Jump to content
RockWare Support Forum


  • Content count

  • Joined

  • Last visited

Everything posted by Alison

  1. I agree that it is strange that changing the output settings would affect the simulation run. I think its possible that adding specific print times would do this, as it changes the time steps required by the simulation. Changing the number of time steps output shouldn't change the simulation. I would be happy to take a look at your simulation and see if I can determine what is going on. Please send your SIM file to Alison @ rockware.com. You also might want to post this question on the TOUGH2/T2VOC forum supported by LBNL, as the TOUGH2 authors are really the most qualified to answer questions like this. Best Regards, Alison
  2. These parameters are described in the TOUGHREACT manual starting on page 31. CGUESS: initial guess for the concentration of the individual primary species (not total concentration), in moles/kg H2O (molal) for species other than H2O and in kg for H2O. Input values of CGUESS do not affect results of speciation calculations, but could affect the number of chemical iterations required during initial speciation computations. CTOT: if ICON=1, CTOT is total moles of aqueous species, and total amount (in kg) of liquid water for H2O. Molalities are then internally computed as CTOT/CTOT(h2o). If ICON > 1, refer to the discussion of ICON above for the meaning of CTOT.
  3. The mass fraction is based on units of mass (so grams of salt per grams of water).
  4. Hi Yuki, If you mean production of energy, yes, I would expect that to decrease over time assuming you are pumping cooler water into the model. If you mean the actual amount of water coming out of the wells, then I would think that there are many factors that are coming into play that might effect this (such as the type of production well you are using in the model). For example, are you using a constant mass out production well, or a well on deliverability production well? Best Regards, Alison
  5. It sounds like you are running into convergence problems, but its hard to know what could be causing these problems without looking at your model. I would recommend that you look at the output file to determine what model element is causing the problem. The simulation includes information about the element with the highest residual for each time step. You might try rerunning your model with the print options for this cell turned on, so that you can look more closely at the state of the cell as the model proceeds. Its possible that it is reaching a pressure/temperature not supported by the simulator. If you continue to have problems, you can send the SIM file directly to me at Alison @ rockware.com. Best Regards, Alison
  6. create a new layer - from file

    Julia, I can't think of a faster way to do this. I believe that the DXF and contours still require that the program re-triangulate the surface, so it probably wouldn't speed up the process. I do think that removing internal boundaries could possibly help, as the creation of regions can be pretty time consuming. Best Regards, Alison
  7. Inquiry about the Extra Cell in PetraSim

    You are correct, "Dist. to" is the distance between the center of the connecting (geometric) cell and the connection to the extra cell. Dist. This is the distance of the connection in the extra cell. Depending on what you are trying to represent with an extra cell, these distances could be very large (for example, if you are trying to represent the atmosphere) or very small (for example, if you are trying to represent a well bore).
  8. Dear Lang, Unfortunately, I don't know the answer to this question. My guess is that it is the ratio of the diffusivity in the free space to the diffusivity in the porous medium, but you may want to ask the authors of TOUGH2 this question on the LBNL TOUGH2 forum. Best Regards, Alison
  9. I'm sorry, but I'm not sure how to answer your question. If you would like to send your SIM file to Alison @ rockware.com, I would be happy to look into this further. Best Regards, Alison
  10. Limit on numbers of restats in a row

    I reviewed the SAVE files and found that this is a problem with restarting a TOUGHREACT model where the number of times steps has exceeded 99,999. The number of time steps is listed at the bottom of the SAVE file. For models with more than 99,999 time steps, the files lists a series of * symbols, as there are only 5 spaces allowed to store this information. Changing this to an actual number such as 00001 resolves the problem and the model will run correctly.
  11. Scale Drawing in RockWorks

    Another option is to use ReportWorks (which is included with RockWorks). You can scale your map in paper space and have more options with respect to the format of the scale bar.
  12. Logarithmic scale

    We have added this as a bug in our database and will try to fix it in the next maintenance release. In the meantime, you would have to take a screen shot of the linear legend and manually change the legend labels in a graphics program if you need a linear legend with logarithmic labels. We have put together an excel file that will help you calculate the appropriate labels for a given data range. If anyone else needs this, please contact me at alison @rockware.com and I will send this to you. Thanks, Alison
  13. Dimension of the inclined flow model

    Dear Lang Zou, You cannot create a grid with non-vertical boundaries in PetraSim. You would have to rotate the gravity vector about a flat grid to get the model you want. To do this, go to Properties / Global Properties and click on the Misc. tab. Best Regards, Alison
  14. Again, the answer to this question depends on how you've modified the model input. Keep in mind that flux through a model is always across cell boundaries. If you have modified the gravity vector to simulate an inclined model, then the flux would be across cell connections that are at an angle to the gravity vector and the majority of the flux would probably be across the x and y connections. If you have kept the gravity vector as a vertical vector, then flux will be divided between the x/y and z connections.
  15. The answer to this question really depend on whether you've made this an inclined model by using the layer editor to control the elevation of the cell tops and bases, or whether you have modified the angle of the gravity vector. If you've modified the gravity vector, then cells have essentially been rotated with respect to the gravity vector and kx would be like the second image. If you've modified the cell elevations, then the x and y connections are still perpendicular to the gravity vector and the kx and ky connections are vertical. Keep in mind that if you take this approach you are introducing a small amount of error into the model since the top and base cell connections are no longer horizontal.
  16. Iinjection rate and resulting water velocity

    Jaeshik, I would not expect the velocity be exactly proportional in a 2D model but I would still expect it to increase with increased pumping. If you want to send me your SIM file, I'll take a quick look and let you know if I see anything strange. Regards, Alison

    After looking at the user's SIM file we realized that the program wasn't acting as expected because the wells in the SIM file were non-vertical. The program typically ignores non-vertical wells during meshing unless they are close to vertical within a threshold. The tolerance can be imagined as a vertical cylinder with the center axis going through the top point of the well. The cross-sectional area of the cylinder is the "Max area near wells" input. The solution is to use the option to add additional refinement points during mesh generation.

    It would be helpful if we could take a look at your model - our tests show that the meshing tools get fairly close to the specified areas. For example I entered a value of 500 meters squared and the area of the cell intersecting the well was between 600 and 700. You can send your SIM file to Alison @ rockware.com. Best Regards, Alison
  19. Permeability modifier

    Could you tell me which simulator (TOUGH2, TOUGHREACT, etc.) and which EOS module you are using? Each fluid property module is slightly different, so it would be helpful if we could try to reproduce the problem ourselves. If you would like to send us your SIM file, that would probably be helpful as well. You can send it to Alison @ rockware.com. Best Regards, Alison
  20. Jaeshik, Unfortunately, I don't think there is a list of changes published. You would have to take into account changes implemented in both TOUGHREACT v2 and v3. PetraSim v2015 outputs TOUGHREACT v2 input files and also can be license with a TOUGHREACT v2 executable if you own the simulator. PetraSim v2015 also outputs TOUGHREACT v3 input files that should work with your TOUGHREACT v3 simulator (although keep in mind that this support is still really a beta version that has not been fully tested). Best Regards, Alison
  21. concentration output

    The EOS7C simulator doesn't give you an option to output data as a concentration, but the simulator does output a density for gas. If you want to send us an example of the model you've described, we can take a look at it and let you know if we see any reason why you would get such similar results with such drastically different initial conditions. My guess is that the additional mass is showing up in the model in the aqueous phase. Thanks, Alison
  22. free exit boundary (outflow boundary)

    Hi Amy, I think that most people would simply use a fixed state boundary cell to represent an open boundary. When a cell is specified as "fixed state", fluid (and gas) can flow in and out of the cell without effecting the pressure, temperature and phase state of the cell. Best Regards, Alison
  23. Another option (although maybe not quite as exact as you were looking for) would be to plot the 3D well trace with depth labels, along with the fault plane. You can create the fault plane as a grid through the Grid / Initialize program or as a ribbon through the Fault menu in the Utilities. The depth labels should give you a ballpark estimate of the measured depth at which the well trace crosses the plane.
  24. The key here is that TOUGH2 does not use element numbers, but rather uses “names” composed of 3 characters and 2 numbers. Because FORTRAN does not easily output leading zeros on integers, you end up with spaces in the number portion of the ELEME name. Thanks, Alison
  25. There is no way to load a savechem file like you can a save file, so that the results are loaded directly into the interface as initial conditions. Instead, you would need to use the TOUGHREACT / Restart Options menu to restart the model as documented in the TOUGHREACT manual. Best Regards, Alison