search

OPM / opm-simulators

OPM Flow and experimental simulators, including components such as well models etc.
http://www.opm-project.org
GNU General Public License v3.0
123 stars 121 forks source link

Mathematical explanation of #396 #1835

Open sogoogos opened 5 years ago

sogoogos commented 5 years ago

Now sure if this is a right place to ask, but does anyone has a reference that explains

396 (the way the well equation is solved separately) mathematically?

I personally would like to know how it is done.

GitPaean commented 5 years ago

It is the same well equations. Solving the well equations separately means solving the well equations without updating reservoir state (based on fixed reservoir state). The purpose of it is to provide a hopefully (most of cases it does) better initial guess for the well state. Basically, there is no trivial way to give an reasonable initial guess for the well state.

In the future, it might be depreciated or improved, since the well does change the reservoir state nearby the well in a pretty significant way. Fixing the reservoir might result in some unsolvable situation while it might be solvable by incorporating the change in the reservoir.

sogoogos commented 5 years ago

Oh, I see. I asked because I was not sure if this is a common thing that other reservoir simulators do.

It looks like there is the control option with --solve-welleq-initially for it. But what about --matrix-add-well-contributions and --preconditioner-add-well-contributions? Do they do the same thing?