0

Changing inner grid boundary condition with time

Dears,

I have been trying to model CO2 injection in a saline aquifer reservoir where the injection rate and injection temperature are constant during the injection period. I have implied high density and heat capacity for the innermost grid block to simulate the Dirichlet boundary condition at wellbore gridlock (i.e. constant injection temperature). However, when injection stops this implication makes temperature at wellbore gridlock constant too which is not a realistic result during the shut-in period. Is it possible to change rock density and heat capacity to the actual values after injection stopes to remove the Dirichlet BC assumption? 

7 replies

null
    • Mikey_Hannon
    • 4 yrs ago
    • Reported - view

    It sounds like you'll need to do a restart.  Please look at section 6.1 of the TOUGH3 manual.  You'll need to change the input files to reduce the density and heat capacity of the wellbore block back to normal values after the injection period.

    • yqzhang
    • 4 yrs ago
    • Reported - view

    Please use the TIMBC block. Check it out in user's guide.

    Yingqi

      • Refaat_G_Hashish
      • 4 yrs ago
      • Reported - view

      Yingqi Zhang I have the following questions with TIMBC. The first question is which format has to be used for TIMBC1, TIMBC2, TIMBC3, and TIMBC4. The second question is how to define temperature as the time dependant BC; NBCPV refers to the number of BC, How Tough identifies that temperature is the primary variable that I mean. Finally, which temperature value that should be entered for time > injection time. Please see the following input data for my case.

      TIMBC
      1
      4        1
      A1  2
      0.            31536000
      35            35

    • Mikey_Hannon
    • 4 yrs ago
    • Reported - view

    Yeah, I take back what I said... Yingqi's answer is better. : )

    • Finsterle GeoConsulting
    • Stefan_Finsterle
    • 4 yrs ago
    • Reported - view

    Short clarification:

    TIMBC can be used if you know what the temperature (or pressure or saturation or mass fraction) is during your shut-in phase, and you want to prescribe it (i.e., TIMBC is for time-dependent Dirichlet boundary conditions). I presume, however, you want to have TOUGH3 calculate how the temperature recovers once you stop injection. As TIMBC does not allow you to change material properties or element volumes, you have to use Mikey's initial suggestion: Run the simulation to the end of the injection period, then restart it (using the SAVE file as the INCON block), but change the volume and grain specific heat in the input file for the recovery phase.

    Short notes:

    Having the volume large fixes both the pressure and temperature, i.e., you would not need to also increase SPHT. Increasing SPHT is only needed if you want to fix temperature, but not any of the other primary variables.

    You could elegantly run an injection/shut-in test in a single simulation using features provided by iTOUGH2.

      • Refaat_G_Hashish
      • 4 yrs ago
      • Reported - view

      Stefan Finsterle Thanks for your clarification. I think you understand my problem, I don't know the temperature at the shut-in period. I think doing restart will be useful.

    • yqzhang
    • 4 yrs ago
    • Reported - view

    Sorry I think I confused the TIMBC with the Restart feature (in which you can just change rock properties or element volume) in iTOUGH2

    Yingqi

Content aside

  • 4 yrs agoLast active
  • 7Replies
  • 120Views
  • 4 Following