Martin Diehl
|
f090a1b216
|
generalized dimension of calls to lapack, for performance reason special 3x3 variants will follow
|
2016-01-12 11:00:23 +00:00 |
|
Martin Diehl
|
2eafefe652
|
removed norm functions from math in favor of intrinsic and simplified (mostly by using existing functions, merge intrinsic and array constructors)
|
2016-01-10 13:34:26 +00:00 |
|
Martin Diehl
|
519cd29c6f
|
using norm2 function as L2 norm (vector, math_norm3) and Frobenius norm (tensor, math_norm33)
|
2016-01-09 16:01:30 +00:00 |
|
Tias Maiti
|
53d07215a9
|
reverting back to converged Lp as the initial guess for next iteration
|
2016-01-09 15:03:18 +00:00 |
|
Martin Diehl
|
885b9095c4
|
option not supported by slightly older gfortran versions. thx to Luv for pointing out
|
2016-01-09 12:33:33 +00:00 |
|
Martin Diehl
|
5378dbdcc9
|
cleaned
|
2016-01-09 12:12:31 +00:00 |
|
Martin Diehl
|
b5ab94a0f1
|
added new plasticity module
|
2016-01-09 12:12:05 +00:00 |
|
Tias Maiti
|
7ba031adb3
|
fixed wrong Li and dLi_dTstar_3333 initialization when Tstar is zero.
|
2016-01-09 03:41:56 +00:00 |
|
Tias Maiti
|
a30b596e7f
|
combined J2 and I2 isotropic mechanical response. Default is identical to old plastic_j2. /dilatation/ flag adds (spherical) Li calculation based on hydrostatic pressure.
|
2016-01-08 19:45:20 +00:00 |
|
Tias Maiti
|
6cf92913a3
|
use current Lp for next iteration instead of converged Lp inherited from end of former increment as starting guess
|
2016-01-08 19:01:53 +00:00 |
|
Tias Maiti
|
1d1672ebfc
|
added math_spherical33 to calculate hydrostatic part of a 2nd order tensor
|
2016-01-08 18:57:37 +00:00 |
|
Martin Diehl
|
cbb6b7a689
|
adding fpe report for gfortran, some compiler opions are now used for linking (Debug + Optimization)
|
2016-01-07 12:58:19 +00:00 |
|
Martin Diehl
|
1857e47f75
|
div by zero in output
|
2016-01-07 11:48:30 +00:00 |
|
Martin Diehl
|
86885cb2ba
|
wrong initialization order, thanks to N Grilli for reporting the bug
|
2016-01-06 16:46:37 +00:00 |
|
Martin Diehl
|
268c7f359c
|
wrong index in output
|
2015-12-23 21:15:49 +00:00 |
|
Martin Diehl
|
74086fd0fb
|
correct initialization
|
2015-12-23 16:13:08 +00:00 |
|
Martin Diehl
|
7127403bd9
|
usused variables
|
2015-12-23 11:25:25 +00:00 |
|
Martin Diehl
|
dfe09f11f5
|
cleaner handling of NaN
|
2015-12-22 10:03:15 +00:00 |
|
Martin Diehl
|
ee47c9302a
|
small readability improvement
|
2015-12-22 09:22:57 +00:00 |
|
Martin Diehl
|
e127bd8db4
|
should reflect name of parameter
|
2015-12-21 07:27:21 +00:00 |
|
Chen Zhang
|
2a2f558416
|
update physics in phenoplus module
|
2015-12-17 16:11:13 +00:00 |
|
Martin Diehl
|
5f37ade722
|
followed some suggestions from gfortran
|
2015-12-16 22:42:08 +00:00 |
|
Martin Diehl
|
5901af1212
|
removing code fragments after introducing pointer state
|
2015-12-16 06:15:21 +00:00 |
|
Martin Diehl
|
9baec1a0ff
|
get rid of overly long file names 1/2
|
2015-12-15 20:45:54 +00:00 |
|
Martin Diehl
|
c710709b1a
|
example Makefile typically does not work, deleted
|
2015-12-15 14:34:03 +00:00 |
|
Martin Diehl
|
37aa559bd4
|
fixed missing structure type in material.config, small changes in the other code
|
2015-12-15 14:30:17 +00:00 |
|
Martin Diehl
|
88999c847a
|
reverted handling of NaN in converged reason, seems to be a compiler bug
|
2015-12-14 20:04:59 +00:00 |
|
Martin Diehl
|
dc990a88d6
|
unused variables
|
2015-12-14 19:48:11 +00:00 |
|
Pratheek Shanthraj
|
3a7f4bf43a
|
added new set of accelerated spectral methods based on derivative approximations. use spectral_derivative to set the kind of derivative you like. valid options are:
‘continuous’: computed continuous derivatives (standard spectral method)
‘central_difference’: central difference approximation of derivatives
‘fwbw_difference’: forward difference for gradient and backward difference divergence
|
2015-12-14 18:12:09 +00:00 |
|
Martin Diehl
|
da913ec3c7
|
bug in dot state, thx to Haiming for pointing out
|
2015-12-11 18:36:58 +00:00 |
|
Martin Diehl
|
76fe37e496
|
commented and respected compiler warning
|
2015-12-08 18:10:06 +00:00 |
|
Martin Diehl
|
4174e6f9e2
|
removed dislokmc, use disloucla for tungsten simulations
|
2015-12-05 22:07:03 +00:00 |
|
Martin Diehl
|
6ecbd3ae56
|
makefile had problems with static flag, mentat spectral Box was still using table.croak
|
2015-12-02 20:02:19 +00:00 |
|
Martin Diehl
|
0e76ef85c1
|
checking return code in util.exececute, cleaning makefile, deleted dummy docs
|
2015-12-02 13:21:17 +00:00 |
|
Martin Diehl
|
22a9d65825
|
spectral solver accepting NaN in SNES converged reason as this might happen for homogeneous microstructures
|
2015-12-01 22:36:19 +00:00 |
|
Martin Diehl
|
3365edc6cb
|
simplified and updated abaqus
|
2015-12-01 15:13:33 +00:00 |
|
Martin Diehl
|
4eb43908c3
|
some simplifications as discussed with IBF colleagues
|
2015-11-25 20:55:17 +00:00 |
|
Su Leen Wong
|
a6cc799375
|
Removed unnecessary variables
|
2015-11-17 16:55:23 +00:00 |
|
Su Leen Wong
|
51059abaf0
|
Martensite volume fraction evolution
|
2015-11-17 16:30:06 +00:00 |
|
Su Leen Wong
|
a8b157a87c
|
Did not initialize a variable
|
2015-11-17 15:20:36 +00:00 |
|
Su Leen Wong
|
4a8477c2c3
|
Characteristic transformation shear
|
2015-11-17 14:50:59 +00:00 |
|
Su Leen Wong
|
cf84ceddd8
|
Added Ndot0 for trans
|
2015-11-16 10:15:18 +00:00 |
|
Martin Diehl
|
302178ebb4
|
some trouble since clean was already in PETSc rules
|
2015-11-14 09:47:18 +00:00 |
|
Martin Diehl
|
9d9367ae11
|
still having trouble, incorporation of PETSC_ARCH not needed
|
2015-11-13 10:16:35 +00:00 |
|
Martin Diehl
|
4a52c59854
|
corrected/removed PETSc variables path
|
2015-11-12 23:13:25 +00:00 |
|
Philip Eisenlohr
|
900b072781
|
removed unused math_qDot (deprecated anyway)
|
2015-11-10 21:56:06 +00:00 |
|
Philip Eisenlohr
|
45fb209658
|
respect PETSC_ARCH
|
2015-11-10 21:18:53 +00:00 |
|
Philip Eisenlohr
|
c5b28a22d4
|
small polishing
|
2015-11-10 21:18:13 +00:00 |
|
Su Leen Wong
|
aefd8fa390
|
Added a few more trans dependent variables
|
2015-11-10 15:23:15 +00:00 |
|
Su Leen Wong
|
2778331d7b
|
Added trans interaction matrices
|
2015-11-10 13:30:34 +00:00 |
|
Su Leen Wong
|
4e4136cd58
|
Consistent variable names
|
2015-11-10 13:01:03 +00:00 |
|
Su Leen Wong
|
9c91ba4608
|
Trans interaction matrices
|
2015-11-09 15:57:20 +00:00 |
|
Su Leen Wong
|
2b10625f0c
|
Typo
|
2015-11-09 08:51:05 +00:00 |
|
Martin Diehl
|
ff7cd17b51
|
not needed, PETSc is doing the job
|
2015-11-08 09:43:43 +00:00 |
|
Chuanlai Liu
|
4d9a2f8f6b
|
introduce pointer referencing of plastic state for better readability
|
2015-11-06 17:00:00 +00:00 |
|
Martin Diehl
|
7b0c130d6f
|
started to introduce alias name via pointer for plastic state in phenopowerlaw for more comfortable access
|
2015-10-30 15:48:30 +00:00 |
|
Su Leen Wong
|
2ae4d24fb3
|
Use rotated trans elasticity matrices
|
2015-10-27 12:32:03 +00:00 |
|
Martin Diehl
|
2ba15e462a
|
corrected values and typo
|
2015-10-26 19:27:57 +00:00 |
|
Martin Diehl
|
c91788d721
|
new intel compiler was compiling about *** in float output
|
2015-10-26 17:11:36 +00:00 |
|
Martin Diehl
|
1f356a6833
|
added parameters from indentation results
|
2015-10-22 17:45:28 +00:00 |
|
Su Leen Wong
|
04841cb727
|
List of dependent trans states
|
2015-10-21 13:24:26 +00:00 |
|
Su Leen Wong
|
20b0429517
|
Calculation of equilibrium separation of partials for twin and trans
|
2015-10-21 09:37:45 +00:00 |
|
Su Leen Wong
|
565d68edd7
|
Added some TRIP inputs
|
2015-10-20 15:54:01 +00:00 |
|
Su Leen Wong
|
b9fcd9482b
|
Renamed some input variables to denote twin
|
2015-10-20 13:45:06 +00:00 |
|
Franz Roters
|
fc8a11f027
|
corrected typo: alias for first Piola stress should read 1stpiola NOT 1piola
|
2015-10-19 06:57:23 +00:00 |
|
Chen Zhang
|
f2b29e4844
|
correct communication between
elements with different
phases
|
2015-10-14 21:04:40 +00:00 |
|
Chen Zhang
|
484a34b7f1
|
added pheno+ module
modify crystallite microstructure call
to pass orientations
|
2015-10-14 18:36:19 +00:00 |
|
Martin Diehl
|
ea0b16c726
|
pheno plus for commercial FEM
|
2015-10-13 21:06:59 +00:00 |
|
Chen Zhang
|
d6abc00218
|
add pheno+ module in
|
2015-10-13 18:52:01 +00:00 |
|
Martin Diehl
|
8fac635c15
|
made hybridIA stuff working again
|
2015-10-13 17:02:07 +00:00 |
|
Martin Diehl
|
6de8c821f8
|
improved consistency check on loadcase
|
2015-10-12 19:02:42 +00:00 |
|
Luv Sharma
|
9ec7c08bb6
|
gfortran was complaining abound mismatch
|
2015-10-07 13:07:21 +00:00 |
|
Martin Diehl
|
c1a53bc8c0
|
introduced dummy target to check compilation
|
2015-10-06 18:00:17 +00:00 |
|
Martin Diehl
|
0f0b4eaa6b
|
cleaned
|
2015-09-28 17:47:00 +00:00 |
|
Martin Diehl
|
ad16162eb8
|
simplified
|
2015-09-24 17:50:11 +00:00 |
|
Martin Diehl
|
c50a522dd8
|
some shapes not 100% correct, removed inverse laplace
|
2015-09-24 17:38:49 +00:00 |
|
Martin Diehl
|
2e70e6fded
|
make clear that this isNaN is an intrinsic extension
|
2015-09-24 08:45:44 +00:00 |
|
Aritra Chakraborty
|
b528088653
|
added all 13 slip families for bct(beta)-Sn
increased the allowable columns for lattice.f90 to allow for the large interaction matrix.
|
2015-09-22 18:42:23 +00:00 |
|
Martin Diehl
|
42ac7902f7
|
indents wrong in fortran, seeds_fromDistribution bugfix and using ASCII table facilities
|
2015-09-20 21:03:10 +00:00 |
|
Martin Diehl
|
41f09dd0f8
|
should make no difference, but test with ifort was failing. compierler buig?
|
2015-09-12 18:26:25 +00:00 |
|
Tias Maiti
|
03896fcd09
|
redefined required debug variables removed in last commit
|
2015-09-11 18:20:37 +00:00 |
|
Tias Maiti
|
840fd0e9fa
|
report absolute value of accumulated_shear_slip in dot state
|
2015-09-11 17:44:26 +00:00 |
|
Martin Diehl
|
8716752bf8
|
fixed fourier convolution and div/curl calculation for even/odd grid according to Johnston 2011 (MIT, FFTW)
|
2015-09-11 13:05:46 +00:00 |
|
Martin Diehl
|
8f32d03a9e
|
curl calculation overestimated RMS due to factor 2 instead of one for DC component and Nyquist component
|
2015-09-11 09:19:13 +00:00 |
|
Martin Diehl
|
e88cedc6ae
|
clearer naming, debug options for spectral do not work for MPI
|
2015-09-11 08:52:03 +00:00 |
|
Martin Diehl
|
751d1d7582
|
function statement is more clear
|
2015-09-10 13:01:33 +00:00 |
|
Martin Diehl
|
580bb76b4c
|
pInt missing
|
2015-09-10 10:26:09 +00:00 |
|
Martin Diehl
|
427b5f4bc1
|
wrong results for restart
|
2015-09-10 10:06:14 +00:00 |
|
Martin Diehl
|
bbe37c842e
|
name conflict
|
2015-09-09 21:52:00 +00:00 |
|
Martin Diehl
|
34980a1d44
|
more logical structure when reading in
|
2015-09-08 19:44:32 +00:00 |
|
Martin Diehl
|
7d996cfa11
|
fixed new IO_stringPos for Abaqus
|
2015-09-05 16:26:55 +00:00 |
|
Martin Diehl
|
caf0ac7e8d
|
more verbose
|
2015-09-01 16:53:48 +00:00 |
|
Martin Diehl
|
7a3a67601f
|
realloc lhs was not always working, fixed now and added test for new string pos function
|
2015-08-31 16:30:04 +00:00 |
|
Pratheek Shanthraj
|
197ae53553
|
lots of memory savings (~ stateside + 250 pReals per integration point) when using analytic jacobian
|
2015-08-28 10:55:38 +00:00 |
|
Martin Diehl
|
6848d83d13
|
behavior of IO_stringPos was unsafe in some situations, also took the chance to rename the variables used in its context to a more meaningful name.
functions in the core module are not used, including kdTree
|
2015-08-28 07:38:48 +00:00 |
|
Martin Diehl
|
1171dc4344
|
MPI has a 2GB limit for writing at once, now chunking
|
2015-08-21 17:51:05 +00:00 |
|
Martin Diehl
|
8474da26d9
|
empty line
|
2015-08-18 16:31:47 +00:00 |
|
Martin Diehl
|
bae5cfcf89
|
named if
|
2015-08-18 16:07:01 +00:00 |
|
Tias Maiti
|
53c574033a
|
size of MAXCHUNKS same as maximum allocated size for “heat_time” & “heat_rate”
|
2015-08-18 03:20:16 +00:00 |
|
Martin Diehl
|
f64a5e7552
|
output safe for 64bit integer
|
2015-08-14 05:03:54 +00:00 |