feat: well time step selector based on rates/bhp tables and clarify well rates logic #3427

paveltomin · 2024-11-06T00:23:52Z

Two things in this PR:

Implement setNextDt in well solver which looks at all rates/bhp/status tables and makes sure time step is selected so that we honor table intervals. Can be enabled by timeStepFromTables flag in well solver.
Change logic for well rates - currently inputted and outputted from t_n+dt (end of time step) - switch to t_n (beginning of time step) - this should be more consistent with lower interpolation method that is used and prevent grabbing "future" rate from tables and avoid weird behavior when time step cut can lead to shutting of wells.

paveltomin · 2024-11-06T00:27:54Z

src/coreComponents/physicsSolvers/fluidFlow/wells/WellSolverBase.cpp

@@ -377,4 +377,73 @@ WellControls const & WellSolverBase::getWellControls( WellElementSubRegion const
  return this->getGroup< WellControls >( subRegion.getWellControlsName());
 }

+real64 WellSolverBase::setNextDt( real64 const & currentTime, const real64 & lastDt, geos::DomainPartition & domain )


main thing is in this function

I noticed in test case that the writeCSV report is a little different ... Before ,first report was end of first DT, so no rates report at t=0, but there is report at last time, Now report starts at T=0, which is good, but no report at last timestep. So one needs to add dt to get the full picture, along with manually adding the line if in excel. Maybe on destruction or if component got end of sim call, the file record could be written..

Also how does this affect other output reporting...

i added time step size in the output, i think that should be enough to represent full picture
no effect on other output reporting is expected

…/well-dt

src/coreComponents/functions/TableFunction.cpp

tjb-ltk · 2024-11-12T23:38:39Z

src/coreComponents/physicsSolvers/fluidFlow/wells/WellSolverBase.cpp

@@ -377,4 +377,73 @@ WellControls const & WellSolverBase::getWellControls( WellElementSubRegion const
  return this->getGroup< WellControls >( subRegion.getWellControlsName());
 }

+real64 WellSolverBase::setNextDt( real64 const & currentTime, const real64 & lastDt, geos::DomainPartition & domain )


I noticed in test case that the writeCSV report is a little different ... Before ,first report was end of first DT, so no rates report at t=0, but there is report at last time, Now report starts at T=0, which is good, but no report at last timestep. So one needs to add dt to get the full picture, along with manually adding the line if in excel. Maybe on destruction or if component got end of sim call, the file record could be written..

tjb-ltk · 2024-11-12T23:46:50Z

src/coreComponents/physicsSolvers/fluidFlow/wells/WellSolverBase.cpp

@@ -377,4 +377,73 @@ WellControls const & WellSolverBase::getWellControls( WellElementSubRegion const
  return this->getGroup< WellControls >( subRegion.getWellControlsName());
 }

+real64 WellSolverBase::setNextDt( real64 const & currentTime, const real64 & lastDt, geos::DomainPartition & domain )


Also how does this affect other output reporting...

tjb-ltk · 2025-02-18T20:39:59Z

src/coreComponents/functions/TableFunction.hpp

+    //   - Nearest returns the value of the closest table vertex
+    //   - Upper returns the value of the next table vertex
+    //   - Lower returns the value of the previous table vertex
+    if( interpolationMethod == TableFunction::InterpolationType::Nearest )


Looks like there are 4 interpolation types... Linear, Nearest, Upper, Lower... Using the find method defaults to Upper ? then overridden if method is nearest or lower... Is there any chance of method being called with Linear

calling for Linear would not make much sense i think, should i add a check and throw an error for Linear?

added

GEOS_ASSERT_MSG( interpolationMethod != InterpolationType::Linear, GEOS_FMT( "{}: TableFunction::getCoord should not be called with {} interpolation method", getDataContext(), EnumStrings< InterpolationType >::toString( interpolationMethod )));

src/coreComponents/physicsSolvers/fluidFlow/wells/WellControls.cpp

tjb-ltk · 2025-02-18T21:35:07Z

src/coreComponents/physicsSolvers/PhysicsSolverBase.hpp

Changing BasedOnNewton tier changed to just IterNumber looks to support workflows beyond newton solvers? for better understanding. what is an example

simply the coupled sequential solver, it's not Newton iterations but the solver uses same logic for time step selection

paveltomin · 2025-02-25T21:48:01Z

@CusiniM, @VidarStiernstrom, @castelletto1, @cssherman, @frankfeifan, @jhuang2601, @matteofrigo5, @rrsettgast, @wrtobin please review

CusiniM · 2025-03-07T00:37:09Z

src/coreComponents/physicsSolvers/fluidFlow/wells/WellSolverBase.cpp

+  this->registerWrapper( viewKeyStruct::timeStepFromTablesFlagString(), &m_timeStepFromTables ).
+    setApplyDefaultValue( 0 ).
+    setInputFlag( dataRepository::InputFlags::OPTIONAL ).
+    setDescription( "Choose time step to honor rates/bhp tables time intervals" );


shouldn't this always be true? I am guessing we always want the timestep to respect the intervals, don't we?

If I remember correctly, @dkachuma requested this behavior to be an option and I was hesitant to make it default

MelReyCG

(Minor remarks)
Nice things in this PR!

MelReyCG · 2025-03-07T14:42:33Z

src/coreComponents/functions/TableFunction.hpp

+     * @brief ...
+     * @param[in] input vector of input value
+     * @param[in] dim table dimension
+     * @param[in] interpolationMethod interpolation method
+     * @return coordinate value


Maybe unify the documentation with the other getCoord signature?

MelReyCG · 2025-03-07T14:44:11Z

src/coreComponents/functions/TableFunction.hpp

+  /**
+   * @brief Method to get coordinates
+   * @param input a scalar input
+   * @param dim the table dimension
+   * @param interpolationMethod the interpolation method
+   * @return the coordinate
+   */
+  real64 getCoord( real64 const * const input, localIndex dim, InterpolationType interpolationMethod ) const;


I think you should annotate the docs that this method is assuming createKernelWrapper() has been called (because it is based on the m_kernelWrapper data).
Maybe add an error in it?

added
GEOS_ASSERT( dim >= 0 && dim < m_coordinates.size() );

well time step selector based on rates/bhp tables

fd5d21a

paveltomin added type: feature New feature or request flag: ready for review labels Nov 6, 2024

paveltomin requested review from rasimHZ and tjb-ltk November 6, 2024 00:23

paveltomin self-assigned this Nov 6, 2024

paveltomin requested review from CusiniM, castelletto1, rrsettgast, klevzoff, francoishamon, matteofrigo5, cssherman and wrtobin as code owners November 6, 2024 00:23

remove debug

a724165

paveltomin commented Nov 6, 2024

View reviewed changes

clean up logic

2ed34d6

paveltomin changed the title ~~feat: well time step selector based on rates/bhp tables~~ feat: well time step selector based on rates/bhp tables and clarify well rates logic Nov 6, 2024

paveltomin and others added 3 commits November 6, 2024 15:28

Update WellSolverBase.cpp

42827b1

Merge remote-tracking branch 'origin/develop' into pt/well-dt

f70fb2d

Merge branch 'pt/well-dt' of https://github.com/GEOS-DEV/GEOS into pt…

05d5168

…/well-dt

This comment was marked as resolved.

Sign in to view

paveltomin and others added 2 commits November 7, 2024 08:52

Merge branch 'develop' into pt/well-dt

433498b

code style

6c87a96

paveltomin removed request for klevzoff and francoishamon November 7, 2024 22:46

paveltomin commented Nov 8, 2024

View reviewed changes

src/coreComponents/functions/TableFunction.cpp Outdated Show resolved Hide resolved

Merge branch 'develop' into pt/well-dt

11bb54b

paveltomin requested review from frankfeifan and ryar9534 as code owners November 11, 2024 17:28

Pavel Tomin and others added 6 commits January 22, 2025 17:25

missing file

03170a3

Merge branch 'develop' into pt/well-dt

4cf1a3e

Merge branch 'develop' into pt/well-dt

777f05f

build fix

842ae59

Merge branch 'develop' into pt/well-dt

42a49ff

Merge branch 'develop' into pt/well-dt

31fcd8c

tjb-ltk approved these changes Feb 18, 2025

View reviewed changes

paveltomin and others added 3 commits February 18, 2025 17:08

Update PhysicsSolverBase.cpp

638bf71

Merge branch 'develop' into pt/well-dt

0451f0a

merge fix

75e38a7

paveltomin removed the request for review from ryar9534 March 4, 2025 15:33

Merge branch 'develop' into pt/well-dt

111eea8

CusiniM approved these changes Mar 7, 2025

View reviewed changes

MelReyCG approved these changes Mar 7, 2025

View reviewed changes

paveltomin and others added 3 commits March 7, 2025 09:03

Merge branch 'develop' into pt/well-dt

7a8e7eb

review comments

494eb93

Merge branch 'develop' into pt/well-dt

0c52b88

paveltomin added ci: run CUDA builds Allows to triggers (costly) CUDA jobs ci: run integrated tests Allows to run the integrated tests in GEOS CI and removed flag: ready for review labels Mar 10, 2025

paveltomin and others added 7 commits March 10, 2025 12:20

Merge branch 'develop' into pt/well-dt

f0bce1b

remove std

f641234

try this

c2d84ac

Update .integrated_tests.yaml

7eca216

Update BASELINE_NOTES.md

dcd9121

mpi fix

34d1de8

Update .integrated_tests.yaml

a30fdbe

paveltomin merged commit 82aea9d into develop Mar 11, 2025
24 of 25 checks passed

paveltomin deleted the pt/well-dt branch March 11, 2025 03:35

feat: well time step selector based on rates/bhp tables and clarify well rates logic #3427

feat: well time step selector based on rates/bhp tables and clarify well rates logic #3427

Uh oh!

Conversation

paveltomin commented Nov 6, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

This comment was marked as resolved.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

paveltomin commented Feb 25, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

MelReyCG left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

MelReyCG Mar 7, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

paveltomin commented Nov 6, 2024 •

edited

Loading

MelReyCG left a comment •

edited

Loading

MelReyCG Mar 7, 2025 •

edited

Loading