|
!> Initializes FFTW.
|
|
!--------------------------------------------------------------------------------------------------
|
|
subroutine Utilities_init()
|
|
use, intrinsic :: iso_fortran_env ! to get compiler_version and compiler_options (at least for gfortran >4.6 at the moment)
|
|
use numerics, only: &
|
|
DAMASK_NumThreadsInt, &
|
|
fftw_planner_flag, &
|
|
fftw_timelimit, &
|
|
memory_efficient
|
|
|
|
use debug, only: &
|
|
debug_level, &
|
|
debug_spectral, &
|
|
debug_levelBasic, &
|
|
debug_spectralDivergence, &
|
|
debug_spectralRestart, &
|
|
debug_spectralFFTW
|
|
|
|
use mesh, only : &
|
|
virt_dim
|
|
|
|
implicit none
|
|
integer(pInt) :: i, j, k
|
|
integer(pInt), dimension(3) :: k_s
|
|
!$ integer(pInt) :: ierr
|
|
type(C_PTR) :: tensorField ! field in real and fourier space
|
|
type(C_PTR) :: scalarField_realC, scalarField_fourierC
|
|
type(C_PTR) :: divergence
|
|
|
|
|
|
write(6,'(a)') ''
|
|
write(6,'(a)') ' <<<+- DAMASK_Utilities init -+>>>'
|
|
write(6,'(a)') ' $Id$'
|
|
#include "compilation_info.f90"
|
|
write(6,'(a)') ''
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! set debugging parameters
|
|
debugGeneral = iand(debug_level(debug_spectral),debug_levelBasic) /= 0
|
|
debugDivergence = iand(debug_level(debug_spectral),debug_spectralDivergence) /= 0
|
|
debugRestart = iand(debug_level(debug_spectral),debug_spectralRestart) /= 0
|
|
debugFFTW = iand(debug_level(debug_spectral),debug_spectralFFTW) /= 0
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! allocation
|
|
allocate (xi (3,res1_red,res(2),res(3)), source = 0.0_pReal) ! start out isothermally
|
|
tensorField = fftw_alloc_complex(int(res1_red*res(2)*res(3)*9_pInt,C_SIZE_T)) ! allocate continous data using a C function, C_SIZE_T is of type integer(8)
|
|
call c_f_pointer(tensorField, field_real, [ res(1)+2_pInt,res(2),res(3),3,3]) ! place a pointer for a real representation on tensorField
|
|
call c_f_pointer(tensorField, field_fourier, [ res1_red, res(2),res(3),3,3]) ! place a pointer for a complex representation on tensorField
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! general initialization of fftw (see manual on fftw.org for more details)
|
|
if (pReal /= C_DOUBLE .or. pInt /= C_INT) call IO_error(error_ID=808_pInt) ! check for correct precision in C
|
|
!$ if(DAMASK_NumThreadsInt > 0_pInt) then
|
|
!$ ierr = fftw_init_threads()
|
|
!$ if (ierr == 0_pInt) call IO_error(error_ID = 809_pInt)
|
|
!$ call fftw_plan_with_nthreads(DAMASK_NumThreadsInt)
|
|
!$ endif
|
|
call fftw_set_timelimit(fftw_timelimit) ! set timelimit for plan creation
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! creating plans
|
|
plan_forward = fftw_plan_many_dft_r2c(3,[ res(3),res(2) ,res(1)],9,& ! dimensions , length in each dimension in reversed order
|
|
field_real,[ res(3),res(2) ,res(1)+2_pInt],& ! input data , physical length in each dimension in reversed order
|
|
1, res(3)*res(2)*(res(1)+2_pInt),& ! striding , product of physical lenght in the 3 dimensions
|
|
field_fourier,[ res(3),res(2) ,res1_red],&
|
|
1, res(3)*res(2)* res1_red,fftw_planner_flag)
|
|
|
|
plan_backward =fftw_plan_many_dft_c2r(3,[ res(3),res(2) ,res(1)],9,&
|
|
field_fourier,[ res(3),res(2) ,res1_red],&
|
|
1, res(3)*res(2)* res1_red,&
|
|
field_real,[ res(3),res(2) ,res(1)+2_pInt],&
|
|
1, res(3)*res(2)*(res(1)+2_pInt),fftw_planner_flag)
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! depending on (debug) options, allocate more memory and create additional plans
|
|
if (debugDivergence) then
|
|
divergence = fftw_alloc_complex(int(res1_red*res(2)*res(3)*3_pInt,C_SIZE_T))
|
|
call c_f_pointer(divergence, divergence_real, [ res(1)+2_pInt,res(2),res(3),3])
|
|
call c_f_pointer(divergence, divergence_fourier, [ res1_red, res(2),res(3),3])
|
|
allocate (divergence_post(res(1),res(2),res(3),3)); divergence_post = 0.0_pReal
|
|
plan_divergence = fftw_plan_many_dft_c2r(3,[ res(3),res(2) ,res(1)],3,&
|
|
divergence_fourier,[ res(3),res(2) ,res1_red],&
|
|
1, res(3)*res(2)* res1_red,&
|
|
divergence_real,[ res(3),res(2) ,res(1)+2_pInt],&
|
|
1, res(3)*res(2)*(res(1)+2_pInt),fftw_planner_flag)
|
|
endif
|
|
|
|
if (debugFFTW) then
|
|
scalarField_realC = fftw_alloc_complex(int(res(1)*res(2)*res(3),C_SIZE_T)) ! do not do an inplace transform
|
|
scalarField_fourierC = fftw_alloc_complex(int(res(1)*res(2)*res(3),C_SIZE_T))
|
|
call c_f_pointer(scalarField_realC, scalarField_real, [res(1),res(2),res(3)])
|
|
call c_f_pointer(scalarField_fourierC, scalarField_fourier, [res(1),res(2),res(3)])
|
|
plan_scalarField_forth = fftw_plan_dft_3d(res(3),res(2),res(1),& !reversed order
|
|
scalarField_real,scalarField_fourier,-1,fftw_planner_flag)
|
|
plan_scalarField_back = fftw_plan_dft_3d(res(3),res(2),res(1),& !reversed order
|
|
scalarField_fourier,scalarField_real,+1,fftw_planner_flag)
|
|
endif
|
|
|
|
if (debugGeneral) write(6,'(a)') 'FFTW initialized'
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! calculation of discrete angular frequencies, ordered as in FFTW (wrap around)
|
|
do k = 1_pInt, res(3)
|
|
k_s(3) = k - 1_pInt
|
|
if(k > res(3)/2_pInt + 1_pInt) k_s(3) = k_s(3) - res(3)
|
|
do j = 1_pInt, res(2)
|
|
k_s(2) = j - 1_pInt
|
|
if(j > res(2)/2_pInt + 1_pInt) k_s(2) = k_s(2) - res(2)
|
|
do i = 1_pInt, res1_red
|
|
k_s(1) = i - 1_pInt
|
|
xi(1:3,i,j,k) = real(k_s, pReal)/virt_dim
|
|
enddo; enddo; enddo
|
|
|
|
if(memory_efficient) then ! allocate just single fourth order tensor
|
|
allocate (gamma_hat(3,3,3,3,1,1,1), source = 0.0_pReal)
|
|
else ! precalculation of gamma_hat field
|
|
allocate (gamma_hat(3,3,3,3,res1_red ,res(2),res(3)), source =0.0_pReal) ! singular point at xi=(0.0,0.0,0.0) i.e. i=j=k=1
|
|
endif
|
|
|
|
end subroutine Utilities_init
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
!> @brief updates references stiffness and potentially precalculated gamma operator
|
|
!> @details Sets the current reference stiffness to the stiffness given as an argument.
|
|
!> If the gamma operator is precalculated, it is calculated with this stiffness.
|
|
!> In case of a on-the-fly calculation, only the reference stiffness is updated.
|
|
!> The gamma operator is filtered depening on the filter selected in numerics
|
|
!--------------------------------------------------------------------------------------------------
|
|
subroutine Utilities_updateGamma(C)
|
|
|
|
use numerics, only: &
|
|
memory_efficient
|
|
|
|
implicit none
|
|
real(pReal), dimension(3,3,3,3), intent(in) :: C
|
|
real(pReal), dimension(3,3) :: temp33_Real, xiDyad
|
|
real(pReal) :: filter
|
|
integer(pInt) :: i, j, k, l, m, n, o
|
|
|
|
C_ref = C
|
|
if(.not. memory_efficient) then
|
|
do k = 1_pInt, res(3); do j = 1_pInt, res(2); do i = 1_pInt, res1_red
|
|
if(any([i,j,k] /= 1_pInt)) then ! singular point at xi=(0.0,0.0,0.0) i.e. i=j=k=1
|
|
forall(l = 1_pInt:3_pInt, m = 1_pInt:3_pInt) &
|
|
xiDyad(l,m) = xi(l, i,j,k)*xi(m, i,j,k)
|
|
forall(l = 1_pInt:3_pInt, m = 1_pInt:3_pInt) &
|
|
temp33_Real(l,m) = sum(C_ref(l,m,1:3,1:3)*xiDyad)
|
|
temp33_Real = math_inv33(temp33_Real)
|
|
filter = Utilities_getFilter(xi(1:3,i,j,k))
|
|
forall(l=1_pInt:3_pInt, m=1_pInt:3_pInt, n=1_pInt:3_pInt, o=1_pInt:3_pInt)&
|
|
gamma_hat(l,m,n,o, i,j,k) = filter*temp33_Real(l,n)*xiDyad(m,o)
|
|
endif
|
|
enddo; enddo; enddo
|
|
gamma_hat(1:3,1:3,1:3,1:3, 1,1,1) = 0.0_pReal ! singular point at xi=(0.0,0.0,0.0) i.e. i=j=k=1
|
|
endif
|
|
end subroutine Utilities_updateGamma
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
!> @brief forward FFT of data in field_real to field_fourier with highest freqs. removed
|
|
!> Does an unweighted FFT transform from real to complex.
|
|
!> In case of debugging the FFT, also one component of the tensor (specified by row and column)
|
|
!> is independetly transformed complex to complex and compared to the whole tensor transform
|
|
!--------------------------------------------------------------------------------------------------
|
|
subroutine Utilities_FFTforward(row,column)
|
|
use mesh, only : &
|
|
virt_dim
|
|
use math, only: &
|
|
math_divergenceFFT
|
|
|
|
implicit none
|
|
integer(pInt), intent(in), optional :: row, column
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! copy one component of the stress field to to a single FT and check for mismatch
|
|
if (debugFFTW) then
|
|
if (.not. present(row) .or. .not. present(column)) stop
|
|
scalarField_real(1:res(1),1:res(2),1:res(3)) =& ! store the selected component
|
|
cmplx(field_real(1:res(1),1:res(2),1:res(3),row,column),0.0_pReal,pReal)
|
|
endif
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! call function to calculate divergence from math (for post processing) to check results
|
|
if (debugDivergence) &
|
|
divergence_post = math_divergenceFFT(virt_dim,field_real(1:res(1),1:res(2),1:res(3),1:3,1:3)) ! padding
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! doing the FT
|
|
call fftw_execute_dft_r2c(plan_forward,field_real,field_fourier)
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! comparing 1 and 3x3 FT results
|
|
if (debugFFTW) then
|
|
call fftw_execute_dft(plan_scalarField_forth,scalarField_real,scalarField_fourier)
|
|
write(6,'(a,i1,1x,i1)') 'checking FT results of compontent ', row, column
|
|
write(6,'(a,2(es11.4,1x))') 'max FT relative error = ',&
|
|
maxval( real((scalarField_fourier(1:res1_red,1:res(2),1:res(3))-&
|
|
field_fourier(1:res1_red,1:res(2),1:res(3),row,column))/&
|
|
scalarField_fourier(1:res1_red,1:res(2),1:res(3)))), &
|
|
maxval(aimag((scalarField_fourier(1:res1_red,1:res(2),1:res(3))-&
|
|
field_fourier(1:res1_red,1:res(2),1:res(3),row,column))/&
|
|
scalarField_fourier(1:res1_red,1:res(2),1:res(3))))
|
|
endif
|
|
!--------------------------------------------------------------------------------------------------
|
|
! removing highest frequencies
|
|
field_fourier ( res1_red,1:res(2) , 1:res(3) ,1:3,1:3)&
|
|
= cmplx(0.0_pReal,0.0_pReal,pReal)
|
|
field_fourier (1:res1_red, res(2)/2_pInt+1_pInt,1:res(3) ,1:3,1:3)&
|
|
= cmplx(0.0_pReal,0.0_pReal,pReal)
|
|
if(res(3)>1_pInt) &
|
|
field_fourier (1:res1_red,1:res(2), res(3)/2_pInt+1_pInt,1:3,1:3)&
|
|
= cmplx(0.0_pReal,0.0_pReal,pReal)
|
|
end subroutine Utilities_FFTforward
|
|
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
!> @brief backward FFT of data in field_fourier to field_real
|
|
!> Does an inverse FFT transform from complex to real
|
|
!> In case of debugging the FFT, also one component of the tensor (specified by row and column)
|
|
!> is independetly transformed complex to complex and compared to the whole tensor transform
|
|
!> results is weighted by number of points stored in wgt
|
|
!--------------------------------------------------------------------------------------------------
|
|
subroutine Utilities_FFTbackward(row,column)
|
|
|
|
implicit none
|
|
integer(pInt), intent(in), optional :: row, column
|
|
integer(pInt) :: i, j, k, m, n
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! comparing 1 and 3x3 inverse FT results
|
|
if (debugFFTW) then
|
|
scalarField_fourier = field_fourier(1:res1_red,1:res(2),1:res(3),row,column)
|
|
do i = 0_pInt, res(1)/2_pInt-2_pInt ! unpack fft data for conj complex symmetric part
|
|
m = 1_pInt
|
|
do k = 1_pInt, res(3)
|
|
n = 1_pInt
|
|
do j = 1_pInt, res(2)
|
|
scalarField_fourier(res(1)-i,j,k) = conjg(scalarField_fourier(2+i,n,m))
|
|
if(n == 1_pInt) n = res(2) + 1_pInt
|
|
n = n-1_pInt
|
|
enddo
|
|
if(m == 1_pInt) m = res(3) + 1_pInt
|
|
m = m -1_pInt
|
|
enddo; enddo
|
|
endif
|
|
|
|
|
|
call fftw_execute_dft_c2r(plan_backward,field_fourier,field_real) ! back transform of fluct deformation gradient
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! comparing 1 and 3x3 inverse FT results
|
|
if (debugFFTW) then
|
|
write(6,'(a,i1,1x,i1)') 'checking iFT results of compontent ', row, column
|
|
call fftw_execute_dft(plan_scalarField_back,scalarField_fourier,scalarField_real)
|
|
write(6,'(a,es11.4)') 'max iFT relative error = ',&
|
|
maxval((real(scalarField_real(1:res(1),1:res(2),1:res(3)))-&
|
|
field_real(1:res(1),1:res(2),1:res(3),row,column))/&
|
|
real(scalarField_real(1:res(1),1:res(2),1:res(3))))
|
|
endif
|
|
field_real = field_real * wgt
|
|
|
|
end subroutine Utilities_FFTbackward
|
|
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
!> @brief doing convolution gamma_hat * field_real with average value given by fieldAim
|
|
!--------------------------------------------------------------------------------------------------
|
|
subroutine Utilities_fourierConvolution(fieldAim)
|
|
|
|
use numerics, only: &
|
|
memory_efficient
|
|
|
|
implicit none
|
|
real(pReal), dimension(3,3), intent(in) :: fieldAim
|
|
real(pReal), dimension(3,3) :: xiDyad, temp33_Real
|
|
real(pReal) :: filter
|
|
integer(pInt) :: i, j, k, l, m, n, o
|
|
complex(pReal), dimension(3,3) :: temp33_complex
|
|
|
|
|
|
write(6,'(/,a)') '... doing convolution .....................................................'
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! to the actual spectral method calculation (mechanical equilibrium)
|
|
if(memory_efficient) then ! memory saving version, on-the-fly calculation of gamma_hat
|
|
do k = 1_pInt, res(3); do j = 1_pInt, res(2) ;do i = 1_pInt, res1_red
|
|
if(any([i,j,k] /= 1_pInt)) then ! singular point at xi=(0.0,0.0,0.0) i.e. i=j=k=1
|
|
forall(l = 1_pInt:3_pInt, m = 1_pInt:3_pInt) &
|
|
xiDyad(l,m) = xi(l, i,j,k)*xi(m, i,j,k)
|
|
forall(l = 1_pInt:3_pInt, m = 1_pInt:3_pInt) &
|
|
temp33_Real(l,m) = sum(C_ref(l,m,1:3,1:3)*xiDyad)
|
|
temp33_Real = math_inv33(temp33_Real)
|
|
filter = Utilities_getFilter(xi(1:3,i,j,k))
|
|
forall(l=1_pInt:3_pInt, m=1_pInt:3_pInt, n=1_pInt:3_pInt, o=1_pInt:3_pInt)&
|
|
gamma_hat(l,m,n,o, 1,1,1) = filter*temp33_Real(l,n)*xiDyad(m,o)
|
|
forall(l = 1_pInt:3_pInt, m = 1_pInt:3_pInt) &
|
|
temp33_Complex(l,m) = sum(gamma_hat(l,m,1:3,1:3, 1,1,1) * field_fourier(i,j,k,1:3,1:3))
|
|
field_fourier(i,j,k,1:3,1:3) = temp33_Complex
|
|
endif
|
|
enddo; enddo; enddo
|
|
else ! use precalculated gamma-operator
|
|
do k = 1_pInt, res(3); do j = 1_pInt, res(2); do i = 1_pInt,res1_red
|
|
forall( m = 1_pInt:3_pInt, n = 1_pInt:3_pInt) &
|
|
temp33_Complex(m,n) = sum(gamma_hat(m,n,1:3,1:3, i,j,k) * field_fourier(i,j,k,1:3,1:3))
|
|
field_fourier(i,j,k, 1:3,1:3) = temp33_Complex
|
|
enddo; enddo; enddo
|
|
endif
|
|
|
|
field_fourier(1,1,1,1:3,1:3) = cmplx(fieldAim*real(mesh_NcpElems,pReal),0.0_pReal,pReal) ! singular point at xi=(0.0,0.0,0.0) i.e. i=j=k=1
|
|
|
|
end subroutine Utilities_fourierConvolution
|
|
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
!> @brief calculate root mean square of divergence of field_fourier
|
|
!--------------------------------------------------------------------------------------------------
|
|
real(pReal) function Utilities_divergenceRMS()
|
|
implicit none
|
|
integer(pInt) :: i, j, k
|
|
real(pReal) :: err_div_RMS, err_real_div_RMS, err_post_div_RMS,&
|
|
err_div_max, err_real_div_max
|
|
complex(pReal), dimension(3) :: temp3_complex
|
|
|
|
write(6,'(/,a)') '... calculating divergence ................................................'
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! calculating RMS divergence criterion in Fourier space
|
|
Utilities_divergenceRMS = 0.0_pReal
|
|
do k = 1_pInt, res(3); do j = 1_pInt, res(2)
|
|
do i = 2_pInt, res1_red -1_pInt ! Has somewhere a conj. complex counterpart. Therefore count it twice.
|
|
Utilities_divergenceRMS = Utilities_divergenceRMS &
|
|
+ 2.0_pReal*(sum (real(math_mul33x3_complex(field_fourier(i,j,k,1:3,1:3),& ! (sqrt(real(a)**2 + aimag(a)**2))**2 = real(a)**2 + aimag(a)**2. do not take square root and square again
|
|
xi(1:3,i,j,k))*TWOPIIMG)**2.0_pReal)& ! --> sum squared L_2 norm of vector
|
|
+sum(aimag(math_mul33x3_complex(field_fourier(i,j,k,1:3,1:3),&
|
|
xi(1:3,i,j,k))*TWOPIIMG)**2.0_pReal))
|
|
enddo
|
|
Utilities_divergenceRMS = Utilities_divergenceRMS & ! Those two layers (DC and Nyquist) do not have a conjugate complex counterpart
|
|
+ sum( real(math_mul33x3_complex(field_fourier(1 ,j,k,1:3,1:3),&
|
|
xi(1:3,1 ,j,k))*TWOPIIMG)**2.0_pReal)&
|
|
+ sum(aimag(math_mul33x3_complex(field_fourier(1 ,j,k,1:3,1:3),&
|
|
xi(1:3,1 ,j,k))*TWOPIIMG)**2.0_pReal)&
|
|
+ sum( real(math_mul33x3_complex(field_fourier(res1_red,j,k,1:3,1:3),&
|
|
xi(1:3,res1_red,j,k))*TWOPIIMG)**2.0_pReal)&
|
|
+ sum(aimag(math_mul33x3_complex(field_fourier(res1_red,j,k,1:3,1:3),&
|
|
xi(1:3,res1_red,j,k))*TWOPIIMG)**2.0_pReal)
|
|
enddo; enddo
|
|
|
|
Utilities_divergenceRMS = sqrt(Utilities_divergenceRMS) *wgt ! RMS in real space calculated with Parsevals theorem from Fourier space
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
! calculate additional divergence criteria and report
|
|
if (debugDivergence) then ! calculate divergence again
|
|
err_div_max = 0.0_pReal
|
|
do k = 1_pInt, res(3); do j = 1_pInt, res(2); do i = 1_pInt, res1_red
|
|
temp3_Complex = math_mul33x3_complex(field_fourier(i,j,k,1:3,1:3)*wgt,& ! weighting P_fourier
|
|
xi(1:3,i,j,k))*TWOPIIMG
|
|
err_div_max = max(err_div_max,sum(abs(temp3_Complex)**2.0_pReal))
|
|
divergence_fourier(i,j,k,1:3) = temp3_Complex ! need divergence NOT squared
|
|
enddo; enddo; enddo
|
|
|
|
call fftw_execute_dft_c2r(plan_divergence,divergence_fourier,divergence_real) ! already weighted
|
|
|
|
err_real_div_RMS = sqrt(wgt*sum(divergence_real**2.0_pReal)) ! RMS in real space
|
|
err_post_div_RMS = sqrt(wgt*sum(divergence_post**2.0_pReal)) ! RMS in real space
|
|
err_real_div_max = sqrt(maxval(sum(divergence_real**2.0_pReal,dim=4))) ! max in real space
|
|
err_div_max = sqrt( err_div_max) ! max in Fourier space
|
|
|
|
write(6,'(1x,a,es11.4)') 'error divergence FT RMS = ',err_div_RMS
|
|
write(6,'(1x,a,es11.4)') 'error divergence Real RMS = ',err_real_div_RMS
|
|
write(6,'(1x,a,es11.4)') 'error divergence post RMS = ',err_post_div_RMS
|
|
write(6,'(1x,a,es11.4)') 'error divergence FT max = ',err_div_max
|
|
write(6,'(1x,a,es11.4)') 'error divergence Real max = ',err_real_div_max
|
|
endif
|
|
|
|
end function Utilities_divergenceRMS
|
|
|
|
|
|
!--------------------------------------------------------------------------------------------------
|
|
!> @brief calculates mask compliance
|
|
!--------------------------------------------------------------------------------------------------
|
|
function Utilities_maskedCompliance(rot_BC,mask_stressVector,C)
|
|
|
|
implicit none
|
|
real(pReal), dimension(3,3,3,3) :: Utilities_maskedCompliance
|
|
real(pReal), dimension(3,3,3,3), intent(in) :: C
|
|
integer(pInt) :: j, k, m, n
|
|
real(pReal), dimension(3,3), intent(in) :: rot_BC
|
|
logical, dimension(9), intent(in) :: mask_stressVector
|
|
real(pReal), dimension(3,3,3,3) :: C_lastInc
|
|
real(pReal), dimension(9,9) :: temp99_Real
|
|
integer(pInt) :: size_reduced = 0_pInt
|
|
real(pReal), dimension(:,:), allocatable :: s_reduced, c_reduced, sTimesC ! reduced compliance and stiffness (only for stress BC)
|
|
logical :: errmatinv
|
|
character(len=1024):: formatString
|
|
|
|
size_reduced = count(mask_stressVector)
|
|
if(size_reduced > 0_pInt )then
|
|
allocate (c_reduced(size_reduced,size_reduced), source =0.0_pReal)
|
|
allocate (s_reduced(size_reduced,size_reduced), source =0.0_pReal)
|
|
allocate (sTimesC(size_reduced,size_reduced), source =0.0_pReal)
|
|
|
|
C_lastInc = math_rotate_forward3333(C*0.75_pReal+0.05_pReal*C_ref,rot_BC) ! calculate stiffness from former inc
|
|
temp99_Real = math_Plain3333to99(C_lastInc)
|
|
k = 0_pInt ! build reduced stiffness
|
|
do n = 1_pInt,9_pInt
|
|
if(mask_stressVector(n)) then
|
|
k = k + 1_pInt
|
|
j = 0_pInt
|
|
do m = 1_pInt,9_pInt
|
|
if(mask_stressVector(m)) then
|
|
j = j + 1_pInt
|
|
c_reduced(k,j) = temp99_Real(n,m)
|
|
endif; enddo; endif; enddo
|
|
call math_invert(size_reduced, c_reduced, s_reduced, errmatinv) ! invert reduced stiffness
|
|
if(errmatinv) call IO_error(error_ID=400_pInt)
|
|
temp99_Real = 0.0_pReal ! build full compliance
|
|
k = 0_pInt
|
|
do n = 1_pInt,9_pInt
|
|
if(mask_stressVector(n)) then
|
|
k = k + 1_pInt
|
|
j = 0_pInt
|
|
do m = 1_pInt,9_pInt
|
|
if(mask_stressVector(m)) then
|
|
j = j + 1_pInt
|
|
temp99_Real(n,m) = s_reduced(k,j)
|
|
endif; enddo; endif; enddo
|
|
sTimesC = matmul(c_reduced,s_reduced)
|
|
do m=1_pInt, size_reduced
|
|
do n=1_pInt, size_reduced
|
|
if(m==n .and. abs(sTimesC(m,n)) > (1.0_pReal + 10.0e-12_pReal)) errmatinv = .true.
|
|
if(m/=n .and. abs(sTimesC(m,n)) > (0.0_pReal + 10.0e-12_pReal)) errmatinv = .true.
|
|
enddo
|
|
enddo
|
|
if(debugGeneral .or. errmatinv) then
|
|
write(formatString, '(I16.16)') size_reduced
|
|
formatString = '(a,/,'//trim(formatString)//'('//trim(formatString)//'(2x,es9.2,1x)/))'
|
|
write(6,trim(formatString),advance='no') 'C * S', transpose(matmul(c_reduced,s_reduced))
|
|
write(6,trim(formatString),advance='no') 'S', transpose(s_reduced)
|
|
endif
|
|
if(errmatinv) call IO_error(error_ID=400_pInt)
|
|
deallocate(c_reduced)
|
|
deallocate(s_reduced)
|
|
deallocate(sTimesC)
|
|
else
|
|
temp99_real = 0.0_pReal
|
|
endif
|
|
Utilities_maskedCompliance = math_Plain99to3333(temp99_Real)
|
|
|
|
end function Utilities_maskedCompliance
|
|
|
|
subroutine Utilities_constitutiveResponse(coordinates,F_lastInc,F,temperature,timeinc,&
|
|
P,C,P_av,ForwardData,rotation_BC)
|
|
use debug, only: &
|
|
debug_reset, &
|
|
debug_info
|
|
use CPFEM, only: &
|
|
CPFEM_general
|
|
use FEsolving, only: restartWrite
|
|
|
|
implicit none
|
|
real(pReal), dimension(res(1),res(2),res(3)) :: temperature
|
|
real(pReal), dimension(res(1),res(2),res(3),3), intent(in) :: coordinates
|
|
|
|
real(pReal), dimension(3,3,res(1),res(2),res(3)), intent(in) :: F,F_lastInc
|
|
real(pReal), dimension(3,3,res(1),res(2),res(3)) :: P
|
|
real(pReal) :: timeinc
|
|
logical, intent(in) :: forwardData
|
|
integer(pInt) :: i, j, k, ielem
|
|
integer(pInt) :: calcMode, collectMode
|
|
real(pReal), dimension(3,3,3,3) :: dPdF, C
|
|
real(pReal), dimension(6) :: sigma ! cauchy stress
|
|
real(pReal), dimension(6,6) :: dsde
|
|
real(pReal), dimension(3,3), intent(in) :: rotation_BC
|
|
real(pReal), dimension(3,3) :: P_av
|
|
|
|
write(6,'(/,a,/)') '... evaluating constitutive response ......................................'
|
|
if (forwardData) then
|
|
calcMode = 1_pInt
|
|
collectMode = 4_pInt
|
|
else
|
|
calcMode = 2_pInt
|
|
collectMode = 3_pInt
|
|
endif
|
|
if (cutBack) then
|
|
calcMode = 2_pInt
|
|
collectMode = 5_pInt
|
|
endif
|
|
print*, 'collect mode', collectMode
|
|
print*, 'calc mode', calcMode
|
|
ielem = 0_pInt
|
|
do k = 1_pInt, res(3); do j = 1_pInt, res(2); do i = 1_pInt, res(1)
|
|
ielem = ielem + 1_pInt
|
|
call CPFEM_general(collectMode,& ! collect cycle
|
|
coordinates(i,j,k,1:3), F_lastInc(1:3,1:3,i,j,k),F(1:3,1:3,i,j,k), &
|
|
temperature(i,j,k),timeinc,ielem,1_pInt,sigma,dsde,P(1:3,1:3,i,j,k),dPdF)
|
|
collectMode = 3_pInt
|
|
enddo; enddo; enddo
|
|
|
|
P = 0.0_pReal ! needed because of the padding for FFTW
|
|
C = 0.0_pReal
|
|
ielem = 0_pInt
|
|
call debug_reset()
|
|
do k = 1_pInt, res(3); do j = 1_pInt, res(2); do i = 1_pInt, res(1)
|
|
ielem = ielem + 1_pInt
|
|
call CPFEM_general(calcMode,& ! first element in first iteration retains CPFEM_mode 1,
|
|
coordinates(i,j,k,1:3),F_lastInc(1:3,1:3,i,j,k), F(1:3,1:3,i,j,k), & ! others get 2 (saves winding forward effort)
|
|
temperature(i,j,k),timeinc,ielem,1_pInt,sigma,dsde,P(1:3,1:3,i,j,k),dPdF)
|
|
calcMode = 2_pInt
|
|
C = C + dPdF
|
|
enddo; enddo; enddo
|
|
call debug_info()
|
|
|
|
P_av = math_rotate_forward33(sum(sum(sum(P,dim=5),dim=4),dim=3) * wgt,rotation_BC) !average of P rotated
|
|
restartWrite = .false.
|
|
cutBack = .false.
|
|
|
|
write(6,'(a,/,3(3(2x,f12.7,1x)/))',advance='no') 'Piola-Kirchhoff stress / MPa =',&
|
|
math_transpose33(P_av)/1.e6_pReal
|
|
|
|
C = C * wgt
|
|
end subroutine Utilities_constitutiveResponse
|
|
|
|
|
|
function Utilities_calculateRate(delta_aim,timeinc,timeinc_old,guessmode,field_lastInc,field)
|
|
implicit none
|
|
real(pReal), intent(in), dimension(3,3) :: delta_aim
|
|
real(pReal), intent(in) :: timeinc, timeinc_old, guessmode
|
|
real(pReal), intent(in), dimension(3,3,res(1),res(2),res(3)) :: field_lastInc,field
|
|
real(pReal), dimension(3,3,res(1),res(2),res(3)) :: Utilities_calculateRate
|
|
|
|
if (guessmode == 1.0_pReal) then
|
|
Utilities_calculateRate = (field-field_lastInc) / timeinc_old
|
|
else
|
|
Utilities_calculateRate = spread(spread(spread(delta_aim,3,res(1)),4,res(2)),5,res(3))/timeinc
|
|
endif
|
|
|
|
end function Utilities_calculateRate
|
|
|
|
|
|
function Utilities_forwardField(timeinc,aim,field_lastInc,rate)
|
|
implicit none
|
|
real(pReal), intent(in) :: timeinc
|
|
real(pReal), intent(in), dimension(3,3) :: aim
|
|
real(pReal), intent(in), dimension(3,3,res(1),res(2),res(3)) :: field_lastInc,rate
|
|
real(pReal), dimension(3,3,res(1),res(2),res(3)) :: Utilities_forwardField
|
|
real(pReal), dimension(3,3) :: fieldDiff
|
|
|
|
Utilities_forwardField = field_lastInc + rate*timeinc
|
|
fieldDiff = sum(sum(sum(Utilities_forwardField,dim=5),dim=4),dim=3)*wgt - aim
|
|
Utilities_forwardField = Utilities_forwardField - &
|
|
spread(spread(spread(fieldDiff,3,res(1)),4,res(2)),5,res(3))
|
|
|
|
end function Utilities_forwardField
|
|
|
|
real(pReal) function Utilities_getFilter(k)
|
|
|
|
use numerics, only: &
|
|
myfilter
|
|
|
|
implicit none
|
|
|
|
real(pReal), dimension(3),intent(in) :: k
|
|
|
|
select case (myfilter)
|
|
|
|
case ('none')
|
|
Utilities_getFilter = 1.0_pReal
|
|
|
|
case ('cosine')
|
|
Utilities_getFilter = (1.0_pReal + cos(pi*k(3)/res(3))) &
|
|
*(1.0_pReal + cos(pi*k(2)/res(2))) &
|
|
*(1.0_pReal + cos(pi*k(1)/res(1)))/8.0_pReal
|
|
|
|
case default
|
|
call IO_error(error_ID = 892_pInt, ext_msg = trim(myfilter))
|
|
|
|
end select
|
|
|
|
end function Utilities_getFilter
|
|
|
|
|
|
subroutine Utilities_destroy()
|
|
|
|
implicit none
|
|
if (debugDivergence) call fftw_destroy_plan(plan_divergence)
|
|
|
|
if (debugFFTW) then
|
|
call fftw_destroy_plan(plan_scalarField_forth)
|
|
call fftw_destroy_plan(plan_scalarField_back)
|
|
endif
|
|
|
|
call fftw_destroy_plan(plan_forward)
|
|
call fftw_destroy_plan(plan_backward)
|
|
|
|
end subroutine Utilities_destroy
|
|
|
|
|
|
end module DAMASK_spectral_Utilities
|