Search

Realizable CO2 storage potential for saline formations without closed lateral boundaries depends on the combined effects of physical and chemical trapping mechanisms to prevent long-term migration out of the defined storage area. One such mechanism is the topography of the caprock surface, which may retain CO2 in structural pockets along the migration path. Past theoretical and modeling studies suggest that even traps too small to be accurately described by seismic data may play a significant role. In this study, we use real but scarce seismic data from the Gassum Formation of the Norwegian Continental shelf to estimate the impact of topographical features of the top seal in limiting CO2 migration. We seek to estimate the amount of macro- and sub-scale trapping potential of the formation based on a few dozen interpreted 2D seismic lines and identified faults. We generate multiple high-resolution realizations of the top surface, constructed to be faithful to both large-scale topography and small-scale statistical properties. The structural trapping and plume retardation potential of these top surfaces is subsequently estimated using spill-point (static) analysis and dynamical flow simulation. By applying these techniques on a large ensemble of top surface realizations generated using a combination of stochastic realizations and systematic variation of key model parameters, we explore the range of possible impacts on plume advancement, physical trapping and migration direction. The stochastic analysis of trapping capacity and retardation efficiency in statistically generated, sub-seismic resolution features may also be applied for surfaces generated from 3D data. [ABSTRACT FROM AUTHOR]

Fluid mechanics simulation of steady state flow in complex geometries has many applications, from the micro-scale (cell membranes, filters, rocks) to macro-scale (groundwater, hydrocarbon reservoirs, and geothermal) and beyond. Direct simulation of steady state flow in such porous media requires significant computational resources to solve within reasonable timeframes. This study outlines an integrated method combining predictions of fluid flow (fast, limited accuracy) with direct flow simulation (slow, high accuracy) is outlined that reduces computation time by an order of magnitude without loss of accuracy. A convolutional neural network (CNNs) is trained with various configurations on simulations in 2D and 3D porous media to estimate steady state velocity fields. Permeability estimation (as an average of the field) is accurate, but the velocity fields themselves are error prone, unsuitable for further transport studies. This estimate can either be used as an indicative prediction, or as initial conditions in direct simulation to reach a fully accurate result in a fraction of the compute time. Using Deep Learning predictions (or potentially any other approximation method) to accelerate flow simulation to steady state in complex structures shows promise as a technique to push the boundaries fluid flow modelling. Article Highlights: Steady State velocity fields predicted in 2D and 3D using CNNs Permeability estimation with predicted fields over 95% accurate in most cases Fine scale velocity field prediction is error-prone, limited by CNN performance Fast, low accuracy CNN prediction is combined with slow, high accuracy simulation Accelerated technique produces fully accurate results in 10x less time [ABSTRACT FROM AUTHOR]

The lattice Boltzmann method is increasingly utilized in the simulation of flow-induced wall shear stress needed in various applications. In image-based flow simulations, the simulation geometry is usually based on a three-dimensional reconstruction of the true structure of the pore space obtained, for example, by X-ray tomography. The geometry is then given in a voxel-based representation, which complicates an accurate determination of the surface-normal vectors that are necessary in the computation of the wall shear stress. To avoid this problem, we introduce here a method for the determination of surface-normal vectors directly from a greyscale image instead of its segmented binary image version. The proposed method is fast and automatic, and it can be used for an arbitrary pore space geometry provided in a greyscale form by any imaging modality. We show that this method can produce accurate surface-normal vectors even for binary images and that their accuracy is further increased when the original greyscale images are used instead. We compute wall shear stresses for generated benchmark geometries and then demonstrate the utility of the method for soil samples with ‘random’ pores imaged by X-ray tomography. [ABSTRACT FROM AUTHOR]

The prediction of the free-surface seepage flow behavior in fractured rock mass is of significance in geotechnical engineering. There are two major issues in solving the seepage flow in complicated fractured rock mass based on the fractured porous medium (FPM) flow model, in which groundwater is assumed to flow simultaneously in both rock matrix and embedded fractures: One is the mesh generation of rock mass in the presence of the fracture network, especially when there exist a large number of stochastic fractures; the other is that a robust iteration algorithm is required since the free surface is unknown at the beginning of solution. Aiming at these two issues, this paper proposes a novel numerical method by coupling radial point interpolation method (RPIM) and finite element method (FEM), in which RPIM is utilized to model the rock matrix and FEM is utilized to model the fractures. On the basis of the variational inequality (VI) theory for free-surface seepage analysis, the computation formulations of the numerical method are derived and the corresponding computation program is developed. Three examples are solved with the present method. It is found that the VI theory can be extended to solve the free-surface seepage problem based on the FPM flow model. A crucial advantage of the present method is that the mesh generation can be greatly simplified. The present method has been verified to be a robust, efficient and reliable method for modeling the groundwater flow in complicated fractured rock mass. [ABSTRACT FROM AUTHOR]