0

Not to achineve the end time and no error

Hello everyone,

I am trying to run a geothermal reservoir model with EOS1 to predict a production rate for 10 years and the calculation did not achieve the setting end time. (stopped at only for several hours.) When I checked out file, there isn't a particular error.

Do you have any idea or tips what I should check?

Thanks and best regards,

Kodama

8 replies

null
    • Senior Geothermal Scientist - Vatnaskil Consulting Engineers
    • Andri_Arnaldsson
    • 3 yrs ago
    • Reported - view

    Can you share your files?

      • Kodama
      • 3 yrs ago
      • Reported - view

      Andri Arnaldsson 
       

      Thank you for your replying.
      I'm sharing you .out file and I hope you could comfirm the detail about results of the calculation. If you will need more information or other files, please reply again.

      Best regards,

      Kodama 

    • Keurfon_Luu
    • 3 yrs ago
    • Reported - view

    Your time steps seem to be quite small. Have you reached the maximum number of cycles?

      • Kodama
      • 3 yrs ago
      • Reported - view

      Keurfon Luu 

       

      Thank you for your replying.
      I set time steps like the attached image for the first time and thought thet time steps seemed to be quite small from the results. So I've tried before to change time step to 1,000,000s but it didn't work unfortunately.

      Best regard,

      Kodama

    • Staff Scientist
    • Christine_Doughty
    • 3 yrs ago
    • Reported - view

    The code seems to be pretty unhappy, reducing time step frequently.  To know more about what is happening, I recommend setting MOP(1)=1 (to see iteration information), MOP(5)=3 (to see phase changes), and MOP(7)=1 (to printout some of your input information, to verify that the code is reading it correctly), and rerunning the code.

      • Kodama
      • 3 yrs ago
      • Reported - view

      Christine Doughty 

      Thank you for your advice and I tried to set MOP(1),(5) and (7) to see detail about what is happened as you said. I'm sharing the out file under these setting.

      ※I develop the model using Petrasim and MOP(1) /MOP(7) are Automatically determined from selections on output control setting so there are deifferences of the MOP numbers I set from your specified one.

      Best regard,

      Kodama

    • Staff Scientist
    • Christine_Doughty
    • 3 yrs ago
    • Reported - view

    Whenever a code is not running well, it is helpful to set these MOPs to learn more about what is happening.  Look at your output file and look for the string " ST ".  This will show you each timestep the code takes.  The code seems to be running fine through step 11.  Starting at step 12 there are many messages about phase changes back and forth between two-phase and single-phase at element 2494.  This is the so-called "oscillating phase-change problem" and it is not that rare in TOUGH simulations involving propagating fronts of phase change.  Check your grid for sudden changes in grid resolution or big contrasts in properties near element 2494.  In general, changes in grid resolution should be as gradual as possible.  Sudden contrasts in properties are sometimes necessary to represent the actual problem, but sometimes you can make property changes gradual also.  I notice that capillary pressure is zero for your problem.  Try specifying a non-zero capillary pressure - this has the tendency to smooth out phase-change fronts  (make them broader), and this can alleviate the oscillating phase-change problem.  You can also tweak your relative permeability function to see if this helps.  If you are still having trouble, next time post both your input and output files.  This will make it easier for those reading the forum to make suggestions.

    Good luck!  Christine

      • Kodama
      • 3 yrs ago
      • Reported - view

      Christine Doughty 

      Thank you for your detail advise.

      I found that there were big contrasts in properties especially permeability to reproduce high enthalpy area near 2494 and I'm trying to change big contrasts in properties around the error cell as you said.

      Also, I set zoro capillary pressure because I didn't understand well the use of having capillary pressure so I will try various values and check the effect to the model they have.

      Thank you so much again.

      Best regard,

      Kodama

Content aside

  • 3 yrs agoLast active
  • 8Replies
  • 89Views
  • 5 Following