A General Framework for Predicting Permeability in Porous Structures Using Convolutional Neural Networks with Error Estimation
Published in Transport in Porous Media, 2025
This is an original publication, which goes into detail into permeability prediction of porous media using a convolutional neural network. This manuscript delves deep into topics such as data pre-processing, CNN architectures, and dataset diversity, trying to establish explainable links between design choices and CNN accuracy.
For more details, check out the publication. The original code is available on GitHub. There is also a dataset related to this publication, which is available on Mendeley Data.
Recommended citation: Adam, A., Stallard, S.L., Fang, H. et al. A General Framework for Predicting Permeability in Porous Structures Using Convolutional Neural Networks with Error Estimation. Transp Porous Med 152, 100 (2025). https://doi.org/10.1007/s11242-025-02239-4
Download Paper