How to input Injection rate in reservoir condition

[ahmadreza shojaee]

Dear all,

Would you please let me know whether it is possible to input the injection/production rate in the well structure at the reservoir condition? Based on things I found in the MRST books, the unit for 'rate' condition is sm^3/sec. I could not find anything about how to input them at reservoir condition.

Kind regards,

Ahmad

[ahmadreza shojaee]

Dear All,

I think there is a problem within the compositional module in setting well condition. when I input the 'rate' condition in a compositional simulation for a producer well, I think it is interpreted as constant mass flow rate condition.

The interesting is that the qGs output in wellsol structure equals to the summation of gas component flux.

Have any experience in this issue?

Kind regards,

Ahmad

[Olav Møyner]

Dear Ahmad,

I think it is a bit difficult to say without looking at the code you are running, but it sounds like maybe the calculated surface density is around 1 kg/m^3 for the produced gas? If so, the two would be very similar. 

Best regards,

Olav

---

From: sinte...@googlegroups.com <sinte...@googlegroups.com> on behalf of ahmadreza shojaee <ashoj...@gmail.com>
Sent: Friday, June 14, 2024 16:40
To: MRST-users: The Matlab Reservoir Simulation Toolbox User Group <sinte...@googlegroups.com>
Subject: [MRST Users] Re: How to input Injection rate in reservoir condition

 

--
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/7fb4a4ac-0d7c-47c8-b994-abc14c191baan%40googlegroups.com.

[ahmadreza shojaee]

Dear Olav,

Thanks for your reply. Would you please take a look at the attached code. Here is a brief description about the code:

1. Reservoir is saturated with water and CH4. 

2. 10 days of H2 injection.

3. 10 days of Shut-in period (by defining a well with zero rate - is it correct? is there any other way to close the well?)

4. 10 days (day 21 to day 30) of back production by a rate condition. The rate is set to 0.1 of the injection rate.

I think the wellSol.qGs is the summation of wellSol.H2 and wellSol.CH4. Also, wellSol.qWs equals to wellSol.Water. In this case if water density is assumed to be 1000kg/m^3, then wellSol.Water should be 0.001 of WellSol.qWs.

Furthermore, wellSol.val for the production period is the summation of component fluxes! I think it happens when we set the rate condition for a producer well.

Looking forward to hearing from you.

Best,

Ahmad
.

[Olav Møyner]

Dear Ahmad,

Thank you for your script. The reason why you are seeing that these components mass rates match up with the volumetric rates is that the fluid has a defaulted surface density of 1 kg/m^3.

MRST uses your inputs directly for surface rates. To get a better prediction, you will have to specify a separator with conditions that you want to flash at. At the moment, this is limited to oil-gas problems, which should be identical in behavior. I have attached a modification of your script that demonstrates this, and the mass and phase rates match has disappeared.

For shutting wells, setting W.status = false is the best option as that will disable the well completely, but I think it figures out that from a zero rate as well.

Best regards,

Olav

---

From: sinte...@googlegroups.com <sinte...@googlegroups.com> on behalf of ahmadreza shojaee <ashoj...@gmail.com>

Sent: Monday, June 17, 2024 11:24

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

Subject: Re: [MRST Users] Re: How to input Injection rate in reservoir condition

 

To view this discussion on the web visit https://groups.google.com/d/msgid/sintef-mrst/593e6290-2d16-433d-a341-065be25b1f58n%40googlegroups.com.

[ahmadreza shojaee]

Thank you, Olav!

So, is this modification is usable for gas-water compositional system? Now the model does not predict any water flow rate while there is a non-zero qOs flowrate in the well Solution.

My goal is that to re-produce the injected gas with a rate that equals to the injection rate. Is there anyway to do that? I do not need the volumetric flow rates; I only need the mass flow rates but I need to set the production rate the same as injection rate.

Best,

Ahmad

[ahmadreza shojaee]

Dear Olav,

Would you please let me know if there is any way or trick to manage the mentioned problem for compositional gas-water systems?

In the example you provided, you changed the phases from 'wg' to 'og' to connect the reservoir to the separator. Are the calculations still correct in a Water-CH4-H2 system? Is there an easy fix to apply that on 'wg' systems?

One idea comes to mind is that we can input a rate value based on density of 1kg/m^3 that represent the injected rate value and its density as follows:

Q2.Rho2 = Q1.Rho1 ==> Rho2 = 1 ==> Q2 = Q1.Rho1

It is not exact as in the back production all the components start to be produced and as a result the produced volume is not the same as injected volume. It is like setting a constant mass production condition.

Ahmad

[Olav Møyner]

Hi Ahmad,

This is currently not supported for water-gas systems. Note that there is no difference between gas-water and oil-gas other than the names of the labels. You can think of the qOs as the water rate, as it is just a question of labels in the code, and that you are a bit careful when setting up saturation functions (swapping SWOF with SGOF and checking the sign of the capillary pressure table)

 The separator assumes that the water phase is immiscible, but this can probably be fixed in a future MRST release.

If you don't use the separator you can specify the mass rates as you would like, I think, by setting equal densities for rhoS for both phases. Then, it will act as a mass rate condition. As you indeed note it is not feasible to match both mass and volumetric rates when producing since the produced composition influences the density prediction.

Best regards,

Olav

---

From: sinte...@googlegroups.com <sinte...@googlegroups.com> on behalf of ahmadreza shojaee <ashoj...@gmail.com>

Sent: Tuesday, June 18, 2024 10:56

To view this discussion on the web visit https://groups.google.com/d/msgid/sintef-mrst/9eb3b5d3-dd96-4faa-9eed-e7c20ce07107n%40googlegroups.com.

[ahmadreza shojaee]

Thanks for the clarification, Olav. It would be great if this approach works.

I faced with another inconsistency in the well data. I defined a same scenario as I provided earlier with more components:

components = H2O - H2 - CH4 - CO2 - N2 - H2S

From the well data, I tried to calculate water density by dividing H2O mass production rate to qOs which represents water production rate under this labeling. The calculated values don't equal the the value is predicted by EOS for water phase. How can we justify that? Would you please take a look at the given script?

Also, another question is that which one is primary variable from the wolver? component mass flow rate or phase volume flow rate? I think component mass flow rate is primary variable (derived from global mole fraction and density at reservoir condition) and it must be correct.

Kind regards,

Ahmad