-
Hi all, Firstly, thank you for creating such an amazing module - I am in need of some advice. I have an extracted and labelled network. I am wanting to identify clusters of pores and throats that are continually connected from one side of a domain, say the 'left', all the way through the thickness of the network, to another domain, say the 'right', i.e. through the scaffold in an x direction, or y or z. I would then like to trim all other pore clusters that are not connected through the entire domain in the specified direction. I note that in the percolation example, csg.connected_components from csgraph is almost used to this affect, however, I cannot figure out how I may apply this to a network object. Does anyone by chance have any advice around this? I would sincerely appreciate it. |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment
-
Hi @andtoni Also, maybe this example will be helpful: https://openpnm.org/examples/reference/networks/managing_clustered_networks.html |
Beta Was this translation helpful? Give feedback.
Hi @andtoni
We actually have a function which does exactly this:
remove_isolated_clusters
. Just useinlets
as the boolean sum of both inlet and outlet faces.Also, maybe this example will be helpful: https://openpnm.org/examples/reference/networks/managing_clustered_networks.html