Rate constraints for Compositional models

[xcu...@yahoo.com]

Hi There,

I have a compositional reservoir model using the NaturalVarialsCompositionalModel /addWell with a rate constraint and a pbh limit. When the model was run with MRST2018a, there was no switch of well control from rate to pbh, but when running the model with MRST2019b, it switched rate control to bhp constraint at first time step, likely due to the high rate specified and the reservoir is undersaturated at the beginning. But when I tested the reservoir model with simple bhp control using both MRST2018a and 2019b, ithey produced the same results. My question is that what possible changes introduced to MRST2019b cause the two versions to produce different results / model behavior for the rate constraint? 

Another question: it seems the EOS models (e.g., PR) in MRST does not consider volume shift for fluid density computation. If i like to add volume shift, is it enough to just change the functions in the CompositionalPropertyModel.m as volumeshift does not affect the VLE calculations?

Best regards,

Albert

[Olav Møyner]

Hi Albert,

The change in behavior for rate injection is due to a change in how injection rates are computed. Previously, the rate would be interpreted as a surface rate according to the surface rates in the fluid (e.g. fluid.rhoGS and fluid.rhoOS). In order to have better behavior for gas injection, the simulator now flashes the injection mixture at surface conditions and tries to figure out the surface density that way. This behavior can be adjusted by setting

W(wellNo).rhoS = [rhoWS, rhoOS, rhoGS] (or [rhoOS, rhoGS] if the model is two-phase) where rhoXS is the desired surface density for a phase when injected for wellNo.

The developer version of MRST supports volume shift, so it will come in the next release in a few months. If you want to use it early, our development repositories are available here: https://bitbucket.org/mrst/mrst-core/wiki/Home

Here is a pointer to the relevant commit: https://bitbucket.org/mrst/mrst-autodiff/commits/da89adaaeaa43a9fba9fcd0405ca6391c9b1730a

Best regards,

Olav

---

From: 'xcu...@yahoo.com' via MRST-users: The Matlab Reservoir Simulation Toolbox User Group <sinte...@googlegroups.com>
Sent: Wednesday, March 18, 2020 15:45
To: MRST-users: The Matlab Reservoir Simulation Toolbox User Group <sinte...@googlegroups.com>
Subject: [MRST Users] Rate constraints for Compositional models

 

--
You received this message because you are subscribed to the Google Groups "MRST-users: The Matlab Reservoir Simulation Toolbox User Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sintef-mrst...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sintef-mrst/08cf4cfd-94e9-4e04-a964-782f43539542%40googlegroups.com.

[xcu...@yahoo.com]

Thanks Olav for you prompt reply.

Does the flash at surface condition also apply to production wells?  My well is a producing well, i did the flash calculations based on the well mass rates, the volume rate is much larger than the "rate" constraint I specified  for well control.  If the MRST 2019b model is what as you described, there wouldn't be any changes in well control in this case.

Thanks for the development version for volume shift!

Best regards,

Albert

To unsubscribe from this group and stop receiving emails from it, send an email to sinte...@googlegroups.com.

[Olav Møyner]

Hi,

The default separation of components into vapor and liquid streams simply assumes that the K-values at surface conditions are independent of molar concentration. In other words, this means that if pure CO2 is purely vapor at surface conditions, any CO2 in the produced mixture is placed into the gas stream. Obviously, this is not always reliable (esp. for mixtures with lots of intermediate weighted components that are partly dissolved). In this case, you will have to set up a separator. If you use the generic models (e.g. GenericOverallCompositionModel) you can do something like this:

s = EOSSeparator('pressure', 1*atm, 'T', 300); % Set conditions for surface

sg = SeparatorGroup(s); % Group made up of single surface separator

sg.mode = 'moles'; % Use mole mode for compositional

model.FacilityModel.SeparatorGroup = sg; % Set separator group

Then, the surface production streams are flashed during the simulation and the results should be as expected. You can even specify a full tree for separators, as well as different EOS-es (or volume shifts) for each of them. There is some overhead to this, as an extra flash is called per separator.

Some of these features are not very documented, but we hope to add an example for this soon.

Best regards,

Olav

---

From: sinte...@googlegroups.com <sinte...@googlegroups.com> on behalf of karw...@tamu.edu <karw...@tamu.edu>
Sent: Monday, July 20, 2020 18:02

To: MRST-users: The Matlab Reservoir Simulation Toolbox User Group <sinte...@googlegroups.com>

Subject: Re: [MRST Users] Rate constraints for Compositional models

 

I am sorry If I am not allowed to add the thread under this email. I think my question is related to Albert's previous questions. 

I met the same problem to control the production with the (volumetric) rate control for compositional module. For BHP control, I could get the results, which are the same to those with Eclipse 300 but I have the huge difference when using rate control. I also calculated the produced amounts through mass balance over the reservoirs and my own flash calculation. 

In another discussion, entitled "Constant Volume Rate instead of Constant Mass Rate for Compositional Module"

Commented that "the constant volume rate control is not supported for the well" at that time. It has been a long while since the disucssion, and I wonder if there is any modification made for the constant volume rate control for compositional module. 

I would appreciate any help,

Thanks,

To unsubscribe from this group and stop receiving emails from it, send an email to sintef-mrst...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sintef-mrst/9c3e3e40-d63b-48cd-bd75-bcd7a4236acdn%40googlegroups.com.