A method for in-situ analysis of micro- and nano-scale components of mudstone and shale

Abstract

This invention relates to the field of shale permeability analysis technology, specifically to an in-situ analysis method for shale micro-components at the micron-nano scale. The invention first obtains typical micro-components from the microscopic scanning images of the sample to be analyzed; further, it obtains scanning electron microscope images of the typical micro-components and extracts pore regions; based on the spatial distribution of pore regions within local image blocks, it obtains the local pore density; it analyzes the average width of pore regions within each image block, and combined with the local pore density, obtains the influence of wide-pore flow and screens out high-permeability regions; finally, based on the spatial distribution of high-permeability regions and combined with the influence of wide-pore flow, it classifies the permeability effect of shale, effectively solving the problem of inaccurate permeability capacity assessment caused by neglecting pore heterogeneity in traditional methods.

Description

Technical Field

[0001] This invention relates to the field of shale permeability analysis technology, specifically to an in-situ analysis method for shale micro-components at the micrometer-nano scale. Background Technology

[0002] In existing technologies, studies have employed a method combining field emission scanning electron microscopy (FESEM) and optical microscopy for localized observation, enabling the identification and observation of specific microscopic components under scanning electron microscopy ("Application of Argon Ion Polishing-Field Emission Scanning Electron Microscopy Analysis Method in Identifying Organic Microscopic Components," Petroleum Experimental Geology, March 2021). This method first performs FESEM analysis on the sample and manually calibrates the positions of all observed organic matter. Subsequently, based on the calibrated positions, the corresponding organic matter is identified one by one under an optical microscope to complete the identification of microscopic components. The organic porosity in shale is then analyzed based on the distribution of organic matter.

[0003] On the one hand, manual calibration and one-by-one comparison of pore distribution are inefficient, prone to subjective errors, and difficult to achieve high-throughput and repeatable analysis. On the other hand, existing methods are mostly based on traditional parameters such as pore volume and surface area for permeability analysis, failing to fully consider the heterogeneity of pore distribution in shale and the significant influence of pore width on seepage behavior. This makes it difficult for existing methods to accurately reflect the true seepage characteristics of shale, resulting in biases in the assessment of reservoir seepage capacity. Summary of the Invention

[0004] To address the problem of inaccurate permeability assessment caused by neglecting the heterogeneity of pore distribution in existing technologies, the present invention aims to provide an in-situ analysis method for the micro- and nano-scale microstructures of shale. The specific technical solution adopted is as follows:

[0005] Obtain typical microscopic components in the vertical layering plane of the sample to be analyzed using microscopic scanning images;

[0006] Scanning electron microscope (SEM) images of typical microscopic components are acquired and pore regions are extracted. The SEM images are then uniformly divided into image blocks using preset segmentation parameters. Based on the spatial distribution of pore regions within each image block and its preset neighborhood, the local pore density of each image block is obtained. The average width of the pore regions within each image block is analyzed, and combined with the local pore density, the influence of wide-pore seepage is obtained, and high-seepage regions are selected.

[0007] Based on the spatial distribution of high-permeability areas and the influence of wide-pore seepage, the seepage effect of mudstone and shale is classified.

[0008] Furthermore, the method for obtaining the typical microscopic components includes:

[0009] A pre-trained neural network model is used to automatically identify all microscopic components and their relative percentage content. Microscopic components whose relative percentage content exceeds a preset typical threshold are marked as typical microscopic components. The number of marked components is obtained based on the percentage of the typical microscopic components.

[0010] The geometric center points of the pixel regions of the typical microscopic components are extracted as candidate marker points. A spatial optimization algorithm is used, combined with the number of markers, to uniformly mark the typical microscopic components.

[0011] Furthermore, the method for obtaining the number of markers includes:

[0012] The ratio of the relative percentage content of each typical micro-component to the constant 0.1 is rounded up, and the rounded result is used as the number of markers.

[0013] Furthermore, the method for obtaining the local pore density includes:

[0014] The average coordinates of all pore pixels within each image block are obtained as the pore center point. Based on the Euclidean distance between the pore center point of each image block and the pore center points of other image blocks in a preset neighborhood, and combined with the proportion of the pore region occupying the image block, the local pore density of each image block is obtained.

[0015] Furthermore, the method for obtaining the influence degree of wide-gap seepage includes:

[0016] By fusing the side length of the image block, the average width of all the pore regions, and the local pore density, the wide-slot seepage influence degree of the corresponding image block is obtained.

[0017] Furthermore, methods for classifying the seepage effects of mudstone and shale include:

[0018] Obtain the center coordinates of the high-permeability region with the largest wide-gap seepage influence, and the Euclidean distance between the center coordinates of the other high-permeability regions, as the distribution distance; fuse all the distribution distances, the side length of the scanning electron microscope image, and the wide-gap seepage influence of all the high-permeability regions to obtain the heterogeneous seepage influence.

[0019] The seepage effect of mudstone and shale is classified based on the aforementioned heterogeneous seepage influence.

[0020] Furthermore, the method for classifying the seepage effect of shale based on the heterogeneous seepage influence degree includes:

[0021] When the degree of heterogeneous seepage influence is within a preset low threshold range, the seepage effect of shale is determined to be weak; when the degree of heterogeneous seepage influence is within a preset medium threshold range, the seepage effect of shale is determined to be medium; when the degree of heterogeneous seepage influence is within a preset high threshold range, the seepage effect of shale is determined to be strong.

[0022] Furthermore, the method for obtaining the porous region includes:

[0023] The OTSU algorithm is used to obtain binary images of scanning electron microscope images, with the black areas representing pore regions.

[0024] Furthermore, the method for obtaining the average width of the pore region includes:

[0025] The boundary of each pore region is obtained by using the two-sided scanning method. The Euclidean distance between the two farthest points on the boundary is taken as the maximum side length. The ratio of the total number of pixels in the pore region to the maximum side length is taken as the average width of the corresponding pore region.

[0026] Furthermore, the method for obtaining the high permeability region includes:

[0027] The average value of all the wide-gap seepage influence is obtained as the seepage screening threshold, and the image blocks with wide-gap seepage influence greater than the seepage screening threshold are marked as high seepage areas.

[0028] The present invention has the following beneficial effects:

[0029] This invention first acquires typical microscopic components from the microscopic scanning images of the sample to be analyzed, clarifying the targets for subsequent searching and analysis under an electron microscope, ensuring the efficiency of pore analysis. It then acquires scanning electron microscope images of the typical microscopic components and extracts the pore regions, eliminating interference from other areas. The scanning electron microscope images are uniformly divided into image blocks using preset segmentation parameters, facilitating the capture of local spatial distribution characteristics of pores. Furthermore, based on the spatial distribution of pore regions within local image blocks, the local pore density is obtained, characterizing the local density of pores and providing a basis for subsequently obtaining the influence of wide-pore seepage. Based on this, the average width of the pore region within each image block is further analyzed to quantify the inherent conductivity of the pore structure in that local area. Combined with the local pore density, the possibility of pores forming a connected network within the local area is analyzed to obtain the influence of wide-pore seepage, comprehensively reflecting the seepage efficiency of the local area and screening out high-permeability areas, laying the foundation for accurate subsequent assessment of the seepage effect of shale. Finally, based on the spatial distribution of high-permeability areas and combined with the influence of wide-pore seepage, the seepage effect of shale is classified, achieving a precise and reliable evaluation of the seepage capacity of shale reservoirs. By extracting typical microscopic components of the sample, quantifying the pore distribution density and average width, and fusing them to generate the influence of wide-pore seepage, the seepage effect of shale is accurately classified based on the spatial distribution of high-permeability areas, effectively solving the problem of inaccurate seepage capacity assessment caused by neglecting pore heterogeneity in traditional methods. Attached Figure Description

[0030] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 A flowchart illustrating an in-situ analysis method for micro- and nano-scale microstructures of shale provided in an embodiment of the present invention;

[0032] Figure 2 This is a labeled image of a microscopic scanning image provided in one embodiment of the present invention. Detailed Implementation

[0033] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a micron-nano-scale in-situ analysis method for microscopic components of shale according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0035] The following description, in conjunction with the accompanying drawings, details the specific scheme of the in-situ analysis method for micro- and nano-scale micro-components of shale provided by this invention.

[0036] Please see Figure 1 The diagram illustrates a flowchart of an in-situ analysis method for micro- and nano-scale microstructures of shale provided by an embodiment of the present invention, specifically including:

[0037] Step S1: Obtain typical microscopic components in the vertical layering plane of the sample to be analyzed using a microscopic scanning image.

[0038] In one embodiment of the present invention, the sample to be analyzed is cut into sections perpendicular to the bedding plane. Small square blocks, 0.2-0.4 cm thick, are prepared and their vertical stratification surfaces are sanded smooth using 50-600 grit sandpaper. Next, the small blocks are inlaid using a mounting machine and epoxy resin, ensuring that one side of the vertical stratification surface is exposed. Then, the exposed surface of the prepared inlay is finely sanded using 800-1200 grit sandpaper, and finally mechanically polished using micron-level polishing fluid.

[0039] The prepared optical slide is placed on the automatic stage of an optical microscope. After setting the relevant parameters, the optical microscope and image acquisition software automatically perform a surface scan of the optical slide to obtain a set of microscopic molecular images and stitch them together to form a microscopic scanning image of the vertical layering plane of the entire surface scan area.

[0040] Optical microscope setup parameters: A 20x or 50x oil immersion lens should be selected as the objective. In this embodiment, a 20x oil immersion lens is selected. Reflected white light is selected as the light source. Adjust the focal length and imaging software parameters to reproduce the true appearance under the microscope in the acquired image. The scanning area is set to cover the entire exposed polished surface of the rock sample on the light slide. The scanning direction is set to alternately scan line by line along the bedding plane. After completing the image scanning and stitching, switch to the fluorescence light source and repeat the above scanning steps to obtain the corresponding fluorescence sub-image set and stitched image.

[0041] It should be noted that the sample processing methods, the use and settings of the optical microscope, and image stitching are all well-known technologies and will not be elaborated further.

[0042] First, obtain typical microscopic components from the vertical layering planes of the sample to be analyzed, locate the abundant and statistically representative organic matter in the sample, clarify the target for subsequent searching and analysis under electron microscopy, and ensure the efficiency of pore analysis.

[0043] Preferably, in one embodiment of the present invention, considering that the identification and labeling of existing microscopic components heavily rely on the subjective experience of experts, the identification results of different operators may be biased, making it difficult to achieve standardized and batch analysis, a neural network model is trained, the pre-trained neural network model is used to obtain the component identification results and perform automatic identification of typical microscopic components, and a spatial optimization algorithm is combined for uniform labeling, thereby enabling fast, objective and repeatable automatic identification, overcoming the subjectivity and low efficiency of manual identification.

[0044] The construction method of pre-trained neural network models is as follows:

[0045] Model Training: A large number (e.g., 500) of shale microscopic scan images identified by domain experts according to industry standards (e.g., SY / T6414) are collected as training samples. For each training sample image, experts annotate the precise pixel regions of different microscopic components (e.g., lamellar algae, sporophytes, filamentous bodies, etc.) and assign corresponding microscopic component category labels to each region.

[0046] A training dataset is constructed using this dataset, and deep learning semantic segmentation models (such as U-Net and DeepLab) are used for training. The core function of the model is to perform pixel-level classification, that is, to take a microscopic scan image as input and output a segmentation map of the same size. Each pixel in the map is predicted to belong to a specific microscopic component, and the relative percentage of each microscopic component is determined, specifically the proportion of the number of pixels of the corresponding microscopic component to the total number of pixels of all microscopic components.

[0047] The pre-trained neural network model is used to automatically identify all microscopic components and their relative percentage content in the microscopic scanning image of the sample to be analyzed.

[0048] Please refer to Table 1, which shows a component identification result table provided by an embodiment of the present invention;

[0049] Table 1 Component Identification Results

[0050]

[0051] In Table 1, the percentage of microscopic components is the relative percentage content of the microscopic components.

[0052] Microscopic components with a relative percentage content exceeding a preset typical threshold are marked as typical microscopic components, focusing on key features and reducing interference from components with a relatively small percentage content.

[0053] To ensure statistical representativeness, an appropriate number of observation markers are assigned to each typical microscopic component, so the number of markers is obtained based on the percentage of typical microscopic components;

[0054] As an example, the typical threshold is preset to 1%. The ratio of the relative percentage content of each typical micro-component to the constant 0.1 is rounded up, and the rounded result is used as the number of labels.

[0055] For example, the relative percentage content of lamellar algae in Table 1 is 86.9%. Dividing by 0.1 (10%) gives a quotient of 8.69, which is rounded up to 9. The physical meaning of dividing by 0.1 here is to set a benchmark, that is, to assign one marker point for every 10% of the relative percentage content, so that more marker points are automatically assigned to components with high content.

[0056] Furthermore, considering that the statistical results obtained by uniform labeling are more representative of the overall properties of the entire sample, the geometric center points of the pixel regions of typical microscopic components are extracted as candidate labeling points. A spatial optimization algorithm is used in conjunction with the number of labels to uniformly label the typical microscopic components.

[0057] As an example, a two-dimensional Cartesian coordinate system is established in the microscopic scan image. The geometric center point is the average coordinate point of all pixels in the pixel region (the point corresponding to the average of the horizontal and vertical coordinates). The spatial optimization algorithm is a uniform distribution algorithm based on Poisson disk sampling. Taking all candidate marker points as input and the number of markers as a constraint, a set of final marker points that are distributed as evenly as possible in space are automatically selected, and then the markers are marked at the final marker points.

[0058] It should be noted that the preset typical threshold and the relative percentage content setting corresponding to one marker point can be adjusted by the implementer; the training of the neural network model and the uniform distribution algorithm based on Poisson disk sampling are well known to those skilled in the art and will not be described in detail here; in other embodiments of the present invention, the implementer may also round the ratio of the relative percentage content of each typical microscopic component to the constant 0.1 to the nearest integer, and take 1 if it is less than 1.

[0059] Please see Figure 2 The diagram shows a labeled result of a microscopic scan image provided by an embodiment of the present invention. Figure 2The scale bar is 2mm. The marked areas numbered 1-9 are lamellar algae, the marked area numbered 10 is a pollen body, the marked area numbered 11 is a filamentous body, and the marked area numbered 12 is a detritus body.

[0060] In another embodiment of the present invention, a conventional expert labeling method is used to obtain typical microscopic components in a microscopic scanning image.

[0061] Step S2: Obtain scanning electron microscope (SEM) images of typical microscopic components and extract pore regions. Divide the SEM images uniformly using preset segmentation parameters to obtain image blocks. Based on the spatial distribution of pore regions within each image block and the preset neighborhood image blocks, obtain the local pore density of each image block. Analyze the average width of pore regions within each image block, and combine it with the local pore density to obtain the influence of wide-pore seepage and screen out high-seepage regions.

[0062] After the samples with completed component identification and labeling were wire-cut, the epoxy resin was manually peeled off, the bottom was flattened, and then argon-ion polished. Based on the marked positions on the microscopic scanning images, typical micro-components were searched one by one along the bedding direction. First, a low-magnification lens was used to determine the approximate location of the marks, and then a high-magnification lens was used to pinpoint the location. Scanning electron microscopy (SEM) images of each location were obtained to analyze the spatial distribution of pores in the shale.

[0063] It should be noted that the analysis process for the scanning electron microscope (SEM) images at each location is the same. Here, we will only analyze one example and will not repeat the explanation. The SEM images are grayscale images.

[0064] Scanning electron microscope images contain a variety of information such as pores, organic matter, and minerals, so the pore region is extracted first to eliminate interference from other regions;

[0065] For the same pore volume at the nanoscale, densely distributed, unevenly distributed pores have larger local throat radii and shorter connectivity paths, resulting in lower flow resistance to fluids and higher permeability in shale. Therefore, the local spatial distribution of pores in shale determines its permeability. Thus, uniformly segmenting the scanning electron microscope image using preset segmentation parameters to obtain image blocks facilitates the capture of the local spatial distribution characteristics of pores.

[0066] Preferably, in one embodiment of the present invention, since various organic materials in shale have a certain reflective effect and the boundaries between different organic materials have relatively obvious dividing lines, the dividing lines are pores, making the pores darker and closer to black than the various organic materials. Therefore, there is a relatively obvious difference in color between the pores and the various organic materials.

[0067] Based on this, the OTSU algorithm (Otsu method, a known technique) is used to obtain binary images of scanning electron microscope images. The OTSU algorithm is used to determine the optimal segmentation threshold. The pixel values ​​of pixels with gray values ​​less than or equal to the optimal segmentation threshold are set to 0 (black), and the pixel values ​​of pixels with gray values ​​greater than the optimal segmentation threshold are set to 255 (white). The black area is the pore area, and the white area is various organic matter and minerals.

[0068] As an example, with the preset segmentation parameter N set to 10, the scanning electron microscope image is uniformly segmented into... Each image is divided into blocks.

[0069] It should be noted that since only the spatial distribution of pores is analyzed, and detailed textures and other areas do not need to be analyzed, the analysis can be performed in a binary image, and the binary image can be divided into corresponding blocks; the preset segmentation parameters can be adjusted by the implementer according to actual needs.

[0070] Considering the distribution of pore regions in image blocks and neighboring image blocks, which reflects the density of pores in local areas, the local pore density of each image block is obtained based on the spatial distribution of pore regions in each image block and the image blocks in the preset neighborhood. This characterizes the local density of pores and provides a basis for subsequently obtaining the influence of wide-gap seepage.

[0071] In a preferred embodiment of the present invention, a two-dimensional rectangular coordinate system is established in the scanning electron microscope image or the corresponding binary image. First, the average coordinate point (the point corresponding to the average horizontal coordinate and the average vertical coordinate) of all pore pixels in each image block is obtained as the pore center point, which is used to characterize the central distribution position of the pores in the image block.

[0072] Considering that the smaller (closer) the Euclidean distance between the center points of pores in a local area, the denser the spatial distribution of the pore region, and the larger the proportion of the pore region occupying the image block, the wider the distribution of the pore region and the stronger the local density; therefore, based on the Euclidean distance between the center point of the pores of each image block and the center points of the pores of other image blocks in the preset neighborhood, combined with the proportion of the pore region occupying the image block, the local pore density of each image block is obtained.

[0073] As an example, assuming an eight-neighborhood, the formula for calculating local porosity includes:

[0074] ;

[0075] in, This represents the local porosity of the i-th image block; This represents the number of other image blocks (called neighborhood image blocks) within the preset neighborhood of the i-th image block; Let L represent the Euclidean distance between the aperture center point of the i-th image block and the aperture center point of the corresponding j-th neighboring image block; L represents the side length of the image block. This represents the sum of the proportion of the pore region within the i-th image block and the proportion of the pore region within the j-th neighboring image block.

[0076] In the formula for calculating local pore density, the ratio of the side length of an image block to the distance between the center points of the pores represents the relative distribution density of pores between two image blocks. The proportion of the pore area within two image blocks represents the distribution intensity of the pores. These are then fused through multiplication. The smaller The larger the value, the more concentrated the distribution of pore regions and the greater the local pore density in the two image blocks; L is used to eliminate the influence of block size, preventing the denominator from being zero and also eliminating the dimension of Euclidean distance. The units of side length and Euclidean distance are pixels, and the proportion occupied by the pore region is the proportion of the number of pixels in the pore region to the total number of pixels in the image block.

[0077] It should be noted that if at least one of the two image blocks has no pore region, it is skipped and ignored; if the image block itself has no pore region, the local pore density is set to 0.

[0078] For pores in shale, under the premise of the same pore volume, the larger the pore width, the smaller the seepage resistance; conversely, the smaller the width, the exponentially increase the resistance. Therefore, analyzing the average width of the pore region within each image block is used to quantify the inherent conductivity of the pore structure in that local area. Combined with the local pore density, the possibility of pores forming a connected network in the local area is analyzed to obtain the influence of wide-pore seepage, resulting in a composite index that can comprehensively reflect the seepage efficiency of the local area. High-permeability areas are then screened, and the "critical path" or "dominant channel" that contributes the most to the seepage capacity of the entire sample is identified, laying the foundation for subsequent accurate evaluation of the seepage effect of shale.

[0079] Preferably, in one embodiment of the present invention, considering that the pore channels are curved and irregular, it is necessary to approximate the hydraulic diameter of the complex pore shape;

[0080] First, the boundary of each pore region is obtained using the two-sided scanning method, so that each independent pore can be statistically analyzed and measured separately.

[0081] Further, the Euclidean distance between the two farthest points on the boundary is obtained as the maximum side length, the feature length in the main extension direction of the aperture is captured, and the ratio of the total number of pixels in the aperture region to the maximum side length is used as the average width of the corresponding aperture region.

[0082] In this calculation, the maximum side length corresponds to the denominator, and the unit of average width is pixels. A short and thick pore has a large area and a small maximum side length, resulting in a large average width, which is consistent with the characteristics of high flow conductivity. A long and thin pore may have a considerable area, but a large maximum side length, resulting in a small average width, which is consistent with the characteristics of high flow resistance. This aligns with physical laws, so the average width is used to approximate the hydraulic diameter.

[0083] Considering that the smaller the side length of an image block, the larger the average width of all pore regions inside, indicating that the seepage resistance of the pores is smaller and the local pore density is greater, indicating that the pores are more likely to form a connected network and the permeability is greater, the side length of the image block, the average width of all pore regions, and the local pore density are fused to obtain the wide-slot seepage influence of the corresponding image block.

[0084] As an example, in each image block, the sum of the average widths of all pore regions is used as the numerator, the side length of the image block is used as the denominator, and the ratio of the fractions is used as the permeation coefficient. The units of the numerator and denominator are pixels. The permeation coefficient is dimensionless. The product of the permeation coefficient and the local pore density of the same image block is used as the wide-slot permeation influence degree of the corresponding image block.

[0085] The larger the permeability coefficient, the wider the internal pores of the image block. The greater the permeability of the pores in shale at the nanoscale, the greater the local pore density, the easier it is to achieve synergistic permeation, the greater the influence of wide-pore permeation, and the stronger the permeability of the material.

[0086] Considering that the average value reflects the overall spatial distribution level of the data, the average value of all wide gap seepage influence is obtained as the seepage screening threshold, and the image blocks with wide gap seepage influence greater than the seepage screening threshold are marked as high seepage areas.

[0087] It should be noted that, in one embodiment of the present invention, the seepage screening threshold is defined by the average value of the seepage influence of all wide-pore samples in the currently analyzed sample; in other embodiments of the present invention, data of historically analyzed samples can also be collected in advance to obtain the average value of the seepage influence of all wide-pore samples for determination.

[0088] It should be noted that the method of obtaining the boundary of the image region and the pixels contained in the image region by the two-sided scanning method is a well-known technology and will not be described in detail here.

[0089] Step S3: Based on the spatial distribution of high permeability areas and the influence of wide-pore permeability, classify the permeability effect of mudstone and shale.

[0090] The composition of mudstone and shale varies at different locations in the geological strata, resulting in different pore distribution patterns among the organic matter in the mudstone and shale. These pores may be homogeneous or heterogeneous. Homogeneous pores have greater seepage resistance than heterogeneous pores. Under the same pore volume, heterogeneous pores will exhibit a concentrated spatial distribution, making the width of heterogeneous pores larger, thus resulting in better seepage performance of mudstone and shale.

[0091] Therefore, it is necessary to analyze the pore distribution of the region that contributes the most to the seepage capacity of the sample based on the spatial distribution of the high-permeability area. The influence of wide-pore seepage characterizes the seepage efficiency of the local area. Therefore, the seepage effect of mudstone and shale is divided by combining the two, so as to effectively overcome the distortion of seepage effect assessment caused by the traditional method ignoring the heterogeneity of pore distribution, and realize the accurate and reliable evaluation of the seepage capacity of mudstone and shale reservoirs.

[0092] Preferably, in one embodiment of the present invention, the high-permeability region, considering the influence of the maximum wide-gap seepage, is the core and starting point (the main channel of seepage) of the entire seepage network. If other high-permeability regions are closely clustered around the maximum seepage region, it indicates that they can easily connect with the main channel to form an efficient and interconnected seepage network. Fluid can quickly flow from small channels into large channels, resulting in low overall seepage resistance, strong heterogeneity, and favorable flow characteristics.

[0093] Based on this, the spatial location of the central coordinate point of each high seepage zone is represented by the central coordinate point of the high seepage zone with the largest wide gap seepage influence, and the Euclidean distance between the central coordinate points of the other high seepage zones is obtained as the distribution distance between the two high seepage zones.

[0094] Further integrate all distribution distances, the side lengths of scanning electron microscope images, and the wide-gap seepage influence of all high-permeability regions to obtain the heterogeneous seepage influence.

[0095] As an example, the average value of all distribution distances is used as the denominator, the side length of the scanning electron microscope image is used as the numerator, and the fractional ratio is used as the heterogeneity coefficient. By using the side length of the scanning electron microscope image to eliminate the influence of the side length of different images (with the same resolution), the dimension of the distribution distance is eliminated. The product of the average value of the wide-pore seepage influence degree of all high-seepage regions and the heterogeneity coefficient is used as the heterogeneity seepage influence degree.

[0096] Finally, the seepage effect of mudstone and shale was classified based on the degree of influence of heterogeneous seepage.

[0097] Specifically, when the influence of heterogeneous seepage is within the preset low threshold range, it indicates that the pores of mudstone and shale are likely to be relatively uniformly distributed and the pore distribution area is small, thus the seepage effect of mudstone and shale is judged to be weak.

[0098] When the influence of heterogeneous seepage is within the preset threshold range, it indicates that the shale pores are likely to have pore concentration or large pore area, and the seepage effect of the shale is judged to be medium.

[0099] When the influence of heterogeneous seepage is within the preset high threshold range, it indicates that the shale pores are likely to have both pore concentration and large pore area at the same time, and the seepage effect of the shale is judged to be strong.

[0100] As an example, the influence of heterogeneous seepage is normalized by the hyperbolic tangent function, and the preset low threshold range, preset medium threshold range, and preset high threshold range are [0, 1 / 3), [1 / 3, 2 / 3), and [2 / 3, 1), respectively.

[0101] It should be noted that when a scanning electron microscope image contains only one high-permeability area, the distribution distance cannot be calculated. In this case, because the basic conditions for forming a connected permeability network are lacking, the permeability of mudstone and shale is directly judged to be weak.

[0102] It should be noted that in other embodiments of the present invention, the implementer may collect historical sample data and use linear normalization to normalize the influence of heterogeneous seepage. At the same time, based on the manual evaluation of the seepage effect of historical samples (such as dividing it into three levels: weak, medium, and strong), the empirical threshold range of the influence of heterogeneous seepage can be divided.

[0103] In summary, to address the technical problem of inaccurate permeability assessment caused by neglecting the heterogeneity of pore distribution in existing technologies, this invention provides an in-situ analysis method for shale microstructures at the micron-nano scale. This invention first obtains typical microstructures from the microscopic scanning images of the sample to be analyzed; further, it obtains scanning electron microscopy images of the typical microstructures and extracts pore regions; based on the spatial distribution of pore regions within local image blocks, it obtains the local pore density; it analyzes the average width of pore regions within each image block, and combined with the local pore density, obtains the wide-pore permeability influence and filters out high-permeability regions; finally, based on the spatial distribution of high-permeability regions and the wide-pore permeability influence, it classifies the permeability effect of shale. By extracting typical microstructures, quantifying pore distribution density and average width, and fusing them to generate the wide-pore permeability influence, the permeability effect of shale is accurately classified based on the spatial distribution of high-permeability regions, effectively solving the problem of inaccurate permeability assessment caused by neglecting pore heterogeneity in traditional methods.

[0104] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0105] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for in-situ analysis of micro-components of mudstone and shale at the micrometer-nano scale, characterized in that, The method includes: Obtaining typical microscopic components in the vertical layered surface of the sample to be analyzed includes: automatically identifying all microscopic components and their relative percentage content using a pre-trained neural network model, marking microscopic components with a relative percentage content exceeding a preset typical threshold as typical microscopic components; and obtaining the number of markings based on the percentage of typical microscopic components. Scanning electron microscope (SEM) images of typical microscopic components are acquired and pore regions are extracted. The SEM images are then uniformly divided into image blocks using preset segmentation parameters. Based on the spatial distribution of pore regions within each image block and its preset neighborhood, the local pore density of each image block is obtained. The average width of the pore regions within each image block is analyzed, and combined with the local pore density, the influence of wide-pore seepage is obtained, and high-seepage regions are selected. Based on the spatial distribution of high-permeability areas and the influence of wide-pore seepage, the seepage effect of mudstone and shale is classified. The method for obtaining local pore density includes: obtaining the average coordinates of all pore pixels in each image block as the pore center point; and obtaining the local pore density of each image block based on the Euclidean distance between the pore center point of each image block and the pore center points of other image blocks in a preset neighborhood, combined with the proportion of the pore region occupying the image block. The method for classifying the seepage effect of shale includes: obtaining the center coordinates of the high seepage zone with the maximum wide-pore seepage influence, and the Euclidean distance between the center coordinates of the other high seepage zones, as the distribution distance; integrating all distribution distances, the side length of the scanning electron microscope image, and the wide-pore seepage influence of all high seepage zones to obtain the heterogeneous seepage influence; and classifying the seepage effect of shale based on the heterogeneous seepage influence.

2. The in-situ analysis method for micro- and nano-scale microstructures of mudstone and shale according to claim 1, characterized in that, The method for obtaining the typical microscopic components also includes: The geometric center points of the pixel regions of the typical microscopic components are extracted as candidate marker points. A spatial optimization algorithm is used, combined with the number of markers, to uniformly mark the typical microscopic components.

3. The in-situ analysis method for micro- and nano-scale microstructures of mudstone and shale according to claim 2, characterized in that, The method for obtaining the number of markers includes: The ratio of the relative percentage content of each typical micro-component to the constant 0.1 is rounded up, and the rounded result is used as the number of markers.

4. The in-situ analysis method for micro- and nano-scale microstructures of mudstone and shale according to claim 1, characterized in that, The method for obtaining the influence of wide-gap seepage includes: By fusing the side length of the image block, the average width of all the pore regions, and the local pore density, the wide-slot seepage influence degree of the corresponding image block is obtained.

5. The in-situ analysis method for micro- and nano-scale microstructures of mudstone and shale according to claim 1, characterized in that, The method for classifying the seepage effect of shale based on the heterogeneous seepage influence degree includes: When the degree of heterogeneous seepage influence is within a preset low threshold range, the seepage effect of shale is determined to be weak; when the degree of heterogeneous seepage influence is within a preset medium threshold range, the seepage effect of shale is determined to be medium; when the degree of heterogeneous seepage influence is within a preset high threshold range, the seepage effect of shale is determined to be strong.

6. The in-situ analysis method for micro- and nano-scale microstructures of mudstone and shale according to claim 1, characterized in that, The method for obtaining the porous region includes: The OTSU algorithm is used to obtain binary images of scanning electron microscope images, with the black areas representing pore regions.

7. The in-situ analysis method for micro- and nano-scale microstructures of mudstone and shale according to claim 1, characterized in that, The method for obtaining the average width of the pore region includes: The boundary of each pore region is obtained by using the two-sided scanning method. The Euclidean distance between the two farthest points on the boundary is taken as the maximum side length. The ratio of the total number of pixels in the pore region to the maximum side length is taken as the average width of the corresponding pore region.

8. The in-situ analysis method for micro- and nano-scale microstructures of mudstone and shale according to claim 1, characterized in that, The method for obtaining the high permeability region includes: The average value of all the wide-gap seepage influence is obtained as the seepage screening threshold, and the image blocks with wide-gap seepage influence greater than the seepage screening threshold are marked as high seepage areas.

Images