Commit Graph

5033 Commits

Author SHA1 Message Date
Tor Harald Sandve
2a66aa23a9 Merge pull request #1198 from dr-robertk/PR/fix-densewell-for-2p
[bugfix] Fix invalid read in WellDenseStandard when used with 2P Model.
2017-06-07 13:57:44 +02:00
Atgeirr Flø Rasmussen
e5732ea0a7 Merge pull request #1204 from atgeirr/all-messages-to-terminal-option
Logging flexibility improvements.
2017-06-06 07:13:52 +02:00
Atgeirr Flø Rasmussen
b62d84f7b8 Logging flexibility improvements.
Added "all_messages_to_terminal" parameter (defaulting to false),
it true, all messages (including note and debug) will be sent to
the terminal.

In addition, parser messages are tagged, to make the terminal output
less likely to be overwhelmed.
2017-06-02 15:26:23 +02:00
Atgeirr Flø Rasmussen
efc553b175 Merge pull request #1203 from atgeirr/debug-time-report
Add DebugTimeReport utility.
2017-06-02 14:14:28 +02:00
Atgeirr Flø Rasmussen
ee63732b62 Merge pull request #1202 from dr-robertk/PR/fix-no-output-run
[bugfix] Only create asyncOutput when output is enabled globally.
2017-06-02 14:13:36 +02:00
Atgeirr Flø Rasmussen
5f66a8ff4f Add DebugTimeReport utility. 2017-06-02 13:48:24 +02:00
Robert Kloefkorn
46f66e34b9 [bugfix] Only create asyncOutput when output is enabled globally. 2017-06-02 13:09:48 +02:00
Atgeirr Flø Rasmussen
9a32e2baca Merge pull request #1201 from atgeirr/fix-warning
Fix minor warning (returning a const double).
2017-06-02 12:47:55 +02:00
Atgeirr Flø Rasmussen
c8e1f31f1d Fix minor warning (returning a const double). 2017-06-02 11:17:37 +02:00
Atgeirr Flø Rasmussen
37cbdbced2 Merge pull request #1199 from atgeirr/multiphase-upwind-refactor
Multiphase upwind refactor
2017-06-02 11:13:14 +02:00
Atgeirr Flø Rasmussen
4c70c831d4 Merge pull request #1176 from totto82/impl_solvent
Implement solvent model in flow_ebos
2017-06-02 10:08:49 +02:00
Atgeirr Flø Rasmussen
f5795322fa Add missing include directive for std::sort(). 2017-06-02 08:52:16 +02:00
Atgeirr Flø Rasmussen
7e8ed40714 Use free function for multi-phase upwinding. 2017-06-01 14:15:31 +02:00
Atgeirr Flø Rasmussen
33025d752e Clarified notation in test. 2017-06-01 14:07:57 +02:00
Atgeirr Flø Rasmussen
f707abae36 Add unit test for connectionMultiPhaseUpwind(). 2017-06-01 13:55:11 +02:00
Atgeirr Flø Rasmussen
80c38d5a1a Add connectionMultiPhaseUpwind() free function. 2017-06-01 11:01:25 +02:00
Robert Kloefkorn
3db16ede73 [bugfix] Fix invalid read in WellDenseStandard when used with 2P Model. 2017-05-31 16:17:59 +02:00
Tor Harald Sandve
f671af6cd6 Clean-up of the solvent implementation 2017-05-30 14:33:17 +02:00
Tor Harald Sandve
8b75e2eedc Minor fixes solvent model
1) Fix GRAT controlled wells
2) Fix seg_fault in computeWellPotentials for solvent simulations
3) Fix mem_issus for RESV combinded with solvent
2017-05-30 14:22:19 +02:00
Tor Harald Sandve
683ba7e929 Fix fallout from rebase 2017-05-30 14:22:19 +02:00
Tor Harald Sandve
441a8895ac Improvments for solvent model
- add dss to appleyard chopping
- support for bhp injectors with solvent
- copy perfSolventRates between the time steps.
- fix bug in well access indicies when numComponents ~= numPhases
2017-05-30 14:22:19 +02:00
Tor Harald Sandve
50c1a1404a Remove BlackoilSolventState
The solvent saturation is added to BlackoilState and the
BlackoilSolventState is thus redundant.
2017-05-30 14:22:19 +02:00
Tor Harald Sandve
b987e4b324 Implement solvent model in flow_ebos
1) Extends the well model to account for solvent surface volumes
2) Add solvent to updateState
3) Add solvent to well and field output

The solvent parts is encapsled in if (has_solvent_) and should not effect
the standard runs.
2017-05-30 14:22:19 +02:00
Markus Blatt
138eb2c91b Merge pull request #1193 from dr-robertk/PR/fix-mpi-deadlock-in-output
Bugfix: fix MPI deadlock in output when asyncOutput is enabled.
2017-05-30 11:33:54 +02:00
Atgeirr Flø Rasmussen
1e5652ac3b Merge pull request #1191 from totto82/fix_output_new_timestep
FIX output the correct next timestep after convergece failure
2017-05-30 10:39:21 +02:00
Robert Kloefkorn
e46810d2dc [bugfix] 1) create asyncOutput object on all ranks to avoid deadlock in
MPI_Bcast call in writeTimeStepWithCellProperties.
2) ThreadHandle waits on destruction until all objects have been dealt with.
2017-05-29 18:08:18 +02:00
Tor Harald Sandve
434f96db0a FIX output the correct next timestep after convergece failure 2017-05-29 09:48:34 +02:00
Joakim Hove
a61d8ab14a Merge pull request #1187 from joakim-hove/cmake-no-dev
Added -Wno-dev to Eigen cmake args.
2017-05-23 15:14:10 +02:00
Tor Harald Sandve
095c580933 Merge pull request #1174 from totto82/FIX_flowsolvent
BUGFIX Don't try to compute effective properties for FIP if no solvent
2017-05-22 15:02:09 +02:00
dr-robertk
1304e1f2f1 Merge pull request #1185 from blattms/allow_amg_with_flow_ebos_umfpack
Allow usage of AMG if flow_ebos is used and UMFPack is available.
2017-05-22 13:27:17 +02:00
Markus Blatt
79799a5c5c Use pressure for guiding coarsening in flow_ebos, too.
For flow_legacy the first component a block is used, which is the
oil pressure. As flow_ebos uses different indices this commit
explicitly uses BlackoilIndices::pressureSwitchIdx to tell the AMG
at which index the pressure is stored.
2017-05-22 12:09:58 +02:00
Markus Blatt
5a6ace9063 Resort to setting one define for AMG support in the main simulator 2017-05-22 11:34:20 +02:00
Markus Blatt
98e2eeb4dc Allow usage of AMG if flow_ebos is used and UMFPack is available.
This was previously deactivated because of the single precision support
of flow_legacy and missing single-precision support for UMFPack.

As flow_ebos does not use single precision this commit allows using amg
at least for flow_ebos.
2017-05-22 11:32:02 +02:00
Joakim Hove
deab255468 Added -Wno-dev to Eigen cmake args. 2017-05-21 16:30:27 +02:00
Andreas Lauser
c3555a21d2 Merge pull request #1019 from andlaus/flow_ebos-remove-geoprops
flow_ebos: remove the legacy geologic properties object
2017-05-19 18:49:21 +02:00
Atgeirr Flø Rasmussen
27c9585529 Merge pull request #1178 from totto82/dont_output_SSOL
Don't output SSOL if no solvent
2017-05-19 10:02:51 +02:00
Atgeirr Flø Rasmussen
644d1c7bb5 Merge pull request #1175 from nairr/prtheader
PRT header
2017-05-18 14:32:06 +02:00
Joakim Hove
71c22d1268 Merge pull request #1184 from joakim-hove/travis-cmake
Using ppa for cmake in travis.
2017-05-18 12:05:40 +02:00
Joakim Hove
7b7d451619 Using ppa for cmake in travis. 2017-05-18 10:46:21 +02:00
Rohith Nair
59ec8a1bf5 edit based on comments 2017-05-17 17:46:10 +02:00
Tor Harald Sandve
a5fed26486 BUGFIX Don't try to compute effective properties for FIP if no solvent 2017-05-15 10:13:22 +02:00
Andreas Lauser
48b7d6ea56 improve writing of the INIT file
now, the dune APIs are used whereever possible and the data is
computed for the global grid, i.e. for parallel runs it does not need
to be gathered across the processes anymore. Also, the INIT file is
now only written once instead of twice.

I've verified that the sequential and the parallel INIT files stay
identical for the Norne case and that the INIT file does not change
w.r.t. before this patch.
2017-05-12 15:44:55 +02:00
Andreas Lauser
7cbea4be41 fix a bug in the transmissibility output code
the fact that faces may point into the opposite Cartesian direction
was not considered and these output fields are specified
w.r.t. logically Cartesian cell indices. (not compressed ones!)
2017-05-12 15:43:51 +02:00
Andreas Lauser
adb2715c8d flow_ebos: also write the non-input NNCs to the init file
the corresponding code was shamelessly lifted from the DerivedGeology
class. it has been substantially modified to adapt it to the flow_ebos
specifics, though.
2017-05-12 15:43:51 +02:00
Andreas Lauser
e2e0e3290d flow_ebos: do no longer use the legacy object for geologic properties
it was already almost unused (except for output). Besides making the
overall flow_ebos code leaner because it reduces redundancies, this
patch also implies a small reduduction of memory consumption and a
minor performance improvement. the latter is due to the fact that the
transmissibilities now do not need to be calculated more often than
necessary anymore.
2017-05-12 15:43:51 +02:00
Tor Harald Sandve
4ecc01d30a Don't output SSOL if no solvent 2017-05-12 14:45:00 +02:00
Joakim Hove
9dda677a28 Merge pull request #1177 from totto82/dummy_pr
Dummy commit
2017-05-12 13:39:02 +02:00
Atgeirr Flø Rasmussen
e07ddfccee Merge pull request #1170 from akva2/2p
Add 2 phase regression test for flow_ebos and flow_legacy
2017-05-12 13:02:46 +02:00
Tor Harald Sandve
2af3019299 Dummy commit 2017-05-10 15:36:20 +02:00
Joakim Hove
8b36daa7aa Merge pull request #1173 from joakim-hove/use-libecl
Use libecl in jenkins build.
2017-05-10 13:37:05 +02:00