Dynamic Pore Network Modeling Of Two Phase Flow And Solute Transport In Disordered Porous Media And Rough Walled Fractures

Dynamic Pore Network Modeling of Two-phase Flow and Solute Transport in Disordered Porous Media and Rough-walled Fractures

Reliable, physics-based predictions of two-phase flow and solute transport properties (e.g., relative permeability and dispersion coefficient) of porous media and fractures are crucial for reducing the uncertainty in performance predictions of many subsurface applications such as carbon dioxide (CO2) sequestration, enhanced oil recovery, and remediation of contaminated groundwater. However, problems of multi-phase flow in disordered porous media are exceedingly difficult due to, for instance, complicated pore space geometries and topologies, wettability conditions, and fluid–fluid displacements. Furthermore, even though the majority of subsurface reservoirs have natural or induced fractures, there is still no consensus regarding two-phase fluid flow behaviors in rough-walled fractures because of the intricate aperture-field geometries and their impacts on fluid–fluid interactions. Therefore, the development of physics-based models that can accurately predict flow and transport properties of disordered porous media and rough-walled fractures is warranted. In this study, we developed two new heavily-parallelized, dynamic pore-scale modeling platforms that can better account for the underlying displacement and transport physics and thereby predict the corresponding macroscopic properties. Both modeling frameworks incorporate many advanced features that have not been collectively used in previous dynamic models. For instance, they rigorously solve for the capillary, viscous, and buoyancy forces, integrate detailed descriptions of pore-scale fluid displacement dynamics, and explicitly account for flow through wetting layers forming in corners and rough surfaces of pore spaces. These modeling platforms further benefit from heavy parallelization and advanced domain decomposition techniques to achieve computational efficiency. We utilize the dynamic models to carry out various two-phase displacement simulations in core-sized pore networks constructed from high-resolution x-ray images of sandstone rock samples and a Berea sandstone fracture. Measured two-phase relative permeability data and fluid occupancy maps for both primary drainage and imbibition displacements are employed to validate our network models and probe their predictive capabilities. Afterward, the validated models are utilized to investigate a series of critical two-phase flow and solute transport properties including fluid trapping behaviors in porous media and rough-walled fractures under a broad range of fluid properties and flow conditions, relative permeability curves of single fractures with different geometric features, and dispersive solute transport behaviors in a sandstone rock sample under both single- and two-phase fluid configurations. These findings are of fundamental importance to understanding the physics governing fluid flow and solute transport in subsurface systems as well as field-scale predictions of two-phase flow in porous media and fractures.

Direct Pore-level Modeling of Fluid Flow in Porous Media

A three-dimensional fully parallel particle-based model for direct pore-level simulation of incompressible viscous fluid flow in disordered porous media is presented. The model is capable of simulating flow in porous media, taking directly as input three-dimensional high-resolution microtomography images of naturally-occurring or man-made porous systems. These images provide the most faithful representations of the pore space in the porous medium. In this method, the entire medium, i.e., solid and fluid, is discretized using particles. The model is based on the Moving Particle Semi-implicit (MPS) technique and modified to improve its stability. The model handles highly irregular fluid-solid boundaries effectively, accounts for viscous pressure drop in addition to gravity forces, conserves mass, and can automatically detect any false connectivities with fluid particles in the neighboring pores and throats. The model also includes a sophisticated algorithm to automatically split and merge fluid particles to maintain hydraulic connectivity of extremely narrow conduits. Finally, it uses novel methods to handle particle inconsistencies and open boundaries. To handle the computational load, a fully parallel version of the model is presented that runs on distributed memory computer clusters and exhibits excellent scalability. The accuracy and reliability of the model in predicting the true flow in porous systems was validated rigorously against analytical, numerical, and experimental data available in the literature. The validated model was then used to simulate flow in naturally-occurring sandstones to compute their absolute permeabilities and their variations with sample size and flow direction. The model was extended to handle the pore-level transport of solutes in random porous media. Two different sandstones possessing different topologies were used to investigate pre-asymptotic and asymptotic longitudinal dispersion coefficients, the impact of pore-scale topologies on the transient behavior of convective-diffusive solute transport at the pore-level, and the impact of inertial forces on dispersion coefficients at very high Peclet numbers. The asymptotic dispersion coefficients were then successfully compared against the experimental data available in the literature for a wide range of Peclet number. Furthermore, a solute adsorption module was developed and integrated with the model allowing us to study the adsorptive-diffusive-convective solute transport in a rough-walled single fracture. These studies provide new insights into the pore-scale behavior of fluid flow and solute transport in random porous media. The model serves as a reliable platform for studies of different pore-level phenomena that can improve our understanding of the pore-level physics of flow and transport.

Dynamic Pore Level Modeling of Two-phase Flow Through Porous Media