Pore‐Scale Reconstruction and Simulation of Non‐Darcy Flow in Synthetic Porous Rocks

Abstract

We image synthetic porous rocks of varied porosity and pore size by micro–computed tomography with pore‐scale finite element modeling representing the pore space for single‐phase fluid flow. The simulations quantify the key features of microscale flow behavior in the synthetic cores. The smaller the permeability, the greater the critical pressure gradient required for the onset of non‐Darcy fluid flows, and the easier the emergence of nonlinear seepage within the tested cores. The relationship between permeability and porosity from different methods shows a power law correlation with pressure. Structural heterogeneity and anisotropy of the pore systems are shown to have a significant impact on transport through the three‐dimensional pore model—exhibiting irregular flow fields. The simulated permeabilities of the tested cores vary by a factor of 2–5 depending on the fluid flow directions. With the increase of flow seepage velocity, the flow regime deviates from Darcy linear relationship and non‐Darcy behavior (inertia) leads to a reduction in the effective permeability of the core. Both experiments and modeling demonstrate that the larger the porosity and permeability, the stronger the non‐linear phenomenon of the seepage within the pore space. A method is proposed to estimate the non‐Darcy coefficient β based on simulation results, which provides a good prediction for all the tested cores. A new equation is established to predict the transition of flow patterns as a function of the apparent permeability K* and Reynolds number Re. The K*‐Re model provides a theoretical approach to dynamically describe the transition from Darcy to Forchheimer flow.

Equivalent Pore Channel Model for Fluid Flow in Rock Based on Microscale X-ray CT Imaging

“…However, in DNS, if the computational domain was chosen with a 4 mm domain length, which was the target area for construction of the pore channel model as shown in Figure 2, it required a considerable working time of more than six hours and a large computational memory. These difficulties have also been encountered in previous studies [1,8]. It was therefore necessary to designate the ROI as a sufficiently small area (less than 2 mm 2 in this study), which in turn led to a problem of dependence on the chosen region.…”

“…Through the pore geometry after being meshed, the flow characterization yield insights into the effect of pore morphology on both microscopic and macroscopic flow properties [4][5][6]. Key features of the fluid velocity and pressure field in mesh domain have been examined by the finite volume method [7,8] or finite element method [9]. In general, these direct pore-scale models should represent the microstructures of porous media and accurately reflect the original images.…”

A massively multi-scale approach to characterising tissue architecture by synchrotron micro-CT applied to the human placenta

“…To evaluate the size and connectivity of the porous region (maternal intervillous space), the 'Separate objects' module (Avizo 2020.1) that uses the watershed method (61,62) was employed on the labelled 3D image volume to separate the porous regions ( Figure 2F).…”

Estimation of the Non‐Darcy Coefficient Using Supercritical CO2 and Various Sandstones

“…Pore-scale simulation of non-Darcy flow in porous rock has been made possible as a result of improvements in computed tomography technology. After a pore structure model is constructed through image processing and mesh generation, non-Darcy flow behavior can be implemented by various numerical methods (Bird et al, 2014;Muljadi et al, 2016;Wang et al, 2018;Zhao et al, 2018). Some research in this field has considered the impact of fluid properties on the non-Darcy coefficient.…”