1

Simulated CO2 injection into aquifers using T2WELL/ECO2M did not run successfully

Dear TOUGH community users,

  I am new to T2WELL/ECO2M and I am simulating CO2 injection into the aquifer, establishing a model of 0~-900m, the aquifer is located at -750~-800m, the injection rate is 8kg/s, and the specific enthalpy is set at 0.247102e6 J/kg. I tried to run it but the calculations didn't converge.Please see the attachment for the input files . What am I doing wrong? Thank you! 

10 replies

null
    • kenny
    • 10 mths ago
    • Reported - view

    It is very difficult to tell what causes problem.  Bad covergence is from the energy balance equation at the  element "* B41" which is the element of well mouth. There are several versions of T2WELL. Handeling the energy balance equations is slightly different  between the versions. I noticed you are using "ECO2F" (from th output file). I have no any idea of ECO2F.   

    I would suggest playing around with this element by such as  turning on or off the heat exchange with surrounfing, changing the potential energy (adjusting Z), or settng it as constant temperature. 

      • happy12688
      • 10 mths ago
      • Reported - view

       Thank you Kenny. I very much appreciate your help!

    • Reservoir Engineer
    • Alfredo_b
    • 10 mths ago
    • Reported - view

    It is difficult to understand what it is not going on properly.

    I just noticed that the injection element reaches a low negative temperature of -17.96°C at about 20.4 bar. This mainly depends on the injection enthalpy assigned. Do you want to inject at this low temperature? Or chosen enthalpy should be revised?

    The T value is also very close to the minimum T supported by CO2TAB file used.

    Remember that CO2 enthalpy value should be congruent with the reference state used in the CO2TAB file.

    About the EOS version, I think ECO2F was the provisional name of the 3-phase ECO2M module implementing high temperature thermodynamic correlations up to 300°C. That version became ECO2M V.2.

    Regards,

    Alfredo

      • happy12688
      • 10 mths ago
      • Reported - view

       Thank you very much for your reply!

      I also wanted to ask you for help. The version I use is T2WELL-ECO2M. When liquid CO2 is injected into the reservoir, the simulated initial temperature and initial pressure are 20℃ and 8MPa respectively, and the enthalpy value corresponds to 0.247089e6 J/Kg. I'm not quite sure why the injection element reaches a low negative temperature of -17.96°C at about 20.4 bar.Can you give me some advice?

      Thanks again.

      • Reservoir Engineer
      • Alfredo_b
      • 10 mths ago
      • Reported - view

      @happy12688

      To inject CO2 at WH there are basically two approaches:

      - inject at constant T;

      - inject at constant enthalpy.

      If you inject at constant enthalpy, the enthalpy you assign should be that of the CO2 at the conditions of the pipeline supplying the CO2. As the WHP changes during the injection, the T corresponding to the local WHP and assigned enthalpy will vary accordingly (along an iso-enthalpic process). If you have an expansion of CO2, because the WHP is lower than pipeline P (this may happen when injecting in depleted reservoirs), then the CO2 will cool down and can give negative WHT. You can also have flashing of the CO2 and the generation of gaseous CO2 will further ruduces the mixture enthalpy.

      You write the enthalpy value corresponds to 20°C and 8 MPa, but the initial WHP seems to be 0.4185MPa. 

      Another aspect to consider is the value assigned to CO2 injection enthalpy. Let's assume you know the P&T of the CO2 available for injection (20°C and 8 MPa in yor case?).

      The corresponding enthalpy MUST be determined using ECO2M, not from external sources as the reference conditions for CO2 enthalpy might be different.

      You need to run a simulation in which one element has the given P&T conditions and it is saturated by CO2 only. Then you need to read the CO2 enthalpy in the output file. 

      I guess ECO2M does not print the enthalpy of CO2 phases. You may take the enthalpy from the content of PARAM array if you ask for the printout of secondary variables. Or you may add a MASS extraction with a tiny flow rate to read the enthalpy of extracted fluid.

      Hoping it may help.

      Regards,

      Alfredo

      • happy12688
      • 10 mths ago
      • Reported - view

       I'm glad you got back to me.

      The approach of CO2 injection at WH is constant enthalpy injection.The following grapgh is the parameters set by the GENER data block in the input file.The three lines represent the time, rate and enthalpy values respectively.

      Enthalpy Settings I am running a simulation, the following are the simulation in the input file INCON data block and the output file results.

       

      When I ran ECO2M the output file showed there was indeed a gaseous CO2 present as you said.  I'm simulating an injection of liquid CO2 from the surface.The first unit in the center of the surface is selected as the wellhead unit. I don't know where there is a problem in setting the enthalpy value, and the running result can't converge.Can you tell me what is wrong? Thanks very much.

    • Reservoir Engineer
    • Alfredo_b
    • 10 mths ago
    • Reported - view

    Have you checked if the enthalpy value you assign to injected CO2 corresponds to the intended P&T conditions in ECO2M as I suggested before?

    If you inject liquid CO2 with enthalpy corresponding to high P in a wellbore initially at low WHP, flashing of CO2 is likely to occur with cooling driven by the flash process. If cooling persists, the WHT can go well below 0°C and down to the lower limit of T handled by ECO2M. It is possible that the injection conditions you are trying to simulate are not supported by the present version of ECO2M.

    In addition, I would not start injecting at full rate from the beginning. I would try injecting at an increasing rate from time 0 to let say a couple of minutes (120 s) to reach the final target rate. This might reduce the initial transients giving so many troubles and would be probably closer to what is done at the field. This may alleviate the convergence issues at the beginning of the simulation, but will not solve the low T issue.

    I would also suggest to change the wellbore spatial discretization. You inject in element * B41 whose node is at -32.5 m. Thus, the conditions computed at element * B41 do not represent the WH conditions, but those at 32.5 m below ground level.

    **********Well/fracture cell geometry********
     ID   Name   SideArea   Diameter   Height
       1 * B41     0.1222E+030.5984E+000.6501E+02
       2   B42     0.1222E+030.5984E+000.6500E+02
       3   B43     0.1222E+030.5984E+000.6500E+02

    Regards,

    Alfredo

      • happy12688
      • 10 mths ago
      • Reported - view

        Thank you for your partience.

      I'd like to consult you something. In the process of learning T2WELL-ECO2M, I saw a simulation case that I did not understand very well. 

      In this case, a new data block REGFX appears in the input file for the simulation and an explanation in the output file.

       

      By looking at the input file, the simulation is performed in non-isothermal mode (NEQ = 4 in data block MULTI). There are some questions about how it determines the injection wellhead unit and enthalpy setting.

      Based on your reply, I have modified the GENER data block in the input file, but there still seems to be some problems. Could you please tell me whether this setting method meets what you said?  Thank you kindly!

    • happy12688
    • 10 mths ago
    • Reported - view

    I tried to simulate it with the REGFX data block in the input file, and the result was abnormal wellhead pressure rise. The GENER data block was also used in the input file to simulate it, resulting in incorrect wellhead temperature changes. Could someone please help me see where the problem is, I need to adjust. Thank you very much !!

    • Yongyi_li
    • 7 mths ago
    • Reported - view

    Hi, i am also a new student about tough2,could you please tell me how to solve this problem ?

Content aside

  • 1 Likes
  • 7 mths agoLast active
  • 10Replies
  • 196Views
  • 5 Following