Simulation method for micro-pore structure evolution in crushing process of porous fragile particles

Abstract

The invention belongs to the technical field of granular materials, and relates to a method for simulating micro-pore structure evolution in the crushing process of porous fragile particles. Comprising the following steps of sample preparation, high-resolution X-CT scanning, image processing to obtain an analysis body, three-dimensional reconstruction to generate a three-dimensional rendering model, multi-scale simulation to establish the model, pore extraction, porosity analysis, pore connectivity analysis, connected porosity analysis and pore structural parameter analysis. On the basis of an X-CT technology and AVIZO software, high-resolution tomography data are utilized, the denoising, segmentation and modeling functions of the AVIZO software are combined, pore structure changes in the particle crushing process are accurately extracted, and through three-dimensional reconstruction and pore structure analysis of single particles in the micro scale and overall simulation of particle group crushing behaviors in the micro scale, the crushing behavior of the particle groups in the micro scale is analyzed. And dynamically quantifying the porosity and the evolution law of key parameters of the porosity, and establishing a quantitative correlation model of the crushing degree and the change of the porosity.

Description

Technical Field

[0001] The present invention belongs to the technical field of granular materials, and specifically relates to a method for simulating the evolution of microscopic pore structure during the crushing process of porous and fragile particles, which is suitable for analyzing the porosity change and pore structure evolution caused by particle crushing. Background Art

[0002] Porous and fragile rock and soil granular materials, such as carbonate granular materials, volcanic rock granular materials, moraine granular materials, expansive soil granular materials, etc., are widely present in nature and engineering fields. For example, carbonate particles represented by coral coarse particles are commonly found in tropical and subtropical coastal areas and terrestrial sedimentary strata. They have a unique biological framework structure, so they are rich in internal pores and fragile skeletons. This type of material is usually prone to particle crushing due to external forces such as earthquakes, collapses, and floods in the natural environment due to its high porosity and low compressive strength, which in turn causes significant evolution of the pore structure, ultimately leading to weakening of mechanical properties and changes in permeability characteristics. In the field of engineering, taking materials according to local conditions can minimize economic costs and reduce environmental pollution. Porous and fragile rock and soil granular materials existing in different environments are often important components of local strata. In reality, these materials are widely used in projects such as embankments, roadbeds, and fills. However, they are prone to particle crushing under engineering loads, and changes in pore structure during the crushing process often become a key factor affecting engineering stability and safety. Therefore, studying the pore structure evolution of this type of material is of great significance for understanding its performance evolution and optimizing engineering applications.

[0003] At present, the technical methods for studying the pore structure evolution of porous and fragile materials mainly include scanning electron microscopy (SEM), mercury intrusion (MIP) and nuclear magnetic resonance imaging (NMR). Among them, SEM technology can finely observe the pore morphology and microscopic characteristics of the particle surface, with high resolution and suitable for characterizing the local geometric details of the pores, but its field of view is small, and it can only observe the local area of ​​the sample, which cannot reflect the overall pore structure characteristics, and can only be used for static monitoring. MIP technology can quantify the total porosity and pore size distribution by measuring the penetration characteristics of mercury under different pressures, and is particularly suitable for describing the macroscopic pore characteristics of the material. However, its testing process is destructive, causing irreversible damage to the pore structure of the sample, and it cannot capture the dynamic changes of the pore structure during the particle crushing process. NMR technology uses magnetic resonance signals to non-destructively detect the pore distribution inside the sample, which has the advantages of non-destructive and high efficiency, but the resolution is low, and it is difficult to accurately characterize complex pore networks and multi-scale pore structures. In general, these traditional techniques are mainly limited to static structural analysis and lack the ability to continuously monitor the dynamic evolution of porosity during particle crushing. They are also unable to reveal the correlation mechanism between pore structure changes and material mechanical properties and permeability behavior, which limits their application potential in dynamic process research.

[0004] In recent years, X-ray computed tomography (X-CT) technology has shown significant advantages in the study of porous materials, and can achieve high-resolution three-dimensional non-destructive reconstruction. However, the existing research on the evolution of pore structure during the crushing of porous and fragile rock and soil granular materials is still mainly based on static analysis, lacking a systematic framework for multi-scale dynamic monitoring and correlation modeling. This makes it an important issue to be solved in current research on how to accurately characterize the dynamic changes of pore structural parameters such as porosity, pore equivalent diameter and pore connectivity during particle crushing, and further reveal their influence on mechanical behavior and permeability. Summary of the invention

[0005] Based on the deficiencies in the prior art research on the evolution of porosity during the crushing of porous particles, the present invention provides a simulation method for the evolution of microscopic pore structure during the crushing of porous and fragile particles, aiming to solve the problem in the prior art that it is difficult to achieve continuous dynamic monitoring and quantitative analysis of porosity during particle crushing.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions:

[0007] A method for simulating the evolution of microscopic pore structure during the crushing process of porous and fragile particles comprises the following steps:

[0008] 1. Sample preparation: Dry the sample in an oven at 110°C. When the moisture content is 0%, take out the dried sample and pass it through a sieve with a pore size of 2mm to select particles with a diameter of d>2mm. Use a fine brush to remove dust and impurities on the surface of the sample to obtain the sample to be tested;

[0009] 2. X-CT scanning: stick the sample on the test bench of a high-resolution X-CT scanner with double-sided tape, make the long axis of the sample perpendicular to the test bench, and use an X-CT scanner to scan the sample with a scanning voltage of 120 kV, a current of 150 μA, and a resolution of 20 μm to obtain a particle tomographic image data file in TIFF format. The resolution of the TIFF format file must exceed 1000 pixels;

[0010] 3. Image processing: Import the TIFF format file into AVIZO (2022 version) software, use the "Non-LocalMeans" algorithm to remove the noise generated during the scanning process, cut the image area through the "Volume Edit" function, remove the redundant space voxels outside the solid interface, and obtain the analysis volume;

[0011] 4. 3D reconstruction: Use the "Thresholding" tool on the analysis volume obtained in step 3 to set a threshold to separate the pores from the particle body; use the "Watershed" algorithm to further refine the boundaries of the pore area; use the "VolumeRendering" function to generate a 3D rendering model of the particle body based on the processed image sequence;

[0012] 5. Multi-scale simulation: To simulate the reduction of particle volume caused by particle crushing, the initial 3D model in step 4 is divided into multiple volume bodies of different scales through the "Extract Subvolume" function, and multiple 3D rendering models of different scale bodies are established through the "VolumeRendering" function;

[0013] 6. Pore extraction: For the three-dimensional rendering models of the multiple main bodies of different scales obtained in step 5, use the "Thresholding" tool to perform pore threshold segmentation, and obtain the three-dimensional pore rendering model through the "Volume Rendering" function;

[0014] 7. Porosity analysis: Use the "Volume Fraction" module to calculate the total porosity of the model and the surface ratio of the two-dimensional slice. The calculation method is: total porosity (Φ v ) is the voxel volume (V pore ), and the voxel volume occupied by each subject model in step 5 (V model ) to get the percentage between: Φ v The closer it is to 1, the greater the porosity of the model, and the closer it is to 0, the smaller the porosity of the model; the porosity of the two-dimensional slice (Φ s ) is the pore area of ​​the slice (S pore ), and the total area of ​​the slice (S s ) percentage to get: φ s The closer it is to 1, the greater the face rate of the slice, and the closer it is to 0, the smaller the face rate of the slice;

[0015] 8. Pore connectivity analysis: Use the "Axis Connectivity" function to perform connectivity analysis on the pore 3D rendering model of each volume scale in step 6. If "Data Info" in the analysis result shows "min-max: 0...1", it means that the corresponding model has connected pores, that is, connectivity exists; if "Data Info" in the analysis result shows "min-max: 0...0", it means that the corresponding model does not have connected pores, that is, connectivity does not exist;

[0016] 9. Connected porosity analysis: If there are connected pores in the 3D pore rendering model in step 8, the connected porosity is further calculated based on the connected pore data volume obtained in step 8 through the "Volume Rendering" function. The calculation method is: use the voxel volume (V tp ), and the voxel volume occupied by the 3D rendering model of each subject in step 5 (V model ) percentage to obtain the interconnected porosity (φ t ): φ t The closer it is to 1, the stronger the connectivity of the model, and the closer it is to 0, the weaker the connectivity of the model;

[0017] 10. Pore structural parameter analysis: Based on the 3D rendering model of pores at each volume scale obtained in step 6, the “Label Analysis” function is used to calculate the pore structural parameters according to the output results. The pore structural parameters are the volume of a single pore (V pi ), equivalent diameter (D e ), shape factor (G p ). The calculation method is: pore volume (V pi ) is directly derived from the volume of the space voxel occupied by a single pore; the equivalent diameter (D e ) is determined by the volume of a single pore (V pi ) formula yields: Shape factor (G p ) is defined by the pore cross-sectional area (S) and the pore perimeter (P):

[0018] Preferably, in step 1, the particle size of the porous rock and soil granular material is recommended to be 20-40 mm.

[0019] Preferably, in step 2, the high-resolution X-CT scanner is recommended to have a resolution of 10-30 μm.

[0020] Preferably, in step 5, the initial three-dimensional model is divided into its volume Scale, that is scale, further, the volume scale is Scale, that is The three-dimensional model of the scale needs to be taken from the volume scale Inside the three-dimensional model, and so on, the volume scale is Right now The three-dimensional model of the scale needs to be taken from the volume scale Inside the three-dimensional model, the volume scale is Right now The three-dimensional model needs to be taken from a volume scale of The step-by-step scale division method ensures that the simulation is based on the same fragmented block, thus ensuring the accuracy of the analysis.

[0021] Preferably, in step 7, further, using the obtained initial subject model Total porosity at volume scale (φ v ), establish φ v The corresponding relationship curves with different scale volumes are obtained to obtain the influence of multi-scale particle crushing on the total porosity, and to find out whether there is a scale effect in the invention method. The specific method is: for homogeneous granular materials, under ideal conditions, φ at volume scale v Should be consistent.

[0022] Preferably, in step 9, further, using the obtained initial subject model The connected porosity (φ t ), establish the connected porosity (φ t ) and the corresponding curves of different scale volumes, and then the changes in the pore structure connectivity during the multi-scale crushing of particles are obtained.

[0023] Preferably, in step 10, further, using the single pore volume (V pi ), equivalent diameter (D e ), shape factor (G p ), establish the pore volume, equivalent diameter distribution frequency diagram, and average equivalent diameter (D e ), shape factor (G p ) The change curve between it and the volume at each scale is used to achieve the purpose of the effect of particle crushing on the pore structure.

[0024] Based on X-CT technology and AVIZO software, this paper proposes a simulation method for the evolution of microscopic pore structure during the crushing process of porous and fragile particles. By using high-resolution tomography data, combined with the denoising, segmentation and modeling functions of AVIZO software, the pore structure changes during the particle crushing process are accurately extracted. Through the three-dimensional reconstruction and pore structure analysis of single particles at the microscale, and the overall simulation of the crushing behavior of particle groups at the microscale, the evolution law of porosity and its key parameters (such as pore size distribution, connectivity, etc.) is dynamically quantified, and a quantitative correlation model between the degree of crushing and the change of porosity is established, providing data support for the performance prediction and engineering application of porous granular materials.

[0025] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0026] 1. The combination of X-CT technology and AVIZO software can realize high-precision non-destructive monitoring of the dynamic changes of porosity during particle crushing, without the need for macroscopic mechanical tests, which improves work efficiency and provides high accuracy of test results.

[0027] 2. The present invention establishes a multi-scale simulation framework of microscopic scale (pore evolution inside a single particle) and microscopic scale (overall behavior of a particle group), which can comprehensively characterize the effect of particle crushing on porosity and other key parameters (such as pore size distribution, connectivity, shape factor, etc.), with the advantages of clear principles and simultaneous acquisition of multiple test parameters;

[0028] 3. The process of the present invention is standardized, automated, easy to operate, the principle of the method is clear, the operation is simple, and the cost is low. Researchers do not need to master complex professional knowledge to use it, which reduces the analysis threshold and labor costs. At the same time, compared with traditional experimental methods, sample preparation is simple and the experimental cost is significantly reduced;

[0029] 4. The present invention has a wide range of applications and is applicable to a variety of porous and fragile granular materials, such as carbonate particles, coral coarse particles, etc., and has significant advantages, especially in engineering research that requires accurate characterization of porosity evolution laws and prediction of particle properties;

[0030] 5. The present invention fills the gap in the existing technology that cannot dynamically quantify the evolution law of particle porosity. By combining a multi-scale framework with high-resolution dynamic analysis, and utilizing X-CT technology in combination with AVIZO software, the three-dimensional pore structure of particles can be non-destructively and rapidly reconstructed, and the multi-scale changes in porosity can be accurately quantified, providing an efficient and reliable solution for the performance research and engineering application of porous and fragile granular materials; it provides new tools and new paths for the performance research and engineering application of porous granular materials. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A flow chart of a method for simulating the evolution of microscopic pore structure during the crushing process of porous and fragile particles;

[0032] Figure 2 Schematic diagram of multi-scale simulation in step 5 of a method for simulating the evolution of microscopic pore structure during the crushing process of porous and fragile particles;

[0033] Figure 3 The present invention provides a method for simulating the evolution of microscopic pore structure during the crushing process of porous and fragile particles, and provides a three-dimensional main body model and a three-dimensional pore model at an initial volume scale;

[0034] Figure 4It is a comparative analysis diagram of total porosity variation with particle simulated volume scale of Examples 1, 2, and 3 of a simulation method for the evolution of microscopic pore structure during the crushing process of porous and fragile particles and a comparative test based on mercury intrusion test (MIP);

[0035] Figure 5 A comparative analysis diagram of surface porosity results of two-dimensional slices of simulation methods 1, 2, and 3 of the microscopic pore structure evolution during the crushing process of porous and fragile particles and a comparative test based on scanning electron microscopy (SEM) combined with image digitization technology;

[0036] Figure 6 A comparative analysis diagram of the average pore equivalent diameter versus particle simulated volume scale variation results of embodiments 1, 2, and 3 of a simulation method for the evolution of microscopic pore structure during the crushing process of porous and fragile particles and a comparative test based on scanning electron microscopy (SEM) combined with image digitization technology;

[0037] Figure 7 This is a simulation method for the evolution of microscopic pore structure during the crushing process of porous and fragile particles, and the results of connected porosity changes with particle volume scale in Examples 1, 2, and 3. DETAILED DESCRIPTION

[0038] The following describes in detail the simulation method of the evolution of microscopic pore structure during the crushing process of porous fragile particles in three embodiments and comparative tests of the present invention in conjunction with the accompanying drawings.

[0039] Embodiment 1:

[0040] The test object of this embodiment is a cylindrical carbonate coarse particle (actual picture as shown in Figure 2 ), taken from a stratum at a construction site in Sanya, Hainan Province, and is a common granular material in the local stratum. For ease of understanding, a flow chart of a simulation method for the evolution of microscopic pore structure during the crushing process of a porous and fragile particle in this embodiment is shown in Figure 1 shown.

[0041] according to Figure 1 , 2 It can be seen that a method for simulating the evolution of microscopic pore structure during the crushing process of porous and fragile particles comprises the following steps:

[0042] 1. Sample preparation: Dry the sample in an oven at 110°C. When the moisture content is 0%, take out the dried sample and pass it through a sieve with a pore size of 2mm to select particles with a diameter of d>2mm. Use a fine brush to remove dust and impurities on the surface of the sample to obtain the sample to be tested;

[0043] 2. X-CT scanning: stick the sample on the test bench of a high-resolution X-CT scanner with double-sided tape, make the long axis of the sample perpendicular to the test bench, and use an X-CT scanner to scan the sample with a scanning voltage of 120 kV, a current of 150 μA, and a resolution of 20 μm to obtain a particle tomographic image data file in TIFF format. The resolution of the TIFF format file must exceed 1000 pixels;

[0044] 3. Image processing: Import the TIFF format file into AVIZO (2022 version) software, use the "Non-LocalMeans" algorithm to remove the noise generated during the scanning process, cut the image area through the "Volume Edit" function, remove the redundant space voxels outside the solid interface, and obtain the analysis volume;

[0045] 4. 3D reconstruction: Use the "Thresholding" tool to set the threshold to separate the pores from the particle body. Use the "Watershed" algorithm to further refine the boundaries of the pore area. Use the "VolumeRendering" function to generate a 3D rendering model of the particle body based on the processed image sequence, such as Figure 3 As shown;

[0046] 5. Multi-scale simulation: To simulate the reduction in particle volume caused by particle crushing, the initial 3D model in step 4 is divided into volumetric subvolumes through the “Extract Subvolume” function. scale, the initial 3D model is divided into its volume scale, further, the volume scale is The three-dimensional model needs to be taken from a volume scale of Inside the three-dimensional model, and so on, the volume scale is The three-dimensional model needs to be taken from a volume scale of Inside the three-dimensional model, the volume scale is The three-dimensional model needs to be taken from a volume scale of The three-dimensional model is inside. And four main three-dimensional rendering models are established through the "Volume Rendering" function. The three-dimensional main model under the initial volume is as follows Figure 3 As shown;

[0047] 6. Pore extraction: For the four main 3D rendering models of different scales obtained in step 5, use the "Thresholding" tool to perform pore threshold segmentation, and use the "Volume Rendering" function to obtain a 3D pore rendering model. The 3D pore model under the initial volume is as follows: Figure 3 As shown;

[0048] 7. Porosity analysis: Use the "Volume Fraction" module to calculate the total porosity of the model and the surface ratio of the two-dimensional slice. The calculation method is: total porosity (φ v ) is the voxel volume (V pore ), and the voxel volume occupied by each subject model in step 5 (V model ) to get the percentage between: φ v The closer it is to 1, the greater the porosity of the model, and the closer it is to 0, the smaller the porosity of the model; the porosity of the two-dimensional slice (φ s ) is the pore area of ​​the slice (S pore ), and the total area of ​​the slice (S s ) percentage to get: φ s The closer it is to 1, the greater the face rate of the slice, and the closer it is to 0, the smaller the face rate of the slice. Total porosity at volume scale (φ v ), establish φ v The relationship between the curves of different volume scales and the corresponding Figure 4 As shown; using the two-dimensional slice surface ratio (φ s ) data, establish φ s The relationship between the curve and the number of two-dimensional slices N (from bottom to top) is as follows Figure 5 As shown;

[0049] 8. Pore connectivity analysis: Use the "Axis Connectivity" function to perform connectivity analysis on the pore 3D rendering model of each volume scale in step 6. If "Data Info" in the analysis result shows "min-max: 0...1", it means that the corresponding model has connected pores, that is, connectivity exists; if "Data Info" in the analysis result shows "min-max: 0...0", it means that the corresponding model does not have connected pores, that is, connectivity does not exist;

[0050] 9. Connected porosity analysis: If there are connected pores in the 3D pore rendering model in step 8, the connected porosity is further calculated based on the connected pore data volume obtained in step 8 through the "Volume Rendering" function. The calculation method is: use the voxel volume (V tp ), and the voxel volume occupied by the 3D rendering model of each subject in step 5 (V model ) percentage to obtain the interconnected porosity (Φ t ): Φ tThe closer it is to 1, the stronger the connectivity of the model is, and the closer it is to 0, the weaker the connectivity of the model is. The connected porosity (Φ t ), establish Φ t The relationship between the curves of different volume scales and the corresponding Figure 7 As shown;

[0051] 10. Pore structural parameter analysis: Based on the 3D rendering model of pores at each volume scale obtained in step 6, the “Label Analysis” function is used to calculate the pore structural parameters according to the output results. The pore structural parameters are the volume of a single pore (V pi ), equivalent diameter (D e ), shape factor (G p ). The calculation method is: pore volume (V pi ) is directly derived from the volume of the space voxel occupied by a single pore; the equivalent diameter (D e ) is determined by the volume of a single pore (V pi ) formula yields: Shape factor (G p ) is defined by the pore cross-sectional area (S) and the pore perimeter (P): Using the initial main model obtained Equivalent diameter (D) in volume scale e ), establish the curve relationship between the average equivalent diameter of pores and different volume scales and the corresponding Figure 6 shown.

[0052] The results of this example are shown in Figure 3-7 As shown. Figure 3 In the figure, the three-dimensional model and the schematic diagram of the pore model intuitively show the three-dimensional structure of the cylindrical calcium carbonate coarse particles and the internal pore distribution characteristics. The results show that the simulation results are highly consistent with the actual camera photos, indicating that the model obtained by the simulation method of the present invention has good reduction and intuitiveness. Figure 4 , Figure 6 The results in the paper show that the total porosity (Φ v ) is distributed around 10%, and the average equivalent pore diameter (D e ) are distributed in the range of 10-15μm, and both hardly change with the volume scale of the model, indicating that there is no scale effect in this simulation method. Figure 5 The face ratio of the two-dimensional slice (Φ s ) shows that the pores of the sample are concentrated in the middle of the long axis, which is different from Figure 3 This is consistent with the pore distribution shown in the figure, reflecting the accuracy of the invention. Figure 7Reflects the interconnected porosity (φ t ) increases with the decrease of the simulated volume scale, and the large number of pores in the middle of the sample increase the connectivity due to the internal pores connected to the outside world caused by particle crushing, and its connected porosity changes almost linearly. The multi-scale evolution law of the pore structure during the crushing of porous and fragile particles obtained by the simulation method of the present invention has high accuracy and rationality.

[0053] Embodiment 2:

[0054] The test object of this embodiment is a branch-like coral coarse particle (actual picture as shown in Figure 2 ), taken from a construction site in Sanya, Hainan Province, and is a common granular material in the local strata. For ease of understanding, see the flow chart of this embodiment Figure 1 shown.

[0055] according to Figure 1 , 2 It can be seen that a method for simulating the evolution of microscopic pore structure during the crushing process of porous and fragile particles comprises the following steps:

[0056] 1. Sample preparation: Dry the sample in an oven at 110°C. When the moisture content is 0%, take out the dried sample and pass it through a sieve with a pore size of 2mm to select particles with a diameter of d>2mm. Use a fine brush to remove dust and impurities on the surface of the sample to obtain the sample to be tested;

[0057] 2. X-CT scanning: stick the sample on the test bench of a high-resolution X-CT scanner with double-sided tape, make the long axis of the sample perpendicular to the test bench, and use an X-CT scanner to scan the sample with a scanning voltage of 120 kV, a current of 150 μA, and a resolution of 20 μm to obtain a particle tomographic image data file in TIFF format. The resolution of the TIFF format file must exceed 1000 pixels;

[0058] 3. Image processing: Import the TIFF format file into AVIZO (2022 version) software, use the "Non-LocalMeans" algorithm to remove the noise generated during the scanning process, cut the image area through the "Volume Edit" function, remove the redundant space voxels outside the solid interface, and obtain the analysis volume;

[0059] 4. 3D reconstruction: Use the "Thresholding" tool to set the threshold to separate the pores from the particle body. Use the "Watershed" algorithm to further refine the boundaries of the pore area. Use the "VolumeRendering" function to generate a 3D rendering model of the particle body based on the processed image sequence, such as Figure 3 As shown;

[0060] 5. Multi-scale simulation: To simulate the reduction in particle volume caused by particle crushing, the initial 3D model in step 4 is divided into volumetric subvolumes through the “Extract Subvolume” function. scale, the initial 3D model is divided into its volume scale, further, the volume scale is The three-dimensional model needs to be taken from a volume scale of Inside the three-dimensional model, and so on, the volume scale is The three-dimensional model needs to be taken from a volume scale of Inside the three-dimensional model, the volume scale is The three-dimensional model needs to be taken from a volume scale of The three-dimensional model is inside. And four main three-dimensional rendering models are established through the "Volume Rendering" function. The three-dimensional main model under the initial volume is as follows Figure 3 As shown;

[0061] 6. Pore extraction: For the four main 3D rendering models of different scales obtained in step 5, use the "Thresholding" tool to perform pore threshold segmentation, and use the "Volume Rendering" function to obtain a 3D pore rendering model. The 3D pore model under the initial volume is as follows: Figure 3 As shown;

[0062] 7. Porosity analysis: Use the "Volume Fraction" module to calculate the total porosity of the model and the surface ratio of the two-dimensional slice. The calculation method is: total porosity (Φ v ) is the voxel volume (V pore ), and the voxel volume occupied by each subject model in step 5 (V model ) to get the percentage between: Φ v The closer it is to 1, the greater the porosity of the model, and the closer it is to 0, the smaller the porosity of the model; the porosity of the two-dimensional slice (Φ s ) is the pore area of ​​the slice (S pore ), and the total area of ​​the slice (S s ) percentage to get: Φ s The closer it is to 1, the greater the face rate of the slice, and the closer it is to 0, the smaller the face rate of the slice. Total porosity at volume scale (Φ v ), establish Φ v The relationship between the curves of different volume scales and the corresponding Figure 4 As shown; using the two-dimensional slice surface ratio (Φs ) data, establish Φ s The relationship between the curve and the number of two-dimensional slices N (from bottom to top) is as follows Figure 5 As shown;

[0063] 8. Pore connectivity analysis: Use the "Axis Connectivity" function to perform connectivity analysis on the pore 3D rendering model of each volume scale in step 6. If "Data Info" in the analysis result shows "min-max: 0...1", it means that the corresponding model has connected pores, that is, connectivity exists; if "Data Info" in the analysis result shows "min-max: 0...0", it means that the corresponding model does not have connected pores, that is, connectivity does not exist;

[0064] 9. Connected porosity analysis: If there are connected pores in the 3D pore rendering model in step 8, the connected porosity is further calculated based on the connected pore data volume obtained in step 8 through the "Volume Rendering" function. The calculation method is: use the voxel volume (V tp ), and the voxel volume occupied by the 3D rendering model of each subject in step 5 (V model ) percentage to obtain the interconnected porosity (Φ t ): Φ t The closer it is to 1, the stronger the connectivity of the model is, and the closer it is to 0, the weaker the connectivity of the model is. The connected porosity (Φ t ), establish Φ t The relationship between the curves of different volume scales and the corresponding Figure 7 As shown;

[0065] 10. Pore structural parameter analysis: Based on the 3D rendering model of pores at each volume scale obtained in step 6, the “Label Analysis” function is used to calculate the pore structural parameters according to the output results. The pore structural parameters are the volume of a single pore (V pi ), equivalent diameter (D e ), shape factor (G p ). The calculation method is: pore volume (V pi ) is directly derived from the volume of the space voxel occupied by a single pore; the equivalent diameter (D e ) is determined by the volume of a single pore (V pi ) formula yields: Shape factor (G p ) is defined by the pore cross-sectional area (S) and the pore perimeter (P): Using the initial main model obtained Equivalent diameter (D) in volume scale e ), establish the curve relationship between the average equivalent diameter of pores and different volume scales and the corresponding Figure 6 shown.

[0066] The results of this example are shown in Figure 3 , 4 , 5, 6, and 7. Figure 3 In the figure, the three-dimensional model and the schematic diagram of the pore model intuitively show the three-dimensional structure of the dendritic coral coarse particles and the internal pore distribution characteristics. The results show that the simulation results are highly consistent with the actual camera photos, indicating that the model obtained by the simulation method of the present invention has good reduction and intuitiveness. Figure 4 , Figure 6 The results in the paper show that the total porosity (φ v ) is distributed around 10%, and the average equivalent pore diameter (D e ) are distributed in the range of 10-15μm, and both hardly change with the volume scale of the model, indicating that there is no scale effect in this simulation method. Figure 5 The surface ratio of the two-dimensional slice (φ s ) shows that the pore distribution of the sample is irregular, with multiple peaks, which is consistent with Figure 3 This is consistent with the distribution of pores mainly along their branches, reflecting the accuracy of the invention. Figure 7 Reflects the interconnected porosity (φ t ) increases with the decrease of the simulated volume scale, and the large number of pores existing on the sample branches, due to the internal pores caused by particle crushing, are connected to the outside world, thereby improving the connectivity. The multi-scale evolution law of the pore structure during the crushing of porous and fragile particles obtained by the simulation method of the present invention has high accuracy and rationality.

[0067] Embodiment 3:

[0068] The test object of this embodiment is a calcareous sand coarse particle in a dissolution state (actual picture as shown in FIG. Figure 2 ), taken from a construction site in Sanya, Hainan Province, and is a common granular material in the local strata. For ease of understanding, see the flow chart of this embodiment Figure 1 shown.

[0069] according to Figure 1 , 2 It can be seen that a method for simulating the evolution of microscopic pore structure during the crushing process of porous and fragile particles comprises the following steps:

[0070] 1. Sample preparation: Dry the sample in an oven at 110°C. When the moisture content is 0%, take out the dried sample and pass it through a sieve with a pore size of 2mm to select particles with a diameter of d>2mm. Use a fine brush to remove dust and impurities on the surface of the sample to obtain the sample to be tested;

[0071] 2. X-CT scanning: stick the sample on the test bench of a high-resolution X-CT scanner with double-sided tape, make the long axis of the sample perpendicular to the test bench, and use an X-CT scanner to scan the sample with a scanning voltage of 120 kV, a current of 150 μA, and a resolution of 20 μm to obtain a particle tomographic image data file in TIFF format. The resolution of the TIFF format file must exceed 1000 pixels;

[0072] 3. Image processing: Import the TIFF format file into AVIZO (2022 version) software, use the "Non-LocalMeans" algorithm to remove the noise generated during the scanning process, cut the image area through the "Volume Edit" function, remove the redundant space voxels outside the solid interface, and obtain the analysis volume;

[0073] 4. 3D reconstruction: Use the "Thresholding" tool to set the threshold to separate the pores from the particle body. Use the "Watershed" algorithm to further refine the boundaries of the pore area. Use the "VolumeRendering" function to generate a 3D rendering model of the particle body based on the processed image sequence, such as Figure 3 As shown;

[0074] 5. Multi-scale simulation: To simulate the reduction in particle volume caused by particle crushing, the initial 3D model in step 4 is divided into volumetric subvolumes through the “Extract Subvolume” function. scale, the initial 3D model is divided into its volume scale, further, the volume scale is The three-dimensional model needs to be taken from a volume scale of Inside the three-dimensional model, and so on, the volume scale is The three-dimensional model needs to be taken from a volume scale of Inside the three-dimensional model, the volume scale is The three-dimensional model needs to be taken from a volume scale of The three-dimensional model is inside. And four main three-dimensional rendering models are established through the "Volume Rendering" function. The three-dimensional main model under the initial volume is as follows Figure 3 As shown;

[0075] 6. Pore extraction: For the four main 3D rendering models of different scales obtained in step 5, use the "Thresholding" tool to perform pore threshold segmentation, and use the "Volume Rendering" function to obtain a 3D pore rendering model. The 3D pore model under the initial volume is as follows: Figure 3 As shown;

[0076] 7. Porosity analysis: Use the "Volume Fraction" module to calculate the total porosity of the model and the surface ratio of the two-dimensional slice. The calculation method is: total porosity (φ v ) is the voxel volume (V pore ), and the voxel volume occupied by each subject model in step 5 (V model ) to get the percentage between: φ v The closer it is to 1, the greater the porosity of the model, and the closer it is to 0, the smaller the porosity of the model; the porosity of the two-dimensional slice (Φ s ) is the pore area of ​​the slice (S pore ), and the total area of ​​the slice (S s ) percentage to get: Φ s The closer it is to 1, the greater the face rate of the slice, and the closer it is to 0, the smaller the face rate of the slice. Total porosity at volume scale (φ v ), establish Φ v The relationship between the curves of different volume scales and the corresponding Figure 4 As shown; using the two-dimensional slice surface ratio (φ s ) data, establish Φ s The relationship between the curve and the number of two-dimensional slices N (from bottom to top) is as follows Figure 5 As shown;

[0077] 8. Pore connectivity analysis: Use the "Axis Connectivity" function to perform connectivity analysis on the pore 3D rendering model of each volume scale in step 6. If "Data Info" in the analysis result shows "min-max: 0...1", it means that the corresponding model has connected pores, that is, connectivity exists; if "Data Info" in the analysis result shows "min-max: 0...0", it means that the corresponding model does not have connected pores, that is, connectivity does not exist;

[0078] 9. Connected porosity analysis: If there are connected pores in the 3D pore rendering model in step 8, the connected porosity is further calculated based on the connected pore data volume obtained in step 8 through the "Volume Rendering" function. The calculation method is: use the voxel volume (V tp ), and the voxel volume occupied by the 3D rendering model of each subject in step 5 (V model ) percentage to obtain the interconnected porosity (Φ t ): Φ t The closer it is to 1, the stronger the connectivity of the model is, and the closer it is to 0, the weaker the connectivity of the model is. The connected porosity (Φ t ), establish φ t The relationship between the curves of different volume scales and the corresponding Figure 7 As shown;

[0079] 10. Pore structural parameter analysis: Based on the 3D rendering model of pores at each volume scale obtained in step 6, the “Label Analysis” function is used to calculate the pore structural parameters according to the output results. The pore structural parameters are the volume of a single pore (V pi ), equivalent diameter (D e ), shape factor (G p ). The calculation method is: pore volume (V pi ) is directly derived from the volume of the space voxel occupied by a single pore; the equivalent diameter (D e ) is determined by the volume of a single pore (V pi ) formula yields: Shape factor (G p ) is defined by the pore cross-sectional area (S) and the pore perimeter (P): Using the initial main model obtained Equivalent diameter (D) in volume scale e ), establish the curve relationship between the average equivalent diameter of pores and different volume scales and the corresponding Figure 6 shown.

[0080] The results of this example are shown in Figure 3 , 4 , 5, 6, and 7. Figure 3 In the figure, the three-dimensional model and the schematic diagram of the pore model intuitively show the three-dimensional structure of the dissolved calcareous sand coarse particles and the internal pore distribution characteristics. The results show that the simulation results are highly consistent with the actual camera photos, indicating that the model obtained by the simulation method of the present invention has good reduction and intuitiveness. Figure 4 , Figure 6 The results in the paper show that the total porosity (φ v ), average equivalent pore diameter (D e ) increases with the decrease of the simulation volume scale. This is because there are large areas of interconnected pores in the middle of the sample. This does not mean that there is no scale effect in the simulation method. On the contrary, it is consistent with its internal structure. Figure 5 The face ratio of the two-dimensional slice (Φ s ) shows the characteristics of concentrated distribution in the middle, which is consistent with Figure 3 This is consistent with the situation that the pores shown in the figure are mainly concentrated in the middle, reflecting the accuracy of the invention. Figure 7 Reflects the interconnected porosity (φ t ) increases with the decrease of the simulated volume scale, and the large number of pores in the middle of the sample, due to the internal pores caused by particle crushing, are connected to the outside world, thereby improving the connectivity. The multi-scale evolution law of the pore structure during the crushing of porous and fragile particles obtained by the simulation method of the present invention has high accuracy and rationality.

[0081] Comparative Example:

[0082] The test material used in the comparative test of this comparative example is a piece of coarse carbonate particles, which was taken from the stratum of a construction site in Yulin, Sanya City, Hainan Province. It is a common granular material in the local stratum. The comparative test was carried out by scanning electron microscopy (SEM) and mercury injection test (MIP), and was divided into two groups: comparative experiment (SEM) and comparative experiment (MIP). The image digitization technology (existing technical method) was combined to test the total porosity and pore equivalent radius distribution of porous and fragile rock and soil granular materials. This method is based on "Microscopic Experimental Study on the Porosity Characteristics of Coral Debris Particles" (Zhang Yu, Ding Xuanming, Peng Yu, Jiang Chunyong, Microscopic Experimental Study on the Porosity Characteristics of Coral Debris Particles, Journal of Disaster Prevention and Mitigation Engineering, 2020, 41: 3). The steps of this method are briefly summarized as follows:

[0083] 1. Dry the flaky carbonate coarse particles in an oven at 110°C. When the moisture content is 0%, take out the dried sample and use a fine brush to remove dust and impurities on the surface of the sample to obtain the sample to be tested;

[0084] 2. Use the displacement method to calculate the volume difference before and after the sample is immersed in the measuring cylinder, combined with the density parameter of water (ρ = 1g / cm 3 ), and obtain the total volume of the sample (V v );

[0085] 3. After repeating step 1, the sample was subjected to a scanning electron microscope (SEM) test with a resolution of 1.0 nm and a magnification of 12 to 1,000 times, and non-local mean filtering was performed to retain effective image information and remove noise interference to the greatest extent.

[0086] 4. Using image digitization technology, threshold segmentation and binarization processing methods are used to extract internal pore information and obtain a clear pore distribution map. Based on scanning imaging with different magnifications (500 to 4000 times), the pore size (D e ) and the 2D slice surface ratio (Φ s ) for quantitative measurement. The principle is: the surface ratio of two-dimensional slices (Φ s ) is calculated from the slice pore area (S pore ) and the total area of ​​the slice (S s ) percentage to get: Φ s The closer it is to 1, the greater the face rate of the slice, and the closer it is to 0, the smaller the face rate of the slice.

[0087] 5. Through the mercury intrusion test (MIP), the mercury is gradually pressed to make the mercury invade the pores of the particles. When the surface tension of mercury is balanced with the external pressure, the pore volume invaded by mercury is a function of pressure, that is, the pore volume (V pore The specific calculation formula is: P×r=-2σcosθ, where P is the applied pressure, r is the pore radius, σ is the surface tension coefficient of mercury (taken as 0.48N / m), and θ is the contact angle of mercury to the material (taken as 140°). Further, the total porosity (Φ v ): Φ v The closer it is to 1, the greater the porosity of the model, and the closer it is to 0, the smaller the porosity of the model;

[0088] 6. Wash the sample and crush it step by step in 3 times, and try to control the crushing scale to its volume. Repeat steps 1-5 after each smash and get the corresponding data;

[0089] 7. Using the initial subject model Total porosity at volume scale (Φ v ), 2D slice surface ratio (Φ s ), aperture (D e ), establish Φ v , Φ s , D e The curve of the change with particle volume size is as follows Figure 4 , 5 , as shown in Figure 6.

[0090] The results of the comparative test can be found in Figure 4 , 5 , as shown in Figure 6. Figure 4 , Figure 6 Total porosity (Φ v), average pore diameter (D e ) changes with the volume scale of the sample, but no non-uniform pores were found in the sample during the crushing process, so this is obviously inconsistent with the size effect, indicating that the analysis is unreasonable. Figure 5 The face ratio of the two-dimensional slices (Φ s ) can only display data at four volume scales and cannot be used to explore the pore space distribution characteristics of the sample. Comparative analysis shows that the simulation method of the present invention can construct the sample morphology and pore structure in three dimensions, and the obtained model has good reduction and intuitiveness. The multi-scale evolution law of the pore structure during the crushing process of porous and fragile particles obtained by the simulation method has high accuracy and rationality.

Claims

1. A method for simulating the evolution of microscopic pore structure during the crushing process of porous and fragile particles, characterized in that: The following steps are involved: 1) Sample preparation: Dry the sample at 110°C, take it out, sieve it, select porous rock and soil particles with a diameter of d>2mm, remove surface dust and impurities, and obtain the sample to be tested; 2) X-CT scanning: The sample to be tested obtained in step 1) is scanned using an X-CT scanner, with a scanning voltage of 120 kV, a current of 150 μA, and a resolution of 20 μm to obtain a particle tomographic image data file; 3) Image processing: import the particle tomographic image data file obtained in step 2) into AVIZO software to obtain an analysis volume; preferably, use the "Non-Local Means" algorithm to remove the noise generated during the scanning process, and use the "Volume Edit" function to cut the image area and remove the redundant space voxels outside the solid interface; 4) 3D reconstruction: processing the analysis volume obtained in step 3) to generate a 3D rendering model of the particle body; preferably, using the "Thresholding" tool to set a threshold to separate the pores from the particle body; using the "Watershed" algorithm to further refine the boundaries of the pore area to obtain a processed image sequence; using the "Volume Rendering" function to generate a 3D rendering model of the particle body based on the processed image sequence; 5) Multi-scale simulation: To simulate the reduction in particle volume caused by particle crushing, the initial 3D model in step 4) is divided into volume scales by the "Extract Subvolume" function, and multiple 3D rendering models of different scales are established by the "Volume Rendering" function; 6) Pore extraction: For the three-dimensional rendering models of the multiple different scales obtained in step 5), the "Thresholding" tool is used to perform pore threshold segmentation, and the three-dimensional pore rendering model is obtained through the "Volume Rendering" function; 7) Porosity analysis: Use the "Volume Fraction" module to calculate the total porosity of the model φ v with 2D slice face ratio; 8) Pore connectivity analysis: Use the "Axis Connectivity" function to perform connectivity analysis on the pore 3D rendering model of each volume scale in step 6); 9) Connected porosity analysis: If there are connected pores in the pore 3D rendering model in step 8), the connected porosity is calculated based on the connected pore data volume obtained in step 8) through the "VolumeRendering" function; 10) Analysis of pore structural parameters: Based on the three-dimensional pore rendering models of each volume scale obtained in step 6), the "Label Analysis" function is used to calculate the pore structural parameters according to the output results.

2. The simulation method according to claim 1, characterized in that: In the step 1), the particle size of the porous rock and soil granular material is 20-40 mm.

3. The simulation method according to claim 1, characterized in that: In the step 2), the resolution of the X-CT scanner is 10-30 μm.

4. The simulation method according to claim 1, characterized in that: In step 5), the step-by-step scale division method is as follows: the initial three-dimensional model is divided into its volume Scale, that is scale, further, the volume scale is Scale, that is The three-dimensional model of the scale needs to be taken from the volume scale Inside the three-dimensional model, and so on, the volume scale is Right now The three-dimensional model of the scale needs to be taken from the volume scale The volume scale of the three-dimensional model is Right now The three-dimensional model is taken from the volume scale 3D model interior.

5. The simulation method according to claim 1, characterized in that: The calculation method of step 7) is: total porosity φ v is the voxel volume V occupied by each pore model in step 6) pore , and the voxel volume V occupied by each subject model in step 5) model The percentage between them is: 2D slice surface ratio φ s is the slice pore area S pore , and the total slice area S s The percentage obtained is:

6. The simulation method according to claim 4, characterized in that: The step 7) uses the obtained initial subject model Total porosity Φ at volume scale v , establish Φ v The corresponding relationship curve with different scale volumes is used to obtain the influence of multi-scale particle crushing on the total porosity. For homogeneous granular materials, under ideal conditions, Φ at volume scale v Consistent.

7. The simulation method according to claim 6, characterized in that: The step 9) uses the obtained initial subject model Connected porosity Φ at volume scale t , establish the connected porosity Φ t The relationship between the corresponding curves of different scale volumes is obtained to obtain the change of pore structure connectivity during the multi-scale crushing of particles; preferably, the calculation method is: using the voxel volume V occupied by the three-dimensional rendering model of the connected pores tp , and the voxel volume V occupied by the 3D rendering model of each subject in step 5 model The percentage of connected porosity Φ t :

8. The simulation method according to claim 7, characterized in that: The pore structure parameter used in step 10) is the single pore volume V pi , equivalent diameter D e , shape factor G p ; The calculation method is: pore volume V pi Directly derived from the volume of the space voxel occupied by a single pore; equivalent diameter D e By the volume of a single pore V pi The formula yields: Form Factor G p It is defined by the pore cross-sectional area S and the pore perimeter P:

9. The simulation method according to claim 8, characterized in that: Using the single pore volume V obtained in step 10) pi , equivalent diameter D e , shape factor G p , establish the pore volume, equivalent diameter distribution frequency diagram, and average equivalent diameter D e , shape factor G p The variation curve between it and the volume at each scale.

Images