Effective permeability of an immiscible fluid in porous media determined from its geometric state

Abstract

Based on Darcy's law, two-fluid flow is dependent on a relative permeability function of saturation only that is process/path dependent with an underlying dependency on pore structure and wettability. We develop an Artificial Neural Network (ANN) that relies on fundamental geometrical measures to determine relative permeability. The developed ANN is based on a prescribed set of state variables based on physical insights that predicts the relative permeability of 4,500 unseen pore-scale geometrical states with ${R}^{2}$ = 0.98.