Commit Graph

252 Commits

Author SHA1 Message Date
babrodtk
68f1cff930 Bugfix parallel computation of weighted pressure etc. 2016-10-07 15:11:59 +02:00
babrodtk
b774982878 Initial version of summary data 2016-10-07 15:06:53 +02:00
Markus Blatt
fd24070931 Omit "well converge" on non-root process 2016-10-05 10:33:14 +02:00
Atgeirr Flø Rasmussen
51237cb960 Merge pull request #829 from blattms/parallelize-computeFluidInPlace
Correctly parallelize computeFluidInPlace
2016-09-19 09:31:05 +02:00
babrodtk
5b75649a07 Simplified arguments to computeWaterShearVelocityFaces 2016-09-15 15:56:06 +02:00
Markus Blatt
2c70f05d6b Correctly parallelize computeFluidInPlace.
Its first implementation computed wrong results in parallel. With this commit
we noe have completely parallelized the computations and the results seem correct
for parallel runs with norne.
2016-09-15 15:43:08 +02:00
Markus Blatt
e15f9bfb9c Save space in computeFluidAndSpace.
Both hcpv and res will be used to save only dims elements. As dims
will most likely be much smaller than the number of cells, we only
allocate containers of size dims with this commit.
2016-09-15 15:35:27 +02:00
babrodtk
f7e3185546 Fixed bug where rsSat/rvSat were never set 2016-09-14 15:41:47 +02:00
Bård Skaflestad
dccf75d806 Initialise fields in declaration order 2016-09-08 16:28:43 +02:00
Atgeirr Flø Rasmussen
6962eb070c Merge pull request #813 from qilicun/fix-segfault-in-fip
Remove SolutionState dependency in FIP calculation.
2016-09-08 10:15:08 +02:00
Liu Ming
e0fad32695 remove SolutionState dependency in FIP calculation. 2016-09-08 09:41:02 +08:00
babrodtk
739976535f Added rs and rv output capabilities 2016-09-07 14:55:53 +02:00
babrodtk
40b2b95d87 Extended the support for keywords for restart file output 2016-09-07 14:55:53 +02:00
Liu Ming
4b88d4edd1 use PV weighted pressure when HCPV is zero. 2016-09-07 14:00:30 +08:00
Liu Ming
170607e60c check if the phase is present. 2016-09-05 16:14:27 +08:00
Liu Ming
490dbbf133 correct the formulation for PAV calculations. 2016-09-02 16:53:39 +08:00
Atgeirr Flø Rasmussen
b5e986df5f Remove well state argument from computeFluidInPlace(). 2016-08-30 13:34:18 +02:00
Atgeirr Flø Rasmussen
a2001ca2e1 Avoid variable shadowing. 2016-08-30 13:33:50 +02:00
Liu Ming
c9c7a5c744 compute hydrocarbon weighted pore voulme pressue and pv for fip regions. 2016-08-30 10:30:26 +02:00
Liu Ming
2ac3306c18 drop the for loop over fip regions. 2016-08-30 10:30:26 +02:00
Liu Ming
59a2668152 compute different fip regions. 2016-08-30 10:30:25 +02:00
Liu Ming
90086f690e only compute field fluid in place. 2016-08-30 10:30:25 +02:00
Liu Ming
056708574c compute fluid in place by regions. 2016-08-30 10:30:25 +02:00
Liu Ming
28583e4237 Add computeFluidInPlace function. 2016-08-30 10:30:25 +02:00
Atgeirr Flø Rasmussen
a084d36da6 Merge pull request #752 from andlaus/pass_timer_instead_of_dt
pass the simulation timer object instead of the time step size
2016-08-02 11:05:27 +02:00
Liu Ming
b43cbe0531 drop useages of std::numeric_limits for counting well iterations. 2016-07-14 10:25:32 +08:00
Andreas Lauser
5410d97701 pass the simulation timer object instead of the time step size
models may need a more detailed picture of where they are in the
simulation. Note that since the timer objects are available at every
call site, this is also not a very deep change.
2016-07-12 15:55:12 +02:00
Liu Ming
d3e6a657cd Initial well iterations as a special number, output it if it is valid. 2016-06-30 09:04:44 +08:00
Markus Blatt
79670b1c29 Prevent signed-unsigned-comparison warning. 2016-06-28 15:50:50 +08:00
Markus Blatt
c7cca46b2c Adds missing parallel reduction to euclidianNorm for multiple components 2016-06-28 15:50:50 +08:00
Markus Blatt
72d0b8123b Fixes the parallel Euclidian product for multiple phases.
This is used to compute the Euclidian product for the saturations.
Thes are ordered in an interleaved manner (all saturations for cell
with index 0, the all for index 1, ...). Up to now the implementation
assumed a different ordering: blockwise (all saturations for phase 0 first,
then all saturations phase 1, ...).
With this commit the computation uses the right assumption.
2016-06-28 15:50:50 +08:00
Markus Blatt
68f9180a65 Allow access to wells pointer for wells.
Using &stdwells.wells() throws an assertion for null pointers
without -DNDEBUG, but was used nevertheless. That prevented running
models without wells.
2016-06-28 15:50:50 +08:00
Markus Blatt
4c9d359638 Print all well switching in parallel.
As for each well only one process is responsible, the output process
does not see all wells. Ergo some well switching information was never
printed in a parallel run.

Therefore with this commit the well switching
message is printed regardless on which process it appears.
2016-06-28 15:50:50 +08:00
Liu Ming
ad01934670 initial IterationReport. 2016-06-28 15:48:20 +08:00
Liu Ming
c6586d36c8 return IterationReport for assemble and solver well eq methods. 2016-06-28 14:37:48 +08:00
Liu Ming
66b8a2379b write residuals to PRT file. 2016-06-21 09:27:19 +08:00
Liu Ming
969f6f1d1b return well iteration from simulators. 2016-06-21 08:35:33 +08:00
Atgeirr Flø Rasmussen
3776ccab9e Add use_update_stabilization parameter.
This makes it easier to run without, for example for debugging.
The default is 'true', preserving existing behaviour.
2016-06-17 10:44:53 +02:00
Atgeirr Flø Rasmussen
bf5fecfe9a Make prepareStep() parameters const. 2016-06-17 08:41:02 +02:00
Markus Blatt
82d6d949db Fix the parallel L-infinity norm calculation.
It was computing a global maximum before, which obviously is not the same thing.
2016-06-07 15:10:25 +02:00
Kai Bao
879f127ae8 moving variableStateExtractWellsVars() from models to Well Models. 2016-05-23 15:11:52 +02:00
Kai Bao
846ff890de refactoring the interface of computeWellPotentials()
to reduce the cost of makeConstantState when not calculating the
potentials.
2016-05-23 15:11:52 +02:00
Kai Bao
19a256dce0 some pure cleaning up. 2016-05-23 15:11:52 +02:00
Kai Bao
f2b7f2a56c moving numWellVars from models to Wells. 2016-05-23 15:11:52 +02:00
Kai Bao
a102e934ac removing baseSolveWellEq from BlackoilMultiSegmentModel
to solve the different interfaces of computeWellConnectionPressures for
StandardWells and MultisegmentWells, a function
computeWellConnectionPressures was introduced for the models.
2016-05-23 15:11:52 +02:00
Kai Bao
01328559e2 removing the updateWellState from the models. 2016-05-23 15:11:52 +02:00
Kai Bao
184bff95ba using well_model_ to represent the WellModel member 2016-05-23 15:11:51 +02:00
Kai Bao
06775ee02c Make BlackoilMultisegmentModel constructured from MultisegmentWells
Instead of being coustructed based StandardWells and having a
MultisegmentWells member.
2016-05-23 15:11:51 +02:00
Kai Bao
4254b48b57 unifying interface of a few functions of Wells classes
updateWellControls()
updateWellState()
addWellControlEq()

The change of function computeWellConnectionPressures() is not done
completely. Should find a solution later.
2016-05-23 15:11:51 +02:00
Kai Bao
be165a26e0 adding a few more members to the Wells classes.
gravity, vfp_properties_ and depth_.

to further simplify the interfaces.
2016-05-23 15:11:51 +02:00