0

Relative permeability curves in tough2

Hello Every one

I faced problem with the relative permeability curve while I working with tough2.1 to build a 3D reservor model for CO2 storage by using both the source code and also the petrasim interface. The problem is I can not find any of the correlation can match the experemental curves. Actaully The proble is for all of these curve the maxilmum Kr liquid is = 1. while just one correlation can finished below the 1 but also not represent the actual case for example for water wet reservoir the maximum value of Kr water is depend on the value of sgr and finish stight with it. I attached the field or the actual curves which I want to use it in my model. also attached examples of all corelations in tough. Could any one help me how I can use my own curves in my model. I Have the source code for ECO2M and also Petrasim interface.

Kind Regards

Emad

5 replies

null
    • Jan_Tecklenburg
    • 1 yr ago
    • Reported - view

    Hi all,

    is there a solution for this problem? I have the same question.

    I am working with ECO2N and would like to model the relative permeability curves measured by Bennion and Bachu (2016). These curves have endpoints Kr_max < 1. All relative permeability models in TOUGH3 seems to have fixed endpoints at Kr_max = 1.

    Kind regards,

    Jan Tecklenburg

    • Staff Scientist
    • Christine_Doughty
    • 1 yr ago
    • Reported - view

    If you are using TOUGH3 (or iTOUGH I think), the option IRP=12, ICP=12 invokes hysteretic relative permeability and capillary pressure, which includes the option for Krg_max<1 (with RP(5)).  If you do not actually want hysteresis, you can set Sgrmax=0 (with RP(3)). 

    If you know how to program in Fortran and have the TOUGH source code, you can create your own relative permeability functions.  The general procedure is described in the TOUGH2 and TOUGH3 User's guides in the appendix with the relative permeability functions.

    • Reservoir Engineer
    • Alfredo_b
    • 1 yr ago
    • Reported - view

    Dear Emad,

    I would try with the VG model for the water phase and the Corey model for the gas phase using the set of parameters listed below. The two figures show a reasonable match of your data. 

    You may try to get something better. I used the excel file you can download from the Petrasim web site.

    You have to check if the negative SWR, which has no physical meaning but is used only for matching purposes, gives you any kind of problem when used inside TOUGH2. I have not tried.

    Be careful with your data, as at SW=0.5 and 0.4 the overall mobility of your mixture is really low.

    About the KRGmax<1 I would reasoning about the definition of relative permeability. In this case it means that your medium has an effective permeability to the gas phase at full gas saturation which is lower than the effective permeability to the aqueous phase.  In practice rock permeability is higher for the water phase.

    As mentioned by Christine, other relationships can be coded for specific purposes.

    Using:  Krg= krgmax  (1- S*)^ nG

    with:  S*=( SW-SWr)/(1-SWr-SGr) 

    could probably allows you to match your KRg data.

    Regards,

    Alfredo

     

    RELP-1

    RP(1)= 0.55 (lambda)
    RP(2)= 0.03 (Slr)
    RP(3)= 1 (Sls)
    RP(4)= 0.4 (Sgr)

     RELP-2

    RP(1)= 0.615 (lambda)
    RP(2)= -0.13 (Slr)
    RP(3)= 1 (Sls)
    RP(4)= 0.42 (Sgr)
    • Nematollah_Zamani
    • 1 yr ago
    • Reported - view

    Hi, 

    Regarding to this issue, I am wondering how I can define different Kr for liquid and gas in the same rock type?

    Regards 

    • Staff Scientist
    • Christine_Doughty
    • 1 yr ago
    • Reported - view

    If none of the existing relative permeability options in TOUGH2 provide Krl and Krg that you like, then you will have to create your own.  You could take the Krl function from one IRP option and the Krg function from a different IRP option, but it is up to you to make sure they make sense together (for example, you should have Krl+Krg<=1 for all values of saturation).  You will need the TOUGH source code and some rudimentary knowledge coding FORTRAN.  In TOUGH3, the hysteretic choice IRP=12 (must be used with ICP=12) probably has the most degrees of freedom for setting Krl and Krg independently, and you can turn off hysteresis by setting Sgrmax=0.

Content aside

  • 1 yr agoLast active
  • 5Replies
  • 746Views
  • 6 Following