[Opm] Convergence Problems - New Shale Well Model (Resend w/o attach)

Sun Mar 22 17:39:29 UTC 2020

PS: I should add that I think you put your finger on the problem with the comment about high contrast permeabilities. I don't recall having this much trouble with my Eclipse models. Hence, my suspicions that it is a difference in the solver logic that is causing my trouble. Cheers! - NR

________________________________
From: Opm <opm-bounces at opm-project.org> on behalf of Neale ROBERTS <ncroberts4 at msn.com>
Sent: Sunday, March 22, 2020 10:59 AM
To: OPM User <opmuser0 at gmail.com>; opm at opm-project.org <opm at opm-project.org>
Subject: Re: [Opm] Convergence Problems - New Shale Well Model (Resend w/o attach)

OPM User,

Thanks for your reply. Unfortunately, the things you are questioning tend to be revisions I've made trying to resolve the solver issues. Water generally does not flow in tight reservoirs. Since it has compressibility similar to the rock, eliminating it can be a reasonable simplification. I've had good success matching tight oil wells using RVCONST to continue some liquid production after Sorg, and have actually observed in the field where the liquid production evolves from a dark oil into a light/clear condensate. I extended the PVT tables to cover the increasing pressures, hoping that the extensions would correct the oscillations. They did not. I am unable to look at your run as it was blocked by the administrator, but if it ran slowly with oscillations,  I'm afraid it will contain the same problems as my original runs. If you look at the pressure grid at the end of the run, I think you'll still see the high pressures adjacent to the high perm fracture cells that are obviously incorrect. I appreciate the comment about the restart files, and will definitely act on that as the files are filling up my drive! Thanks again! - NR

________________________________
From: Opm <opm-bounces at opm-project.org> on behalf of OPM User <opmuser0 at gmail.com>
Sent: Sunday, March 22, 2020 12:16 AM
To: opm at opm-project.org <opm at opm-project.org>
Subject: Re: [Opm] Convergence Problems - New Shale Well Model (Resend w/o attach)

Neale,

The email with the results has been automatically held until the list
moderator can review it for approval. So I have just sent the text
instead in this email.

OPM User
opmuser0 at gmail.com

------------------------------------------------------------------------
Neale,

I had a quick look at this and I'm a bit confused on what you are trying
to do, for example, why is there no water in the model? Now I understand
that you probably making changes just to get the model working, as per
the commented out keywords, so I need to get a better understanding of
what you want the final model to look like.

Note that OPM Flow does not have the same level of error checking as the
commercial simulator, but this is improving with each release, so one
needs to be more careful in checking the decks. Note that:

 1.

    There was a few errors on the TABDIMS keyword but that does not
    effect OPM Flow as it automatically dimensions the tables as it
    reads the keywords.

 2.

    RVCONST keyword is not needed for runs containing dry gas.

 3.

    On the RPTSCHED keyword BASIC is not a valid sub keyword.

 4.

    OPM Flow correctly reports that BASIC is not supported on the RPTRST
    keyword, that means the simulator is going to write a restart record
    every report time step. I don't think you want this.

I ran you model as is using 2019-10 release and it runs, it is slow with
lots of oscillating behavior messages, but it does run. The obvious
candidates for poor numerical performance are the relative permeability
curves and the PVT data. Your relative permeability curves are nice and
smooth so I don't think they are the problem. It looks as though the PVT
data set has been modified to account for high pressures, is that
correct? If so that could be part of the problem. Some times this can
result in consistencies between the oil and gas formation volume factors
and compressibilities. The commercial simulator normally gives the
dreaded "Negative Compressibility Found" warning message; however, OPM
Flow (as far as I'm aware) does not check for this. So I have another
look at the PVT. For a quick fix you good just set a lower reservoir
pressure so that you stay below the extrapolated PVT data.

Note also that these type of models with high contract permeabilities
are always challenging models to run.

I have attached results from running your case, except for the 2GB
restart files. You can load the summary data into OPM ResInsight to
review the results.

Let me know how you get on.

OPM User
opmuser0 at gmail.com

On 21-Mar-20 20:00, opm-request at opm-project.org wrote:
> Send Opm mailing list submissions to
>        opm at opm-project.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
>        https://opm-project.org/cgi-bin/mailman/listinfo/opm
> or, via email, send a message with subject or body 'help' to
>        opm-request at opm-project.org
>
> You can reach the person managing the list at
>        opm-owner at opm-project.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Opm digest..."
>
>
> Today's Topics:
>
>    1. Convergence Problems - New Shale Well  Model (Neale ROBERTS)
>
>
> _______________________________________________
> Opm mailing list
> Opm at opm-project.org
> https://opm-project.org/cgi-bin/mailman/listinfo/opm

--
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
_______________________________________________
Opm mailing list
Opm at opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm
_______________________________________________
Opm mailing list
Opm at opm-project.org
https://opm-project.org/cgi-bin/mailman/listinfo/opm