Robert K
f38b76abe9
AdaptiveSimulatorTimer: initialization of first time step size follows the same rule as
...
for later steps.
PIDTimeStepControl: added maxgrowth factor which indicates the maximum allow groth of
the time step from one to the next value.
2015-01-16 16:08:59 +01:00
Atgeirr Flø Rasmussen
f7bd6076e6
Merge pull request #720 from blattms/add-facetag-access-2
...
Adds a free function to UgGridHelpers to access the face tag.
2015-01-15 14:19:42 +01:00
Atgeirr Flø Rasmussen
75ae3c33a2
Merge pull request #713 from dr-robertk/PR/EclipseWriter-revision-to-write-substeps
...
Introduce interface for SimulatorTimers and support writing of substeps with EclipseWriter.
2015-01-15 13:52:22 +01:00
Robert K
d97f622f91
EclipseWriter: remove leftovers of WriterTimer.
...
AdaptiveSimulatorTimer: use back instead of rbegin.
2015-01-15 11:34:59 +01:00
Markus Blatt
9cf21d87ee
Skip the pointer arithmetic.
...
Bard was right that the iterator_range returns the underlying
native pointers. Therefore can write the computation much more
natural.
2015-01-14 21:58:37 +01:00
Joakim Hove
a5f00298e0
Protecting against use of unassigned deck values.
2015-01-14 11:41:41 +01:00
Markus Blatt
1b6bbb035e
Fixed bogus documentation of vanished cell_index.
2015-01-14 10:09:01 +01:00
Markus Blatt
cba54977bb
Adds a free function to UgGridHelpers to access the face tag.
...
To identify the face tag we use the iterator over the cell faces.
This provides all the information needed both for UG and CpGrid.
2015-01-13 20:36:33 +01:00
Atgeirr Flø Rasmussen
7f8ae2a878
Merge pull request #718 from blattms/fixes-mulisym-minpvprocessor
...
Avoid multiple symbol definitions for MinPvProcessor.
2015-01-12 16:45:18 +01:00
Markus Blatt
f7622f0e71
Avoid multiple symbol definitions for MinPvProcessor.
...
The non-template class MinpvProcessor contains all function
definitions inside of the header file without inlining. The
results in the symbols being in the library of opm-core and
in case of using CpGrid an addtional occurence in other
libraries (opm-autodiff?).
This commit changes this by simply inlining all these functions.
Note that another (better?) option would be to move at least
the bigger once into a cpp file.
2015-01-12 16:35:03 +01:00
Robert K
01fe222c2b
Merge remote-tracking branch 'upstream/master' into PR/EclipseWriter-revision-to-write-substeps
2015-01-12 12:47:40 +01:00
Joakim Hove
f9bce38852
Wellsmanager will check completion status
...
With this commit the WellsManager will check the status of completions
before adding them to the internal struct wells
datastructure. Completions can be in the four states:
OPEN, SHUT, AUTO, POPN
Completions with state == SHUT will be ignored, wheras the wellsmanager
will throw if the states AUTO or POPN are encountered. The WELOPEN
keyword can also have the value 'STOP'; for completions that is
translated to 'SHUT' by Schedule object.
2015-01-12 12:43:45 +01:00
Joakim Hove
634c70a090
Pure white-space change
2015-01-12 12:43:44 +01:00
Joakim Hove
461e53e891
Added missing enum rename
2015-01-12 12:41:56 +01:00
Robert K
0ecff82940
Merge remote-tracking branch 'upstream/master' into PR/EclipseWriter-revision-to-write-substeps
2015-01-12 12:23:48 +01:00
Joakim Hove
5d0a76e728
Added WellCompletion namespace
2015-01-12 12:10:51 +01:00
Robert K
65a314db92
cleanup: reportStepIdx --> writeStepIdx.
...
startDate --> startDateTime
removal of debug output.
2015-01-09 16:22:03 +01:00
Robert K
6eeecbb02b
- adjust OutputWriter to SimulatorTimerInterface
...
- remove WriterTimer from EclipseWriter and use SimulatorTimerInterface
- adjust to the above in AdaptiveTimeStepping.
2015-01-09 16:22:03 +01:00
Robert K
4d1a2ad918
Introduce an interface for SimulatorTimer and AdaptiveSimulatorTimer.
...
currentDateTime and currentPosixTime are default implementations.
2015-01-09 16:21:20 +01:00
Robert K
10cffa770b
EclipseWriter: allow for writing of substeps in addition to report steps.
2015-01-09 16:21:03 +01:00
Tor Harald Sandve
ed0854bc88
Fix comments.
2015-01-09 09:10:10 +01:00
Tor Harald Sandve
a3f658f3b1
Add warning to c solvers when stopped wells occur
...
Stopped wells are treated as shut wells in the c solvers in opm-core.
I.e they are completly isolated from the domain.
2015-01-09 09:10:10 +01:00
Tor Harald Sandve
cf142969c8
Renaming well is shut to well is stopped.
...
Rename the the meaning for shut as whats used in Eclipse.
STOP: Well stopped off above the formation. I.e. allow for flow in the
well.
SHUT: Well completely isolated from the formation. The well is removed
from the well list.
2015-01-09 09:10:10 +01:00
Tor Harald Sandve
9f64449aa8
Support stopped wells
...
The given control is applied to the stopped well, and the status is set
to not open. (i.e closed, but currently named shut)
2015-01-09 09:10:10 +01:00
Atgeirr Flø Rasmussen
ab567ba4e9
Put reformatted error message in anonymous namespace.
2015-01-07 10:16:13 +01:00
Atgeirr Flø Rasmussen
7ed65b8ae0
Make compilation of class depend on Boost.Heap availablility.
...
If boost version is too old, the API is still the same and the
class will build, but throws upon construction.
2015-01-07 09:49:47 +01:00
Atgeirr Flø Rasmussen
27e88a54f2
Fix anisotropy closeness test.
...
The old test was simply wrong: it computed the M-distance and compared
to the grid radius, which becomes dependent on the scaling of the
metric M. The corrected test in isClose() depends on the anisotropy
ratio of M and the grid radius.
2015-01-02 15:22:02 +01:00
Atgeirr Flø Rasmussen
ad4de2af0f
Make isClose() use actual grid and metric data.
...
Note that current implementation is somewhat ad-hoc, and not
in line with the algorithm of the paper.
2014-12-29 12:28:30 +01:00
Atgeirr Flø Rasmussen
2e2ad711e5
Whitespace fix.
2014-12-29 12:28:29 +01:00
Atgeirr Flø Rasmussen
8cc68a231c
Add computeValueUpdate() to avoid recomputing.
2014-12-29 12:28:29 +01:00
Atgeirr Flø Rasmussen
1415384146
Change order of steps in algorithm to avoid rework.
2014-12-29 12:28:29 +01:00
Atgeirr Flø Rasmussen
eeae67c647
Follow renaming of function.
2014-12-29 12:28:29 +01:00
Atgeirr Flø Rasmussen
c10e28478f
Fix bug in distance derivative function.
2014-12-29 12:28:29 +01:00
Atgeirr Flø Rasmussen
86655f1962
Work in progress.
2014-12-29 12:28:29 +01:00
Atgeirr Flø Rasmussen
1ddb0abc89
Initial versions of compute functions done.
...
Still has bugs.
2014-12-29 12:28:29 +01:00
Atgeirr Flø Rasmussen
89b4ecb193
Solver logic finished.
...
Methods isClose(), computeFromTri() and computeFromLine() are still dummies.
2014-12-29 12:28:29 +01:00
Atgeirr Flø Rasmussen
c0fbb8ea03
Work in progress on AnisotropicEikonal.
2014-12-29 12:28:29 +01:00
Atgeirr Flø Rasmussen
3997d1c462
Added simple test, work in progress.
2014-12-29 12:28:29 +01:00
Atgeirr Flø Rasmussen
1f79502f10
Add skeleton of AnisotropicEikonal class and test.
2014-12-29 12:28:29 +01:00
Atgeirr Flø Rasmussen
060a2fea0f
Merge pull request #702 from jorgekva/linsolver_petsc
...
PETSc support in OPM
2014-12-19 11:25:58 +01:00
Jørgen Kvalsvik
31fbb2f58d
Sets initial guess to zero.
...
Setting it to nonzero means starting to approximate from memory garbage values.
2014-12-10 18:15:33 +01:00
Jørgen Kvalsvik
ed18b283f1
Adds missing VecCreate.
2014-12-10 18:15:13 +01:00
Jørgen Kvalsvik
e2e6299ee6
Adds missing preconditoner destruction.
2014-12-10 18:09:13 +01:00
Jørgen Kvalsvik
33c7b48d06
Renamed variables to match Ax = b
2014-12-10 18:09:01 +01:00
Jørgen Kvalsvik
ef59c5afdb
Vector construction is now value oriented.
...
No longer initialised and constructed in two places.
2014-12-10 18:07:45 +01:00
Jørgen Kvalsvik
2db194aca6
Removes unecessary whitespace.
2014-12-10 18:06:28 +01:00
Jørgen Kvalsvik
c6f42a63bd
Fixes performance bug in matrix construction.
...
At the same time changes the construction to be value-oriented, not
constructing it in OEM_DATA, but rather in to_petsc_mat
2014-12-10 18:05:04 +01:00
Joakim Hove
2b9b09865c
Revert signature of writeECLData()
2014-12-08 15:21:56 +01:00
chflo
29dd303eaa
OPM-139: Minor changes after review
2014-12-08 12:04:07 +01:00
chflo
0f97f947e6
OPM-139: Cleaned out compiler warnings
2014-12-08 12:04:07 +01:00