zero pore pressure boundary condition
Can TOUGH3 define pore pressure Dirichlet boundary condition as zero values?
In order to validate the coupled THM code, I have to compare the numerical solution to analytical solution. However, I see many problems with analytical solutions have zero pore pressure boundary conditions (e.g., the Terzaghi 1D consolidation problem).
When I try to define 0 pressure boundary condition in EOS3, the program will not work.
5 replies
-
Many fluid properties are calculated from functions of pressure and temperature. These functions are effect only at certain range of pressure. I do not think you can define 0 pressure boundary condition in TOUGH simulation.
-
I cannot run EOS3 under TOUGH3, so I'm referring to EOS3 supported by TOUGH2. EOS3 assumes that if the first primary variable X1<374.15 (the critical temperature of water for IFC67), then X1 is interpreted as the temperature. It seems no P lower than 374.15 Pa can be assigned.
Assigning for instance X1=400 Pa, T=10°C and Xair=1, EOS3 is able to run. 400 Pa is a tiny P with respect to atmospheric P and even if not 0, it would probably allow a comparison with the analytical solution.
You need also to consider the thermodynamic conditions you want to assign. No problem with 400 Pa and a single-gas phase with air, but at 400 Pa liquid water cannot exist as P is lower than the triple point P of water ( about 612 Pa). For instance, using P=660 Pa, T=1°C and Xair=0, EOS3 is able to run with a single liquid water phase. You need to consider the water saturation P as function of T.
Regards,
Alfredo
-
Alfredo Battistelli Much thanks for your detailed reply! Indeed the temperature and pressure we set should conform to the thermodynamic conditions.
Since I see many THM problems with analytical solutions always have zero-pressure boundary conditions (e.g., the 1D consolidation problem, the upper bound is set to be drained with 0 Pa), I wonder how to reproduce such conditions in TOUGH3-FLAC3D coupling simulation. Could I just set the boundary to atmospheric pressure (1e5 Pa) and then subtract the atmospheric pressure from the final result?
Thanks advance fou your time :)
-
-
I guess that if geomechanical properties are not function of P, and system changes are dependent on P changes only, you my probably shift the 0 P boundary to atmospheric P.
I'm not a geomechanical expert, so better to ask for some additional advice.
Alfredo
