Search

At present, researchers predict permeability through core experiments that require specific experimental conditions and methods, which are difficult and time-consuming. Conventional simulation methods for predicting permeability require considerable computational resources. Therefore, deep learning can be used as a pore-scale simulation prediction method. In this study, we established a workflow for directly predicting permeability from images. Considering that the mineral properties of the nanopore wall of shale oil have a large influence on the flow, a core dataset with organic distribution was constructed with random circles, and the slip influence of organic pores was considered. From our dataset, we found that the average permeability with organic distribution was 32.3% higher than that without organic distribution. Therefore, to simulate the microscopic flow and predict the permeability of shale oil, considering the differences in the pore flow mechanisms of different minerals is necessary. We designed a convolutional network for the dataset, adopted the structure of SE-ResNet, added the squeeze-and-excitation (SE) module to the double-layer residual module of ResNet18, and combined the characteristics of the SE block with the attention mechanism and ResNet to effectively obtain the information between channels and avoid the problem of gradient disappearance or explosion. Using SE-ResNet for directly predicting the apparent permeability from images, the accuracy of the test set reached 88.5%. The model had strong generalization ability, and the SE-ResNet could map the image of the core to the apparent permeability, which was approximately 100 times faster than the direct flow simulation. [ABSTRACT FROM AUTHOR]

Accurate pore-scale modeling demands high-quality digital rock images, which should possess a broad imaging field of view (FOV) and high resolution to characterize multi-scale rock components. However, achieving both conditions simultaneously is challenging due to hardware constraints. Super-resolution techniques can mitigate this issue by reconstructing high-resolution details from low-resolution images captured with a wide FOV. To reconstruct high-quality 3D digital rocks, we propose a novel Efficient Attention Super-Resolution Transformer (EAST) model. It integrates self-attention and channel attention mechanisms and undergoes structural optimization. Evaluation demonstrates that EAST achieves superior reconstruction quality with a 1.85 × speedup while reducing parameters by 78 % over the advanced model. Additionally, to tailor to the characteristics of digital rocks and the constrained dataset, we employ a hybrid loss function along with two data augmentation techniques. Visualizations reveal that EAST can resist noise and blur interference, highlighting valuable features such as pore edges and textures. Ultimately, we introduce an approach based on self-supervised fine-tuning to enhance the model robustness. Direct flow simulation verifies that the reconstructed results closely align with high-resolution images in terms of physical accuracy. EAST reduces the relative error of the absolute permeability by 18.5 % and 33 % over the Tricubic method on two external samples. • The novel Transformer super-resolution model has enhanced digital rocks quality. • Our model is efficient, lightweight and outperforms advanced models. • Our model effectively resists blurring and noise in digital rocks. • A self-supervised fine-tuning method has been introduced to boost model robustness. • Spatial statistics, topology, and permeability have been accurately measured. [ABSTRACT FROM AUTHOR]

Fractal characteristic has been considered as the essential parameter to determine transport properties of porous media. Fluid flow in the real porous media is more influenced by the narrow throat system than the relatively large pore system. In this work, the concept of throat fractal dimension is proposed and applied to predict permeability in porous media. First, pore system and throat system of six sandstone rock samples and three carbonate rock samples are divided from the pore network models extracted from rock CT images. Then, fractal dimensions based on the total pore space, pore system and throat system are calculated, respectively, based on the three-dimensional (3D) box-counting method. Three groups of permeabilities are estimated, respectively, using the fractal theory based on the pore space fractal parameter, pore system fractal parameter and the proposed throat system fractal parameter. The results indicate that permeabilities estimated by the proposed throat system fractal parameter are closer to the accurate permeabilities of rock samples compared with that estimated by the pore space fractal parameter and pore system fractal parameter. Our study results reveal that throat fractal dimension should be used to accurately predict fluid transport properties in porous media. [ABSTRACT FROM AUTHOR]

• A DAGP model is proposed based on shale geostatistical and petrophysical data. • A representative model is constructed by fractal organic matter particles. • DAGP exhibits different patterns, depending on a high or low matrix Young's modulus. • The impact of OM pores development degree on DAGP can be ignored at TOC < 5%. Current shale gas apparent permeability models often fail to integrate morphological properties of multiple storage spaces and various transport mechanisms. To deal with this, we propose a comprehensive model for estimating the dynamic apparent gas permeability (DAGP) during depressurization based on a representative model. First, dual fractal characteristics of organic matter (OM) particles and inorganic matter (IOM)/OM pores along with morphological information of OM particles from scanning electron microscope (SEM) images are considered to establish the representative model conditioned to a total organic carbon (TOC) content. Next, multiple transport mechanisms and a pore size change due to a poromechanical response and desorption-induced volumetric strain are incorporated in the DAGP calculation. Results reveal that the controlling factor of the DAGP comes from pore size shrinkage owing to a poromechanical response to gas transport mechanisms during depressurization at a high Young's modulus, while the DAGP is totally controlled by a poromechanical response at a low Young's modulus. The critical TOC content is determined to be about 5% at which the influence of the degree of OM pore development can be neglected. This work provides a practical method to estimate the DAGP from both geostatistical and petrophysical perspectives, which effectively decreases the uncertainty during productivity prediction in shale gas reservoirs and other unconventional reservoirs. [ABSTRACT FROM AUTHOR]

Due to the complex conditions of carbonate reservoir, in this paper, a unified lattice Boltzmann method was used to study the rule of the flow in carbonate reservoir as the foundation. Two group models were designed to simulate the influences of fractures and vugs. In experiments, at first, the model for carbonate reservoir considering different amounts and lengths of fractures was considered. Then, the model was improved by taking the influences of fractures and vugs into consideration. The result from the first group shows that the whole permeability enhanced a lot when multiple fractures form a big one which connects to two boundaries. At the same time, the main result of the other group shows that the main capability of vugs is flow accumulation. Through a series of experiments, the flow rule in carbonate reservoir with vugs and fractures is proved based on LBM theory, which has a huge impact on the study in LBM and researches on carbonate reservoir. [ABSTRACT FROM AUTHOR]

• The region of heterogeneous formation first touched by downward CO 2 fingers plays a dominant role in convective mixing. • For heterogeneous porous media, the relationship between Sh and Ra is not power-law. • High permeability reverse rhythm reservoir which have higher density-driven CO 2 convection rate is suitable for CO 2 storage. Underground storage of CO 2 in saline aquifers has emerged as a promising strategy for mitigating global warming. The dissolution of CO 2 in saline aquifers induces the descent of high-density CO 2 -saturated brine to the reservoir bottom, driven by gravity-induced natural convection. The manifestation of this convection is linked to the heterogeneity of the reservoir. We thus analyzed the impact of heterogeneous permeability distribution, which represent various geological structures, on the CO 2 plume morphology and CO 2 storage efficiency. In a homogeneous high-permeability core, high convection rate resulting in wider CO 2 fingers. For heterogeneous cores, the convective mixing rate depends on average permeability, while the morphology of fingers is determined by local permeability. Abrupt permeability changes (from low permeability to high permeability) led to CO 2 finger channeling, resulting in reduced sweep efficiency. In heterogeneous layered core, the permeability of the upper layer dominated convection, while vertically decreasing permeability (reversed-rhythm) was more conducive to the dissolution capture of CO 2. However, increasing the number of rhythm layer, the larger permeability distribution range and the number of vertical bedding layers enhanced the convective instability, which was unconducive to efficient CO 2 sequestration. Besides, a power relationship between the Rayleigh number and the Sherwood number was identified. The power-law exponent b = 1.07 was higher than the value (b ≈0.984) obtained by two-dimensional plate experiments. In conclusion, reversed-rhythm reservoirs with high permeability exhibit a preference for CO 2 dissolution trapping. Our study provides detailed insights into the processes of CO 2 dissolution for heterogenous reservoir. The results should further facilitate the extensive implementation of Carbon Capture and Storage projects. [ABSTRACT FROM AUTHOR]

• Coupled porous media and free flow model of fluid flow in fractured karst reservoirs. • Analytical expression of connectivity for discrete fracture-cave networks. • Flow channeling dominated by network connectivity. • Permeability scaling with connectivity in a power law for connected systems. Fractured karst reservoirs play a significant role in hydrocarbon reserves and groundwater storage. They often exhibit complex multiscale heterogeneities involving pores, fractures and caves, whose length scales range from microns to tens or hundreds of meters. Thus, the study of fractured karst reservoirs is faced with a significant unresolved challenge in quantitatively characterizing the geometrical connectivity and hydraulic conductivity as well as their interrelationships in such strongly heterogeneous, multicomponent systems. In this paper, we propose an analytical formulation to characterize the connectivity of discrete fracture-cave networks building upon the excluded area concept of the percolation theory. By implementing a state-of-the-art computational model solving coupled Navier-Stokes (free flow) and Darcy (porous media flow) equations, we numerically derive the permeability of a fractured and karstified porous media, such that the relationship between the connectivity and permeability is further explored. The high-fidelity numerical model also permits us to elucidate the process of flow channelization within the fracture-cave network. The results show that the fracture-cave network connectivity correlates to the permeability via a power law scaling for connected systems. Significant flow channeling occurs around the percolation threshold where the flow is dominated by a limited number of preferential pathways. Caves play a crucial role in the flow due to the fact that caves could globally enhance the network connectivity and locally serve as hotspots for high fluid velocity. Finally, a semi-analytical permeability model for fractured karst reservoirs is developed. The results of our research and insights obtained have important implications for understanding the subsurface fluid flow in fractured karst reservoirs. [ABSTRACT FROM AUTHOR]

Reservoir modeling data could be divided into two categories: spatial data (i.e. permeability, effective grids, crack, irregular boundaries) and vector data (fluid properties i.e. relative permeability, density, viscosity). This paper is mainly interested in considering the permeability and relative permeability, which are the representatives of the two data types, to construct a proxy model for forecasting saturation and pressure maps in heterogeneous reservoirs during water flooding. The proxy model is built on the dense encoder-decoder network, to learn the reservoir dynamic states at different time steps. Results indicate that the trained proxy model could predict fluid saturation and pressure fields accurately. This paper presents a calibration method, which is adding a constraint to well-blocks. After calibration, the trained proxy model is utilized to calculate reservoir production. The comparison results illustrate that the proxy model can forecast well rates with relatively high accuracy. Compared with traditional reservoir numerical simulators, the proxy model could predict fluid saturation, pressure and well rates with similar accuracy and less time-cost. • This paper constructs a proxy model for forecasting saturation and pressure maps in heterogeneous reservoirs. • The paper considers permeability and relative permeability, which are the representatives of spatial data and vector data. • This paper presents a calibration method, which is adding a constraint to well-blocks. [ABSTRACT FROM AUTHOR]

Water flooding is the most widely used method to enhance oil recovery. Simulating the influence of particle content on fluid flow in water flooding process can provide instructive significance for oilfield development. In this paper, a pore network model was established to simulate the transport process of particles in injected water and fallen particles from the reservoir. It was used to study the influence of particle blockage, deposition and migration on the clogged throats and changes of oil-water relative permeability curves. In the model, four parameters were set, fallen particle diameter, injected particle diameter, fallen particle number and injected particle number. Other parameters were fixed and only a variable was changed to explore the changeable rule of relative permeability curves. It can be generally noted that fallen particle changes have great influence on the relative permeability curves of oil phase and injected particle changes mainly affect the relative permeability curves of water phase. It is of great significance for oilfield development to study the influence of particle transportation on fluid seepage processes. • A pore network model containing particles was established. • A transport process of injected particles and fallen particles from the reservoir during water flooding was simulated. • The influence of particle blockage, deposition and migration on the clogged throats and oil-water relative permeability curves was studied. [ABSTRACT FROM AUTHOR]