Multi-scale quantitative characterization method for pore structure of coal
By combining multi-scale analysis methods using scanning electron microscopy, computed tomography, and X-ray scattering, the problem of large errors in the characterization of coal pore structure in existing technologies has been solved, enabling multi-scale quantitative characterization of coal pore structure and improving the accuracy and comprehensiveness of information acquisition.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- LIAO NING GONG CHENG JI SHU DA XUE E ER DUO SI YAN JIU YUAN
- Filing Date
- 2024-11-06
- Publication Date
- 2026-05-08
AI Technical Summary
Existing methods for characterizing coal pore structure, such as mercury intrusion porosimetry, nuclear magnetic resonance, and gas adsorption analysis, have limitations. They are difficult to fully demonstrate the multi-scale characteristics of coal pores, resulting in large measurement errors and an inability to accurately obtain pore structure information.
By combining scanning electron microscopy, computed tomography, and X-ray scattering, macroscopic, micrometer-scale, and nanometer-scale features of coal pore structure are obtained through multi-scale analysis, including techniques such as three-dimensional reconstruction, filtering and noise reduction, and image segmentation, to achieve multi-scale quantitative characterization.
It achieves a comprehensive and accurate characterization of coal pore structure at different scales, breaks through the limitations of single-scale analysis, and improves the accuracy of obtaining coal pore structure feature information.
Smart Images
Figure CN121994668A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of gas extraction and coal engineering, and specifically relates to a multi-scale quantitative characterization method for coal pore structure. Background Technology
[0002] As a complex porous medium, the formation, distribution, and structural characteristics of coal pores are crucial to its gas storage and transportation performance. While some traditional characterization methods, such as mercury intrusion porosimetry, nuclear magnetic resonance (NMR), and gas adsorption analysis, are widely used, they each have limitations and cannot comprehensively demonstrate the multi-scale characteristics of coal pores. Mercury intrusion porosimetry measures pore structure based on the principle of mercury penetrating coal pores under pressure. However, this method assumes that mercury intrusion into pores is an ideal capillary phenomenon. In reality, the physicochemical properties of coal pore surfaces are complex, with surface roughness and uneven wettability. This makes the mercury intrusion process not entirely consistent with the ideal assumptions, and it is difficult to accurately obtain microstructural information such as the connectivity between pores, leading to measurement biases. In nuclear magnetic resonance (NMR), the fluids (such as water and gas) in different pores and the coal matrix itself affect the NMR signal. The molecular motion states of fluids in pores of different sizes vary greatly, making accurate interpretation of NMR signals extremely difficult, resulting in significant errors in the quantitative analysis of pore structure. Commonly used adsorption models in gas adsorption methods are based on idealized assumptions and differ significantly from the coal structure, thus also leading to substantial errors in analysis. Therefore, there is an urgent need for a method that integrates multiple advanced analytical techniques to comprehensively understand the pore structure and characteristics of coal. Summary of the Invention
[0003] To address the shortcomings of existing technologies, this application proposes a multi-scale quantitative characterization method for coal pore structure. This method provides a detailed quantitative characterization of the pore characteristics of coal structure at multiple scales, including macroscopic, micrometer, and nanometer scales, thereby enabling efficient gas extraction and rational utilization of coal energy.
[0004] This application proposes a multi-scale quantitative characterization method for coal pore structure, including:
[0005] Collect coal samples and preprocess them;
[0006] The pretreated coal samples were scanned using a scanning electron microscope to obtain macroscopic images of the coal samples;
[0007] The pre-processed coal samples were scanned using a computed tomography scanner to obtain micron-level images of the coal samples;
[0008] The pretreated coal samples were scanned using an X-ray scattering instrument to obtain nanoscale images of the coal samples;
[0009] Extracting macroscopic features of the pore structure of coal samples from macroscopic images of coal samples;
[0010] Based on macroscopic and micron-level images of coal samples, micron-level features of the pore structure of coal samples are extracted.
[0011] Nanoscale features of the pore structure of coal samples are extracted from nanoscale images of coal samples. The macroscopic features, micrometer-level features, and nanoscale features of the pore structure of coal samples are used as the results of multi-scale quantitative characterization of the pore structure of coal samples.
[0012] The step of extracting micron-level features of the pore structure of coal samples based on macroscopic and micron-level images includes:
[0013] A reconstruction algorithm was used to perform 3D reconstruction of macroscopic and micron-level images of coal samples to obtain 3D reconstructed images.
[0014] Median filtering was used to filter and reduce noise in the 3D reconstructed image to obtain the filtered 3D image.
[0015] Image segmentation was performed on the filtered 3D image to obtain the micron-level features of the coal sample's pore structure.
[0016] The method employs a reconstruction algorithm to perform three-dimensional reconstruction of macroscopic and micron-level images of coal samples, resulting in three-dimensional reconstructed images, including:
[0017] The projection parameters were calculated in both the macroscopic image and the micrometer-level image of the coal sample to obtain the first projection parameter and the second projection parameter.
[0018] The first projection parameter and the second projection parameter are weighted and summed to obtain the final projection parameter.
[0019] The final projection parameters are then back-projected to reconstruct the image, resulting in a 3D reconstructed image.
[0020] The step of performing image segmentation on the filtered 3D image to obtain micron-level features of the coal sample's pore structure includes:
[0021] Image segmentation is performed on the filtered 3D image to obtain pore differentiation results, coal matrix differentiation results, and mineral differentiation results;
[0022] Pore spatial characteristics were analyzed based on the pore differentiation results, coal matrix differentiation results, and mineral differentiation results to obtain representative volume units, pore volume fraction, pore size distribution, and porosity.
[0023] Based on the results of pore differentiation, coal matrix differentiation, and mineral differentiation, the maximum sphere algorithm is adopted to characterize the pore structure using an interconnected equivalent pore network model. In the interconnected equivalent pore network model, pore structures with pores larger than a preset threshold are equivalent to spheres, and pore structures with pores smaller than or equal to the preset threshold are equivalent to throats, thus obtaining pore-throat characteristics and coordination numbers.
[0024] The micron-level characteristics of the pore structure of the coal sample include: representative volume unit, pore volume fraction, pore size distribution, porosity, pore throat characteristics, and coordination number.
[0025] The macroscopic features of the pore structure of the coal sample include pore connectivity and fracture expansion.
[0026] The nanoscale characteristics of the pore structure of the coal sample include: scattering intensity, radius of gyration, pore distribution, and fractal dimension.
[0027] Beneficial effects:
[0028] This application proposes a multi-scale quantitative characterization method for coal pore structure. This innovative method combines three different techniques—SEM (Scanning Electron Microscopy), CT (Computed Tomography), and SAXS (Small Angle X-ray Scattering)—to complement and validate each other, enabling quantitative characterization of coal pore structure at multiple scales, including macroscopic, micrometer, and nanometer scales. This multi-scale comprehensive analysis approach overcomes the limitations of traditional methods that focus only on a single scale, and can comprehensively and accurately obtain characteristic information of coal pore structure at different scales. Attached Figure Description
[0029] Figure 1 A flowchart of a multi-scale quantitative characterization method for coal pore structure according to an embodiment of this application;
[0030] Figure 2 A schematic diagram of the multi-scale quantitative characterization method for coal pore structure according to an embodiment of this application;
[0031] Figure 3 A schematic diagram of a CT scanning system according to an embodiment of this application;
[0032] Figure 4 A schematic diagram of the physical structure of the three-dimensional tomographic imaging system according to an embodiment of this application;
[0033] Figure 5 The principle of the FDK reconstruction algorithm in this application embodiment;
[0034] Figure 6 The coal sample pore network model of this application embodiment. Detailed Implementation
[0035] The specific implementation methods of this application will be further described in detail below with reference to the accompanying drawings and embodiments.
[0036] Example:
[0037] This application proposes a multi-scale quantitative characterization method for coal pore structure, such as... Figure 1 , Figure 2 As shown, it includes:
[0038] Step S1: Collect coal samples and preprocess them;
[0039] In this embodiment, coal samples are collected, then dried, ground, and cut into blocks, columns, or powder.
[0040] Step S2: Scan the pretreated coal sample using a scanning electron microscope to obtain a macroscopic image of the coal sample;
[0041] In this embodiment, the processed coal sample is fixed on the test frame, and the surface of the coal sample is observed at high resolution using SEM (Scanning Electron Microscope). This allows for observation of the surface morphology of the coal sample and provides a macroscopic understanding of the characteristics of the coal sample surface, such as the connectivity of pores and the propagation of cracks.
[0042] Step S3: Scan the pre-processed coal sample using a computed tomography scanner to obtain micron-level images of the coal sample;
[0043] In this embodiment, to obtain the micron-level pore structure characteristics of the coal sample, X-ray computed tomography (CT) is performed on the coal sample on the test stand. The X-rays emitted by the X-ray source attenuate after penetrating the rotating sample, are then received and processed by the detector as an electrical signal. This electrical signal is converted into a digital signal and then displayed as a CT image of the object under test using the image processing software Avizo, thus obtaining a micron-level image of the coal sample. X-ray computed tomography generally consists of an X-ray source, a turntable, an X-ray detector, and a computer and software system. Figure 3 As shown, in CT images, each pixel has a corresponding gray value, thus forming a gray value matrix.
[0044] Step S4: The pretreated coal sample is scanned using an X-ray scattering instrument to obtain nanoscale images of the coal sample;
[0045] In this embodiment, in order to obtain the nanoscale pore structure characteristics of the coal sample, small-angle X-ray scattering (SAXS) was used to test the pore structure of the coal powder. The nanoscale pore structure of the coal sample was quantitatively analyzed by analyzing parameters such as scattering intensity, radius of gyration, pore distribution, and fractal dimension obtained from the data.
[0046] Step S5: Extract the macroscopic features of the pore structure of the coal sample from the macroscopic image of the coal sample;
[0047] Step S6: Based on the macroscopic image and micron-level image of the coal sample, extract the micron-level features of the pore structure of the coal sample;
[0048] Step S7: Extract the nanoscale features of the pore structure of the coal sample from the nanoscale image of the coal sample, and use the macroscopic features, micrometer-level features, and nanoscale features of the pore structure of the coal sample as the results of multi-scale quantitative characterization of the pore structure of the coal sample.
[0049] In this embodiment, the micron-level characteristics of the coal sample pore structure include: representative volume unit, pore volume fraction, pore size distribution, porosity, pore throat characteristics, and coordination number. The macroscopic characteristics of the coal sample pore structure include: pore connectivity and fracture propagation. The nanoscale characteristics of the coal sample pore structure include: scattering intensity, radius of gyration, pore distribution, and fractal dimension.
[0050] Scattering intensity refers to the sample's ability to scatter X-rays, reflecting the degree of response of the sample's internal structure to X-rays. Scattering intensity is related to the difference in electron density distribution within the sample; the greater the difference in electron density between different regions within the sample, the stronger the scattering intensity. Radius gyration is a parameter describing particle shape; it represents the root-mean-square distance between each point within the particle and its center of mass, reflecting the particle's size and shape. In SAXS experiments, the radius of gyration can be calculated using the Guinier approximation, thus inferring the particle's size. A larger radius of gyration indicates a more complex pore shape and spatial structure. Fractal dimension is a parameter describing the characteristics of complex structures, reflecting the degree of structural complexity. In SAXS experiments, the fractal dimension can be calculated by analyzing the relationship between scattering intensity and scattering angle, thus inferring the sample's internal structural characteristics. A larger pore fractal dimension indicates a more complex pore structure.
[0051] The step of extracting micron-level features of the pore structure of coal samples based on macroscopic and micron-level images includes:
[0052] Step S6.1: Use a reconstruction algorithm to perform three-dimensional reconstruction on the macroscopic image and the micron-level image of the coal sample to obtain a three-dimensional reconstructed image;
[0053] In this embodiment, the boundary information of each substance in the test sample may not be clear in the two-dimensional image, which will affect the quantitative analysis based on the image. Therefore, using the FDK (Feldkamp-Davis-Kress) reconstruction algorithm to perform three-dimensional reconstruction of the two-dimensional images obtained from CT scans and SEM can more clearly reflect the true structure of the coal sample's pore space. After the three-dimensional reconstruction of the coal sample's pore space structure, Avizo software is used for three-dimensional display.
[0054] Step S6.2: Median filtering is used to filter and reduce noise in the 3D reconstructed image to obtain the filtered 3D image;
[0055] In this embodiment, due to the presence of image noise in the CT scan images, the boundaries of pores and coal matrix are relatively blurred, reducing image quality. To improve the accuracy of data analysis, median filtering is used to filter and reduce noise in the scanned images. Common noise reduction methods include mean filtering, median filtering, and Gaussian filtering. Mean filtering cannot effectively preserve image details, Gaussian filtering is suitable for noise that follows a normal distribution, while median filtering can smooth noise, ensuring clear image boundaries and preserving the image's contour features.
[0056] Step S6.3: Perform image segmentation on the filtered 3D image to obtain the micron-level features of the coal sample's pore structure.
[0057] In this embodiment, the CT scan images of the coal sample can clearly distinguish the different components inside the coal, mainly including the coal matrix, pores, and minerals. Since the goal is to study the pore characteristics of the coal structure, image segmentation is required to clearly distinguish the pores, coal matrix, and minerals on the CT scan images.
[0058] In this embodiment, the image segmentation uses a thresholding method. Thresholding methods are divided into single-threshold methods and multi-threshold methods. The basic principle of single-threshold image segmentation is as follows:
[0059] In an image of size M×N with L gray levels, f(x, y) represents the gray level of the pixel at coordinates (x, y), where x∈[1, M], y∈[1, N]. Single threshold segmentation involves determining a threshold t and mapping the gray levels of all pixels accordingly.
[0060]
[0061] The segmented image has only two gray levels, 0 and L-1, so this method is suitable for image segmentation with significantly different gray levels.
[0062] The principle of multi-threshold segmentation is similar. Let the number of thresholds be n, then the mapping of all pixel gray levels is:
[0063]
[0064] Where, l0, l1, ..., l n For the n+1 gray levels after image segmentation, t1…t n These are the preset thresholds for different gray levels.
[0065] For targets with different gray levels, the multi-threshold segmentation method works well. Since the study focuses on the porosity characteristics of coal, a binary method is used to segment the image and extract the pores.
[0066] The method employs a reconstruction algorithm to perform three-dimensional reconstruction of macroscopic and micron-level images of coal samples, resulting in three-dimensional reconstructed images, including:
[0067] Step S6.1.1: Calculate the projection parameters in the macroscopic image and the micron-level image of the coal sample respectively to obtain the first projection parameter and the second projection parameter;
[0068] Step S6.1.2: Perform a weighted summation of the first projection parameters and the second projection parameters to obtain the final projection parameters;
[0069] Step S6.1.3: Perform backprojection reconstruction on the final projection parameters to obtain a 3D reconstructed image.
[0070] The principle of the FDK reconstruction algorithm is as follows:
[0071] Figure 4 This is a schematic diagram of the physical structure of a three-dimensional tomographic imaging system, where the distance from the radiation source to the rotation axis is d, and the distance from the radiation source to the detector plane is D = d + d'. Figure 5 This is a schematic diagram of the FDK reconstruction algorithm. The detector system is represented here by the projection a-a' onto a straight line that passes through the origin and is parallel to the actual detector line. The intersection of the coordinate axes and the detector is taken as the origin. The rotation angle of the object under test is Φ, the value of the specified detection point along the detector plane is Y, the distance from the radiation source to the rotation axis is d, and the perpendicular distance from the origin to the intersecting ray detector plane is l.
[0072] l=Yd / (d 2 +Y 2 ) 1 / 2
[0073] The angle ξ between the x-axis and the perpendicular line:
[0074] ξ = Φ + π / 2 + α
[0075] in:
[0076] α = tan -1 (Y / d)=tan -1 [l / (d 2-l 2 ) 1 / 2 ]
[0077] The projection value is represented by p(l,ξ) in cylindrical coordinates. After Fourier transform, we finally obtain:
[0078]
[0079] Where f(r,ξ) is the projected value of the object density distribution function in the 3D space to be reconstructed, φ is the angle related to the projection geometry, ω is the frequency variable, and P is used to represent the frequency component in the Fourier transform. Φ (Y) Projection data, used to describe the data collected on the plane, the position vector in the Cartesian coordinate system, representing a specific position inside the object, and the final reconstruction result is in complex form, where Re (real part) represents the specific value.
[0080] In step S6.3, the image segmentation of the filtered three-dimensional image to obtain the micron-level features of the coal sample's pore structure includes:
[0081] Step S6.3.1: Perform image segmentation on the filtered 3D image to obtain pore differentiation results, coal matrix differentiation results, and mineral differentiation results;
[0082] Step S6.3.2: Perform pore space characteristic analysis on the pore differentiation results, coal matrix differentiation results, and mineral differentiation results to obtain representative volume units, pore volume fraction, pore size distribution, and porosity;
[0083] In this embodiment, to study the spatial structure of coal sample pores, a representative elementary volume (REV) is extracted, and the porosity of different voxels is calculated (where voxel, short for volume pixel, is the smallest unit in three-dimensional spatial segmentation, similar to a pixel in a two-dimensional image). The final REV voxel is selected considering that the porosity of the coal sample tends to stabilize with changes in voxels. Since pores in coal are mostly irregular in shape, the volume of the pores can be determined in image processing software based on the voxels they occupy. Combined with the sphere volume calculation formula, the equivalent diameter of the pores can be calculated, thus obtaining the pore volume fraction, pore size distribution, and porosity.
[0084] The formula for calculating the volume of a sphere is as follows:
[0085]
[0086] Among them, D Eq V is the equivalent diameter of the pores, in μm. p The actual volume of the pores, in μm 3 .
[0087] The Labeling module was used to label the pores in the coal samples. Based on the pore labeling results, the Label and Analysis module was used to perform statistical analysis on the equivalent diameter, pore volume and pore surface area of the pores.
[0088] Step S6.3.3: Based on the pore differentiation results, coal matrix differentiation results, and mineral differentiation results, the maximum sphere algorithm is used to characterize the pore structure using an interconnected equivalent pore network model. In the interconnected equivalent pore network model, pore structures with pores larger than a preset threshold are equivalent to spheres, and pore structures with pores smaller than or equal to the preset threshold are equivalent to throats, thus obtaining pore-throat characteristics and coordination numbers.
[0089] In this embodiment, to further analyze the connectivity characteristics of the pores, based on the maximum sphere algorithm, the complex pore structure can be represented by an interconnected equivalent pore network model (PNM). In this model, larger pores can be equivalent to spheres, while smaller pores are equivalent to throats. The coordination number of the pores is analyzed. The principle of the interconnected equivalent pore network model is as follows: Figure 6 As shown in the diagram, in the left figure: holes x and y represent spheres, and hole t represents a throat; in the right figure: spheres of different colors represent different pores, and rods of different diameters represent different throats. The equivalent length of a throat is the distance between two larger pores; the radius of the smaller pore represents the equivalent radius of the throat; the pore coordination number refers to the number of pores connected to a given pore. The coordination number reflects the connectivity of the pore network; the larger the coordination number, the better the connectivity of the pores.
[0090] This embodiment proposes a multi-scale quantitative characterization method for coal pore structure, abandoning the traditional mindset of single-technology or single-scale analysis. Previously, researchers often focused on a single characterization method, attempting to improve the accuracy of coal pore structure characterization by refining that method, but with limited results. This embodiment innovatively combines three different technologies—SEM (Scanning Electron Microscopy), CT (Computed Tomography), and SAXS (Small Angle X-ray Scattering)—to complement and verify each other, quantitatively characterizing coal pore structure at multiple scales, including macroscopic, micrometer, and nanometer scales. This multi-scale comprehensive analysis approach breaks through the limitations of traditional methods that only focus on a single scale, enabling the comprehensive and accurate acquisition of characteristic information of coal pore structure at different scales.
[0091] The various embodiments in this application 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.
[0092] The scope of protection of this application is not limited to the embodiments described above. Obviously, those skilled in the art can make various modifications and variations to this disclosure without departing from the scope and spirit of this disclosure. If such modifications and variations fall within the scope of the claims of this disclosure and their equivalents, then the intent of this disclosure also includes such modifications and variations.
Claims
1. A multi-scale quantitative characterization method for coal pore structure, characterized in that, include: Collect coal samples and preprocess them; The pretreated coal samples were scanned using a scanning electron microscope to obtain macroscopic images of the coal samples; The pre-processed coal samples were scanned using a computed tomography scanner to obtain micron-level images of the coal samples; The pretreated coal samples were scanned using an X-ray scattering instrument to obtain nanoscale images of the coal samples; Extracting macroscopic features of the pore structure of coal samples from macroscopic images of coal samples; Based on macroscopic and micron-level images of coal samples, micron-level features of the pore structure of coal samples are extracted. Nanoscale features of the pore structure of coal samples are extracted from nanoscale images of coal samples. The macroscopic features, micrometer-level features, and nanoscale features of the pore structure of coal samples are used as the results of multi-scale quantitative characterization of the pore structure of coal samples.
2. The multi-scale quantitative characterization method for coal pore structure according to claim 1, characterized in that, The step of extracting micron-level features of the pore structure of coal samples based on macroscopic and micron-level images includes: A reconstruction algorithm was used to perform 3D reconstruction of macroscopic and micron-level images of coal samples to obtain 3D reconstructed images. Median filtering was used to filter and reduce noise in the 3D reconstructed image to obtain the filtered 3D image. Image segmentation was performed on the filtered 3D image to obtain the micron-level features of the coal sample's pore structure.
3. The multi-scale quantitative characterization method for coal pore structure according to claim 2, characterized in that, The method employs a reconstruction algorithm to perform three-dimensional reconstruction of macroscopic and micron-level images of coal samples, resulting in three-dimensional reconstructed images, including: The projection parameters were calculated in both the macroscopic image and the micrometer-level image of the coal sample to obtain the first projection parameter and the second projection parameter. The first projection parameter and the second projection parameter are weighted and summed to obtain the final projection parameter. The final projection parameters are then back-projected to reconstruct the image, resulting in a 3D reconstructed image.
4. The multi-scale quantitative characterization method for coal pore structure according to claim 2, characterized in that, The step of performing image segmentation on the filtered 3D image to obtain micron-level features of the coal sample's pore structure includes: Image segmentation is performed on the filtered 3D image to obtain pore differentiation results, coal matrix differentiation results, and mineral differentiation results; Pore spatial characteristics were analyzed based on the pore differentiation results, coal matrix differentiation results, and mineral differentiation results to obtain representative volume units, pore volume fraction, pore size distribution, and porosity. Based on the results of pore differentiation, coal matrix differentiation, and mineral differentiation, the maximum sphere algorithm is adopted to characterize the pore structure using an interconnected equivalent pore network model. In the interconnected equivalent pore network model, pore structures with pores larger than a preset threshold are equivalent to spheres, and pore structures with pores smaller than or equal to the preset threshold are equivalent to throats, thus obtaining pore-throat characteristics and coordination numbers.
5. The multi-scale quantitative characterization method for coal pore structure according to claim 1, characterized in that, The micron-level characteristics of the pore structure of the coal sample include: representative volume unit, pore volume fraction, pore size distribution, porosity, pore throat characteristics, and coordination number.
6. The multi-scale quantitative characterization method for coal pore structure according to claim 1, characterized in that, The macroscopic features of the pore structure of the coal sample include pore connectivity and fracture expansion.
7. The multi-scale quantitative characterization method for coal pore structure according to claim 1, characterized in that, The nanoscale characteristics of the pore structure of the coal sample include: scattering intensity, radius of gyration, pore distribution, and fractal dimension.