OilfieldToolsNet/opm-simulators
4
0 Fork 0
mirror of https://github.com/OPM/opm-simulators.git synced 2026-10-13 09:22:05 -05:00
Code Issues Packages Projects Releases Wiki Activity
5,170 Commits 31 Branches 117 Tags
338828fdce89231126cd6034f219ea4d1470b5d9
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Kai Bao 338828fdce fixing the running of the vfp related for flow_ebos 
the output of the THP is still not yet correct.
 2017-08-07 11:39:11 +02:00
Kai Bao ab1b57b994 fixing the if condition for 2p + dummy phase 
from if (numComp < numEq) to if (numComp < numWellEq)
 2017-07-31 12:43:50 +02:00
Kai Bao e0e1fc65e2 removing ebosCompToFlowPhaseIdx from StandardWellsDense 2017-06-21 11:44:43 +02:00
Kai Bao aaa66e0982 not switching the order of well equations. 2017-06-21 11:44:43 +02:00
Kai Bao 70e193696d removing the order switiching of the well primary variables. 
Order swtiching of the well primary variables make the code more
complicated, and harder to understand and extend.
 2017-06-21 11:44:43 +02:00
Kai Bao 908b2bdf81 making wellMod_ in WellModelMatrixAdapter is actually a well model 
instead of the BlackoilModelEbos, which reduces some confusion and also
removes a few unnecessary wrapping functions.
 2017-06-07 14:26:00 +02:00
Kai Bao 0700c8cf67 adding more checking for the potential values 
during the well well potential calculation.
 2017-04-11 16:51:16 +02:00
Kai Bao ce0e580cee fixing comments. 
no functional change.
 2017-04-11 16:51:16 +02:00
Kai Bao dd95114e6d fixing well potentials calculation after introducing the saturation table 
The calculation of well potentials of newly added well failed after using the saturation
table. Here, we fix it using much smaller parameters to stablize the
calculation, while it slows down the iteration process.

TODO: improves the convergence rate by either introducing better
paramters or better iteration strategy.
 2017-04-11 16:51:16 +02:00
Kai Bao 972b911495 fixing two warnings. 
no function change.
 2017-04-11 16:51:16 +02:00
Kai Bao 72e2b7c6c3 fixing errors resulting from the rebasing procedure. 2017-04-11 16:51:16 +02:00
Kai Bao 7a4d0bb8aa slightly re-organizing prepareTimeStep() 
should not impact the simulation results.
 2017-04-11 16:51:16 +02:00
Kai Bao 7038b6eb57 recovering the running of flow_multsigement without group control 
flow_multisegment should not be able to handel group control yet.
 2017-04-11 16:51:16 +02:00
Kai Bao e0d96b3410 removing compute_well_potentials_ from BlackoilModelParameters 2017-04-11 16:51:16 +02:00
Kai Bao 4ed95fcbf0 calculate well potentials based on requireWellPotentials() 
from well_colllection. Not based on command line paramters anymore.
 2017-04-11 16:51:16 +02:00
Kai Bao 181112491b recovering group control for legacy_flow. 2017-04-11 16:51:16 +02:00
Kai Bao 8fa3c4b324 removing one unnecessary if condition 2017-04-11 16:51:16 +02:00
Kai Bao e627f35176 removing well_potential_ from WellStateFullyImplicitBlackoil 2017-04-11 16:51:16 +02:00
Kai Bao 72d0d4ddb8 revising the computeWellPotentials in StandardWells 
so it compute the potentials for each well instead of each perforation.
 2017-04-11 16:51:16 +02:00
Kai Bao 8d96d835b3 removing the well potential from WellsManager constructor. 2017-04-11 16:51:16 +02:00
Kai Bao 1aff98c4f4 adding the extra template type from recent merged PR 2017-04-11 16:51:16 +02:00
Kai Bao 22b2217dd3 removing setWellSolutionsFromPrevState from WellStateFullyImplicitBlackoilDense 
Not used anymore.
 2017-04-11 16:51:16 +02:00
Kai Bao 032f9803be cleaning up unuseful comments and code fragments. 2017-04-11 16:51:16 +02:00
Kai Bao 279fbad17c removing the comments from the NonlinearSolver_impl 
which is from the development process.
 2017-04-11 16:51:16 +02:00
Kai Bao ef7769b77e tracking if wells are newly added in WellStateFullyImplicitBlackoil 
For a newly added well, the initialized rate cans cause big problem when
applied to VFP interploation when THP control is involved.
 2017-04-11 16:51:16 +02:00
Kai Bao a8ae9276c5 adding computeWellPotentialWithTHP() to compute well potential 
in an iterative way. When VFP interpolation is performed, the bhp and
rates are coupled together. Some iteration will be required to get the
consistent bhp and well potential.
 2017-04-11 16:51:16 +02:00
Kai Bao 182c5897c8 adding leastStrictBhpFromBhpLimits() 
to pick the least strict bhp limits in the well controls. It is used to
calculate the well potential when there is no thp limits there.

If there is thp limits there, it will be used for further chosing of the
bhp value for well potential calculation.
 2017-04-11 16:51:16 +02:00
Kai Bao 402a5a125a adding computeWellRatesWithBhp() to calculate well rates 
for one well based on the give Bhp.
 2017-04-11 16:51:16 +02:00
Kai Bao 7b2e6760e1 adding wellHasTHPConstraints() to decide if there is THP control 
for a specific well.
 2017-04-11 16:51:16 +02:00
Kai Bao 86dc231cc3 adding more things related to group control when restarting 
with a new time step due to convergence failure.
 2017-04-11 16:51:16 +02:00
Kai Bao e39ffefb2e performing prepareTimeStep in assemble for StandardWellsDense 
instead of model->preppareStep()

to avoid segmenttation error from eclSimulator and also not intervening
with NonlinearSolver
 2017-04-11 16:51:16 +02:00
Kai Bao 4a6dbbe0a7 updating the control index for both well state and well_controls struct 
when applying VREP group controls.
 2017-04-11 16:51:16 +02:00
Kai Bao cb85319a72 set thp() value to zero when no THP constraints exist 
in updateWellState()
 2017-04-11 16:51:16 +02:00
Kai Bao 7a7dcdf3ba correcting a comment in computeWellFlux() in StandardWellsDense 2017-04-11 16:51:16 +02:00
Kai Bao d02f028260 not resetWellControlFromState in assemble() in StandardWellsDense 
other places should gurantee the control indices from the well_state and
well_controls struct are the same.

And also, since the first element in wellSolutions() depends on the type
of the control, this function is complete yet.
 2017-04-11 16:51:16 +02:00
Kai Bao 34385ee58f synchronizing the control index at the beginning of assemble() 
in StandardWellsDense class.

It takes effects after restarting from a previous unconverged solution.
 2017-04-11 16:51:16 +02:00
Kai Bao 2bc1c8702e copying the thp value from prevState. 2017-04-11 16:51:16 +02:00
Kai Bao 77ec45cdd2 recovering the well_controls if solveWellEq not convergent 
since we are using the control index from both WellState and
well_controls.
 2017-04-11 16:51:16 +02:00
Kai Bao ea3adb3c37 using reference of WellState in resetWellControlFromState() 2017-04-11 16:51:16 +02:00
Kai Bao 0acb646a4d when there is group control, we should update WellState 
since the group targets are updated.
TODO: maybe it is not necessary, need to investigate the design of getBhp() and
getQs().
 2017-04-11 16:51:16 +02:00
Kai Bao 087b9471c7 completing prepareTimeStep() for StandardWellsDense 
updating all the group control related.
 2017-04-11 16:51:16 +02:00
Kai Bao b83f37dcc0 considering the distr has zero values for non-injecting phases 
to avoid NaN resulting from devided by zero.
 2017-04-11 16:51:16 +02:00
Kai Bao b21f577989 when there is VREP control, updateRateConverter in prepareStep 
in BlackoilModelEbos.
 2017-04-11 16:51:16 +02:00
Kai Bao 68eaf6ac1f Not copying the wellSolutions from previous WellState 
in WellStateFullyImplicitBlackoilDense.
 2017-04-11 16:51:16 +02:00
Kai Bao ed4e87ac24 not copying the contrl_index from the previous WellState 
when there is new control keywords enter, the combination of control /
constraints change. Under this kind of situation, copying the control
index does not mean you copy the same types of control.

Using the control index from the previous WellState for many situation
is a good practice, while it also risks copying a random and rather bad
choice. It is a major problem when we have VFP, VFP can have very
damaging values for certain situation to show it is not desirable to
be in this kind of situation.
 2017-04-11 16:51:16 +02:00
Kai Bao 653baae039 making the computeWellPotentials calculate potential for each well 
Not store the well potentials to WellState anymore.
 2017-04-11 16:51:16 +02:00
Kai Bao 4b5e1dfadc more treatment for zero rate wells in updateWellStateWithTarget 2017-04-11 16:51:16 +02:00
Kai Bao f70bb960dd updating rates of multiphase with updateWellStateWithTargets 
for StandardWellsDense model.
 2017-04-11 16:51:16 +02:00
Kai Bao 819aa90d84 adding prepareTimeStep() to the StandardWellsDense 
to handle the well potential related calculation
 2017-04-11 16:51:16 +02:00
Kai Bao 57f7650228 adding a function computeWellPotentials for StandardWellsDense 
for each well. It is a temporay approach for testing without breaking
any compilation before changing the interface for WellsManager.
 2017-04-11 16:51:16 +02:00
Kai Bao 2f3d142b38 a little more generalized implementation of BHP determination 
for well potentials calculation, not caring much about how many THP/BHP constraints exist.
 2017-03-14 15:07:10 +01:00
Kai Bao 18a75dcabc more careful determination of the most restrictive bhp value 
when computing the well potentials
 2017-03-14 14:58:21 +01:00
Kai Bao 3a89f07f3c not using comp_frac for producers in getQs 2017-03-08 11:02:47 +01:00
Kai Bao f48de7f146 using wellSurfaceVolumeFraction() in a few places 
instead of using wellVolumeFraction() to give more correct physical
meaning.
 2017-03-08 11:01:46 +01:00
Kai Bao 9c62d6d26c adding support of three phase suraface rate in getQs() 
and a few comments.
 2017-03-07 15:49:45 +01:00
Kai Bao 01d8813f53 A function to calulate the surface volume fraction of the rates 
for StandardWellsDense.
 2017-03-07 13:31:40 +01:00
Kai Bao b5e2c5262f using rs rv from grid cells when calculating the volume ratio 
during computeWellFlux for StandardWellsDense.
 2017-03-02 14:21:00 +01:00
Kai Bao 4129928be4 when no THP control/limit, thp value will be set to zero. 
It provides a strategy to reset the zero thp value when no thp control
is involved anymore.
 2017-02-17 17:12:52 +01:00
Kai Bao 71d777f839 update the thp value for flow_ebos 
otherwise the thp constraint will not work.
 2017-02-15 20:50:43 +01:00
Kai Bao 8b38b7b8a3 finishing separating the StandardWellsDense.hpp implementations. 2017-02-15 10:07:22 +01:00
Kai Bao 498f40f896 fourth part in separating the StandardWellsDense.hpp implementations. 2017-02-15 10:07:22 +01:00
Kai Bao 2a3fe58ac2 third part in separating the StandardWellsDense.hpp implementations. 2017-02-15 10:07:22 +01:00
Kai Bao 8de7795629 the second part in separating the StandardWellsDense.hpp implementations. 2017-02-15 10:07:22 +01:00
Kai Bao 8354f3600f the first part in separating the StandardWellsDense.hpp implementations. 
split into commits for easy debugging purpose.
 2017-02-15 10:07:22 +01:00
Kai Bao 4159542186 setting tolerance_wells to be 1.e-4. 
for better convergence behavoir.
 2017-01-31 10:50:10 +01:00
Kai Bao c3f3fed47a applying the efficiency factor to all the Jacobians 
of the cq_s when entering the reservoir mass balance equations.
 2017-01-25 15:56:15 +01:00
Kai Bao 68558bec76 moving the update the well_state at the end of the updatewellCotnrols 
in StandardWellsDense.

The reason is that the update the well controls can change the well
controls. It is better to update the well states after all the control
checking is done.
 2017-01-25 14:11:27 +01:00
Kai Bao 26785597b9 putting the update of the well states to one function 
in StandardWellsDense to simplify the updateWellControls() function.
 2017-01-25 14:11:27 +01:00
Kai Bao 19eb0d96c8 adding applyVREPGroupControl() to StandardWellsDense. 2017-01-25 14:11:27 +01:00
Kai Bao 849dae8ec5 adding computeWellVoidageRates() to StandardWellsDense 2017-01-25 14:11:27 +01:00
Kai Bao f53bf6f957 adding a pointer to rate_converter in BlackoilModelEbos 
in StandardWellsDense.
 2017-01-25 14:11:27 +01:00
Kai Bao af26b70fc9 adding a rate converter to BlackoilModelEbos 
and also function updateRateConverter() to update its state.
 2017-01-25 14:11:27 +01:00
Kai Bao 2dd4fa5bb2 fixing the LRAT control for StandardWellsDense 2017-01-23 11:30:23 +01:00
Kai Bao 31e7dec018 removing a few blank lines. 
and adding a assertion that bhp != 0 when calculating well potentials.
 2017-01-17 15:33:45 +01:00
Kai Bao 9e93ebcc64 check whether the group targets are converged. 2017-01-16 16:18:35 +01:00
Kai Bao 65a3fff6f2 moving computeWellPotentials to be earlier in assemble() 
in StandardWellsDense.

solveWellEq then have access to the updated well potentials.

does not change the result.
 2017-01-16 16:18:35 +01:00
Kai Bao 01322b740e fixing the mistakes in the code due to rebasing. 2017-01-16 16:18:35 +01:00
Kai Bao a95977e716 making the group control works for StandardWellsDense 2017-01-16 16:18:35 +01:00
Kai Bao b0e2bd9221 adding well perforation efficiency factor to StandardWellsDense 
And also calculateEfficiencyFactor to initialize the values.
 2017-01-16 16:18:35 +01:00
Kai Bao 4b47fa52ca adding well collection to StandardWellsDense 
to make StandardWellsDense have access to the group information.
 2017-01-16 16:18:35 +01:00
Kai Bao 941689722f adding functioin computeWellPotentials to StandardWellsDense 2017-01-16 16:18:35 +01:00
Kai Bao 92a4d8687f adding bhp as a parameter of computeWellFlux in StandardWellsDense 
To make the function usable for computing well potentionals.
 2017-01-16 16:18:35 +01:00
Kai Bao b8e0cda83a putting update of water injection mobility to a separate function 
computeInjectionMobility() in BlackoilPolymerModel
 2016-12-16 08:56:35 +01:00
Kai Bao 4f3ce37b8e correcting one not helpful comments. 
which results from the re-organization of the code.
 2016-12-16 08:56:35 +01:00
Kai Bao 894bffd1b3 handling the injection mobility when mixing paramter not 1 
The treatment with shear-thinning is not well tested yet.
 2016-12-16 08:56:35 +01:00
Kai Bao 4f8ae4be73 putting the re-calculating wellflux inside if plyshlog 
it only happens when shear effect calculation is needed. No point to
re-calculate the well flux when no shear thinning needed.
 2016-12-16 08:56:35 +01:00
Kai Bao 9307a41fd5 using the polymer concentration in the polymer transport 
since we already introduce the polymer viscosity there, we should apply
the water and polymer viscosity ratio again.
 2016-12-16 08:56:35 +01:00
Kai Bao 8148105ae3 Revert "changing polymer production concentration to be the grid block concnetration." 2016-12-05 15:05:56 +01:00
Kai Bao 81a73650de polymer production concentration to be the grid block concnetration. 
To be more consistent with the injection well.
 2016-12-05 12:52:21 +01:00
Kai Bao 69d5ec94c9 adding localWellsActive() checking for two MSW functions 
updatePerfPhaseRatesAndPressures() and addWellFluxEq()
 2016-11-25 13:45:38 +01:00
Kai Bao e57da11fbd making the updateWellControls an iterative process for StandardWells 
Seperating the updateWellStateWithTarget from updateWellControls, will
update the controls when any of the control constraints gets broken.

The reason is that the updated well control can still break some of
other the contstaints. We need to choose one control to make sure all the
constraints are honored.
 2016-11-24 11:27:59 +01:00
Kai Bao 0ab04b04cd checking whethter group control is active 
When applying group control related functions.
 2016-11-17 16:01:59 +01:00
Kai Bao d38a6eb459 adapting the change of the function findWellNode(). 2016-11-16 14:28:35 +01:00
Kai Bao d06acaa4b5 adapting to the change in opm-core 2016-11-16 14:06:20 +01:00
Kai Bao f85da3768e cleaning up and updating MultisegmentWells 2016-11-16 13:27:30 +01:00
Kai Bao 6bbc726bc2 checking and outputing the residuals for the well equations. 
Before, the tolerance for the norm of the residual of the well control
equations was 1bar, which makes no sense for rate control.
For some rate controls, we are not able to update the well state with
the target rates, we have to enforce the convergence throught the
residual checking.
 2016-11-16 13:26:54 +01:00
Kai Bao 002439c5b1 WIP in testing new group update strtegy. 
results with min_iter=2 works okay, while not min_iter=1.

It means the iteration stop at the wrong situation.
 2016-11-16 13:26:28 +01:00
Kai Bao 753da2dd36 putting the VREP control update after updateWellControls 
which makes it almost should be put inside updateWellControls function.

it does not change the result.
 2016-11-16 13:25:35 +01:00
Kai Bao a5aaf9b802 easier way to update whether the well under group control 
potentially can be more consistent way.
It does not change the result.
 2016-11-16 13:25:35 +01:00
Kai Bao 8a1e1e010a using findWellNode() to avoid dynamic_casting 
form WellGroupInterface* to WellNode*.
 2016-11-16 13:25:35 +01:00