Method for characterizing and permeability prediction of multi-scale pore network model of dense rock-soil medium

Abstract

The application discloses a method for characterizing and predicting permeability of a dense rock-soil medium multi-scale pore network model, which is characterized in that: a non-regular pore network model is established based on micron-scale void structure information; a first regular pore network model is established based on the pore diameter distribution information of the scale where the void ratio dominates the void; a pore set located between the corresponding scales of the non-regular pore network model and the first regular pore network model is taken as a connecting channel, and is embedded into the non-regular pore network model according to embedding criteria; the non-regular pore network model and the first regular pore network model are connected, and the cross-scale pore model connection is performed under the constraint of the double parameters of porosity and permeability, so as to obtain a multi-scale pore network model, and the seepage flow simulation analysis is performed on the multi-scale pore network model, so as to obtain the permeability. The method fully gives play to the advantages of the non-regular pore network model in reflecting the anisotropy and heterogeneity of the dense rock-soil medium and the advantages of the regular pore network model in low characterization data quality requirement and high calculation efficiency.

Description

Technical Field

[0001] This invention relates to the field of deep energy engineering, and in particular to a method for characterizing multi-scale pore network models and predicting permeability of dense rock and soil media. Background Technology

[0002] The statements in this section are merely background information relating to this disclosure and do not necessarily constitute prior art.

[0003] Dense soil and rock media are characterized by low permeability and low porosity. Their microstructure is complex and they usually exhibit typical dual-media characteristics: discontinuous pores (pores, microcracks) are distributed at the micrometer scale (mainly 1-100 micrometers). Although the space of pores (pores, microcracks) at this scale accounts for a small proportion of the total porosity of dense soil and rock media, their contribution to the permeability of dense soil and rock media as local dominant flow channels cannot be ignored; nanoscale pores (especially <100 nanometers) are the dominant pores in dense soil and rock media, accounting for the main porosity (usually more than 80%), and are the main components of the pore network of dense soil and rock media.

[0004] In deep energy projects such as oil and gas extraction, high-level radioactive waste disposal, and carbon dioxide geological storage, accurate prediction of the permeability of dense soil and rock media such as shale, compacted bentonite, and dense sandstone is crucial for project safety and efficiency.

[0005] Pore ​​network models, as a mesoscopic numerical simulation method, have been widely used in predicting the permeability of dense soil and rock media. However, existing multi-scale pore network models developed for the multi-scale structural characteristics of dense soil and rock media mainly rely on high-quality imaging corresponding to different structural scales (such as obtaining high-resolution images of nanoscale pore structures of dense soil and rock media through nano-CT, SEM imaging, FIB-SEM imaging, TEM imaging, etc., and obtaining lower-resolution images of micrometer-scale pore structures of dense soil and rock media through micron-CT). This method of constructing multi-scale pore network models is expensive and has strict requirements on the operating environment and sample preparation, which to some extent restricts the application and promotion of multi-scale pore network model methods. Furthermore, when superimposing models of different scales based on images of different resolutions, the number of computational units in the resulting multi-scale pore network model increases dramatically, causing an exponential decrease in model computational efficiency and significantly increasing computational costs. This, to some extent, contradicts the modeling principle of pore network models to simplify the structure to ensure reliability while improving computational efficiency. Furthermore, existing multi-scale pore network models for dense soil and rock media often neglect the contribution of micron-scale discontinuous pores (pores, microcracks) to the permeability of dense soil and rock media, or only focus on the influence of microcracks in micron-scale discontinuous pores. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, this invention provides a method for characterizing and predicting the permeability of dense soil and rock media using a multi-scale pore network model. A regular pore network model is used to characterize the nanoscale pore structure of dense soil and rock media, while an irregular pore network model is used to characterize the micrometer-scale void (pore, microcrack) structure. Simultaneously, the fusion and superposition of pore network models at different scales are achieved through dual-parameter constraints of porosity and permeability, thereby realizing multi-scale characterization and permeability prediction of dense soil and rock media.

[0007] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0008] In a first aspect, the present invention provides a method for characterizing and predicting the permeability of multi-scale pore network models in dense soil and rock media, including:

[0009] Indirect characterization techniques and micron-scale CT were used to obtain nanoscale pore size distribution information and micron-scale pore structure information of dense soil and rock media;

[0010] The nanoscale pore size distribution information and the micrometer-scale void structure information are normalized to obtain the full-scale pore size distribution characteristics of micro and nanometer scales; based on the pore size distribution characteristics, the full-scale void structure information of micro and nanometer scales is divided into multiple scales to obtain multiple regions of different scales.

[0011] An irregular pore network model is established based on the micron-scale pore structure information, and a first regular pore network model is established based on the pore size distribution information at the scale where the porosity dominates.

[0012] The set of pores located at the scale between the irregular pore network model and the first regular pore network model is used as a connection channel and embedded into the irregular pore network model according to the embedding criterion.

[0013] By connecting the irregular pore network model with the first regular pore network model, and using porosity and permeability as dual-parameter constraints to connect the cross-scale pore models, a multi-scale pore network model is obtained.

[0014] The permeability of the dense rock and soil medium was obtained by performing seepage simulation analysis on the multi-scale pore network model.

[0015] A further technical solution involves establishing a second regular pore network model based on the pore size distribution information of the dominant pore size if there exists a pore size smaller than the dominant pore size. The second regular pore network model is then connected across scales to the first regular pore network model using porosity and permeability as dual-parameter constraints.

[0016] A further technical solution involves connecting the second regular pore network model across scales to the first regular pore network model. The specific steps are as follows:

[0017] Seepage simulation analysis was performed on the second regular pore network model to obtain the apparent permeability of the second regular pore network model;

[0018] The small-scale second-order pore network model is abstracted into simplified cylindrical pores and added to the first-order pore network model;

[0019] In the geometric network of the first regular pore network model, the pores of the second regular pore network model are simplified based on the porosity ratio, and the pores of the corresponding scales of the first regular pore network model are assigned. The two types of pores are regarded as a series relationship in the first regular pore network model.

[0020] Based on the porosity and pore size distribution data of the pores at the corresponding scale of the first regular pore network model, and referring to the adjustment formula in the establishment of the regular pore network model, the actual physical size corresponding to the first regular pore network model to be connected across scales is determined.

[0021] Based on the total length of the simplified pores allocated to the second regular pore network model in the first regular pore network model and the porosity of the corresponding scale pores in the second regular pore network model, the equivalent radius of the simplified pores in the second regular pore network model in the first pore network model is calculated.

[0022] Based on the apparent permeability and porosity of the corresponding scale pores in the second regular pore network model, the equivalent permeability of the simplified pores in the second regular pore network in the first pore network model is calculated.

[0023] A further technical solution involves the following steps in establishing the regular porous network model:

[0024] A regular pore network model is constructed based on regular geometric units, and the pore radius is generated using statistical and programming methods based on pore size distribution data.

[0025] Porosity is introduced, and the actual physical dimensions of the model are given by adjusting the formula.

[0026] A further technical solution involves establishing the irregular pore network model using the following steps:

[0027] Box counting was used to perform volume analysis of representative units to determine the appropriate analytical range.

[0028] Threshold segmentation of digital core samples obtained from micron-scale CT characterization images was performed using an image segmentation algorithm to obtain the void structure;

[0029] The topological information of the void structure is extracted using a void network model extraction algorithm. The void body is simplified into a sphere or cylinder, and the topological isomorphic ball-and-stick void model is reconstructed using the spherical void body as the model node.

[0030] A further technical solution is that the embedding criteria are specifically as follows:

[0031] In the irregular pore network model, spherical pores are used as network nodes, and the Delaunay triangulation algorithm and programming methods are used to generate a set of potential connection channels between adjacent pores.

[0032] Referring to the pore radius allocation method in the regular pore network model, a set of pore radii equal to the number of potential connection channels is generated based on the scale pore size distribution data.

[0033] Calculate the length of potential connecting channels, and generate a number of pore channels based on porosity that does not exceed the number of potential connecting channels;

[0034] If the porosity of the potential connection channel set in the irregular porous network model is less than the characteristic porosity of the void set, then the potential connection channel set is updated using optimization criteria so that the new potential connection channel set meets the requirements.

[0035] A further technical solution is provided, wherein the optimization criterion is specifically as follows:

[0036] If the porosity of the potential connection channel set in the irregular pore network model is less than the characteristic porosity of the void set, then the void set is further divided. The large-diameter part of the set is equivalently embedded as an isolated spherical pore at the midpoint of the potential connection channel in the irregular pore network model. Then, the new pore body set in the irregular pore network model is used as a network node, and the Delaunay triangulation algorithm and programming method are used to generate a new potential connection channel set between adjacent pore bodies.

[0037] The remaining small-diameter portion of the void set is constrained by porosity and embedded into a new irregular void network model according to the embedding criterion.

[0038] During the embedding process, if the porosity corresponding to the new set of potential connection channels is still less than the porosity of the remaining small-aperture set in the void set, the above process is repeated to update the set of potential connection channels until the porosity corresponding to the set of potential connection channels in the irregular pore network model meets the requirements.

[0039] A further technical solution involves connecting the irregular pore network model with the first regular pore network model as follows:

[0040] Seepage simulation analysis was performed on the first regular pore network model to obtain the apparent permeability of the first regular pore network model;

[0041] The small-scale first-order pore network model is abstracted into simplified connection channels and assigned to the potential connection channel paths generated by the irregular pore network model;

[0042] Based on the porosity corresponding to the pore size in the first regular pore network model and the potential connection channel length in the irregular pore network model, the equivalent radius of the connection channel in the first regular pore network model is calculated and obtained.

[0043] Based on the apparent permeability of the first regular pore network model and the porosity corresponding to the pore scale, the equivalent permeability of the first regular pore network model as the connecting channel in the irregular pore network model is calculated.

[0044] If large-scale connecting channels already exist between spherical pores in a large-scale irregular pore network model, then the equivalent small-scale connecting channels are considered to be parallel to the original channels and merged into a whole channel.

[0045] The equivalent radius of the original large-scale connecting channel is set as the equivalent radius of the overall channel. The equivalent permeability of the overall channel is calculated based on the equivalent radius and equivalent permeability of the equivalent small-scale connecting channel and the original large-scale connecting channel.

[0046] In a second aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the method for characterizing and predicting the permeability of a multi-scale pore network model of dense soil and rock media as described in the first aspect.

[0047] Thirdly, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the method for characterizing and predicting the permeability of a multi-scale pore network model of dense soil and rock media as described in the first aspect.

[0048] The above one or more technical solutions have the following beneficial effects:

[0049] This invention employs a regular pore network model to characterize the nanoscale pore structure of dense soil and rock media, and an irregular pore network model to characterize the micrometer-scale void (pore, microcrack) structure of dense soil and rock media. At the same time, it uses the dual parameters of porosity and permeability to achieve the fusion and superposition of pore network models of different scales, thereby realizing multi-scale characterization and permeability prediction of dense soil and rock media.

[0050] This invention reduces the reliance on high-quality imaging techniques for constructing multi-scale pore network models of dense soil and rock media. In particular, it eliminates the need for high-resolution imaging techniques (nano-CT, FIB-SEM, etc.) to characterize the microstructure of dense soil and rock media, greatly reducing data acquisition costs. At the same time, it uses indirect characterization methods to obtain nanoscale pore structure information of dense soil and rock media, which has the characteristics of rapid characterization and simple sample preparation, greatly reducing the application cost of multi-scale pore network models.

[0051] The anisotropy and heterogeneity of micron-scale pores (pores, microcracks) in dense soil and rock media significantly affect the overall permeability of these media, while nanoscale pore structures mainly contribute to overall permeability by forming interconnected networks. This invention combines regular and irregular pore network models, fully leveraging the advantages of irregular pore network models in reflecting the anisotropy and heterogeneity of dense soil and rock media, and the advantages of regular pore network models in terms of low data quality requirements and high computational efficiency. This approach effectively balances computational accuracy and efficiency in the multi-scale structural characterization and permeability prediction of dense soil and rock media.

[0052] When connecting models across scales, this invention adopts an equivalent superposition of porosity and permeability dual-parameter constraints, without increasing the number of computational units in the model. This allows the model to consider the influence of more pore volume factors on the permeability of dense soil and rock media (such as the influence of micron-scale discontinuous pores on permeability) and the seepage process under the influence of complex physicochemical processes (such as the seepage process under the influence of adsorption, mineral composition, electric field and other factors) while ensuring computational efficiency.

[0053] This invention, when reflecting the gap between the scale corresponding to the voids (pores, microcracks) characterized by micron-scale CT and the scale corresponding to the voids of the dominant porosity, can, to a certain extent, obtain the connectivity of voids (pores, microcracks) within this scale range in dense soil and rock media through model assumptions and porosity parameter correction. It qualitatively analyzes the void connectivity in dense soil and rock media, and to a certain extent compensates for the inability of indirect characterization techniques (nitrogen adsorption method, mercury intrusion porosimetry) to obtain the void (pore, microcrack) connectivity of dense soil and rock media. Attached Figure Description

[0054] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0055] Figure 1 This is a flowchart of the multi-scale pore network model characterization and permeability prediction method for dense soil and rock media according to an embodiment of the present invention;

[0056] Figure 2 This is a pore size distribution diagram of compacted bentonite according to an embodiment of the present invention, wherein (a) is a diagram showing the relationship between cumulative porosity and equivalent pore size, and (b) is a diagram showing the relationship between frequency and equivalent pore size;

[0057] Figure 3 This is a geometric structure diagram of a regular pore network model according to an embodiment of the present invention, wherein (a) is a geometric network diagram and (b) is a truncated octahedral cell;

[0058] Figure 4 This is a schematic diagram of the construction of the first regular pore network model according to an embodiment of the present invention;

[0059] Figure 5 This is a schematic diagram of the construction of an irregular pore network model based on micron-sized CT characterization images, where (a) is an isolated pore from the CT image and (b) is the reconstructed topological isolated pore model.

[0060] Figure 6 This is a schematic diagram of the construction of a multi-scale pore network model according to an embodiment of the present invention. Detailed Implementation

[0061] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, 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.

[0062] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0063] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0064] Example 1

[0065] like Figure 1 As shown in the figure, this embodiment discloses a method for characterizing and predicting the permeability of a multi-scale pore network model in dense soil and rock media. The method includes the following steps:

[0066] S1: Indirect characterization techniques and micron-scale CT are used to obtain nanoscale pore size distribution information and micron-scale pore structure information of dense soil and rock media;

[0067] In this embodiment, indirect characterization techniques employed include nitrogen adsorption and mercury intrusion porosimetry. Taking compacted bentonite as an example, Gaomiaozi bentonite from Gaomiaozi Town, Inner Mongolia, China, was selected as the research object. Nitrogen adsorption, mercury intrusion porosimetry, and micron-scale CT were used to obtain nanoscale pore size distribution information and micron-scale void (pore, microcrack) structure information of the compacted bentonite.

[0068] No microcracks were found in the characterization results, so the voids in the compacted bentonite samples were all treated as pores.

[0069] S2: Normalize the nanoscale pore size distribution information and the micrometer-scale void structure information to obtain the full-scale pore size distribution characteristics of micro and nanometer scales; based on the pore size distribution characteristics, divide the full-scale void structure information of micro and nanometer scales into multiple scales to obtain multiple regions of different scales.

[0070] In this embodiment, the sample densities differ between the nitrogen adsorption method and the mercury intrusion porosimetry method, and the results cannot be directly used for unified analysis. Therefore, porosity is used as a parameter to normalize the test results, and the conversion formula is expressed as follows:

[0071]

[0072] in, The measured aperture distribution is the first Porosity of each pore size range It measures the density of the test sample. The measured aperture distribution is the first The unit mass volume of each aperture range.

[0073] The nano-to-micron full-scale pore size-porosity distribution of compacted bentonite samples measured by nitrogen adsorption, mercury porosimetry, and micron-CT is as follows: Figure 2 As shown in (a).

[0074] The pore shape within the sample is assumed to be cylindrical, and the pore size-frequency distribution is calculated using the following formula:

[0075]

[0076] in, The measured aperture distribution is the first Frequency of each aperture range; The measured aperture distribution is the first The equivalent aperture of each aperture range.

[0077] The nano-to-micron full-scale pore size-frequency distribution of compacted bentonite samples measured by nitrogen adsorption, mercury porosimetry, and micron-CT is as follows: Figure 2 As shown in (b).

[0078] like Figure 2As shown in (a) and (b), it can be found that the porosity of pores in the 2-200nm pore size range accounts for more than 80% of the total porosity, and pores in this range are the dominant porosity pores; the pore size range with a pore size less than 2nm is the pore size smaller than the dominant pores; the pore size range measured by micron CT is the pore size larger than 6.8μm, and the pores in this pore size range are not interconnected and are discontinuous isolated pores; the pore size range of 200nm-6.8μm is the pore size between the dominant pores and the isolated pores.

[0079] Therefore, based on the pore size distribution analysis results (pore size distribution characteristics) of compacted bentonite, the micro-nano full-scale pore structure information obtained from the characterization of compacted bentonite samples is divided into four scales, with different scale regions being: Pores I (pore size <2nm), Pores II (pore size 2-200nm), Pores III (pore size 200nm-6.8μm), and Pores IV (pore size >6.8μm).

[0080] S3: Based on the micron-scale pore structure information, establish an irregular pore network model, and based on the pore size distribution information at the scale where the porosity dominates, establish a first regular pore network model.

[0081] In this embodiment, if there are pore sizes smaller than the dominant pore size, a second regular pore network model is established based on the pore size distribution information of this pore size. The second regular pore network model is connected across scales to the first regular pore network model using porosity and permeability as dual-parameter constraints. Based on the pore size division in S2 above, a regular pore network model (R-PNM) is established for the dominant pore size (Pores II) and the smaller pore size (Pores I). The basic steps for establishing the R-PNM are as follows:

[0082] R-PNM is constructed based on truncated octahedral elements (regular geometric elements), and its geometry is as follows: Figure 3 As shown. Figure 3 (a) is a porous geometric network constructed based on truncated octahedral elements. Figure 3 (b) represents the potential location of a cylindrical pore channel within a truncated octahedral element, with the body center of the truncated octahedron serving as the pore channel connection node. Based on the pore size distribution data obtained from experimental characterization, statistical and programming methods are used to generate the pore radius in the model:

[0083] The aperture-frequency distribution data is processed and converted into a pore radius function of cumulative frequency. ;

[0084] Using a random number generation function, a set of numbers is generated in the interval [0,1] corresponding to the number of pores in the regular pore network model. The same uniform random number;

[0085] The generated uniform random numbers are processed by the function. Linear interpolation is performed to accurately assign a radius to each pore in the model.

[0086] To establish the dimensional relationship between the regular pore network model and the compacted bentonite sample, measured porosity is introduced, and the actual physical dimensions of the model are assigned by adjusting the formula, as follows:

[0087]

[0088] in, It is the total number of truncated octahedral elements in the model. It is the volume occupied by a truncated octahedral unit. It's porosity. It is the total number of cylindrical pores in the model. It is the pore radius. It is the pore length. In this embodiment, This represents the distance between the two square boundaries of the truncated octahedral element. .

[0089] The second regular pore network model R-PNM I, based on Pores I, was completed according to the above R-PNM establishment steps. Seepage simulation analysis was then performed on R-PNM I, and the specific governing equations are as follows:

[0090] Considering the water sensitivity of bentonite, gas was chosen as the fluid medium for studying the permeability of compacted bentonite. The transport mechanism of gas in porous media and the mean free path of molecules are discussed. and pore channel characteristic length Closely related, through Knudsen numbers The formula for measurement is:

[0091]

[0092] in, The calculation formula is:

[0093]

[0094] in, It is the Boltzmann constant; It's temperature; It is the collision diameter of gas molecules; It is pressure; It is the pore diameter.

[0095] To simplify the complexity of the analysis process and establish an effective framework that can uniformly describe the multi-scale permeability of bentonite, the Knudsen modified Darcy flow equation proposed by Beskok and Karniadakis is adopted:

[0096]

[0097] in, Represents the volumetric flow rate through the pores. It is a dimensionless rarefaction coefficient, and the calculation formula is:

[0098]

[0099] in, It is a constant, taking the value -1. It is the pore radius. It is the pressure drop along the length of the pores. It is dynamic viscosity. It is the pore length.

[0100] After performing a seepage simulation analysis on R-PNMI using the above seepage equation, the apparent permeability of the R-PNMI model was obtained. .

[0101] like Figure 4 As shown, the small-scale R-PNM I model is abstracted into simplified cylindrical pores and added to the first regular pore network model R-PNM II based on PoresII. The specific steps are as follows:

[0102] Based on the above R-PNM establishment steps, establish the R-PNM II model geometric network. Based on the porosity ratio of pores in the corresponding pore size ranges of Pores I and Pores II, the R-PNM II model simplifies the pores in the R-PNM I model and the pores in the corresponding pore size ranges of Pores II. The two types of pores are treated as being in series in the R-PNM II model. The specific steps are as follows:

[0103] Using a random sampling algorithm The pores are divided into two distinct groups: a set reflecting the characteristics of Pores I data. and a set of data that reflects the characteristics of Pores II data In the R-PNM II model, the set Number of pores Determined by the following formula:

[0104]

[0105] in, Indicates no more than The largest integer, It is the total number of model pores in R-PNM II. Pores I represents the porosity of the pores. It is the porosity of the pores corresponding to Pores II.

[0106] Based on the porosity and pore size distribution data of corresponding scale pores in Pores II, a set of Randomly assign pore radii, then create a virtual network. Geometric scaling is performed according to the adjustment formula in the R-PNM establishment steps described above to determine the actual physical dimensions corresponding to the R-PNM II model, thus obtaining a network with actual physical dimensions. .

[0107] from Extract the simplified pore set assigned to the R-PNM I model from the model data of the R-PNM II model. Total pore length The simplified pore set of the R-PNM I model in the R-PNM II model is calculated using the following formula. equivalent pore radius :

[0108]

[0109] in, It is the volume of the R-PNM II model.

[0110] The simplified pore set in the R-PNM I model is used in the R-PNM II model. The radius of all pores is set to . .

[0111] Apparent penetration rate based on R-PNM I model Porosity corresponding to the size of pores The equivalent permeability of the simplified pores in the R-PNM I model in the R-PNM II model was calculated. The formula is shown below:

[0112] .

[0113] In this embodiment, based on the pore scale division in S2 above, an irregular pore network model (T-PNM) is established for the micron-scale pore Pores IV obtained based on micron-scale CT. The basic steps for establishing the T-PNM are as follows:

[0114] The representative unit volume (REV) analysis of compacted bentonite digital cores obtained by micron-CT scanning was performed using the box counting method to determine the appropriate analysis range; for example, the REV size is 400×400×400 voxels.

[0115] Threshold segmentation of digital core samples obtained from micron-scale CT characterization images was performed using image segmentation algorithms (such as the watershed algorithm) to obtain pore structures such as... Figure 5 As shown in (a);

[0116] The pore network model extraction algorithm, the maximum sphere method, is used to extract the topological information of the pore structure (void center spatial coordinates, equivalent radius, etc.) from the digital core. Based on the extracted topological information parameters, the pores are simplified into spherical or cylindrical shapes. A topologically isomorphic ball-and-stick pore model is then reconstructed using spherical pores as model nodes. Figure 5 As shown in (b).

[0117] S4: The set of pores located at the scale between the irregular pore network model and the first regular pore network model is used as a connection channel and embedded into the irregular pore network model according to the embedding criterion.

[0118] In this embodiment, based on the pore scale division in S2 above, the pore set Pores III, located at the pore scale between the dominant pores and isolated pores, is embedded as a connecting channel into the irregular pore network model T-PNM. The specific steps are as follows:

[0119] In the irregular pore network model T-PNM established in S3, spherical pores are used as network nodes. The Delaunay triangulation algorithm and programming methods are employed to generate a set of potential connection channels between adjacent pores. The number of potential connection channels is... ;

[0120] Referring to the method of generating (assigning) pore radii in the regular pore network model in S3, a set of pore radii data equal to the number of potential connection channels is generated based on the pore size distribution data of Pores III. ;

[0121] For adjacent spherical pores in the T-PNM model and Its potential connection channel length Precise calculations are performed using the following formula:

[0122]

[0123] in, Indicates porous body and The Euclidean distance between the centers of mass and Porous bodies and The equivalent radius.

[0124] Through the porosity of Pores III ( Constraints are used to generate pore channels in the T-PNM model using the following formula:

[0125]

[0126] in, It is the actual sample volume (i.e., REV volume) reflected by the T-PNM model. It is the length of the pore channel. It is the equivalent radius of the pore channel. Represents the number of pore channels. This refers to volume tolerance.

[0127] If the porosity corresponding to the potential connectivity channel set in the irregular porous network model is less than the characterization porosity of the void set (porosity refers to the ratio of pore volume to total volume), then the potential connectivity channel set is updated using optimization criteria to ensure that the resulting new potential connectivity channel set meets the requirements. During the pore channel generation process, inequality constraints must be strictly satisfied. Quantitative constraints are used to ensure the authenticity of the structure, thereby ensuring the validity of the formula.

[0128] The specific steps for generating pore channels in the T-PNM model are as follows:

[0129] From the set of pore radii Uniform random selection A sample was selected from the set of potential interconnection channel lengths in the pores using the same method. One sample, Start from 1 and gradually increase to During the sample extraction process, the sets derived from these samples are examined step by step. ( If the above pore channel generation formula is satisfied, then the selected sample set is the required T-PNM model embedded pore channel.

[0130] like At that time, the set The maximum value in S3 is still insufficient to satisfy the above pore channel generation formula, that is, the total volume of potential pore channels in the T-PNM model constructed in S3 is smaller than the measured characterization volume. Therefore, the pore size range of the pore connection channels in Pores III as the pore connection channels in the T-PNM model is redefined according to the following optimization criteria, and the T-PNM model is reconstructed. The specific steps are as follows:

[0131] Based on the characterization results, the left endpoint of the largest aperture interval within the Pores III aperture range is selected as the set boundary point. The Pores III of the original defined void set (in this embodiment, the set of embedded void connection channels) is divided into two. The voids within the larger aperture range are defined as isolated voids. The voids within this range are equivalent to spheres and are assigned to the center of the potential connection channels in the T-PNM model. These spherical voids and the spherical voids in the original T-PNM model constitute new network nodes in the reconstructed T-PNM model. The Delaunay triangulation algorithm and programming method are used to generate a new set of potential connection channels between adjacent voids. The voids within the smaller aperture range in the Pores III set are embedded into the reconstructed T-PNM model as connection channels (i.e., embedded into the new irregular void network model with porosity as a constraint and according to the embedding criterion). During the embedding process, if the total volume of potential pore channels in the reconstructed T-PNM model is still less than the measured characteristic volume of pores in the smaller pore size range in the Pores III set (in this embodiment, it can also be if the porosity corresponding to the new potential connection channel set is still less than the characteristic porosity of the remaining smaller pore size set in the void set), then according to the size descending order principle, the set boundary point of Pores III is reselected, the above process is repeated, and the potential connection channel set is updated until the volume corresponding to the potential connection channel set in the irregular pore network model meets the requirements.

[0132] In this embodiment, pores in the Pores III set with a pore size range of 3-6.8μm are redefined as isolated pores for the reconstruction of the irregular pore network model, while pores with a pore size range of 200nm-3μm are embedded as pore connection channels into the irregular pore network model.

[0133] S5: Connect the irregular pore network model with the first regular pore network model, and connect the cross-scale pore models with porosity and permeability dual parameter constraints to obtain a multi-scale pore network model.

[0134] In this embodiment, a multi-scale pore network model is constructed based on the regular pore network model R-PNM II (reflecting the pore structure information of Pores I and Pores II) established in S3 and the irregular pore network model finally obtained in S4 (reflecting the pore structure information of Pores III and Pores IV). The specific steps are as follows:

[0135] Seepage simulation analysis was performed on the first regular pore network model, R-PNM II model. The governing equations used in the seepage simulation in S3 were selected to obtain the apparent permeability of the R-PNM II model. ;

[0136] like Figure 6 As shown, the small-scale R-PNM II model is abstracted into simplified cylindrical pore connection channels and added to the final irregular pore network model obtained in S4. The specific steps are as follows:

[0137] The small-scale R-PNM II model is abstracted into simplified connection channels and added to the potential connection channel paths generated by the Delaunay triangulation algorithm in the irregular porosity network model. If there are already large-scale connection channels (black channels) in the irregular porosity network model, the simplified channels (gray channels) of the R-PNM II model are regarded as parallel to the original channels and merged into a whole channel (white channel).

[0138] Based on the R-PNM II model, porosity is reflected at the pore size ( By calculating the potential connection channel lengths in the irregular porosity network model, the equivalent radius of the R-PNM II model, abstracted as a connection channel (grey channel) in the irregular porosity network model, is obtained. The formula is shown below:

[0139]

[0140] in, It is the total length of all potential channels in the irregular pore network model.

[0141] The equivalent radius of the overall channel (white channel) is set to the radius of the original larger-scale channel (black channel) in the irregular porous network model. .

[0142] Apparent penetration rate based on R-PNM II model and the porosity corresponding to the pore size ( The equivalent permeability of the R-PNM II model, abstracted as connecting channels (grey channels) in the irregular pore network model, was calculated. The formula is shown below:

[0143]

[0144] If large-scale connecting channels already exist between spherical pores in a large-scale irregular pore network model, then the equivalent small-scale connecting channels are considered parallel to the original channels and merged into a unified channel. The equivalent radius of the original large-scale connecting channels is set as the equivalent radius of the unified channel. Based on the simplified equivalent small-scale R-PNM II model (grey channel) and the equivalent radius and equivalent permeability of the original large-scale connecting channels (black channel) in the irregular pore network model, the equivalent permeability of the unified channel (white channel) is calculated. The formula is shown below:

[0145]

[0146] in, It represents the permeability of the original large-scale connecting channels (black channels) in the irregular porous network model.

[0147] S6: Perform seepage simulation analysis on the multi-scale pore network model to obtain the permeability of the dense rock and soil medium it represents.

[0148] In this embodiment, the seepage simulation analysis of the compacted Gaomiaozi bentonite sample is carried out based on the multi-scale pore network model constructed in S5 and the control equation used in the seepage simulation in S3, and the permeability characteristics of the sample are finally obtained.

[0149] Example 2

[0150] The purpose of this embodiment is to provide a computing device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method of Embodiment 1.

[0151] Example 3

[0152] The purpose of this embodiment is to provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the method of Embodiment 1.

[0153] The steps and methods involved in the apparatuses of Embodiments 2 and 3 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0154] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0155] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0156] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for characterizing and predicting the permeability of dense soil and rock media using a multi-scale pore network model, characterized in that, include: Indirect characterization techniques and micron-scale CT were used to obtain nanoscale pore size distribution information and micron-scale pore structure information of dense soil and rock media; The nanoscale pore size distribution information and the micrometer-scale void structure information are normalized to obtain the pore size distribution characteristics at the full micro- and nanoscale. Based on the pore size distribution characteristics, the micro-nano full-scale void structure information is divided into multiple scales to obtain multiple regions of different scales. An irregular pore network model is established based on the micron-scale pore structure information, and a first regular pore network model is established based on the pore size distribution information at the scale where the porosity dominates. The set of pores located at the scale between the irregular pore network model and the first regular pore network model is used as a connection channel and embedded into the irregular pore network model according to the embedding criterion. By connecting the irregular pore network model with the first regular pore network model, and using porosity and permeability as dual-parameter constraints to connect the cross-scale pore models, a multi-scale pore network model is obtained. The permeability of the dense rock and soil medium was obtained by performing seepage simulation analysis on the multi-scale pore network model.

2. The method for characterizing and predicting permeability of dense soil and rock media using a multi-scale pore network model as described in claim 1, characterized in that, If there is a pore scale smaller than the dominant pore scale, a second regular pore network model is established based on the pore size distribution information of that pore scale. The second regular pore network model is connected across scales to the first regular pore network model with porosity and permeability as dual-parameter constraints.

3. The method for characterizing and predicting permeability of dense soil and rock media using a multi-scale pore network model as described in claim 2, characterized in that, The specific steps for connecting the second regular porous network model across scales to the first regular porous network model are as follows: Seepage simulation analysis was performed on the second regular pore network model to obtain the apparent permeability of the second regular pore network model; The small-scale second-order pore network model is abstracted into simplified cylindrical pores and added to the first-order pore network model; In the geometric network of the first regular pore network model, the pores of the second regular pore network model are simplified based on the porosity ratio, and the pores of the corresponding scales of the first regular pore network model are assigned. The two types of pores are regarded as a series relationship in the first regular pore network model. Based on the porosity and pore size distribution data of the pores at the corresponding scale of the first regular pore network model, and referring to the adjustment formula in the establishment of the regular pore network model, the actual physical size corresponding to the first regular pore network model to be connected across scales is determined. Based on the total length of the simplified pores allocated to the second regular pore network model in the first regular pore network model and the porosity of the corresponding scale pores in the second regular pore network model, the equivalent radius of the simplified pores in the second regular pore network model in the first pore network model is calculated. Based on the apparent permeability and porosity of the corresponding scale pores in the second regular pore network model, the equivalent permeability of the simplified pores in the second regular pore network in the first pore network model is calculated.

4. The method for characterizing and predicting permeability of dense soil and rock media using a multi-scale pore network model as described in claim 1, characterized in that, The steps for establishing the regular porous network model are as follows: A regular pore network model is constructed based on regular geometric units, and the pore radius is generated using statistical and programming methods based on pore size distribution data. Porosity is introduced, and the actual physical dimensions of the model are given by adjusting the formula.

5. The method for characterizing and predicting permeability of dense soil and rock media using a multi-scale pore network model as described in claim 1, characterized in that, The steps for establishing the irregular porous network model are as follows: Box counting was used to perform volume analysis of representative units to determine the appropriate analytical range. Threshold segmentation of digital core samples obtained from micron-scale CT characterization images was performed using an image segmentation algorithm to obtain the void structure; The topological information of the void structure is extracted using a void network model extraction algorithm. The void body is simplified into a sphere or cylinder, and the topological isomorphic ball-and-stick void model is reconstructed using the spherical void body as the model node.

6. The method for characterizing and predicting permeability of dense soil and rock media using a multi-scale pore network model as described in claim 1, characterized in that, The embedding criteria are specifically as follows: In the irregular pore network model, spherical pores are used as network nodes, and the Delaunay triangulation algorithm and programming methods are used to generate a set of potential connection channels between adjacent pores. Referring to the pore radius allocation method in the regular pore network model, a set of pore radii equal to the number of potential connection channels is generated based on the scale pore size distribution data; Calculate the length of potential connection channels and generate a number of pore channels based on porosity that does not exceed the number of potential connection channels. If the porosity of the potential connection channel set in the irregular porous network model is less than the characteristic porosity of the void set, then the potential connection channel set is updated using optimization criteria so that the new potential connection channel set meets the requirements.

7. The method for characterizing and predicting permeability of dense soil and rock media using a multi-scale pore network model as described in claim 6, characterized in that, The optimization criteria are as follows: If the porosity of the potential connection channel set in the irregular pore network model is less than the characteristic porosity of the void set, then the void set is further divided. The large-diameter part of the set is equivalently embedded as an isolated spherical pore at the midpoint of the potential connection channel in the irregular pore network model. Then, the new pore body set in the irregular pore network model is used as a network node, and the Delaunay triangulation algorithm and programming method are used to generate a new potential connection channel set between adjacent pore bodies. The remaining small-diameter portion of the void set is constrained by porosity and embedded into a new irregular void network model according to the embedding criterion. During the embedding process, if the porosity corresponding to the new set of potential connection channels is still less than the porosity of the remaining small-aperture set in the void set, the above process is repeated to update the set of potential connection channels until the porosity corresponding to the set of potential connection channels in the irregular pore network model meets the requirements.

8. The method for characterizing and predicting permeability of dense soil and rock media using a multi-scale pore network model as described in claim 1, characterized in that, The connection between the irregular pore network model and the first regular pore network model is specifically as follows: Seepage simulation analysis was performed on the first regular pore network model to obtain the apparent permeability of the first regular pore network model; The small-scale first-order pore network model is abstracted into simplified connection channels and assigned to the potential connection channel paths generated by the irregular pore network model; Based on the porosity corresponding to the pore size in the first regular pore network model and the potential connection channel length in the irregular pore network model, the equivalent radius of the connection channel in the first regular pore network model is calculated and obtained. Based on the apparent permeability of the first regular pore network model and the porosity corresponding to the pore scale, the equivalent permeability of the first regular pore network model as the connecting channel in the irregular pore network model is calculated. If large-scale connecting channels already exist between spherical pores in a large-scale irregular pore network model, then the equivalent small-scale connecting channels are considered to be parallel to the original channels and merged into a whole channel. The equivalent radius of the original large-scale connecting channel is set as the equivalent radius of the overall channel. The equivalent permeability of the overall channel is calculated based on the equivalent radius and equivalent permeability of the equivalent small-scale connecting channel and the original large-scale connecting channel.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the method for characterizing and predicting permeability of multi-scale pore network models of dense soil and rock media as described in any one of claims 1-8.

10. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the method for characterizing and predicting permeability of multi-scale pore network models of dense soil and rock media as described in any one of claims 1-8.

Images