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/numerics.f90

509 lines
34 KiB
Fortran
Raw Normal View History

changed enconding of all source files to UTF-8 without BOM (signature) Codepage 65001
2011-04-07 12:50:28 +05:30
! Copyright 2011 Max-Planck-Institut für Eisenforschung GmbH
added copyright text to all f90 (free) format files
2011-04-04 19:39:54 +05:30
!
! This file is part of DAMASK,
changed enconding of all source files to UTF-8 without BOM (signature) Codepage 65001
2011-04-07 12:50:28 +05:30
! the Düsseldorf Advanced MAterial Simulation Kit.
added copyright text to all f90 (free) format files
2011-04-04 19:39:54 +05:30
!
! DAMASK is free software: you can redistribute it and/or modify
! it under the terms of the GNU General Public License as published by
! the Free Software Foundation, either version 3 of the License, or
! (at your option) any later version.
!
! DAMASK is distributed in the hope that it will be useful,
! but WITHOUT ANY WARRANTY; without even the implied warranty of
! MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
! GNU General Public License for more details.
!
! You should have received a copy of the GNU General Public License
! along with DAMASK. If not, see <http://www.gnu.org/licenses/>.
!
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! $Id$
!--------------------------------------------------------------------------------------------------
!> @author Franz Roters, Max-Planck-Institut für Eisenforschung GmbH
!> @author Philip Eisenlohr, Max-Planck-Institut für Eisenforschung GmbH
!> @brief Managing of parameters related to numerics
!--------------------------------------------------------------------------------------------------
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
module numerics
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
use prec, only: &
pInt, &
pReal
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
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
implicit none
private
character(len=64), parameter, private :: &
numerics_configFile = 'numerics.config' !< name of configuration file
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
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
integer(pInt), protected, public :: &
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
iJacoStiffness = 1_pInt, & !< frequency of stiffness update
added new cut back parameter for spectral solver and instrumented variables for use with doxygen
2012-09-13 15:18:38 +05:30
iJacoLpresiduum = 1_pInt, & !< frequency of Jacobian update of residuum in Lp
nHomog = 20_pInt, & !< homogenization loop limit (only for debugging info, loop limit is determined by "subStepMinHomog")
nMPstate = 10_pInt, & !< materialpoint state loop limit
nCryst = 20_pInt, & !< crystallite loop limit (only for debugging info, loop limit is determined by "subStepMinCryst")
nState = 10_pInt, & !< state loop limit
nStress = 40_pInt, & !< stress loop limit
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
pert_method = 1_pInt !< method used in perturbation technique for tangent
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
integer(pInt), public :: numerics_integrationMode = 0_pInt !< integrationMode 1 = central solution ; integrationMode 2 = perturbation, Default 0: undefined, is not read from file
integer(pInt), dimension(2) , protected, public :: &
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
numerics_integrator = 1_pInt !< method used for state integration (central & perturbed state), Default 1: fix-point iteration for both states
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
real(pReal), protected, public :: &
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
relevantStrain = 1.0e-7_pReal, & !< strain increment considered significant (used by crystallite to determine whether strain inc is considered significant)
added new cut back parameter for spectral solver and instrumented variables for use with doxygen
2012-09-13 15:18:38 +05:30
defgradTolerance = 1.0e-7_pReal, & !< deviation of deformation gradient that is still allowed (used by CPFEM to determine outdated ffn1)
pert_Fg = 1.0e-7_pReal, & !< strain perturbation for FEM Jacobi
subStepMinCryst = 1.0e-3_pReal, & !< minimum (relative) size of sub-step allowed during cutback in crystallite
subStepMinHomog = 1.0e-3_pReal, & !< minimum (relative) size of sub-step allowed during cutback in homogenization
subStepSizeCryst = 0.25_pReal, & !< size of first substep when cutback in crystallite
subStepSizeHomog = 0.25_pReal, & !< size of first substep when cutback in homogenization
stepIncreaseCryst = 1.5_pReal, & !< increase of next substep size when previous substep converged in crystallite
stepIncreaseHomog = 1.5_pReal, & !< increase of next substep size when previous substep converged in homogenization
rTol_crystalliteState = 1.0e-6_pReal, & !< relative tolerance in crystallite state loop
rTol_crystalliteTemperature= 1.0e-6_pReal, & !< relative tolerance in crystallite temperature loop
rTol_crystalliteStress = 1.0e-6_pReal, & !< relative tolerance in crystallite stress loop
aTol_crystalliteStress = 1.0e-8_pReal, & !< absolute tolerance in crystallite stress loop, Default 1.0e-8: residuum is in Lp and hence strain is on this order
forgot to check in; belongs to rev 1762
2012-09-24 21:52:25 +05:30
numerics_unitlength = 1.0_pReal, & !< determines the physical length of one computational length unit
added new cut back parameter for spectral solver and instrumented variables for use with doxygen
2012-09-13 15:18:38 +05:30
absTol_RGC = 1.0e+4_pReal, & !< absolute tolerance of RGC residuum
relTol_RGC = 1.0e-3_pReal, & !< relative tolerance of RGC residuum
absMax_RGC = 1.0e+10_pReal, & !< absolute maximum of RGC residuum
relMax_RGC = 1.0e+2_pReal, & !< relative maximum of RGC residuum
pPert_RGC = 1.0e-7_pReal, & !< perturbation for computing RGC penalty tangent
xSmoo_RGC = 1.0e-5_pReal, & !< RGC penalty smoothing parameter (hyperbolic tangent)
viscPower_RGC = 1.0e+0_pReal, & !< power (sensitivity rate) of numerical viscosity in RGC scheme, Default 1.0e0: Newton viscosity (linear model)
viscModus_RGC = 0.0e+0_pReal, & !< stress modulus of RGC numerical viscosity, Default 0.0e0: No viscosity is applied
refRelaxRate_RGC = 1.0e-3_pReal, & !< reference relaxation rate in RGC viscosity
maxdRelax_RGC = 1.0e+0_pReal, & !< threshold of maximum relaxation vector increment (if exceed this then cutback)
maxVolDiscr_RGC = 1.0e-5_pReal, & !< threshold of maximum volume discrepancy allowed
volDiscrMod_RGC = 1.0e+12_pReal, & !< stiffness of RGC volume discrepancy (zero = without volume discrepancy constraint)
volDiscrPow_RGC = 5.0_pReal !< powerlaw penalty for volume discrepancy
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
logical, protected, public :: &
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
analyticJaco = .false., & !< use analytic Jacobian or perturbation, Default .false.: calculate Jacobian using perturbations
numerics_timeSyncing = .false. !< flag indicating if time synchronization in crystallite is used for nonlocal plasticity
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!* Random seeding parameters
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
integer(pInt), protected, public :: &
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
fixedSeed = 0_pInt !< fixed seeding for pseudo-random number generator, Default 0: use random seed
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!* OpenMP variable
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
integer(pInt), protected, public :: &
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
DAMASK_NumThreadsInt = 0_pInt !< value stored in environment variable DAMASK_NUM_THREADS, set to zero if no OpenMP directive
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
added matrix multiplication 3333x33 to math.f90 added some parameters for spectral method to numerics.f90 (tolerance) changed error message concerning spectral method in IO.f90 corrected calculation of stress BC in mpie_spectral.f90
2010-10-13 21:34:44 +05:30
!* spectral parameters:
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
#ifdef Spectral
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
real(pReal), protected, public :: &
remove some unused variables, set default value for divergence calculation to 1.e-5 and made calculation independent of dim and res
2013-03-13 00:18:28 +05:30
err_div_tol = 1.0e-5_pReal, & !< Div(P)/avg(P)*meter
added new cut back parameter for spectral solver and instrumented variables for use with doxygen
2012-09-13 15:18:38 +05:30
err_stress_tolrel = 0.01_pReal, & !< relative tolerance for fullfillment of stress BC, Default: 0.01 allowing deviation of 1% of maximum stress
err_stress_tolabs = huge(1.0_pReal), & !< absolute tolerance for fullfillment of stress BC, Default: 0.01 allowing deviation of 1% of maximum stress
remove some unused variables, set default value for divergence calculation to 1.e-5 and made calculation independent of dim and res
2013-03-13 00:18:28 +05:30
err_f_tol = 1.0e-6_pReal, &
err_p_tol = 1.0e-5_pReal, &
added new cut back parameter for spectral solver and instrumented variables for use with doxygen
2012-09-13 15:18:38 +05:30
fftw_timelimit = -1.0_pReal, & !< sets the timelimit of plan creation for FFTW, see manual on www.fftw.org, Default -1.0: disable timelimit
restructured AL solver to resemble polarization scheme and introduced alpha and beta parameters in numerics to determine what kind of solution you want. alpha = beta = 1 results in the AL scheme. alpha = beta = 2 results in the accelerated scheme. safe values are anything in between. changed the solver to use stress calculations on the campatible field which seems to result in faster convergence.
2013-02-28 23:05:02 +05:30
rotation_tol = 1.0e-12_pReal, & !< tolerance of rotation specified in loadcase, Default 1.0e-12: first guess
polarAlpha = 1.0_pReal, & !< polarization scheme parameter 0.0 < alpha < 2.0. alpha = 1.0 ==> AL scheme, alpha = 2.0 ==> accelerated scheme
polarBeta = 1.0_pReal !< polarization scheme parameter 0.0 < beta < 2.0. beta = 1.0 ==> AL scheme, beta = 2.0 ==> accelerated scheme
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
character(len=64), private :: &
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
fftw_plan_mode = 'FFTW_PATIENT' !< reads the planing-rigor flag, see manual on www.fftw.org, Default FFTW_PATIENT: use patient planner flag
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
character(len=64), protected, public :: &
added new cut back parameter for spectral solver and instrumented variables for use with doxygen
2012-09-13 15:18:38 +05:30
myspectralsolver = 'basic' , & !< spectral solution method
myfilter = 'none' !< spectral filtering method
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
character(len=1024), protected, public :: &
made PETSc silent by removing -snes_view from default options. introduced PETSc option for debugging that introduces some debugging options into the petsc options and move PETSc initialization from numerics to DAMASK_spectral_utilities
2012-12-15 23:37:49 +05:30
petsc_options = '-snes_type ngmres &
corrected small issues with PETSc debugging
2013-01-03 21:47:23 +05:30
&-snes_ngmres_anderson '
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
integer(pInt), protected, public :: &
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
fftw_planner_flag = 32_pInt, & !< conversion of fftw_plan_mode to integer, basically what is usually done in the include file of fftw
set default number of iteration for spectral solver to 250 instead of 40, abaqus doxygen now does the correct inclusion (everything appeared twice before)
2013-03-15 04:40:02 +05:30
itmax = 250_pInt, & !< maximum number of iterations
added new cut back parameter for spectral solver and instrumented variables for use with doxygen
2012-09-13 15:18:38 +05:30
itmin = 2_pInt, & !< minimum number of iterations
introduced option for regridding to numerics.f90, working as follows: first cut back is tried if material point model oder BVP solver does not converge. If no regridding is enabled after max cut back, in case of non-converged material point the simulation stops and in case of non-converged BVP solver it continues. set regridMode to 2 to enable regridding if BVP solver OR materialPoint model do not converge, set regridMode to 1 to enable regridding if materialPoint model do not converge, non-converged BVP solver will be ignored as in the standard case. For regridding, the load case need to have a restart freq set. enabled restarting for Basic PETSc variant
2012-12-14 20:48:04 +05:30
maxCutBack = 3_pInt, & !< max number of cut backs
divergence_correction for basic solver variants has now 3 possibilities: 0: uncorrected, slope per sidelength (physical dimension) e = res/dim 1: corrected by sidelength, slope per unitlength e = res/1 2: corrected such that distance between FPs e = 1 alway regarding the medium length of x,y,z direction
2013-01-16 16:10:53 +05:30
regridMode = 0_pInt, & !< 0: no regrid; 1: regrid if DAMASK doesn't converge; 2: regrid if DAMASK or BVP Solver doesn't converge
remove some unused variables, set default value for divergence calculation to 1.e-5 and made calculation independent of dim and res
2013-03-13 00:18:28 +05:30
divergence_correction = 2_pInt !< correct divergence calculation in fourier space 0: no correction, 1: dimension scaled to 1, 2: dimension scaled to Npoints, 3: dimension scaled to sqrt(Npoints)
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
logical, protected , public :: &
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
memory_efficient = .true., & !< for fast execution (pre calculation of gamma_hat), Default .true.: do not precalculate
added new cut back parameter for spectral solver and instrumented variables for use with doxygen
2012-09-13 15:18:38 +05:30
update_gamma = .false. !< update gamma operator with current stiffness, Default .false.: use initial stiffness
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
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
#endif
mpie_spectral and mpie_interface: changed workingdir from pathToLoadFile to pathToGeomFile mpie_spectral and numerics: added switch to prevent pre calculation of gamma_hat. slower, but saves memory 3Dvisualize: started to add support for gmsh (not fully working yet) reconstruct: new version of f2py/Fortran subroutines for output of results from spectral method
2011-02-07 20:05:42 +05:30
corrected reporting of basic PETSc variant and unified reporting of all spectral solvers improved warning and error in IO, now able to report correctly elements up to 9 digits
2013-01-10 03:49:32 +05:30
public :: numerics_init
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
contains
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
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
!> @brief reads in parameters from numerics.config and sets openMP related parameters. Also does
! a sanity check
!--------------------------------------------------------------------------------------------------
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
subroutine numerics_init
fixed usage of OpenMP function library
2012-06-12 15:14:05 +05:30
use, intrinsic :: iso_fortran_env ! to get compiler_version and compiler_options (at least for gfortran 4.6 at the moment)
made PETSc silent by removing -snes_view from default options. introduced PETSc option for debugging that introduces some debugging options into the petsc options and move PETSc initialization from numerics to DAMASK_spectral_utilities
2012-12-15 23:37:49 +05:30
use IO, only: &
IO_error, &
IO_open_file_stat, &
IO_isBlank, &
IO_stringPos, &
IO_stringValue, &
IO_lc, &
IO_floatValue, &
IO_intValue, &
reporting current time for most module inits
2013-02-25 22:04:59 +05:30
IO_warning, &
IO_timeStamp
made PETSc silent by removing -snes_view from default options. introduced PETSc option for debugging that introduces some debugging options into the petsc options and move PETSc initialization from numerics to DAMASK_spectral_utilities
2012-12-15 23:37:49 +05:30
made new spectral solver the default (no need for SOLVER=NEW anymore) and added additional warnings switche to be used with gfortran >= 4.6 added comments
2012-12-15 21:51:10 +05:30
#ifndef Marc
!$ use OMP_LIB, only: omp_set_num_threads ! Use the standard conforming module file for omp if not using MSC.Marc
#endif
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
implicit none
made new spectral solver the default (no need for SOLVER=NEW anymore) and added additional warnings switche to be used with gfortran >= 4.6 added comments
2012-12-15 21:51:10 +05:30
#ifdef Marc
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!$ include "omp_lib.h" ! use the not F90 standard conforming include file to prevent crashes with some versions of MSC.Marc
made new spectral solver the default (no need for SOLVER=NEW anymore) and added additional warnings switche to be used with gfortran >= 4.6 added comments
2012-12-15 21:51:10 +05:30
#endif
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
integer(pInt), parameter :: fileunit = 300_pInt ,&
maxNchunks = 2_pInt
!$ integer :: gotDAMASK_NUM_THREADS = 1
integer(pInt), dimension(1+2*maxNchunks) :: positions
character(len=64) :: tag
character(len=1024) :: line
fixed usage of OpenMP function library
2012-06-12 15:14:05 +05:30
!$ character(len=6) DAMASK_NumThreadsString ! environment variable DAMASK_NUM_THREADS
mpie_cpfem_marc.f90 >>> changing the compilation sequence of modules: numerics.f90 and IO.f90 move up, while math.f90 down. IO.f90 >>> adding "inRad" instead of using inRad from math.f90 module, since math.f90 module is now below IO.f90. numerics.f90 >>> now can read the parameter "fixed_seed" from numerics.config. this parameter is used to set a pseudo-random number generator/fixed seeding. default value is 0 (zero), which will give fully random number. math.f90 >> adding a possibility to fix the seeding (i.e., pseudo random number) numerics.config >>> add a new parameter: "fixed_seed"
2009-08-27 21:00:40 +05:30
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
write(6,'(/,a)') ' <<<+- numerics init -+>>>'
write(6,'(a)') ' $Id$'
reporting current time for most module inits
2013-02-25 22:04:59 +05:30
write(6,'(a16,a)') ' Current time : ',IO_timeStamp()
made code standard conform to Fortran 2008 (ignoring warning concerning comments beyond character 132). Basically, changing "x" format specifier to "Nx" ("1x") plus removing $ format specifier added compiler switches for gfortran and ifort to check for standard conformity old gnu compilers <4.4 are not longer supported because they don't provide the c binding for fftw
2012-02-01 00:48:55 +05:30
#include "compilation_info.f90"
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
merge declaration and initialization with standard values corrected handling of $DAMASK_NUM_THREADS
2012-01-31 01:46:19 +05:30
!$ call GET_ENVIRONMENT_VARIABLE(NAME='DAMASK_NUM_THREADS',VALUE=DAMASK_NumThreadsString,STATUS=gotDAMASK_NUM_THREADS) ! get environment variable DAMASK_NUM_THREADS...
IO_error and IO_warning crashed for string of zero length (or only containing whitespaces), fixed now. Also added BEGIN and END statement to numerics in case of non-presen DAMASK_NUM_THREADS, would prevent this error and is better for debugging
2013-02-21 03:26:59 +05:30
!$ if(gotDAMASK_NUM_THREADS /= 0) &
!$ call IO_warning(35_pInt,ext_msg='BEGIN:'//DAMASK_NumThreadsString//':END')
merge declaration and initialization with standard values corrected handling of $DAMASK_NUM_THREADS
2012-01-31 01:46:19 +05:30
!$ read(DAMASK_NumThreadsString,'(i6)') DAMASK_NumThreadsInt ! ...convert it to integer...
polishing: removed variable names like 'unit' and 'data' that are keywords of fortran and ensured that integer and real precision matches independent of machine standard. removed cut_off parameter for damask_spectral removed outpot of derived divergence measures and added RMS output in brackets added comments and options to the makefile
2012-02-16 00:28:38 +05:30
!$ if (DAMASK_NumThreadsInt < 1_pInt) DAMASK_NumThreadsInt = 1_pInt ! ...ensure that its at least one...
N^2 initialization loop (former spectralPictureMode) rewritten in material.f90 additional output in DAMASK_spectral_interface.f90 132 character cut off in constitutive_nonlocal.f90 rounding error in math.f90 complex number initialization (1.0_pReal)*2.0_pReal*pi new $DAMASK_NUM_THREADS warning in numerics.f90 / IO.f90 polishing in DAMASK_spectral.f90
2012-01-30 19:22:41 +05:30
!$ call omp_set_num_threads(DAMASK_NumThreadsInt) ! ...and use it as number of threads for parallel execution
some polishing concerning openMP initialization number of threads is now written to *.out file
2010-12-02 16:34:29 +05:30
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! try to open the config file
fileExists: if(IO_open_file_stat(fileunit,numerics_configFile)) then
write(6,'(a,/)') ' using values from config file'
reordered (and partly redistributed) error message identifiers, deleted those which are not in use anymore. all constitutive as well numerics now raises an error, if an unknown keyword is found in the respective config file
2012-02-13 23:11:27 +05:30
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! read variables from config file and overwrite default parameters if keyword is present
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
line = ''
do
read(fileunit,'(a1024)',END=100) line
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
if (IO_isBlank(line)) cycle ! skip empty lines
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
positions = IO_stringPos(line,maxNchunks)
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
tag = IO_lc(IO_stringValue(line,positions,1_pInt)) ! extract key
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
select case(tag)
case ('relevantstrain')
relevantStrain = IO_floatValue(line,positions,2_pInt)
case ('defgradtolerance')
defgradTolerance = IO_floatValue(line,positions,2_pInt)
case ('ijacostiffness')
iJacoStiffness = IO_intValue(line,positions,2_pInt)
case ('ijacolpresiduum')
iJacoLpresiduum = IO_intValue(line,positions,2_pInt)
case ('pert_fg')
pert_Fg = IO_floatValue(line,positions,2_pInt)
case ('pert_method')
pert_method = IO_intValue(line,positions,2_pInt)
case ('nhomog')
nHomog = IO_intValue(line,positions,2_pInt)
case ('nmpstate')
nMPstate = IO_intValue(line,positions,2_pInt)
case ('ncryst')
nCryst = IO_intValue(line,positions,2_pInt)
case ('nstate')
nState = IO_intValue(line,positions,2_pInt)
case ('nstress')
nStress = IO_intValue(line,positions,2_pInt)
case ('substepmincryst')
subStepMinCryst = IO_floatValue(line,positions,2_pInt)
case ('substepsizecryst')
subStepSizeCryst = IO_floatValue(line,positions,2_pInt)
case ('stepincreasecryst')
stepIncreaseCryst = IO_floatValue(line,positions,2_pInt)
case ('substepminhomog')
subStepMinHomog = IO_floatValue(line,positions,2_pInt)
case ('substepsizehomog')
subStepSizeHomog = IO_floatValue(line,positions,2_pInt)
case ('stepincreasehomog')
stepIncreaseHomog = IO_floatValue(line,positions,2_pInt)
case ('rtol_crystallitestate')
rTol_crystalliteState = IO_floatValue(line,positions,2_pInt)
case ('rtol_crystallitetemperature')
rTol_crystalliteTemperature = IO_floatValue(line,positions,2_pInt)
case ('rtol_crystallitestress')
rTol_crystalliteStress = IO_floatValue(line,positions,2_pInt)
case ('atol_crystallitestress')
aTol_crystalliteStress = IO_floatValue(line,positions,2_pInt)
case ('integrator')
numerics_integrator(1) = IO_intValue(line,positions,2_pInt)
case ('integratorstiffness')
numerics_integrator(2) = IO_intValue(line,positions,2_pInt)
case ('analyticjaco')
analyticJaco = IO_intValue(line,positions,2_pInt) > 0_pInt
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
case ('timesyncing')
numerics_timeSyncing = IO_intValue(line,positions,2_pInt) > 0_pInt
forgot to check in; belongs to rev 1762
2012-09-24 21:52:25 +05:30
case ('unitlength')
numerics_unitlength = IO_floatValue(line,positions,2_pInt)
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
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! RGC parameters
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
case ('atol_rgc')
absTol_RGC = IO_floatValue(line,positions,2_pInt)
case ('rtol_rgc')
relTol_RGC = IO_floatValue(line,positions,2_pInt)
case ('amax_rgc')
absMax_RGC = IO_floatValue(line,positions,2_pInt)
case ('rmax_rgc')
relMax_RGC = IO_floatValue(line,positions,2_pInt)
case ('perturbpenalty_rgc')
pPert_RGC = IO_floatValue(line,positions,2_pInt)
case ('relevantmismatch_rgc')
xSmoo_RGC = IO_floatValue(line,positions,2_pInt)
case ('viscositypower_rgc')
viscPower_RGC = IO_floatValue(line,positions,2_pInt)
case ('viscositymodulus_rgc')
viscModus_RGC = IO_floatValue(line,positions,2_pInt)
case ('refrelaxationrate_rgc')
refRelaxRate_RGC = IO_floatValue(line,positions,2_pInt)
case ('maxrelaxation_rgc')
maxdRelax_RGC = IO_floatValue(line,positions,2_pInt)
case ('maxvoldiscrepancy_rgc')
maxVolDiscr_RGC = IO_floatValue(line,positions,2_pInt)
case ('voldiscrepancymod_rgc')
volDiscrMod_RGC = IO_floatValue(line,positions,2_pInt)
case ('discrepancypower_rgc')
volDiscrPow_RGC = IO_floatValue(line,positions,2_pInt)
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! random seeding parameters
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
case ('fixed_seed')
fixedSeed = IO_intValue(line,positions,2_pInt)
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! spectral parameters
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
#ifdef Spectral
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
case ('err_div_tol')
err_div_tol = IO_floatValue(line,positions,2_pInt)
case ('err_stress_tolrel')
err_stress_tolrel = IO_floatValue(line,positions,2_pInt)
added missing ":" in mesh.f90, introduced absolute stress tolerance for spectral solver in numerics.config/numerics.f90
2012-04-11 18:27:25 +05:30
case ('err_stress_tolabs')
err_stress_tolabs = IO_floatValue(line,positions,2_pInt)
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
case ('itmax')
itmax = IO_intValue(line,positions,2_pInt)
case ('itmin')
itmin = IO_intValue(line,positions,2_pInt)
added new cut back parameter for spectral solver and instrumented variables for use with doxygen
2012-09-13 15:18:38 +05:30
case ('maxcutback')
maxCutBack = IO_intValue(line,positions,2_pInt)
introduced option for regridding to numerics.f90, working as follows: first cut back is tried if material point model oder BVP solver does not converge. If no regridding is enabled after max cut back, in case of non-converged material point the simulation stops and in case of non-converged BVP solver it continues. set regridMode to 2 to enable regridding if BVP solver OR materialPoint model do not converge, set regridMode to 1 to enable regridding if materialPoint model do not converge, non-converged BVP solver will be ignored as in the standard case. For regridding, the load case need to have a restart freq set. enabled restarting for Basic PETSc variant
2012-12-14 20:48:04 +05:30
case ('regridmode')
regridMode = IO_intValue(line,positions,2_pInt)
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
case ('memory_efficient')
memory_efficient = IO_intValue(line,positions,2_pInt) > 0_pInt
case ('fftw_timelimit')
fftw_timelimit = IO_floatValue(line,positions,2_pInt)
case ('fftw_plan_mode')
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
fftw_plan_mode = IO_lc(IO_stringValue(line,positions,2_pInt))
added filter option to the spectral solver to mitigate spurious oscillations due to gibb's phenomenon. activate by setting myfilter in config file appropriately (currently only 'none' and 'cosine' options coded). more cleaning up of AL code
2012-08-07 22:53:13 +05:30
case ('myfilter')
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
myfilter = IO_lc(IO_stringValue(line,positions,2_pInt))
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
case ('rotation_tol')
rotation_tol = IO_floatValue(line,positions,2_pInt)
case ('divergence_correction')
divergence_correction for basic solver variants has now 3 possibilities: 0: uncorrected, slope per sidelength (physical dimension) e = res/dim 1: corrected by sidelength, slope per unitlength e = res/1 2: corrected such that distance between FPs e = 1 alway regarding the medium length of x,y,z direction
2013-01-16 16:10:53 +05:30
divergence_correction = IO_intValue(line,positions,2_pInt)
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
case ('update_gamma')
update_gamma = IO_intValue(line,positions,2_pInt) > 0_pInt
new solver is now compiling without a PETSc installation, however only the plain basic solver is available then. also removed reporting of PETSc related variables in the case it is not installed/found
2012-08-29 00:49:47 +05:30
#ifdef PETSc
case ('petsc_options')
petsc_options = trim(line(positions(4):))
case ('myspectralsolver')
numerics got some protected statements and is now reading in the new keywords for the solver selection in small letters. spectral solver got cut back facilities + improved output to screen
2012-10-02 20:56:56 +05:30
myspectralsolver = IO_lc(IO_stringValue(line,positions,2_pInt))
new solver is now compiling without a PETSc installation, however only the plain basic solver is available then. also removed reporting of PETSc related variables in the case it is not installed/found
2012-08-29 00:49:47 +05:30
case ('err_f_tol')
err_f_tol = IO_floatValue(line,positions,2_pInt)
case ('err_p_tol')
err_p_tol = IO_floatValue(line,positions,2_pInt)
restructured AL solver to resemble polarization scheme and introduced alpha and beta parameters in numerics to determine what kind of solution you want. alpha = beta = 1 results in the AL scheme. alpha = beta = 2 results in the accelerated scheme. safe values are anything in between. changed the solver to use stress calculations on the campatible field which seems to result in faster convergence.
2013-02-28 23:05:02 +05:30
case ('polaralpha')
polarAlpha = IO_floatValue(line,positions,2_pInt)
case ('polarbeta')
polarBeta = IO_floatValue(line,positions,2_pInt)
new solver is now compiling without a PETSc installation, however only the plain basic solver is available then. also removed reporting of PETSc related variables in the case it is not installed/found
2012-08-29 00:49:47 +05:30
#endif
#ifndef PETSc
restructured AL solver to resemble polarization scheme and introduced alpha and beta parameters in numerics to determine what kind of solution you want. alpha = beta = 1 results in the AL scheme. alpha = beta = 2 results in the accelerated scheme. safe values are anything in between. changed the solver to use stress calculations on the campatible field which seems to result in faster convergence.
2013-02-28 23:05:02 +05:30
case ('myspectralsolver', 'petsc_options','err_f_tol', 'err_p_tol', &
'polaralpha','polarBeta') ! found PETSc parameter, but compiled without PETSc
new solver is now compiling without a PETSc installation, however only the plain basic solver is available then. also removed reporting of PETSc related variables in the case it is not installed/found
2012-08-29 00:49:47 +05:30
call IO_warning(41_pInt,ext_msg=tag)
#endif
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
#endif
#ifndef Spectral
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
case ('err_div_tol','err_stress_tolrel','err_stress_tolabs',& ! found spectral parameter for FEM build
next round in modularization
2012-07-25 19:31:39 +05:30
'itmax', 'itmin','memory_efficient','fftw_timelimit','fftw_plan_mode','myspectralsolver', &
new solver is now compiling without a PETSc installation, however only the plain basic solver is available then. also removed reporting of PETSc related variables in the case it is not installed/found
2012-08-29 00:49:47 +05:30
'rotation_tol','divergence_correction','update_gamma','petsc_options','myfilter', &
restructured AL solver to resemble polarization scheme and introduced alpha and beta parameters in numerics to determine what kind of solution you want. alpha = beta = 1 results in the AL scheme. alpha = beta = 2 results in the accelerated scheme. safe values are anything in between. changed the solver to use stress calculations on the campatible field which seems to result in faster convergence.
2013-02-28 23:05:02 +05:30
'err_f_tol', 'err_p_tol', 'maxcutback','polaralpha','polarbeta')
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
call IO_warning(40_pInt,ext_msg=tag)
#endif
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
case default ! found unknown keyword
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
call IO_error(300_pInt,ext_msg=tag)
endselect
enddo
100 close(fileunit)
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
else fileExists
write(6,'(a,/)') ' using standard values'
endif fileExists
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
#ifdef Spectral
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
select case(IO_lc(fftw_plan_mode)) ! setting parameters for the plan creation of FFTW. Basically a translation from fftw3.f
case('estimate','fftw_estimate') ! ordered from slow execution (but fast plan creation) to fast execution
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
fftw_planner_flag = 64_pInt
case('measure','fftw_measure')
fftw_planner_flag = 0_pInt
case('patient','fftw_patient')
fftw_planner_flag= 32_pInt
case('exhaustive','fftw_exhaustive')
fftw_planner_flag = 8_pInt
case default
call IO_warning(warning_ID=47_pInt,ext_msg=trim(IO_lc(fftw_plan_mode)))
fftw_planner_flag = 32_pInt
end select
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
#endif
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
numerics_timeSyncing = numerics_timeSyncing .and. all(numerics_integrator==2_pInt) ! timeSyncing only allowed for explicit Euler integrator
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! writing parameters to output file
debug.config, debug.f90, DAMASK_abaqus_exp.f, DAMASK_abaqus_std.f: changed to new debug scheme (wasn't working) lattice.f90, FEsolving.f90: explicitly defined public functions and variables, all others are now private numerics.f90: changed output format of real numbers, now instead of 0.1eX 1.0e(X-1) is printed to screen Makefile: now using correct Optimization flags for OPTIMIZATION=AGGRESSIVE DAMASK_spectral_AL.f90: improved, but still testing. Stress BCs now seem to be handled correctly
2012-03-20 23:31:31 +05:30
write(6,'(a24,1x,es8.1)') ' relevantStrain: ',relevantStrain
write(6,'(a24,1x,es8.1)') ' defgradTolerance: ',defgradTolerance
write(6,'(a24,1x,i8)') ' iJacoStiffness: ',iJacoStiffness
write(6,'(a24,1x,i8)') ' iJacoLpresiduum: ',iJacoLpresiduum
write(6,'(a24,1x,es8.1)') ' pert_Fg: ',pert_Fg
write(6,'(a24,1x,i8)') ' pert_method: ',pert_method
write(6,'(a24,1x,i8)') ' nCryst: ',nCryst
write(6,'(a24,1x,es8.1)') ' subStepMinCryst: ',subStepMinCryst
write(6,'(a24,1x,es8.1)') ' subStepSizeCryst: ',subStepSizeCryst
write(6,'(a24,1x,es8.1)') ' stepIncreaseCryst: ',stepIncreaseCryst
write(6,'(a24,1x,i8)') ' nState: ',nState
write(6,'(a24,1x,i8)') ' nStress: ',nStress
write(6,'(a24,1x,es8.1)') ' rTol_crystalliteState: ',rTol_crystalliteState
write(6,'(a24,1x,es8.1)') ' rTol_crystalliteTemp: ',rTol_crystalliteTemperature
write(6,'(a24,1x,es8.1)') ' rTol_crystalliteStress: ',rTol_crystalliteStress
write(6,'(a24,1x,es8.1)') ' aTol_crystalliteStress: ',aTol_crystalliteStress
write(6,'(a24,2(1x,i8))') ' integrator: ',numerics_integrator
Introduced an alternative cutback scheme for the nonlocal model, that allows to keep the results of most of the ips that immediately converged, and only do cutbacks in regions where some ips are in big trouble. Only works for nonlocal model and state integrator 2.
2012-11-28 00:06:55 +05:30
write(6,'(a24,1x,L8)') ' timeSyncing: ',numerics_timeSyncing
forgot to check in; belongs to rev 1762
2012-09-24 21:52:25 +05:30
write(6,'(a24,1x,L8)') ' analytic Jacobian: ',analyticJaco
write(6,'(a24,1x,es8.1,/)')' unitlength: ',numerics_unitlength
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
debug.config, debug.f90, DAMASK_abaqus_exp.f, DAMASK_abaqus_std.f: changed to new debug scheme (wasn't working) lattice.f90, FEsolving.f90: explicitly defined public functions and variables, all others are now private numerics.f90: changed output format of real numbers, now instead of 0.1eX 1.0e(X-1) is printed to screen Makefile: now using correct Optimization flags for OPTIMIZATION=AGGRESSIVE DAMASK_spectral_AL.f90: improved, but still testing. Stress BCs now seem to be handled correctly
2012-03-20 23:31:31 +05:30
write(6,'(a24,1x,i8)') ' nHomog: ',nHomog
write(6,'(a24,1x,es8.1)') ' subStepMinHomog: ',subStepMinHomog
write(6,'(a24,1x,es8.1)') ' subStepSizeHomog: ',subStepSizeHomog
write(6,'(a24,1x,es8.1)') ' stepIncreaseHomog: ',stepIncreaseHomog
write(6,'(a24,1x,i8,/)') ' nMPstate: ',nMPstate
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
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! RGC parameters
debug.config, debug.f90, DAMASK_abaqus_exp.f, DAMASK_abaqus_std.f: changed to new debug scheme (wasn't working) lattice.f90, FEsolving.f90: explicitly defined public functions and variables, all others are now private numerics.f90: changed output format of real numbers, now instead of 0.1eX 1.0e(X-1) is printed to screen Makefile: now using correct Optimization flags for OPTIMIZATION=AGGRESSIVE DAMASK_spectral_AL.f90: improved, but still testing. Stress BCs now seem to be handled correctly
2012-03-20 23:31:31 +05:30
write(6,'(a24,1x,es8.1)') ' aTol_RGC: ',absTol_RGC
write(6,'(a24,1x,es8.1)') ' rTol_RGC: ',relTol_RGC
write(6,'(a24,1x,es8.1)') ' aMax_RGC: ',absMax_RGC
write(6,'(a24,1x,es8.1)') ' rMax_RGC: ',relMax_RGC
write(6,'(a24,1x,es8.1)') ' perturbPenalty_RGC: ',pPert_RGC
write(6,'(a24,1x,es8.1)') ' relevantMismatch_RGC: ',xSmoo_RGC
write(6,'(a24,1x,es8.1)') ' viscosityrate_RGC: ',viscPower_RGC
write(6,'(a24,1x,es8.1)') ' viscositymodulus_RGC: ',viscModus_RGC
write(6,'(a24,1x,es8.1)') ' maxrelaxation_RGC: ',maxdRelax_RGC
write(6,'(a24,1x,es8.1)') ' maxVolDiscrepancy_RGC: ',maxVolDiscr_RGC
write(6,'(a24,1x,es8.1)') ' volDiscrepancyMod_RGC: ',volDiscrMod_RGC
write(6,'(a24,1x,es8.1,/)') ' discrepancyPower_RGC: ',volDiscrPow_RGC
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! Random seeding parameter
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
write(6,'(a24,1x,i16,/)') ' fixed_seed: ',fixedSeed
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! openMP parameter
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
!$ write(6,'(a24,1x,i8,/)') ' number of threads: ',DAMASK_NumThreadsInt
added matrix multiplication 3333x33 to math.f90 added some parameters for spectral method to numerics.f90 (tolerance) changed error message concerning spectral method in IO.f90 corrected calculation of stress BC in mpie_spectral.f90
2010-10-13 21:34:44 +05:30
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! spectral parameters
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
#ifdef Spectral
debug.config, debug.f90, DAMASK_abaqus_exp.f, DAMASK_abaqus_std.f: changed to new debug scheme (wasn't working) lattice.f90, FEsolving.f90: explicitly defined public functions and variables, all others are now private numerics.f90: changed output format of real numbers, now instead of 0.1eX 1.0e(X-1) is printed to screen Makefile: now using correct Optimization flags for OPTIMIZATION=AGGRESSIVE DAMASK_spectral_AL.f90: improved, but still testing. Stress BCs now seem to be handled correctly
2012-03-20 23:31:31 +05:30
write(6,'(a24,1x,es8.1)') ' err_div_tol: ',err_div_tol
write(6,'(a24,1x,es8.1)') ' err_stress_tolrel: ',err_stress_tolrel
added missing ":" in mesh.f90, introduced absolute stress tolerance for spectral solver in numerics.config/numerics.f90
2012-04-11 18:27:25 +05:30
write(6,'(a24,1x,es8.1)') ' err_stress_tolabs: ',err_stress_tolabs
new solver is now compiling without a PETSc installation, however only the plain basic solver is available then. also removed reporting of PETSc related variables in the case it is not installed/found
2012-08-29 00:49:47 +05:30
debug.config, debug.f90, DAMASK_abaqus_exp.f, DAMASK_abaqus_std.f: changed to new debug scheme (wasn't working) lattice.f90, FEsolving.f90: explicitly defined public functions and variables, all others are now private numerics.f90: changed output format of real numbers, now instead of 0.1eX 1.0e(X-1) is printed to screen Makefile: now using correct Optimization flags for OPTIMIZATION=AGGRESSIVE DAMASK_spectral_AL.f90: improved, but still testing. Stress BCs now seem to be handled correctly
2012-03-20 23:31:31 +05:30
write(6,'(a24,1x,i8)') ' itmax: ',itmax
write(6,'(a24,1x,i8)') ' itmin: ',itmin
added new cut back parameter for spectral solver and instrumented variables for use with doxygen
2012-09-13 15:18:38 +05:30
write(6,'(a24,1x,i8)') ' maxCutBack: ',maxCutBack
introduced option for regridding to numerics.f90, working as follows: first cut back is tried if material point model oder BVP solver does not converge. If no regridding is enabled after max cut back, in case of non-converged material point the simulation stops and in case of non-converged BVP solver it continues. set regridMode to 2 to enable regridding if BVP solver OR materialPoint model do not converge, set regridMode to 1 to enable regridding if materialPoint model do not converge, non-converged BVP solver will be ignored as in the standard case. For regridding, the load case need to have a restart freq set. enabled restarting for Basic PETSc variant
2012-12-14 20:48:04 +05:30
write(6,'(a24,1x,i8)') ' regridMode: ',regridMode
debug.config, debug.f90, DAMASK_abaqus_exp.f, DAMASK_abaqus_std.f: changed to new debug scheme (wasn't working) lattice.f90, FEsolving.f90: explicitly defined public functions and variables, all others are now private numerics.f90: changed output format of real numbers, now instead of 0.1eX 1.0e(X-1) is printed to screen Makefile: now using correct Optimization flags for OPTIMIZATION=AGGRESSIVE DAMASK_spectral_AL.f90: improved, but still testing. Stress BCs now seem to be handled correctly
2012-03-20 23:31:31 +05:30
write(6,'(a24,1x,L8)') ' memory_efficient: ',memory_efficient
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
if(fftw_timelimit<0.0_pReal) then
debug.config, debug.f90, DAMASK_abaqus_exp.f, DAMASK_abaqus_std.f: changed to new debug scheme (wasn't working) lattice.f90, FEsolving.f90: explicitly defined public functions and variables, all others are now private numerics.f90: changed output format of real numbers, now instead of 0.1eX 1.0e(X-1) is printed to screen Makefile: now using correct Optimization flags for OPTIMIZATION=AGGRESSIVE DAMASK_spectral_AL.f90: improved, but still testing. Stress BCs now seem to be handled correctly
2012-03-20 23:31:31 +05:30
write(6,'(a24,1x,L8)') ' fftw_timelimit: ',.false.
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
else
debug.config, debug.f90, DAMASK_abaqus_exp.f, DAMASK_abaqus_std.f: changed to new debug scheme (wasn't working) lattice.f90, FEsolving.f90: explicitly defined public functions and variables, all others are now private numerics.f90: changed output format of real numbers, now instead of 0.1eX 1.0e(X-1) is printed to screen Makefile: now using correct Optimization flags for OPTIMIZATION=AGGRESSIVE DAMASK_spectral_AL.f90: improved, but still testing. Stress BCs now seem to be handled correctly
2012-03-20 23:31:31 +05:30
write(6,'(a24,1x,es8.1)') ' fftw_timelimit: ',fftw_timelimit
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
endif
debug.config, debug.f90, DAMASK_abaqus_exp.f, DAMASK_abaqus_std.f: changed to new debug scheme (wasn't working) lattice.f90, FEsolving.f90: explicitly defined public functions and variables, all others are now private numerics.f90: changed output format of real numbers, now instead of 0.1eX 1.0e(X-1) is printed to screen Makefile: now using correct Optimization flags for OPTIMIZATION=AGGRESSIVE DAMASK_spectral_AL.f90: improved, but still testing. Stress BCs now seem to be handled correctly
2012-03-20 23:31:31 +05:30
write(6,'(a24,1x,a)') ' fftw_plan_mode: ',trim(fftw_plan_mode)
new solver is now compiling without a PETSc installation, however only the plain basic solver is available then. also removed reporting of PETSc related variables in the case it is not installed/found
2012-08-29 00:49:47 +05:30
added filter option to the spectral solver to mitigate spurious oscillations due to gibb's phenomenon. activate by setting myfilter in config file appropriately (currently only 'none' and 'cosine' options coded). more cleaning up of AL code
2012-08-07 22:53:13 +05:30
write(6,'(a24,1x,a)') ' myfilter: ',trim(myfilter)
debug.config, debug.f90, DAMASK_abaqus_exp.f, DAMASK_abaqus_std.f: changed to new debug scheme (wasn't working) lattice.f90, FEsolving.f90: explicitly defined public functions and variables, all others are now private numerics.f90: changed output format of real numbers, now instead of 0.1eX 1.0e(X-1) is printed to screen Makefile: now using correct Optimization flags for OPTIMIZATION=AGGRESSIVE DAMASK_spectral_AL.f90: improved, but still testing. Stress BCs now seem to be handled correctly
2012-03-20 23:31:31 +05:30
write(6,'(a24,1x,i8)') ' fftw_planner_flag: ',fftw_planner_flag
write(6,'(a24,1x,es8.1)') ' rotation_tol: ',rotation_tol
divergence_correction for basic solver variants has now 3 possibilities: 0: uncorrected, slope per sidelength (physical dimension) e = res/dim 1: corrected by sidelength, slope per unitlength e = res/1 2: corrected such that distance between FPs e = 1 alway regarding the medium length of x,y,z direction
2013-01-16 16:10:53 +05:30
write(6,'(a24,1x,i8)') ' divergence_correction: ',divergence_correction
debug.config, debug.f90, DAMASK_abaqus_exp.f, DAMASK_abaqus_std.f: changed to new debug scheme (wasn't working) lattice.f90, FEsolving.f90: explicitly defined public functions and variables, all others are now private numerics.f90: changed output format of real numbers, now instead of 0.1eX 1.0e(X-1) is printed to screen Makefile: now using correct Optimization flags for OPTIMIZATION=AGGRESSIVE DAMASK_spectral_AL.f90: improved, but still testing. Stress BCs now seem to be handled correctly
2012-03-20 23:31:31 +05:30
write(6,'(a24,1x,L8,/)') ' update_gamma: ',update_gamma
new solver is now compiling without a PETSc installation, however only the plain basic solver is available then. also removed reporting of PETSc related variables in the case it is not installed/found
2012-08-29 00:49:47 +05:30
#ifdef PETSc
write(6,'(a24,1x,es8.1)') ' err_f_tol: ',err_f_tol
write(6,'(a24,1x,es8.1)') ' err_p_tol: ',err_p_tol
restructured AL solver to resemble polarization scheme and introduced alpha and beta parameters in numerics to determine what kind of solution you want. alpha = beta = 1 results in the AL scheme. alpha = beta = 2 results in the accelerated scheme. safe values are anything in between. changed the solver to use stress calculations on the campatible field which seems to result in faster convergence.
2013-02-28 23:05:02 +05:30
write(6,'(a24,1x,es8.1)') ' polarAlpha: ',polarAlpha
write(6,'(a24,1x,es8.1)') ' polarBeta: ',polarBeta
new solver is now compiling without a PETSc installation, however only the plain basic solver is available then. also removed reporting of PETSc related variables in the case it is not installed/found
2012-08-29 00:49:47 +05:30
write(6,'(a24,1x,a)') ' myspectralsolver: ',trim(myspectralsolver)
write(6,'(a24,1x,a)') ' PETSc_options: ',trim(petsc_options)
#endif
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
#endif
some polishing concerning openMP initialization number of threads is now written to *.out file
2010-12-02 16:34:29 +05:30
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
!--------------------------------------------------------------------------------------------------
! sanity checks
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
if (relevantStrain <= 0.0_pReal) call IO_error(301_pInt,ext_msg='relevantStrain')
if (defgradTolerance <= 0.0_pReal) call IO_error(301_pInt,ext_msg='defgradTolerance')
if (iJacoStiffness < 1_pInt) call IO_error(301_pInt,ext_msg='iJacoStiffness')
if (iJacoLpresiduum < 1_pInt) call IO_error(301_pInt,ext_msg='iJacoLpresiduum')
if (pert_Fg <= 0.0_pReal) call IO_error(301_pInt,ext_msg='pert_Fg')
if (pert_method <= 0_pInt .or. pert_method >= 4_pInt) &
call IO_error(301_pInt,ext_msg='pert_method')
if (nHomog < 1_pInt) call IO_error(301_pInt,ext_msg='nHomog')
if (nMPstate < 1_pInt) call IO_error(301_pInt,ext_msg='nMPstate')
if (nCryst < 1_pInt) call IO_error(301_pInt,ext_msg='nCryst')
if (nState < 1_pInt) call IO_error(301_pInt,ext_msg='nState')
if (nStress < 1_pInt) call IO_error(301_pInt,ext_msg='nStress')
if (subStepMinCryst <= 0.0_pReal) call IO_error(301_pInt,ext_msg='subStepMinCryst')
if (subStepSizeCryst <= 0.0_pReal) call IO_error(301_pInt,ext_msg='subStepSizeCryst')
if (stepIncreaseCryst <= 0.0_pReal) call IO_error(301_pInt,ext_msg='stepIncreaseCryst')
if (subStepMinHomog <= 0.0_pReal) call IO_error(301_pInt,ext_msg='subStepMinHomog')
if (subStepSizeHomog <= 0.0_pReal) call IO_error(301_pInt,ext_msg='subStepSizeHomog')
if (stepIncreaseHomog <= 0.0_pReal) call IO_error(301_pInt,ext_msg='stepIncreaseHomog')
if (rTol_crystalliteState <= 0.0_pReal) call IO_error(301_pInt,ext_msg='rTol_crystalliteState')
if (rTol_crystalliteTemperature <= 0.0_pReal) call IO_error(301_pInt,ext_msg='rTol_crystalliteTemperature')
if (rTol_crystalliteStress <= 0.0_pReal) call IO_error(301_pInt,ext_msg='rTol_crystalliteStress')
if (aTol_crystalliteStress <= 0.0_pReal) call IO_error(301_pInt,ext_msg='aTol_crystalliteStress')
if (any(numerics_integrator <= 0_pInt) .or. any(numerics_integrator >= 6_pInt)) &
call IO_error(301_pInt,ext_msg='integrator')
forgot to check in; belongs to rev 1762
2012-09-24 21:52:25 +05:30
if (numerics_unitlength <= 0.0_pReal) call IO_error(301_pInt,ext_msg='unitlength')
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
if (absTol_RGC <= 0.0_pReal) call IO_error(301_pInt,ext_msg='absTol_RGC')
if (relTol_RGC <= 0.0_pReal) call IO_error(301_pInt,ext_msg='relTol_RGC')
if (absMax_RGC <= 0.0_pReal) call IO_error(301_pInt,ext_msg='absMax_RGC')
if (relMax_RGC <= 0.0_pReal) call IO_error(301_pInt,ext_msg='relMax_RGC')
if (pPert_RGC <= 0.0_pReal) call IO_error(301_pInt,ext_msg='pPert_RGC')
if (xSmoo_RGC <= 0.0_pReal) call IO_error(301_pInt,ext_msg='xSmoo_RGC')
if (viscPower_RGC < 0.0_pReal) call IO_error(301_pInt,ext_msg='viscPower_RGC')
if (viscModus_RGC < 0.0_pReal) call IO_error(301_pInt,ext_msg='viscModus_RGC')
if (refRelaxRate_RGC <= 0.0_pReal) call IO_error(301_pInt,ext_msg='refRelaxRate_RGC')
if (maxdRelax_RGC <= 0.0_pReal) call IO_error(301_pInt,ext_msg='maxdRelax_RGC')
if (maxVolDiscr_RGC <= 0.0_pReal) call IO_error(301_pInt,ext_msg='maxVolDiscr_RGC')
if (volDiscrMod_RGC < 0.0_pReal) call IO_error(301_pInt,ext_msg='volDiscrMod_RGC')
if (volDiscrPow_RGC <= 0.0_pReal) call IO_error(301_pInt,ext_msg='volDiscrPw_RGC')
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
#ifdef Spectral
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
if (err_div_tol <= 0.0_pReal) call IO_error(301_pInt,ext_msg='err_div_tol')
if (err_stress_tolrel <= 0.0_pReal) call IO_error(301_pInt,ext_msg='err_stress_tolrel')
added missing ":" in mesh.f90, introduced absolute stress tolerance for spectral solver in numerics.config/numerics.f90
2012-04-11 18:27:25 +05:30
if (err_stress_tolabs <= 0.0_pReal) call IO_error(301_pInt,ext_msg='err_stress_tolabs')
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
if (itmax <= 1.0_pInt) call IO_error(301_pInt,ext_msg='itmax')
in case of regridding, undeformed configuration is written out correctly. Restart improve by reading in additional data from converged step.
2012-07-24 17:51:41 +05:30
if (itmin > itmax .or. itmin < 1_pInt) call IO_error(301_pInt,ext_msg='itmin')
divergence_correction for basic solver variants has now 3 possibilities: 0: uncorrected, slope per sidelength (physical dimension) e = res/dim 1: corrected by sidelength, slope per unitlength e = res/1 2: corrected such that distance between FPs e = 1 alway regarding the medium length of x,y,z direction
2013-01-16 16:10:53 +05:30
if (divergence_correction < 0_pInt .or. &
forgot to commit
2013-02-08 21:26:58 +05:30
divergence_correction > 3_pInt) call IO_error(301_pInt,ext_msg='divergence_correction')
added new cut back parameter for spectral solver and instrumented variables for use with doxygen
2012-09-13 15:18:38 +05:30
if (maxCutBack <= 1.0_pInt) call IO_error(301_pInt,ext_msg='maxCutBack')
merged code for python and spectral solver interfacing (shared most of it anyway). put functionality for getting header information (dimension, homogenization, resolution) in functions in mesh.f90
2012-04-11 22:58:08 +05:30
if (update_gamma .and. &
.not. memory_efficient) call IO_error(error_ID = 847_pInt)
new solver is now compiling without a PETSc installation, however only the plain basic solver is available then. also removed reporting of PETSc related variables in the case it is not installed/found
2012-08-29 00:49:47 +05:30
#ifdef PETSc
if (err_f_tol <= 0.0_pReal) call IO_error(301_pInt,ext_msg='err_f_tol')
if (err_p_tol <= 0.0_pReal) call IO_error(301_pInt,ext_msg='err_p_tol')
restructured AL solver to resemble polarization scheme and introduced alpha and beta parameters in numerics to determine what kind of solution you want. alpha = beta = 1 results in the AL scheme. alpha = beta = 2 results in the accelerated scheme. safe values are anything in between. changed the solver to use stress calculations on the campatible field which seems to result in faster convergence.
2013-02-28 23:05:02 +05:30
if (polarAlpha <= 0.0_pReal .or. &
polarAlpha > 2.0_pReal) call IO_error(301_pInt,ext_msg='polarAlpha')
if (polarBeta <= 0.0_pReal .or. &
polarBeta > 2.0_pReal) call IO_error(301_pInt,ext_msg='polarBeta')
if (err_p_tol <= 0.0_pReal) call IO_error(301_pInt,ext_msg='err_p_tol')
new solver is now compiling without a PETSc installation, however only the plain basic solver is available then. also removed reporting of PETSc related variables in the case it is not installed/found
2012-08-29 00:49:47 +05:30
#endif
introduced preprocessor identifiers Marc, Abaqus, and Spectral to enable conditional compilation. This allows deleted dummy functions that are used by one solver only. Mainly affected modules are IO and mesh. Most of the changes in mesh result from reordering the functions when grouping them depending on their solver. Further advantage is that FE solver do not need FFTW and kdtree2 anymore. The include files for these two libraries moved to DAMASKROO/lib now as I figured out how to use a include path in the Makefile. Put all the files I got when testing compilation with abaqus in a folder which to become the abaqus compilation test.
2012-06-15 21:40:21 +05:30
#endif
removed unused variables and declared external functions as external
2013-02-11 15:14:17 +05:30
if (fixedSeed <= 0_pInt) &
write(6,'(a,/)') ' No fixed Seed: Random is random!'
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
end subroutine numerics_init
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
removed (in IO.f90) a dangerous initialization statement. Please read http://www.cs.rpi.edu/~szymansk/OOF90/bugs.html#4 for more details. Other files are just a little bit polished
2012-03-07 15:37:29 +05:30
end module numerics