The pre-trained ANN to predict effective permeability

Published: 29 July 2025| Version 1 | DOI: 10.17632/2j9xjf3nzh.1
Contributors:
, Mohammad Ebadi

Description

This MATLAB file contains a pre-trained artificial neural network used to predict relative permeability in porous media based on pore-scale fluid configurations. The network architecture includes four hidden layers with 50 neurons each and was trained on 22,556 fluid configurations generated from lattice Boltzmann simulations of a North Sea sandstone under 36 wettability conditions.

Files

Institutions

  • University of New South Wales

Categories

Engineering, Permeability, Wettability, Multiphase Flow in Porous Medium

Funders

Licence