yade-users team mailing list archive

[Luc Scholtès <question691351@xxxxxxxxxxxxxxxxxxxxx>]

Question #691351 on Yade changed:
https://answers.launchpad.net/yade/+question/691351

Luc Scholtès posted a new comment:
I think I understand the problem.

You want to measure the permeability of a fracture contained inside a
"rock" sample and, for that, you want to relate the flux inside the
fracture to the pressure difference along its extension. Am I right?

Let me know if I got it right:

1 - you set up a cubic sample cut by a persistent fracture plane along the Y direction
2 - you use the blockStuff() function to allow flow only in the cells that are cut by the fracture, all the other cells are thus "impermeable"
3 - you apply the pressure difference on the ymin and ymax boundaries of the cubic sample to generate flow inside the fracture only (you use the no flux condition on all the other boundaries xmin, xmax, zmin, zmax).
4 - you can see the flow along the fracture plane (confirm with the visualization of the vtk file in Paraview) 
5 - when you measure Qin and Qout with the getBoundaryFlux() function on the ymin and ymax boundaries they are not equal

If all points above are right, your problem is the difference you get
between Qin and Qout. I think I (i.e., someone I worked with)
experienced such problem before and It may comes from the
getBoundaryFlux() function and the way it works. In your case, you
should have flow only in the non blocked cells contained inside the
sample but I don't know on which cells the getBoundaryFlux() function
measure the flux (external/virtual cells or cells inside the sample?).
Bruno may help with that, otherwise you'll have to read the source code.

To see if the problem is coming from the getBoundaryFlux() function, you
could check the flow velocity (or the flux) at different points located
inside the fracture. If the velocity is the same everywhere, you could
even use it to compute the permeability of your fracture without using
the Qin and Qout.

BTW, I am not sure you can use this

abs(Qout)*flow.viscosity*Y/(X*Z)

to measure the permeability of the fracture since the flow does not go
through the entire surface X*Z of the sample.

Instead, you should look into the cubic law [1] and use something like
that

Q=(deltaP*Width*h**3)/(12*mu*Length)

to measure the hydraulic aperture h of the fracture before computing its
permeability by

permeability=(h**2)/12

Luc

---

[1]
https://inis.iaea.org/collection/NCLCollectionStore/_Public/26/034/26034259.pdf

-- 
You received this question notification because your team yade-users is
an answer contact for Yade.