OilfieldToolsNet/opm-simulators
4
0
Fork 0
mirror of https://github.com/OPM/opm-simulators.git synced 2024-12-28 02:00:59 -06:00
Code Issues Packages Projects Releases Wiki Activity
7,932 Commits 24 Branches 102 Tags 84 MiB
a42ece61ed
Commit Graph

7932 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Andreas Lauser
a42ece61ed BlackoilModelEbos: set the default output directory to the empty string 
an empty string is interpreted as "write the output files into the
same directory as the input". this is the current 'flow' default
behavior, plain ebos uses "." as the default location for output,
i.e., the current directory from which the simulator is run from.
 2018-03-12 15:17:39 +01:00
Andreas Lauser
71d353326a move the old "ThreadHandle" mechanism back to opm-simulators 
this class is only used by the legacy simulators, `flow` uses the
`EclWriter` class provided by eWoms. In turn, this class uses the
new-and-shiny "tasklet" mechanism.
 2018-03-12 15:17:39 +01:00
Atgeirr Flø Rasmussen
79bc265de8
Merge pull request #1431 from GitPaean/silencing_warning_from_opm_common_pr_348 
adding POLYMW to the injectorType determination
 2018-03-10 13:19:37 +01:00
Kai Bao
b0edfd2f04 adding POLYMW to the injectorType determination 
just to silence a warning resulting from OPM/opm-common#348.
 2018-03-09 14:40:49 +01:00
Atgeirr Flø Rasmussen
14645fb1b8
Merge pull request #1429 from akva2/remove_repo_update_data 
changed: remove repos from regression data update script
 2018-03-09 14:15:24 +01:00
Atgeirr Flø Rasmussen
1380564256
Merge pull request #1425 from totto82/fix_enable_storage_term_false 
Bugfix. enableStorageCache=false
 2018-03-09 14:13:10 +01:00
Atgeirr Flø Rasmussen
cef255b212
Merge pull request #1430 from atgeirr/fix-boost-dyn-link 
Leave BOOST_TEST_DYN_LINK setting to the build system.
 2018-03-09 14:12:09 +01:00
Atgeirr Flø Rasmussen
42c41efa4e Leave BOOST_TEST_DYN_LINK setting to the build system. 2018-03-09 12:56:51 +01:00
Atgeirr Flø Rasmussen
8070e9d229
Merge pull request #1397 from blattms/system-amg-rebased 
Add support for CPR (and variants) for flow_ebos
 2018-03-09 12:46:27 +01:00
Arne Morten Kvarving
a7d1544585 changed: remove repos from regression data update script 2018-03-09 09:35:15 +01:00
Atgeirr Flø Rasmussen
d4096f43f7
Merge pull request #1427 from GitPaean/fixing_trans_pvto_test 
fixing test_transmissibilitymultipliers
 2018-03-08 08:48:38 +01:00
Atgeirr Flø Rasmussen
717a9180e5
Merge pull request #1422 from blattms/resort-to-split-operator 
If requested use matrix with well contributions only for the preconditioner.
 2018-03-07 14:33:09 +01:00
Markus Blatt
e27fa72aee Use extended sparsity pattern also of we add well contributions to the preconditioner. 2018-03-07 13:54:21 +01:00
Kai Bao
75e6bcefbb fixing test_transmissibilitymultipliers 
due to the change of the PVTO table policy from the parser
 2018-03-07 10:01:34 +01:00
Arne Morten Kvarving
a6794063c8
Merge pull request #1424 from akva2/adjust_ecl 
adjustments for eclipse support in opm-common
 2018-03-06 15:35:00 +01:00
Tor Harald Sandve
0c4ae5991f Use advanceTimeLevel() and updatedFailed() methods from model 2018-03-06 13:18:03 +01:00
Arne Morten Kvarving
690ca2cd26 adjustments for eclipse support in opm-common 2018-03-06 09:09:58 +01:00
Markus Blatt
01b25f27ed Allow to add well contributions to both matrix and preconditioner matrix. 
matrix_add_well_contributions=true will add to both the preconditioner matrix
and the matrix of the linear operator. preconditioner_add_well_contributions=true
will only add the contributions to the preconditioner matrix.
 2018-03-05 16:48:10 +01:00
Markus Blatt
dc86d972d2 Removed hack to access StandardWell::addWellContributions. 2018-03-05 16:48:03 +01:00
Tor Harald Sandve
5f33a1a1bc Bugfix. enableStorageTerm=false 
Shift intensiveQuantities each timestep
 2018-03-05 14:53:59 +01:00
Atgeirr Flø Rasmussen
13b7a4a426
Merge pull request #1423 from andlaus/fix_icc 
Fix icc
 2018-03-05 09:20:46 +01:00
Andreas Lauser
999653783b do not name exception objects if they are not used 
ICC rightfully produces a warning here...
 2018-03-03 13:31:17 +01:00
Andreas Lauser
732d86613e do not mess around with GCC's optimizations 
generally, this is a bad idea because GCC is very good at determining
which optimizations should be enabled or not and also, this may
interfere with some compilers. (e.g., it produces a warning on icc)
 2018-03-03 13:31:17 +01:00
Markus Blatt
12d2114bd7 Add well contribution to preconditioner matrix and always use old operator approach for wells. 
It turned out that applying the well part of the full matrix has to be
done after the application of the non-well interactions. Otherwise we
screw up the ordering so much that convergence suffers a lot.

Kudos got Atgeirr for inspiration based on his testing.
 2018-03-02 20:57:43 +01:00
Markus Blatt
f07874cf42 Cleanup code. 
Removed unused boolean switch in WellModelMatrixAdapter and removed commented code.
 2018-03-02 20:57:10 +01:00
Markus Blatt
af3c45dd6a Do not add additional entries to matrix when not adding well contributions. 2018-03-02 20:55:58 +01:00
Atgeirr Flø Rasmussen
b276aaa16b Changes made to test variations of matrix vs matrixless(ish) operator. 2018-03-02 14:22:59 +01:00
Atgeirr Flø Rasmussen
0db0e83843
Merge pull request #1414 from GitPaean/hack_to_fix_weird_well_rate_situation 
trying to fix a problematic situation when updateWellState
 2018-02-28 16:00:42 +01:00
Atgeirr Flø Rasmussen
320ae1b229
Merge pull request #1416 from GitPaean/support_oil_water_polymer 
trying to add the functionality to handle oil/water polymer
 2018-02-28 12:30:37 +01:00
Kai Bao
56caffab67 fixing comments for PR#1416 2018-02-28 10:26:54 +01:00
Kai Bao
ec8dc4f534 trying to fix a problematic situation when updateWellState 
for example, distr = 0 1 0, for rate control, F = 1 0 0, we will get a
zero F_target, then it is wrong to do the division with F_target.

It is okay when target = 0. When target != 0, the result is already very
wrong. Non-linear iteration can send us anywhere, not sure the best
solution for this kind of strange situation.
 2018-02-28 09:47:22 +01:00
Kai Bao
b71949b09e adding a test for polymer oil water 2018-02-28 08:39:38 +01:00
Kai Bao
635f6c7db2 trying to add the functionality to handle oil/water polymer 2018-02-28 08:39:38 +01:00
Atgeirr Flø Rasmussen
10887ddaca
Merge pull request #1183 from blattms/matrix-with-well-connections 
Add possibility to use matrix with well connections for preconditioner.
 2018-02-27 16:25:06 +01:00
Markus Blatt
44d31f9dcc Removed unused variable warning in addWellContributions. 2018-02-26 16:23:13 +01:00
Markus Blatt
8aa0973cfc Removed unsused friend declaration. 
This seemed to be a leftover from the approach taken
before the refactoring of the well code.
 2018-02-26 16:20:41 +01:00
Markus Blatt
247241c037 Added test for multMatrixTransposed. 2018-02-26 16:11:53 +01:00
Markus Blatt
3be954fe98 Move multMatrixTransposed for better testability. 
Otherwise we need to deal with the property system of ewoms.
 2018-02-26 15:49:59 +01:00
Markus Blatt
e53374b8ec Check which each well whether to call apply for well contributions. 
This allows for different implementations of wells in well_container
when adding well contributions to the matrix.
 2018-02-26 15:47:25 +01:00
Markus Blatt
455cf79d8d Fixed comments in StandardWell 2018-02-26 15:16:07 +01:00
Markus Blatt
5ec804d362 Prevent unused variable warning 2018-02-26 15:16:07 +01:00
Markus Blatt
3f0d6f7dca Space 2018-02-26 15:16:07 +01:00
Markus Blatt
96a636f25b Let WellModel decide whether to apply well contributions. 
It queries the Well whether the jacobian also contains well contributions.
If not then it applies them in the operator in addition. Thus the
well knows whether that is needed or not.
 2018-02-26 15:16:07 +01:00
Markus Blatt
afb806bc3e Remove effectless typedef in version check. 
The typedef is the same and has no effect since we
switched to Dune::FMatrixHelp::invertMatrix.
 2018-02-26 15:16:07 +01:00
Markus Blatt
8e78f02230 Use Dune::FMatrixHelp::invertMatrix for inversion 
Thus we always take advantage of the specializations in ISTLSolver.hpp
and do need to take care about the type of the matrix block.
 2018-02-26 15:12:14 +01:00
Markus Blatt
b5d55b9fb2 Fix indentation and whitespace. 2018-02-26 14:35:04 +01:00
Markus Blatt
2e7313ac08 Cater for move of CpGrid header 2018-02-26 14:35:04 +01:00
Markus Blatt
9221bc958b Added missing header includes 2018-02-26 14:35:04 +01:00
Markus Blatt
01b90bce05 Improve well matrix products 2018-02-26 14:35:04 +01:00
Markus Blatt
22dfa6f135 Setup auxiliary module only once with all possible well completions. 
Clearing the auxiliary modules will result in the sparsity pattern being
recomputed. Previously we did this before each nonlinear solve. But for the
master branch the sparsity pattern was only computed once for the whole
simulation.

To get the same behaviour (one sparsity pattern computation) we change the
auxiliary module to include all well perforations that might become active
during the simulation and do this once up front of a simulator run.

Please note that the new matrix entries might also improve convergence for
the ILU if the well is not active.
 2018-02-26 14:35:04 +01:00