86283359f1
PetscFinalize should be the last call before exiting
2015-03-26 13:43:18 +00:00
d44fce4a76
Spectral solver now fully parallel (parallel IO, domain decomposition, FFTs and restart). Working but not extensively tested so please report bugs to me
2015-03-25 16:06:19 +00:00
c5a8543f17
updated petsc
2015-03-18 17:18:43 +00:00
d6080d332d
updated to be compatible with latest version of petsc.
...
Please change your PETSC_DIR to /opt/petsc-dev and PETSC_ARCH to ifort (gfort or ifort-debug for gfortran and debugging version respectively on maws01 also available)
2014-09-11 13:28:15 +00:00
4bb5439994
removed copyright from *.f90 and *.f files, will be added using the deployMe script for release/nightly build. Now based on presence of svn:keywords property set (usually "Id"). Similar thing will be done for .py, .sh, and .config files but more involved because they might be executable
2014-03-29 08:20:36 +00:00
62497b648a
removed forgotten debug statement
2014-03-26 09:09:28 +00:00
2659ee51d4
restart is working for all solvers, but slight deviations still occur. Reason is most probably, that for CPFEM init, a 0s timestep is required which seems to alter the state a tiny bit, leading to small deviations. For AL and Polarization, this is even more severe since the regular call to CPFEM has F-residual as the current deformation gradient, however the init step uses only F.
2014-03-25 15:44:16 +00:00
c7994db5ef
writing out f_aim and f_aim_last inc for restart, was calculated as average before
2014-03-24 09:59:30 +00:00
b9e55b4495
moved setting of CPFEM forwarding flag to the forwarding subroutine
2014-01-27 11:05:36 +00:00
de3fc70a02
moved restart writing to the forwarding routine
2013-12-20 10:49:14 +00:00
61981617d7
separated forwarding and solution subroutines for better control at the load step looping level
2013-12-18 09:35:05 +00:00
c4a592867f
updated to PETSc 3.4.3, will not compile on all workstations except for maws0X
2013-12-17 15:37:14 +00:00
a0f304021f
fixed restarting for Polarisation and AL, updated corresponding test and polished output in python (some statements are screwed up when making them Python3 compatible)
2013-09-20 14:22:37 +00:00
3eec0ecdcd
introduced #EOF# as module wide parameter in IO.f90
...
renamed read/write binary file to read/write real/int
removed suffix job from readFile functions as the name of the model is needed as an argument
2013-09-18 14:07:55 +00:00
e62b760a6e
added F_avg = F_aim in boundary condition convergence check
2013-08-09 16:25:13 +00:00
822fa3d97c
forgotten line continuation
2013-08-09 04:16:16 +00:00
3bb5a5e7b5
line too long in DAMASK_spectral_solverPolarisation.f90, reordered curl calculation for clearer look, fixed test for Marc 2013
2013-08-08 18:10:58 +00:00
7f3c24e1f1
fixed some buggy last minute changes in the last commit
2013-08-08 09:13:29 +00:00
ac92b90e0b
fixed al and pol solver, now checking for div(p) = curl(f) = 0
2013-08-07 17:20:05 +00:00
5b80ef3a4c
fixed wrong error calculation in Polarisation and AL solvers, and strictened tolerances for tests
2013-08-02 13:55:44 +00:00
74791a6686
corrected invalid error for mismatch_p
2013-08-02 11:48:24 +00:00
2dc985f558
set tolerances to useful values and renamed them.
...
allowed "keyword = value" in numerics.config (simply removing "=" from line)
2013-07-31 15:03:38 +00:00
b5a11f9d31
fixed bug in AL, weakened correction slightly
2013-07-30 21:04:41 +00:00
81531097f1
changes on convergence tolerances of AL and Polarisation, switched back to immediate correction of stress bc but only when last two average stresses are close to each other (cosine decay)
2013-07-30 15:32:55 +00:00
03d8f14a98
introduced dynamics for high strain rate simulations. so far only works for materials with homogeneous density. the density is read in as an optional argument in the load file using the keyword 'den' or 'density'. if density > 0 then inertial terms are added to the stress equilibrium. an implicit euler time discretisation is used to calculate the inertial terms.
...
dynamic problems are length scale dependent so make sure the geometry size is meaningful (1 m^3 or N m^3 for example is too big for a RVE!).
not specifying a density (density = 0) will perform a quasi static simulation as before.
2013-07-26 16:25:37 +00:00
216063fe36
added Polarisation scheme as solver
2013-07-24 13:06:16 +00:00
Pratheek Shanthraj