The porosity difference between PNM and Porespy

[morendong]

Hello, everyone,
I used blobs in Porespy to generate images with one porosity,
and extracted the pore network through op.io.network_from_porosity, and set the geometry by geometry.spheres_and_cylinders, finally calculated the porosity by pore_volumn and throat_vlumn.
Compared to the two values of porosity, the porosity of PNM is less than that of Porespy obviously.
How to make two values to be close.
Thanks in advance.

[jgostick]

Hi @morendong, I know you use porespy and openpnm a lot, so I'm sorry if you already know about thse examples, but as a first pass, have you looked at:

this
this
and this

They explain lots of ways you can adjust the porosity of the pore network.

[morendong]

Thanks for your reply.
https://openpnm.org/examples/applications/porosity.html
I do as the example, the porosity is consistent, while I adjust the pore and throat as spheres_and_cylinders, which show "'spheres_and_cylinders' can only be applied when throat diameter is smaller than that of adjacent pores."

[jgostick]

The spheres_and_cylinders model has a mathematical infinite if the spheres are too close. You can use the cones_and_cylinders model, or the hybrid_spheres_and_cylinders.