f2dacb5f40
polished debugging output
2012-12-03 16:29:54 +00:00
f92d34dd22
removed some unused variables
2012-11-29 18:44:00 +00:00
e10000a338
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.
2012-11-27 18:36:55 +00:00
e584c0e6e8
deleted space in front of preprocessor line
2012-11-22 20:04:33 +00:00
38ec462f4a
order of integration now same for all state integrators: first calculate the dotState and then the deltaState
2012-11-22 13:04:19 +00:00
7d196fbb25
polishing
2012-11-22 09:58:36 +00:00
d189a6a7a0
polishing
2012-11-21 16:57:57 +00:00
760af1fbbf
crystallite_requested now masked by execIP, now elemhomogeneous works again
2012-11-19 15:13:48 +00:00
9fb94cd707
breaking crystallite loop in case of terminally ill (this functionality was probably lost since rev1886)
2012-11-19 08:39:48 +00:00
d9a98417ca
switched element library to geomType based.
...
saves to copy same geometry description for different elements that are essentially similar regarding the IP number but differ in total node count.
introduced quadratic tetrahedron (Marc element 127 -- element 157 might also work, but did not perform well in fully elastic calc so far)
2012-11-15 22:45:20 +00:00
11264f3fe4
forgot to check for convergence flag in FPI integrator
2012-11-08 20:15:19 +00:00
60438cdbcc
criterion for crystallite cutback loop changed from crystallite_subStep to crystallite_todo; this avoids problems when the last substep of the crystallite increment is below subStepMinCryst (which is allowed during last wind forward)
...
fixed bug in the adaptive Euler integrator and the RK4 integrator, which could not reach the stateJump function, because of erroneous check for converged flag; also corrected the state counter for the RK4 integrator
2012-11-08 20:00:29 +00:00
c775edaa6d
better avoid initialization of residuum_old with huge, since it posed problems in Abaqus
2012-11-08 13:26:22 +00:00
dad9922f54
fixed bug in crystallite_FPI: stateDamper always has to be defined for each grain
...
added some OMP FLUSH statements were necessary
replaced openmp do by forall construct where possible; this is much safer and perhaps even as fast for small loops
2012-11-07 15:43:29 +00:00
bb033c5fe7
Simplified algorithm of crystallite_integrateStress while preserving (almost) same functionality.
...
Removed "leapfrogging" (increase of step for next guess, when last guess was ok); Replaced Armijo rule testing for step size by simple check if the residuum got better, since the former virtually did not have any effect; consistently using the 2-norm of the residuum rather than infinity-norm for the convergence check throughout the function
2012-11-06 12:35:45 +00:00
639d6e0655
inserted missing flush(crystallite_todo) in parallel section; might have caused trouble in parallel calculations before
2012-11-05 14:45:52 +00:00
086fe138b1
substituted matrix inversion to solve equation by direct solution routine from LAPACK
2012-10-31 09:56:26 +00:00
2be331b74d
In crystallite for state integrators 2 and 3: now that "stateJump" is called before integrating dotState, we have to do state = state + dotState *dt instead of state = subState0 + dotState * dt; otherwise the deltaState is lost
2012-10-22 13:34:15 +00:00
0980920b66
In crystallite: for state integrators 2 and 3 the deltaState always has to be done at the beginning of the integration, not at the end; otherwise the stiffness calculation can be completely wrong, when the state is changed after the stress integration for the central solution.
...
Also changed some debugging levels in crystallite and mainly in constitutive_nonlocal
2012-10-22 07:59:35 +00:00
36676029e1
for state integrator 2 (explicit euler) and 3 (adaptive 2nd order euler): don't do any state evolution during stiffness perturbation, but only stress integration, since for those cases the state is not restored after every perturbation step
2012-10-19 11:20:31 +00:00
dd5f453994
mainly cosmetics. added some comments. changed debug levels for some outputs. corrected (probably non-essential) check for NaN in FPI stressIntegrator.
2012-10-18 09:53:26 +00:00
5f0440fde3
crystallite_integrateStress now raises an error if Fp becomes NaN
2012-10-17 16:19:42 +00:00
0bcb8f59db
added doxygen comments, some polishing, added "protected" statements where applicable
2012-10-02 12:53:25 +00:00
22812c9a91
some minor improvement on precision detection: checking only once (in prec and no longer in math and crystallite), added one more 4/8 switch for LAPACK, as there is no single precision FFTW, stopping compilation for spectral method if FLOAT=4
...
new function in IO to print integers without leading zeros, implemented it at some places in the new spectral solver (reporting still needs some serious polishing)
updated preprocessing for documentation to handle precision correctly
2012-08-30 20:26:28 +00:00
0959ff3299
substituted hand written matrix inversion by LAPACK version with precision selection.
...
also introduced check for inversion into DAMASK_spectral_Utilities.f90 for the stress BC calculation. This part is further improved by using 5% of the reference stiffness to avoid trouble in the fully plastic regime (where the stiffness is underestimated)
Test for Marc 2010 is updated because the new inversion give slightly different results near 0 (order of e-13)
2012-08-28 16:59:45 +00:00
ad7381cfdd
maximum relative step length in stress integration cannot decrease below 1
...
intermediate tensor "A" is calculated only once before the Lp loop in "integrateStress"
2012-08-21 08:53:36 +00:00
c1c5941733
stressAndItsTangent:
...
when winding forward to last substep in increment, also do the calculation if this substep is below "subStepMinCryst";
output of "wind forward" and "cutback" info only for crystallites that are still "todo"
integrateStress:
maximum component in Lp determines absolute convergence criterium:
aTol = max(maxLpComp * rTol_fromConfig , aTol_fromConfig)
this ensures that small values in Lp do not spoil the convergence, similar to the effect of the relevantStrain parameter before (not needed anymore in the stress integration)
2012-08-14 12:07:59 +00:00
4d09ef0648
changed variable name "debug_what" to "debug_level"
2012-07-05 09:54:50 +00:00
380f40561b
first calculation of Tstar still used hard coded Hooke, now usese constitutive_TandItsTangent(...)
2012-06-18 07:26:34 +00:00
fc7b4d6471
new function stateJump, which takes care of immediate changes in the state (deltaState); this introduces discontinuities in the state evolution; therefore this is always and only once done after each integration step, so no evaluation of deltaState for intermediate steps of e.g. the Runge-Kutta integrator; otherwise integration becomes a pain without significant gain in accuracy
...
deltaState now successfully tested for nonlocal model; was not correct for integrators 1,4,5 before
2012-06-06 15:11:30 +00:00
8c849d62ee
small corrections for how to treat the new deltaState in FPI and euler integration; all integrators successfully tested with j2
2012-05-18 13:00:15 +00:00
abbae76c51
instantaneous jumps in the state by constitutive_deltaState are now incorporated for all state integrators. still they (should) not influence the result, since all constitutive laws simply return zero for the deltaState
2012-05-17 15:25:21 +00:00
a54439e3b5
dotState does not have to be reset to zero. this is a remnant from older versions when the dotState for the nonlocal model used to be updated by the neighboring integration point
2012-05-16 15:35:14 +00:00
7ee368f324
check whether there is any constitutive output
2012-05-11 07:38:20 +00:00
91a70b0fb3
before turning terminally ill report at which g,i,e crystallite did not converge
2012-03-30 13:16:45 +00:00
131c9ac93e
dPdF calculations made consistent with constitutive_TandItsTangent
2012-03-21 15:00:36 +00:00
8a2f2c5a95
stress iteration loop now uses generalized elasticity by calling TandItsTangent
...
the anlalytical tangent calculation should now be adopted to also use TandItsTangent
2012-03-21 10:57:27 +00:00
c2f5cebacb
simplified analytic jacobian calculation. removed Lpfrac, time_sensitive. introduced rate_sensitivity flag when calling crystallite_stressAndItsTangent that is currently set to .false. and is to be set according to which dPdF the FE solver is asking for
2012-03-14 13:56:50 +00:00
c9ed989dad
replaced keyword constitution with plasticity in preparation of upcomming generalized elasticity
2012-03-12 14:09:37 +00:00
bd9667bd4b
added new, flexible debugging scheme.
...
now all modules have their own debug specification.
compiles and runs, I hope nothing is broken
did a lot of polishing
2012-03-08 20:25:28 +00:00
17e9698659
added time_sensitive flag to ignore or add time sensitive component to analytic jacobian (should be close to perturbed jacobian when ignored)
...
now uses sub_dt for analytic jacobian calculations
bug fixes
2012-03-01 19:42:43 +00:00
211f66a728
added elastic stiffness matrix as crystallite output
2012-03-01 13:06:09 +00:00
040e244993
Added stress-order terms to analytic stiffness
2012-02-27 17:40:28 +00:00
99d6dcecb5
fixed bug in calculation of analytic jacobian (should work much better now). parallelized analytic jacobian calculation loop
2012-02-24 14:07:46 +00:00
28d654d61e
forgot to remove some print statements in previous commit
2012-02-22 20:22:54 +00:00
b9a82156c9
crystallite.f90: can now calculate analytic jacobian by setting analyticJaco = 1 in numerics.config
...
math.f90: added math_mul3333xx3333…
numerics.f90: to read in analyticJaco and Lp_frac
2012-02-22 20:11:09 +00:00
d68a3d1869
added LF=UNIX to a bunch of files
2012-02-22 13:30:00 +00:00
d8ffc29236
again: adding pInt, removing redundant use statments, chang in dble to real(,pReal)
2012-02-21 16:31:37 +00:00
6c0f9d163b
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
2012-02-15 18:58:38 +00:00
c786336af3
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 17:41:27 +00:00
94100e8d8e
missing line break
2012-02-13 15:21:03 +00:00
156ec4582a
polishing, adding _pInt etc. where applicable
...
post_results now handels zero increment different (like FEM, it is always there even if it is not counted)
2012-02-13 14:18:07 +00:00
fa32a00d56
changed symmetryID initialization to integer (instead of float) value
2012-02-10 17:46:43 +00:00
e7d407c2f6
crystallite_Fp has to be initialized in crystallite_init, cause it is used by constitutive_microstructure
2012-02-03 14:52:09 +00:00
c5128e94eb
some f2008 standard related correction i didn't check in last time
2012-02-01 20:20:05 +00:00
800e291240
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
2012-01-31 19:18:55 +00:00
80583fefb5
added compilation_info.f90, to store compilation date,time and compiler info in each file
...
added #include statement at each init() routine
2012-01-31 14:54:49 +00:00
1330576a01
added new math function "math_deviatoric33" which returns the deviatoric part of a 3x3 tensor
...
renamed some math functions, so that we have a universal naming scheme: for matrix multiplications use an "x" (e.g. math_mul33x3); don't use the "x" to describe the shape of the tensor that the function is applied to (e.g. math_invert33 instead of math_invert3x3)
2012-01-26 13:50:00 +00:00
6fec57bbd7
replaced variable fraction by "timefraction"
2012-01-26 12:48:59 +00:00
4980f57fd5
added total Green-Lagrange strain to crystallite outputs
2012-01-20 10:25:35 +00:00
8739b9522e
dislocation stress does not add to stress field anymore, but is only available as constitutive output for purposes of postprocessing. instead, we now use a backstress term (which is added to the resolved shear stress) that depends on the gradient of excess density on the same slip system.
2012-01-17 10:26:57 +00:00
a7d9d711f6
edited lines exceeding 132 chars (before any comment started)
2012-01-11 16:56:35 +00:00
5ebeb96e85
made DAMASK to work with gfortran:
...
-removed to long lines
-restructured f2py modules and merged make_DAMASK2Python into setup processing
-setup_code.py now sets library path in makefile and asks for compile switches for spectral code
-substituted \ in format strings with $
restructured DAMASK_spectral:
-more logical output and structure of code
-better input for spectral debug parameters
2011-12-06 16:58:17 +00:00
9d1bc584d0
corrected stress integration scheme: now use norm(Lpresiduum) as a target function for Armijo's rule instead of whole tensor Lp; also corrected the guess for the improvement in Lp
2011-11-04 12:57:12 +00:00
ca3d21a3b6
Depending on which state integrator one uses for the stiffness calculation, the initial state has to be chosen accordingly: e.g. for FPI choose last converged state, but for explicit RK choose converged state from start of increment (in case of explicit euler no state integration at all, but only stress integration). For this purpose we also need to remember "Fe" which now follows the cutbacking procedure as it is used for "Fp".
2011-11-04 12:44:50 +00:00
c35ea33f8e
did a lot of polishing:
...
- removed unnecessary "return" before end of subroutine or function:
- changed undetermined array length (:) to (1:3)
To prevent problems with some code analysing tools:
- "3D oneliner loops" (with ";) only for "do" and "enddo" at the same time
- removed line continuation in OMP statements
made the makefile more flexible, removed heap-arrays switch
2011-09-13 15:54:06 +00:00
b300cc7faa
ip volume is now based on the determinant of F. "mesh_ipVolume" represents the initial volume and is multiplied with det(F) wherever the current volume is needed. Since this works for all solver types, the "volume" output in crystallite is now also correct for spectral method and abaqus.
2011-08-19 11:18:56 +00:00
1ffb59a96a
Calculation of current ip volumes now working. Crystallite output also reflects current grain volume, not reference volume. However, this is only available for Marc. Abaqus and spectral method still return the reference ip volume. The ip coordinates though are correctly updated for all solver types.
2011-08-10 16:37:17 +00:00
ef7405fe21
Stress integration now uses Armijo rule to find an appropriate correction of Lp: decreases step in case that residuum does not improve significantly, accelerates as usual in case of good convergence. This turned out to improve convergence behavior.
2011-08-02 11:29:08 +00:00
75c67f53f7
introduced alternative location for material configuration.
...
If present, the >>>JOB.materialConfig<<< file takes precedence over the standard >>>material.config<<<
2011-08-02 10:14:16 +00:00
564eb5009f
shape mismatch in crystallite, invalid names for variables(pos, size) in homogenization corrected. polishing of DAMASK_spectral_interface and makefile
2011-08-01 18:10:55 +00:00
a26d9c844b
max size of crystallite result is now based on actual use within the mesh, not just picking the largest chunk from the material.config file...
2011-06-08 09:16:48 +00:00
8041587a72
changed internal debug verbosity in accord with debug.config listing.
2011-06-06 15:27:35 +00:00
e5a2d829b0
new substructure with 'include' and 'config' directories
...
renaming of mpie_... to DAMASK for main usersubroutines
extension of element outputs from 5 to 8 digits (FFT!!!)
2011-05-11 16:38:45 +00:00
affd383ef8
simplified disorientation calculation. performed FEM check that "grainrotation" reflects the *active* rotation from the starting orientation to the current one with respect to the lab coord system (as expected).
2011-04-14 19:09:44 +00:00
5814262f55
moved some write statements within stateIntegrationEuler
2011-04-14 09:35:56 +00:00
349f022100
reinstantiated calculation of grainrotation (still needs check for correctness, i.e. active or passive rotation output)
2011-04-13 16:08:52 +00:00
7d84a0911e
removed unused variables
2011-04-13 14:16:22 +00:00
c1b8391110
changed enconding of all source files to UTF-8 without BOM (signature) Codepage 65001
2011-04-07 07:20:28 +00:00
fcdb805225
added copyright text to all f90 (free) format files
2011-04-04 14:09:54 +00:00
790dbed1e4
* removed last remnants of old debugger
...
* replaced "dble" intrinsic function by "real" with pReal kind in constitutive_nonlocal.f90
* removed useless line breaks in output of state in CPFEM.f90
2011-04-04 08:34:52 +00:00
314ca3fe7f
Internal stress calculation in nonlocal model: instead of integration of excess gradients (->Bayley) we now sum up contributions from adjacent superdislocations within a certain radius R. When periodicity is used, also periodic images are considered in stress evaluation.
2011-03-29 07:34:33 +00:00
3d51dd36fa
* Introduced preprocessor directives in order to suppress compilation of most write statements when using openmp. This tremendously improves efficiency of parallelization.
...
* Also added some more openmp directives to increase percentage of parallelized code.
* "implicit none" was missing in two subroutines of homogenization and constitutive.
2011-03-29 07:27:19 +00:00
7be2edb10e
added "texture
...
" as possible crystallite output => reports ID for later discrimination...
2011-03-22 13:40:27 +00:00
462eda6736
split parallel region in integrateStateFPI into two
2011-03-21 14:30:10 +00:00
5ee73dee72
age dotState before resetting it to zero, otherwise the statedamping does not work.
2011-03-21 13:06:11 +00:00
fc6f2ae68f
No need to call constitutive_microstructure at the beginning of each crystallite step, since it's already been done at the end of the previous step. Just do it once after initialization to start with correct values for the dependent state variables.
2011-03-21 10:35:42 +00:00
11138e3ee2
debugging output is now controlled by the "verbosity" parameter in the debug.config ranging from 0 (=almost no output) to 8 (=very detailed output)
...
0 : only version infos and all from "hypela2"/"umat"
1 : basic outputs from "CPFEM.f90", basic output from initialization routines, debug_info
2 : extensive outputs from "CPFEM.f90", extensive output from initialization routines
3 : basic outputs from "homogenization.f90"
4 : extensive outputs from "homogenization.f90"
5 : basic outputs from "crystallite.f90"
6 : extensive outputs from "crystallite.f90"
7 : basic outputs from the constitutive files
8 : extensive outputs from the constitutive files
If verbosity is equal to zero, all counters in debug are not set during calculation (e.g. debug_StressLoopDistribution or debug_cumDotStateTicks). This might speed up parallel calculation, because all these need critical statements which extremely slow down parallel computation.
2011-03-21 10:31:17 +00:00
235266b169
openmp parallelization working again (at least for j2 and nonlocal constitutive model).
...
In order to keep it like that, please follow these simple rules:
DON'T use implicit array subscripts:
example: real, dimension(3,3) :: A,B
A(:,2) = B(:,1) <--- DON'T USE
A(1:3,2) = B(1:3,1) <--- BETTER USE
In many cases the use of explicit array subscripts is inevitable for parallelization. Additionally, it is an easy means to prevent memory leaks.
Enclose all write statements with the following:
!$OMP CRITICAL (write2out)
<your write statement>
!$OMP END CRITICAL (write2out)
Whenever you change something in the code and are not sure if it affects parallelization and leads to nonconforming behavior, please ask me and/or Franz to check this.
2011-03-17 10:46:17 +00:00
97a0146672
deactivated one debug statement at the end of crystallite_updateState(g,i,e) as it leads to division by zero if constitutive_state(g,i,e)%p(1:mySize) contains zero values
...
statement should be changed to check for zero values
2011-03-11 10:05:30 +00:00
ad4706673b
* removed calculations for dipole formation/dissociation by stress change, since it is not used anyways; also removed associated constitutive outputs from material.config
...
* removed input variables in constitutive_collectDotState and constitutive_postResults that are not needed anymore (because of recent changes in constitutive_nonlocal)
2011-02-25 09:53:20 +00:00
2f503f5cdb
added "Lp" as crystallite output
2011-02-25 08:15:26 +00:00
96d3682d5e
* new global integer variable "numerics_integrationMode" (1 indicating integration of central solution, 2 indicating integration of perturbed state)
...
* combined "integrator" and "integratorStiffness" in new global variable "numerics_integrator"
2011-02-23 08:29:51 +00:00
466ced911e
* check convergence for nonlocals also in stiffness calculation mode
2011-02-09 08:39:07 +00:00
c00a42a9a4
fixed last remaining tensor outputs to their transposed versions
2011-01-27 14:29:06 +00:00
6ac06af517
sorry, forgot to alter the most important part: post_results..! eventually, the t16 has defgrads and friends in correct format: du_i/dx_j is listed as linear array (1,1),(1,2),(1,3),(2,1),...
2011-01-24 16:23:37 +00:00
4ee40df5ba
transposed writing into t16 file (and screen/debug) did not work with just stating array(1:3,:)...
...
fixed by using math_transpose3x3 (and similar) instead.
2011-01-24 15:02:19 +00:00
794aeb253b
crystallite and CPFEM now print/output tensors in ij notation (used to be, particularly in the t16, ji, i.e. transposed...)
...
math reports random seed used for rnd() generation
2011-01-20 19:25:45 +00:00
b9b226548c
* enclose wall time measurement for LpAndItsTangent and count of leapfrogbreaks by CRITICAL construct
...
* better use SINGLE (having an implicit barrier at the end) instead of MASTER construct
* deleted all explicit BARRIERs after do loops since parallel loop construct implies barrier at the end
2010-11-19 17:29:29 +00:00
4f267f4375
parallelization now works properly (not yet tested extensively though):
...
* had to add some BARRIER constructs
* only the master thread is allowed to increase the state counter
yet parallelization seems not to give a significant decrease in calculation time with nonlocal model (because of too many CRITICAL statements?)
2010-11-11 16:16:05 +00:00
Christoph Kords