TOUGH3; long running time (3D model)
Hi everyone,
I am trying to build 3D model for CO2 injection in underground saline aquifer that is upper bounded by impermeable strata which extends to the surface. The model constitutes an injector and monitoring well. The latter is model using EPM (Equivalent Porous Media) concept (porosity = 0.99 & k = 10^-8 m2& No capillary pressure). The monitoring well extends from reservoir up to the surface. I have initialized the model such that pressure follows hydrostatic pressure gradient and temperature follows geothermal gradient (0.025 degC/m). I spent some time building the model, however it takes long time for first iteration!. I am not sure what is the problem in my model.
Note: Initially, I built the model with no impermeable cap strata and I defined the pressure and temperature at the top of reservoir using hydrostatic pressure gradient and geothermal temperature gradient and the model run successfully.
Thanks in advance,
2 replies
-
Hi Refaat,
I tried your input file, using TOUGH3 with ECO2N and got the same result as you. I tried decreasing the permeability in the monitoring-well material to 10^-10 and it runs the 20 time steps that I asked for. In the past, I have had to carefully choose the "huge" permeabilities I use to represent wells, faults, etc. You want them to be big compared to the other permeabilities in the problem, but I have found that you can't make them too big, just like here. It is a trial and error process for me to find a value that works and represents what I want to be happening.
Good luck, Christine
-
Dear Dr. Christine,
Thank you for your reply. I have tried to use published values of input data while building the model. I totally agree with you that the value of permeability that has to be specified to the monitoring well is unknown and I need to validate the results obtained from EPM model with T2Well to get a representative permeability value that best describes flow in wellbore. This what I am planning to do.
Regards,
Refaat