Digital Core Research Articles

Packing Soft Polygons in a Minimum Height Rectangular Target Domain

The paper studies packing polygons of variable shapes, regarding the stretching coefficient, in a rectangular target domain of minimum height. Packing objects of a variable shape have a wide spectrum of applications, e.g, in biology, materials science, mechanics, land allocation, and logistics. Interest in these problems is also due to the modeling of the structures of porous media under pressure, e.g., for creating test models of artificial digital cores. Elements of porous media can be deformed under the influence of an external force, but the mass of each particle remains unchanged. This corresponds to conservation of area for the two-dimensional case. Polygonal objects must be completely contained within the target domain (containment constraint) and do not overlap (non-overlapping constraint), provided they have free translations, continuous rotations, stretch transformations, and conserve their area. The phi-function technique is used for an analytical description of the placement constraints for variable shape polygons. Quasi-phi-functions for describing non-overlapping constraints and phi-functions for describing containment constraints are defined. The packing problem is presented in the form of a nonlinear programming model. A solution strategy is proposed, which consists of the following stages: generation of feasible starting points; search for local minima of the problem of packing soft polygons for each starting point using the decomposition algorithm; choosing the best local minimum found at the previous stage. To search for smart starting arrangements, an optimization algorithm for packing original polygons using their homothetic transformations is applied. Decomposition of the problem of packing polygons of variable shapes is based on an iterative procedure that allows reducing a large-scale problem to a sequence of smaller nonlinear programming problems (linear to the number of objects). Numerical examples are provided for oriented rectangles and non-oriented regular polygons. Keywords: packing, polygons, stretch transformation, rectangular container, quasi-phi-functions, optimization, decomposition algorithm.

Super-resolution reconstruction of 3D digital rocks by deep neural networks

Digital rock technology provides valuable insights into the pore structure and fluid flow properties of geoenergy resources. Artificial intelligence technology in vision and image processing, especially the image super-resolution, has great potential for digital rock reconstruction and resolution enhancement. However, the analyzed core samples are typically sandstones/carbonates in micro-scale resolutions and in two-dimensional (2D) space, whereas the shale rocks in nano-scale resolutions for unconventional resources or three-dimensional (3D) digital cores are rarely investigated. Additionally, previous studies primarily emphasized image quality from a computer vision perspective, with little consideration given to estimating physical properties of digital rocks using super-resolution techniques.This study presents a very deep super-resolution (VDSR) algorithm, specifically designed to generate high-resolution 3D digital rock images, for nano-scale shale matrix and micro-scale hydraulic fractures. We compare both image quality and permeability accuracy between the original high-resolution images and the super-resolution images reconstructed by the proposed method. The results reveal that the reconstructed images using the proposed method closely resemble the actual images, and effectively reduce errors in permeability computations. This study highlights the applicability of the proposed VDSR algorithm in establishing the detailed structures of 3D nano-scale shale matrix and hydraulic fractured rocks, thus advancing super-resolution techniques in digital core analysis for geoenergy resources development.

OmniSR-M: A Rock Sheet with a Multi-Branch Structure Image Super-Resolution Lightweight Method

With the rapid development of digital core technology, the acquisition of high-resolution rock thin section images has become crucial. Due to the limitation of optical principles, thin section imaging involves a contradiction between resolution and field of view. In order to solve this problem, this paper proposes a lightweight, fully aggregated network with multi-branch structure for super resolution of rock thin section images. The experimental results on the rock thin section dataset demonstrate that the improved method, called OmniSR-M, achieves significant enhancement compared to the original OmniSR method and also surpasses other state-of-the-art methods. OmniSR-M effectively recovers image details while maintaining its lightweight nature. Specifically, OmniSR-M reduces the number of parameters by 26.56% and the computation by 27.66% compared to OmniSR. Moreover, this paper quantitatively analyzes both the facies porosity rate and grain size features in the application scenario. The results show that the images generated by OmniSR-M successfully recover key information about the rock thin section.

Application of new axial power distribution synthesis method using in-core detector signal in digital core protection system

ABSTRACT A new method for the generation of axial power distributions using artificial neural network (ANN) technique and real-time measured planar radial peaking factor, Fxy (hereinafter ‘Live Fxy’) method for digital core protection system of pressurized water reactors is presented. ANN with Simulated Annealing (SA) technique to find global optimum solution was used to calculate core average axial power distribution. In Live Fxy method, axial node-wise Fxys were calculated by multiplying pin/box factors to the measured assembly powers inferred from in-core detector signals without calculating detailed 3D power distributions. The hot pin power distribution for DNBR and LPD was then obtained by multiplying the core average axial power distribution and Fxys at axial nodes. The validation of the method was performed for various core conditions (e.g. core power levels, control rod positions, etc.) of Korean OPR1000 power plant. The result showed a decrease in RMS errors by 2.2% ~ 7.0% (3.6% on average), and the minimum thermal margins for DNBR and LPD were increased by 6.4% and 15.6%, respectively. The application of the method would improve the accuracy of axial power distribution and thermal margin, contributing to the operability and safety of digital core protection system.

Characterization of pore systems in fine-grained carbonate rocks using digital core technology

Characterization of pore systems in fine-grained carbonate rocks using digital core technology

Unlocking the Potential of Digital Twins in Construction: A Systematic and Quantitative Review Using Text Mining

The construction industry has been trying to enhance the level of digitalization and autonomy by adopting various communication and information technologies (ICT), e.g., augmented reality (AR), virtual reality (VR), robotics, drones, or building information modeling (BIM). However, improvement of the safety and productivity in their domains is still a struggle. One of the main reasons for failing to accelerate their digital transformation is ignoring the deep understanding of the concept of digital twin, its usage, and the potential benefits of digital twins in the construction industry. Therefore, this paper investigated the impacts and potentials of digital twins on the construction industry through a quantitative systematic review assisted by the text mining method. The study presented the potential usability of digital twins, leading and core technologies, and applications, revealing their benefits and potential for optimizing project planning, execution, and management process. Through this comprehensive literature review, this study elucidated the distinctive features, advantages, and immense potential that digital twins bring to the construction field. The findings highlight the transformative impact of digital twins, providing critical insights for their broader adoption and groundbreaking applications in the industry. By addressing the challenges of adopting this technology, the article provided valuable insights for advancing research and the broad implementation of digital twins in the sector.

A Multiscale Approach for Assessing Shale Oil Availability: Digital Core, Molecular Simulation, and Machine Learning Analysis

Summary This study presents a novel multiscale approach for assessing the accessibility of shale oil in cores by use of focused ion beam-scanning electron microscopy (FIB-SEM) equipment to build digital core, watershed, and maximum ball methods to extract pore size and shape factor. Then, molecular simulation is used to study the availability of shale oil in individual pores with different shapes and radii. Finally, combining the results of the above two scales, machine learning is used to predict shale oil availability across the entire core. On the core scale, the watershed and maximum ball methods are used to extract the core pore network model, and it is found that square pores occupy the highest proportion among the three pore types, and most of the radii are distributed in the range of 2–3 nm. The molecular-scale dynamic simulation results show that the adsorption forms of shale oil are different in different pores, and the adsorption of shale oil in circular pores is less than that in flat pores. The proportion of shale oil adsorption in square pores is the highest, followed by triangular pores, and the proportion of shale oil adsorption in circular pores is the lowest. The random forest machine learning algorithm is used to predict the availability of shale oil with different pore shapes and obtain the shale oil availability ratio of the whole core. The results show that the pores with a more obvious angular structure show a lower shale oil availability ratio. In general, the impact of pore shapes on shale oil availability is not significant, and the difference between different pore shapes is only 10%. The multiscale evaluation method for shale oil availability proposed in this study is helpful to better understand the availability of shale oil in reservoirs and to optimize recovery strategies.

Leveraging Spatial Metadata in Machine Learning for Improved Objective Quantification of Geological Drill Core

AbstractHere we present a method for using the spatial x–y coordinate of an image cropped from the cylindrical surface of digital 3D drill core images and demonstrate how this spatial metadata can be used to improve unsupervised machine learning performance. This approach is applicable to any data set with known spatial context, however, here it is used to classify 400 m of drillcore imagery into 12 distinct classes reflecting the dominant rock types and alteration features in the core. We modified two unsupervised learning models to incorporate spatial metadata and an average improvement of 25% was achieved over equivalent models that did not utilize metadata. Our semi‐supervised workflow involves unsupervised network training followed by semi‐supervised clustering where a support vector machine uses a subset of M expert labeled images to assign a pseudolabel to the entire data set. Fine‐tuning of the best performing model showed an f1 (macro average) of 90%, and its classifications were used to estimate bulk fresh and altered rock abundance downhole. Validation against the same information gathered manually by experts when the core was recovered during the Oman Drilling Project revealed that our automatically generated data sets have a significant positive correlation (Pearson's r of 0.65–0.72) to the expert generated equivalent, demonstrating that valuable geological information can be generated automatically for 400 m of core with only ∼24 hr of domain expert effort.

Distribution of gas hydrate in fractured reservoirs: implications from anisotropic elastic and electrical numerical simulations

SUMMARY Fractured hydrate-bearing reservoirs are extensively discovered worldwide and show notable anisotropic geophysical properties. Hydrate distribution in fractures significantly affects the anisotropic properties, and hence plays an important role in the accurate assessment of hydrate resources. However, the knowledge about how the hydrate distribution impacts the anisotropic geophysical properties of fractured reservoirs, which is the premise for the identification and quantification of hydrate in fractured reservoirs, is still poorly understood. To obtain such knowledge, we forward study the effects of various hydrate distribution (including floating, bridging and evolving hydrate distribution) in aligned fractures on the anisotropic elastic, electrical and joint elastic–electrical properties of a digital core using validated numerical methods. We show that for all the hydrate distribution, the anisotropic velocities increase, while the conductivities decrease with increasing hydrate saturation, with the effects of the floating and bridging distribution being the least and greatest, respectively. We also show that the anisotropic velocities and conductivities for the floating and bridging distribution vary approximately linearly with hydrate saturation, leading to linear correlations between the elastic and electrical properties. Further investigation illustrates that the difference in the slopes of the linear joint correlations between the two distribution is significantly greater than that of the individual properties as a function of hydrate saturation. The results have revealed the distinct effects of hydrate distribution on the anisotropic elastic and electrical properties of fractured reservoirs, and have confirmed the superiority of the joint elastic–electrical properties for the distinguishment of hydrate distribution in fractures over individual physical properties.

An All-digital Compute-in-memory FPGA Architecture for Deep Learning Acceleration

Field Programmable Gate Array (FPGA) is a versatile and programmable hardware platform, which makes it a promising candidate for accelerating Deep Neural Networks (DNNs). However, FPGA’s computing energy efficiency is low due to the domination of energy consumption by interconnect data movement. In this article, we propose an all-digital Compute-in-memory FPGA architecture for deep learning acceleration. Furthermore, we present a bit-serial computing circuit of the Digital CIM core for accelerating vector-matrix multiplication (VMM) operations. A Network-CIM-deployer ( NCIMD ) is also developed to support automatic deployment and mapping of DNN networks. NCIMD provides a user-friendly API of DNN models in Caffe format. Meanwhile, we introduce a Weight-stationary dataflow and describe the method of mapping a single layer of the network to the CIM array in the architecture. We conduct experimental tests on the proposed FPGA architecture in the field of Deep Learning (DL), as well as in non-DL fields, using different architectural layouts and mapping strategies. We also compare the results with the conventional FPGA architecture. The experimental results show that compared to the conventional FPGA architecture, the energy efficiency can achieve a maximum speedup of 16.1×, while the latency can decrease up to 40% in our proposed CIM FPGA architecture.

Study on the influence of steam state on seepage, production, migration and deposition of offshore heavy oilfields

In this paper, the influence of steam state on seepage, production, migration and deposition are refined, and the superiority of superheated steam is confirmed. Firstly, the relative permeability of oil-water and oil-steam in offshore cores was measured to analyze flowing characteristics. Then the cyclic steam stimulation of long core was conducted with supersaturated, saturated and superheated steam, and the recovery capacity was quantified, the difference in composition of produced oil was analyzed. Finally, the CT scanning and digital core 3D reconstruction were performed to visualize the influence of steam state on migration of heavy oil components. The results show that the temperature has a significantly higher effect on seepage capacity than permeability, which can effectively broaden the co-seepage range and reduce residual oil saturation. The total recovery of superheated steam is 12 % and 7 % higher than the other two steam respectively, and the main recovery cycle is the former four rounds, which accounts for over 85 % of total recovery. The retention of heavy components can reduce porosity by 4 %∼7 %, superheated steam can alleviate blockage and increase the porosity by over 2 %. This work is expected to provide valuable insights into the application of superheated steam in offshore heavy oilfields.

The bitumen development effect on the pore structure and electrical resistivity of carbonate reservoirs in the Sinian Dengying Formation

Abstract In the Sichuan Basin, bitumen widely distributes in the carbonate reservoir, and the bitumen occurrence morphology affects the pore structure and petrophysics. Two types of bitumen, boundary-filled and centre-filled, were reconstructed using a pore morphology-based method based on dolomite carbonate digital cores from X-ray computerized tomography. The pore structures of the reconstructed bitumen-filled digital core models were determined using an extracted pore network model. The finite element method simulated the electrical resistivity of the digital core model. The investigation showed that (i) boundary-developed bitumen increases the pore size whereas centre-developed bitumen decreases the pore size, and bitumen development increases the shape factor while reducing the coordination number; (ii) for electrical resistivity, boundary-filled bitumen increases the porosity exponent with increasing pore size, whereas centre-filled bitumen decreases the porosity exponent. Bitumen development caused a negative correlation between the porosity exponent and pore shape factor, a logarithmic increase between the porosity exponent and coordination number, and a power increase between the porosity exponent and porosity. Subsequently, a new porosity exponent model related to porosity was proposed based on the curve-fitting method. This new porosity exponent combines the Archie equations to calculate the saturation of the carbonate reservoir in the Sinian Dengying Formation in the Penglai gas area on the northern slope of the central Sichuan paleo-uplift, which fits the sealed core plug saturation well and provides the basis for further reservoir evaluation and fluid identification in the Penglai gas area.

Three-Dimensional Spatial Microscopic Characteristics and Developmental Influencing Factors of Tight Gas Layers in Hangjinqi Prospect Area, Ordos Basin, China

The unconventional tight oil and gas resources in the Xinzhao East belt of the Hangjinqi Prospect area in the Ordos Basin of China are abundant. However, the reservoir’s internal storage space is complex, and the microscopic pore throat structural features are not well recognized, which has led to some trouble in the deployment of oil and gas exploration. To reveal the microscopic characteristics of the dense sandstone gas layer in the first member of the Lower Stone Box Formation of the D-well Zone in the Xinzhao East belt of the Hangjinqi Prospect area, a three-dimensional space digital core was built, and the stored set spatial data were extracted, based on rock sheet and coring data and X-CT scanning technology. Quartz grain size was segmented and analyzed based on an adaptive approach. The microscopic characteristics of the gas layer in the studied section and the factors influencing its development were studied, combining the use of a field emission scanning electron microscope, helium porosimeter, and gas permeability meter. We found that in the studied section, the porosity is relatively high, the pore throat size is large, and the pore permeability correlation is good. The reservoir space, which consists of intergranular pores, intragranular pores, and microcracks at the grain edges in the study area, is characterized by a complex distribution pattern. Within the gas layer, isolated pores are connected by microcracks to form a network of reservoir spaces, which increases the pore throat size, enhances the connectivity of the pore throat, and makes the microscopic characteristics of the reservoir space better. The first member of the Lower Stone Box Formation could be an advantageous reservoir. Hole–throat connectivity is poor because of the gas layer having underdeveloped primary pores, the blockage of pores by unstable minerals (kaolinite, etc.), and poorly connected pore throats based on insoluble mud cementation. The high content of quartz brittle minerals and the development of natural microcracks within the gas formation are favorable conditions for fracking development. The quartz grain size within the gas layer is positively correlated with the pore throat size, which suggests that the quartz grain size somewhat influences the microscopic characteristics of the reservoir space. This comprehensive study shows that the methodology of the study is more advantageous than traditional methods in the fine and three-dimensional spatial characterization of the microstructure of dense sandstone reservoirs. The research results of this paper have certain guiding significance for further reservoir evaluation and advantageous reservoir prediction in the Hangjinqi Prospect area in the Ordos Basin. We also provide the basis for the subsequent efficient development of the gas reservoir.

Acoustical-electrical models of tight rocks based on digital rock physics and double-porosity theory

The acoustical-electrical (AE) properties of reservoir rocks are affected by their microstructure (pores, microfractures, cracks and their geometry) and saturating fluids. To aid the interpretation, digital rock physics (DRP) is a useful technique to characterize this microstructure, as well as equivalent petrophysical models (EPM) to obtain joint AE properties. We perform thin-section analysis on rocks from tight-oil reservoirs, along with X-ray diffraction (XRD) and computed tomography (CT) to study rock lithology and minerals. We also perform porosity, permeability, ultrasound, and electrical conductivity experiments as a function of confining pressure to analyze pore structure. First, the 3D digital multicomponent cores are created based on a geology-driven multiphase segmentation workflow and images of the samples, and verified based on the porosity and minerals. Then, the AE properties and permeability are calculated by using numerical simulations (finite difference and finite volume methods). Next, the rigid and microcracked (soft) porosities are determined by using the Shapiro model to create the rock skeleton. We develop a joint EPM based on the effective medium AE and the Biot-Rayleigh equations with double porosity to describe the rock properties. The ultrasonic, sonic and seismic multiple data are used to compare and analyze the two approaches. The results show that the DRP techniques based on real cores are effective in characterizing the microstructure and the proposed EPM can describe the AE properties of real rocks, indicating a potential for quantitative characterization of reservoirs.

Microstructure Imaging and Characterization of Rocks Subjected to Liquid Nitrogen Cooling

Liquid nitrogen (LN2) fracturing is a potential stimulation method in unconventional hydrocarbon recovery, showing its merits in being water free, creating low formation damage and being environmentally friendly. The microstructure evolution of rocks subjected to LN2 cooling is a fundamental concern for the engineering application of LN2 fracturing. In this paper, pore-scale imaging and characterization were performed on two rocks, i.e., tight sandstone and coal specimens subjected to LN2 cooling using computed tomography scanning. The digital core technique was employed to reconstruct the microstructures of rocks and give a quantitative analysis of the pore structure evolution of both dry and water-saturated rocks. The results indicate that LN2 cooling has a great effect on the pores’ morphology and their spatial distribution, leading to a great improvement in pore diameter and aspect ratio. When compared to the sandstone, coal is more sensitive to LN2 cooling and thermal stresses, having a more noticeable growth in pore–throat size. The porosity growth of coal is 291% higher than that of sandstone. There is a growing trend in the irregularity and complexity of pore structures. After LN2 cooling, the fractal dimensions of the pores of sandstone and coal grow by 11.7% and 0.87%, respectively, and the proportion of pores with a shape factor > 100 increases. More bundle-like and strip-shape pores with multiple branches are generated, which causes a significant growth in the throat size and the proportion of connected pores with a coordination number ≥ 1, enhancing the complexity and connectivity of pore structures dramatically. Additionally, pore water plays an important role in aggravating rock damage during LN2 cooling, enhancing the pore space and connectivity. The porosities of the saturated sandstone and coal samples grow by 22.6% and 490.4%, respectively, after LN2 cooling, which are 5.6% and 186.6% higher than dry samples. The generation of macropores ≥ 70 μm is the primary contributor to porosity growth during LN2 cooling, although such pores account for only a small proportion of the total. These findings contribute to our understanding of the microscopic mechanism of LN2 cooling on rock damage and may provide some guidance for the engineering application of LN2 fracturing.

Measurement of the scale and development trend of digital economy core industries in China’s provinces

In order to measure the scale and development trend of China’s inter-provincial digital economy, this paper calculates the scale of digital economy core industries and the added value of four different categories of core industries of 30 provinces in China from 2007 to 2017 based on the division standard of digital economy industries of the National Bureau of Statistics. In order to measure the results more accurately and reflect the characteristics of each province’s industrial structure, this paper uses the input-output tables of more than 100 departments of each province in 2007, 2012 and 2017 and the data of the Economic Census Yearbook to split the digital and non-digital parts of the industry. The results show that the scale of digital economy industry in the eastern region of China is the largest and the development speed is fast. The scale of digital economy in the central region is relatively small, but the development speed is the fastest. The western and northeastern regions have the smallest digital economies and the slowest growth rates. In addition, among the four major industries, the digital economy application industry and digital product manufacturing industry account for the largest proportion, and the digital product service industry accounts for the smallest proportion. The research results of this paper can provide reference for establishing a national unified digital economy core industry measurement system and improving digital industry related data collection.

Digital therapeutics-based lumbar core exercise for patients with low back pain: A prospective exploratory pilot study.

This study aimed to implement a digital therapeutics-based approach based on motion detection technology and analyze the clinical results for patients with chronic low back pain (LBP). A prospective, single-arm clinical trial was conducted with 22 patients who performed mobile app-based sitting core twist exercise for 12 weeks. Clinical outcomes were assessed using the visual analog scale (VAS) for LBP, Oswestry Disability Index-Korean version (K-ODI), and EuroQol-5 dimension 5-level version (EQ-5D-5L) every 4 weeks after the initiation of treatment. Laboratory tests for factors associated with muscle metabolism, plain X-ray for evaluating sagittal balance, and magnetic resonance imaging for calculating cross-sectional area (CSA) of back muscles were performed at pretreatment and 12 weeks post-treatment. The study population included 20 female patients with an average age of 45.77 ± 15.45 years. The clinical outcomes gradually improved throughout the study period in the VAS for LBP (from 6.05 ± 2.27 to 2.86 ± 1.86), K-ODI (from 16.18 ± 6.19 to 8.64 ± 5.58), and EQ-5D-5L (from 11.09 ± 3.24 to 7.23 ± 3.89) (p < 0.001, respectively). The laboratory test results did not show significant changes. Pelvic incidence (from 53.99 ± 9.70° to 50.80 ± 9.20°, p = 0.002) and the mismatch between pelvic incidence and lumbar lordosis (from 8.97± .67° to 5.28 ± 8.57°, p = 0.027) decreased significantly. Additionally, CSA of erector spinae and total back muscles increased by 5.20% (p < 0.001) and 3.08% (p = 0.013), respectively. The results of this study suggest that the efficacy of digital therapy-based lumbar core exercise for LBP is favorable. However, further large-scale randomized controlled studies are necessary.

PORE STRUCTURE AND PERMEABILITY EVOLUTION OF POROUS MEDIA UNDER IN SITU STRESS AND PORE PRESSURE: DISCRETE ELEMENT METHOD SIMULATION ON DIGITAL CORE

In stress-sensitive oil and gas reservoirs, formation rock deformation occurs under in situ stress and pore pressure, affecting the rock's porosity and permeability. Pore deformation is the fundamental mechanism. However, the literature on numerical simulation of rock porosity and permeability at the pore scale is rare. In this paper a numerical simulation framework of pore scale is proposed based on the discrete element method. The pore geometry and permeability evolution of the core are quantitatively analyzed by the digital core method. Firstly, the coupled fluid-discrete element method (CFM-DEM) is used to simulate the samples' deformation under different stress and pore pressures. We then reconstruct the digital core using Avizo. Finally, the pore geometric topological structures are analyzed, and the permeability changes are calculated. The results show that stress can reduce porosity, modify pore shape, and lead to poor porosity connectivity and permeability, while pore pressure can weaken such trends.

Mesoscale migration of oil in tight sandstone reservoirs by multi-field coupled two-phase flow

The oil migration process plays a pivotal role in determining the distribution and scale of oil reservoirs in tight sandstone formations, while also simultaneously influencing their development outcomes. However, the mechanisms governing oil migration in these tight formations remain poorly understood. This study conducted numerical simulations to investigate the oil migration characteristics in tight sandstone at the mesoscale. Multivariate regression analysis was used to characterize the micromechanical parameters of a 3D digital core. In addition, the main equations describing the two-phase flow that couples the stress and seepage fields were proposed, thereby elucidating the oil migration process. Finally, the effects of the reservoir space type, pressure gradient, and stress on the oil migration efficiency were assessed. The results showed that: (1) Compared with the porous-type model, the fractured-type model had a higher level and more concentrated distribution of damage, better connectivity, and higher oil migration efficiency due to the seepage-stress coupling effect; (2) High fluid pressure gradients improved the efficiency of oil migration and enhanced the control of the three-dimensional distribution of the pore domain over oil distribution in the matrix domain. (3) When σ1 >σ3, the oil phase migration efficiency improves with the increasing axial stress. When σ1 <σ3, the oil phase migration efficiency decreases with the increasing confining stress. When σ1 = σ3, high stress leads to low oil phase migration efficiency. This study reveals the mesoscopic-scale mechanism of oil migration in tight sandstone reservoirs and may prove useful in the exploration and development in such reservoirs.

Digital core: study of textural heterogeneities in a rock

Relevance. When real fluid flows through channels of limited dimensions, it experiences a constant loss of mechanical energy (hydraulic resistance) due to the action of viscous friction. Hydraulic drag has two components: drag along the length of the flow and local drag. The drag along the length of the flow arises because the fluid encounters a force per unit area perpendicular to the direction of the flow. Local drag results from the viscous interaction between the fluid and the channel walls. Local hydraulic resistances include sudden and gradual constrictions/expansions (diffusers/confusions) of the fluid channel, angular and choke gaps. The value of the hydraulic resistance coefficient depends on the type of channel geometry, conditions of fluid entry (wettability of the rock texture) and on the flow regime. When fluid flows through channels of different cross-section, as the channel narrows, the flow velocity increases and the pressure, based on Bernoulli's equation, decreases. Objects. Polymictic sandstones of the Tyumen Formation. Methods. The developed algorithms for identifying areas of contraction/expansion (confusers/diffusers) in the rock texture (polymictic sandstone) are based on neural network reconstruction of core data. The study of geometric diversity of geological rock images is based on generalization of real spatial forms in combination with the axiom of hyperplanes. A methodical approach for finding distribution of inertial exciters (confusers/diffusers) affecting the character of multiphase flow front changes in the rock texture is based on the metrics for estimating possible distributions. Results. The paper introduces the initial stages of core digital transformation data. It was determined that the main problem of geological data digital projection on a large scale is the need to take into account the systems of inertial effects, both in the available volume of the rock and in the extrapolated space of it. In order to adequately represent the character of multiphase fluid flow for different scales, the authors have developed the algorithms for segmentation of geological images of fluid space with further identification of distribution laws of inertial exciters (confusers/diffusers) in the rock texture. For training and testing of the developed algorithms we used the computer tomography images of core material and detailed resolution images of rock sections. The source code of neural network algorithms was written in the Python programming language with the additional use of non-commercial libraries. The algorithms were tested on real data, confirming by experimental studies in the laboratory of digital research in oil and gas as part of the technological project "Digital Core" (Industrial University of Tyumen, Tyumen), the laboratory of the V.I. Shpilman Research and Analytical Center for Rational Subsoil Use (Khanty-Mansiysk), Core Research Laboratory (University of Tyumen, Tyumen).