adding a few connection summary keywords#1599
Conversation
|
It is something I am really not familiar with, it is good someone can take a look at it. @bska @joakim-hove |
|
jenkins build this please |
GitPaean
force-pushed
the
adding_some_connection_summary_keywords
branch
from
55c6f7c to
eef67fb
Compare
|
jenkins build this please |
|
I tested it with the example The relevant summary output is set as follows, While when I do |
|
It looks like the generated summary keywords are in line with the E100 reference, even though I do not understand how it works. But most of COFR results are zero.
The purple line is the E100 reference result. |
|
The water cut and gas/oil ratio definitions look reasonable to me, but the flow-rate quantities probably need some adjustments. The flow rates are positive for production and negative for injection, so I would probably define them as // production - injection
{ "CGFR", sub(crate<rt::gas, producer>, crate<rt::gas, injector>) },
{ "COFR", sub(crate<rt::oil, producer>, crate<rt::oil, injector>) },
{ "CWFR", sub(crate<rt::wat, producer>, crate<rt::wat, injector>) },Note that the |
|
The jenkins failure are due to different summary keyword sets in the summary output. This will not be in the way of updating reference. |
Thanks @bska . I will do that. |
GitPaean
force-pushed
the
adding_some_connection_summary_keywords
branch
from
eef67fb to
8027e73
Compare
|
Thanks @bska . The result is better. The difference in the later part are not due to the summary output.
|
|
jenkins build this please |
Glad to hear it. Did you mean to push an update with the revised definitions of the connection phase flow rates or is that still in local developement on your machine? |
COFR CWFR CGFR CWCT CGOR
GitPaean
force-pushed
the
adding_some_connection_summary_keywords
branch
from
8027e73 to
1369a35
Compare
I forgot to commit it. Now it is pushed. |
|
jenkins build this update_data please |
Reason: OPM/opm-common#1599 opm-common = e52a188d861f8344ecd6f34d4b4ef6a06d1ddd4f opm-grid = f7df833275491828545b4aab43a895a272cd3ee2 opm-material = 16fcd186c70f170c80431c97e802eabcc548feae opm-models = 214100686144f801c4ff56c016bf71195c789c25 opm-simulators = e52a188d861f8344ecd6f34d4b4ef6a06d1ddd4f
|
For The result from the commercial simulator is the purple line. It looks like for closed well due to physical reason (probably also for economic limit, not sure about the SHUT well from deck), Basically, it might give us some clues about how the commercial simulator handles closed wells. They probably still have the closed wells in the well model, which is different from OPM-flow's current way. (excuse me if I am wrong at this point. ) However, for the |
|
The following (and also the above comment) is some findings about the summary output. Some are bugs, and some are different behavior from the commercial simulator. One is the historical mode summary values, as stated above, this case is a prediction case, the commercial simulator manages to provide some non-zero historical values, while the behavior is not very consistent. For this
For Same applies to For
Not sure whether the commercial simulator did some rounding off for the very small values, or some other techniques
|
|
The comments above is for providing information for further development. It should not be in the way of merging of this PR. This PR is ready for reviewing and merging. |
|
jenkins build this opm-tests=266 please |
bska
left a comment
bska
left a comment
There was a problem hiding this comment.
Looks good to me. I'll merge this and the downstream data update (OPM/opm-tests#266) into master.
2 participants
COFR CWFR CGFR CWCT CGOR