TOUGHREACT1.2/ECO2N: About the inject rate and maximum pressure

denim_scarf
5 mths ago
10replies

Hi, everyone

I'm using TOUGHREACT1.2/ECO2N with Petrasim. I am simulating the injection of CO2(isothermal) at a constant rate of 6.5kg/s（just one GENER）using a rectangle mesh, but I find that it will be an error that shows after 30 days, but I want to simulate for years. And I finally found that the P around my GENER ELEMNT is about 59MPa which calculates 30 days, so I'm wondering if it's overpressurized. （The manual shows the maximum pressure is 600bar）

if it does， How should I set up my GENER items or change some settings to let my high inject rate run successfully？

Attached is my Petrasim file，I urgently need your help, thank you very much!

- Reservoir Engineer
- Alfredo_b
- 5 mths ago
- Reported - view
Your output file does not show any problem for the 30 days simulated. The P builds up to 596.33 bara, as you mention, for a relatively limited injection rate of 5 kg/s. And you have not linked permeability to the precipitation of solid salt, which actually occurs around the injection well. The error you mention after 30 years of simulation may be linked to P going above 600 bar. I have not checked that by running the model.

You inject in eleme " 1" belonging to domain "sand ", which as permeabilities kx, ky and kz of 5.8E-16, 5.8E-16, and 5.8E-18 m2, respectively.

In my view, you are simply trying to inject a too high rate into a very low permeability formation: less than 0.6 mD ! If the assigned permeability is representative of true formation properties, then you have to reduce your injection rate to levels compatible with the very poor formation properties.

Or considering to inject in a different formation...

Regards,

Alfredo
- denim_scarf
  
  5 mths ago
  
  Reported - view
  Alfredo Battistelli Thanks for your help, Alfredo.
  I changed my permeability to 550mD(the real parameter of the formation) which is equal to 5.8E-13m2, but it didn't work fine, just a little bit longer of the injection（nearly 30days） and I modified the simulation time, which is the maximum time that can be calculated and I found that the P of the injection element up to 59Mpa too quickly, so I would like to know is there some ways to fix this problem? such as changing the volume factor of the injecting element, adding the extra element, changing the scale of the element, or changing the element to the fixed state?
  
  or maybe it has relationship with my parameter of the mud or sandstone？
  
  （The layer is divided into sandstone and mudstone, overlying 80m of mudstone with permeability of 1e-18m2, porosity of 0.05, and a capillary entry pressure of 0.6MPa, and underlain by 500m of sandstone with permeability of 5.8e-13m2, porosity of 0.17, and a capillary entry pressure of 0.02MPa）
  
  For this stratigraphic parameter and injection rate，What should I do to fix the problem of the injection rate？ thank you very much！
  
  flow.inp flow.out
- Luoluo
  
  5 mths ago
  
  Reported - view
  My last reply kept showing up as pending review，so I changed for another account.
  
  Thanks for your help, Alfredo.
  I changed my permeability to 550mD(the real parameter of the formation) which is equal to 5.8E-13m2, but it didn't work fine, just a little bit longer of the injection（nearly 30days） and I modified the simulation time, which is the maximum time that can be calculated and I found that the P of the injection element up to 59Mpa too quickly, so I would like to know is there some ways to fix this problem? such as changing the volume factor of the injecting element, adding the extra element, changing the scale of the element, or changing the element to the fixed state?
  
  or maybe it has relationship with my parameter of the mud or sandstone？
  
  （The layer is divided into sandstone and mudstone, overlying 80m of mudstone with permeability of 1e-18m2, porosity of 0.05, and a capillary entry pressure of 0.6MPa, and underlain by 500m of sandstone with permeability of 5.8e-13m2, porosity of 0.17, and a capillary entry pressure of 0.02MPa）
  
  For this stratigraphic parameter and injection rate，What should I do to fix the problem of the injection rate？ thank you very much！
  
  (The files shows upload failure)
- Reservoir Engineer
- Alfredo_b
- 5 mths ago
- Reported - view
You are using a 2D vertical slice just 1 m wide. Injecting 5 kg/s CO2 in a vertical slice is not comparable to inject the same rate in a radial system. I think you should either use a radial grid or scale the injection rate from the radial injection to that on the vertical slice.

The aquifer domain "sands" has a considerable thickness of some 550 m, but you inject only in the bottom element 50 m thick. I imagine you wanted to simulate the upward migration of CO2 under gravity driven flow.

But you assigned a very low vertical permeability, initially with a kx/kz ratio of 100. Thus, your injected CO2 is constrained into a 1 m wide 50 m thick volume with an almost linear flow. Under these conditions, there might be anything wrong in the simulation, but a too high CO2 injection rate for the system permeability and geometry.

Alfredo
- Luoluo
  
  5 mths ago
  
  Reported - view
  Thank you for your answer.
  
  following the tips, i changed my mesh to radial gird, and I modified my permeability to 5.8E-13 for kx and ky, and 5.8E-14 for kz, this allows the use of higher injection rate up to 8kg/s, but it failure to exhibit a gravity plume(see figures below) which can be represented in a rectangular grid, so I'm still wondering how to decide and adjust between the mesh and the injection rate.
  
  flow.inp
- Reservoir Engineer
  
  Alfredo_b
  
  5 mths ago
  
  Reported - view
  Not sure what you mean for: but it failure to exhibit a gravity plume.
  
  The plot shows the upflow of CO2 under gravity driven flow and its accumulation below the cap-rock. This time kx/kz=10 and it seems that a circular area with a radius of about 100 m is able to let almost all the injected CO2 to flow upward under density driven flow. Because of that, the radial expansion of CO2 plume at the model bottom, where you are injecting, is limited to less than 250 m.
  
  The values and ratio of kx and kz are controlling the shape of CO2 plume for a given injection rate, with additional impact of relative permeabilitie and capillary pressure functions used.
  
  I would add the flow vectors of both gas and aqueous phases to better understand the flow dynamics.
  
  Regards,
  
  Alfredo
- Luoluo
  
  5 mths ago
  
  Reported - view
  thanks for your guide,i will try follow your help
- zhangtuka
- 5 mths ago
- Reported - view
Hello, can you share a TOUGHREACT1.2 source code?Thank you very much
- TOUGHREACT Developer
  
  Eric_Sonnenthal
  
  5 mths ago
  
  Reported - view
  Unfortunately, the University of California License for TOUGHREACT does not allow sharing of source or executable versions.
- Sunny_Dai
- 4 wk ago
- Reported - view
Hello, I have also encountered the same problem. Can you help me take a look at my files?

flow.inp INCON MESH