Pore-scale modeling of multiphase flow in porous media using a conditional generative adversarial network (cGAN)

Multiphase flow in porous media is involved various natural and industrial applications, including water infiltration into soils, carbon geosequestration, underground hydrogen storage. Understanding the invasion morphology at pore scale critical for better prediction of properties continuum partially saturated permeable media. The deep learning method, as a promising technique to estimate transport processes media, has gained significant attention. However, existing works have mainly focused on single-phase flow, whereas capability data-driven techniques yet be applied pore-scale modeling fluid–fluid displacement Here, conditional generative adversarial network multiphase two-dimensional trained based data set generated using particle-deposition with corresponding morphologies after calculated recently developed interface tracking algorithm. results demonstrate predicting both fluid saturation spatial distribution. It also shown that method can generalized distribution under different wetting conditions particle shapes. This work represents first effort application immiscible highlights strength surrogate