Import CSV or VTK file of pore body and throat in openPNM

[Soheil1993]

Dear Dr. @jgostick

I extracted the pore body data in both CSV (x,y,z,r) and VTK formats and pore throat data in both CSV (pore id1,pore id2,r) and VTK formats from Yade. Now I want to use openPNM for obtaining the permeability of the media.

Would you please let me know how I should modify the files and import them into the openPNM?

Should I combine them or just import the pore throat file for permeability?

Best regards,
Soheil

[ma-sadeghi]

You can use openpnm.io.from_csv but that requires you to format your CSV in a special way. There's another option, which is to create an empty network and add your data manually. Here's a pseudo-script (you need to probably modify it since I didn't test it):

import openpnm as op

net = op.network.GenericNetwork(coords=coords, conns=conns)
net["pore.diameter"] = Dp
net["throat.diameter"] = Dt

where coords is the xyz coordinates of your pores (Np by 3 array), conns is pore connectivity (i.e., which pore is connected to which -> Nt by 2 array), Dp is your pore diameter array, and Dt is throat diameter. Once you get this working, you need to add some more models to your network object to calculate other properties like throat length, etc. If you look into our examples, you'll see which models are required to estimate permeability.

[Soheil1993]

Dear Dr. @ma-sadeghi

Thank you very much for your kind reply and valuable information.

I am a bit new to Python. Would you please give me more information in detail? Is there any similar example to follow?
Should I create all of these files separately before importing? In which format?
And how can I import them?

Many thanks in advance.

Best regards,
Soheil

[ma-sadeghi]

Well, Dp and Dt should be numpy arrays. You've mentioned that you have pore radii stored as CSV. You can use pandas.read_csv (see here for docs) to read your CSV file into a bunch of numpy arrays stored column-wise. You can google how to use pandas.read_csv, and you can find tons of examples.

[Soheil1993]

Dear Dr. @ma-sadeghi

Thank you very much for your kind reply and useful link.
I really appreciate your comprehensive description.

Cheers,
Soheil