A method for identifying and extracting dominant paths in rock mass flow and mass transfer simulation considering fracture-matrix interaction

Abstract

The application discloses a kind of rock mass seepage mass transfer simulation advantage path identification and extraction method of fusion fissure-matrix influence.First, the three-dimensional geometric characteristic parameters of fissure network are obtained, a random three-dimensional discrete fissure network model, i.e.the original model Full-DFN of fissure rock mass is generated.Second, the fissure network is structured modeling, the intersection of fissure surface and fissure is respectively abstracted as the two vertex sets of two-part graph, the mapping relationship between "fissure surface-line" is established;Further, the weighted calculation of fissure length, opening, permeability multi-parameter constraint, the advantage path in the process of seepage mass transfer is identified and extracted;Then, the simplified fissure rock mass submodel Sub-DFN is constructed.Finally, the fissure surface of Sub-DFN and Full-DFN is triangulated, the rock matrix is tetrahedral meshing, and the grid element is data conversion, the fissure-matrix rock mass submodel Sub-DFM and the fissure-matrix rock mass full model Full-DFM are obtained, and then seepage and mass transfer simulation analysis is carried out.

Description

Technical Field

[0001] This invention belongs to the technical field of fracture model establishment, and in particular relates to a method for identifying and extracting dominant paths in rock mass seepage mass transfer simulation that integrates fracture-matrix influence. Background Technology

[0002] The seepage and solute transport behavior in fractured rock masses is crucial for groundwater resource assessment, oil and gas extraction, carbon dioxide geological sequestration, and high-level radioactive waste disposal. Fracture networks are the main channels for fluid seepage and solute transport in rock masses, and their spatial distribution exhibits significant heterogeneity and anisotropy, which makes the fluid seepage and solute transport processes in fractured rock masses more complex.

[0003] Currently, mathematical models characterizing complex fractured rock masses are mainly divided into three categories: equivalent continuous medium models, discrete fracture network models, and fracture network rock mass (or fracture-pore dual medium) models. Among them, discrete fracture network models are widely used because they can accurately characterize the geometric features (such as dip angle, length, aperture, roughness, etc.) and hydraulic properties of each fracture, and can realistically reflect the complex structure of the fracture network. They have significant advantages in revealing the seepage and solute transport behavior of fracture networks.

[0004] However, traditional discrete fracture network models only simulate the flow channels within the fracture network, neglecting the role of the rock matrix and failing to describe key physicochemical processes such as molecular diffusion, matrix adsorption, and radioactive decay within the fracture-matrix system. Fractured rock masses typically consist of a fracture network and a rock matrix. Research indicates that fractures themselves are insufficient to fully characterize actual solute transport; the matrix's influence on solute transport processes is significant. Although the matrix has small pores and low permeability, it can affect seepage and solute transport over long timescales. To more accurately reflect the interaction between fractures and the rock matrix, researchers have recently developed fracture-matrix system models. By embedding the characteristics of the rock matrix into a complex fracture network model, the interaction between fractures and the matrix is ​​described through mathematical models. For example, some researchers have proposed a unified network method, constructing a unified network by discretizing fractures into triangular elements and rock masses into tetrahedral elements; others have proposed using solid elements to simulate matrix rock blocks and thickness-free elements to simulate complex fracture networks; still others have proposed a discrete fracture-matrix model that represents fractures and intersections as low-dimensional objects embedded in a three-dimensional matrix. To improve simulation accuracy, these fracture-matrix models require complex mathematical models and the establishment of transport equations between the fracture and matrix models, leading to a significant increase in computational load and limiting their application in practical engineering problems.

[0005] In recent years, to balance computational efficiency and simulation accuracy, researchers have conducted studies on extracting dominant paths from complex fracture networks. For example, by constructing fracture networks with both deterministic and stochastic characteristics, dominant subnetworks that eliminate narrow-width fractures are generated to simulate fluid flow. Furthermore, some researchers have used deep learning methods to identify three-dimensional discrete fracture network models, filter key fractures by calculating their expected correlation scores, and combine iterative results with optimization strategies to remove dead-end fractures.

[0006] However, existing methods fail to adequately consider the geometric and hydraulic characteristics of fractures when identifying and extracting dominant paths in fracture networks. They also neglect the material exchange processes between fractures and the rock matrix, leading to significant discrepancies between predicted and actual solute migration behavior. This is particularly problematic in high-level radioactive waste disposal scenarios, where the composition and pore structure of the rock matrix vary considerably, resulting in long nuclide migration times. Adsorption, diffusion, and decay processes within the matrix have a decisive impact on solute transport. Fracture models considering only fracture networks cannot accurately predict long-term migration behavior; while considering only matrix effects requires simulating processes spanning decades or even centuries due to the long timescales of nuclide migration and the large spatial scales, exacerbating the computational burden.

[0007] Against this backdrop, organically combining the steps of "dominant path extraction" and "considering the rock matrix" presents significant technical challenges and motivational obstacles for researchers in this field. The main reason is that "dominant path extraction" primarily simplifies the model and reduces computational costs, while "considering the rock matrix" requires increasing model complexity to describe the interactions between the fracture network and the rock matrix (such as diffusion, adsorption, and decay), leading to increased computational costs. These two steps are fundamentally conflicting in their technical approaches. Researchers are typically accustomed to simplifying models using discrete fracture network (DFN) models or performing seepage mass transfer simulations in fracture-matrix (DFM) models, without a better way to integrate "dominant path extraction" and "considering the rock matrix" into a tightly interdependent new workflow.

[0008] Therefore, no model has yet been able to improve the overall simulation efficiency while ensuring the accuracy of the seepage and mass transfer simulation of fractured-matrix rock masses, and no research has yet been able to resolve this contradiction. Summary of the Invention

[0009] This application constructs a three-dimensional fracture-matrix rock mass model that integrates dominant path extraction and the influence of the rock matrix. It fully considers the interaction between fractures and the surrounding rock matrix, ensuring the accuracy of solute transport simulation in the fracture network. At the same time, it reduces the number of fractures in the seepage mass transfer simulation, optimizes the model, reduces computational costs, and improves the overall simulation efficiency and practicality. It provides a reliable and practical numerical implementation method for studying fluid transport and pollutant migration behavior in fractured rock masses at the engineering scale.

[0010] The purpose of this invention is to provide a method for identifying and extracting dominant pathways in rock mass seepage mass transfer simulation that integrates the effects of fractures and matrix.

[0011] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0012] Step 1. Determine the three-dimensional geometric characteristic parameters of the fracture network. Use the Monte Carlo simulation method to generate a random three-dimensional discrete fracture network model. After verifying that the fracture density and number of fractures meet the preset requirements, output the original fracture rock mass model Full-DFN.

[0013] Step 2. A structured model of the fracture network is performed based on bipartite graph theory. This structured modeling abstracts the fracture surface and fracture intersection line as two vertex sets of a bipartite graph. By establishing a mapping relationship between the fracture surface and the intersection line, a mathematical expression of the complex fracture geometry and topology is achieved. Then, combined with the edge-non-intersecting path algorithm, weighted calculations of fracture length, aperture, and permeability constraints are performed to systematically identify and extract dominant paths in the seepage or mass transfer process. The number of dominant paths can be set according to the simulation accuracy and time requirements. Finally, based on the extracted dominant paths, a simplified fractured rock mass sub-model (Sub-DFN) is constructed. The structured modeling of the fracture network based on bipartite graph theory includes the following sub-steps:

[0014] Step 2.1 Define the fracture network as a tuple ,in Represents a collection of fracture surfaces. Represents the set of fracture intersections;

[0015] Step 2.2 Establishing a fracture network To bipartite graph mapping ;in, The vertex in the middle corresponds to the fracture surface, that is ; The vertex in the diagram corresponds to the intersection of the cracks, that is... If and only if the vertex The corresponding fracture surface is the vertex. When one of the cracks involved in the intersection line exists, there is an edge. ;

[0016] Step 2.3 involves characterizing the inflow and outflow boundary information in the fracture network model, within the vertex set... Add source vertex and target vertex ; bipartite graph At the fracture surface Intersection of fracture lines By projecting the images onto the surface separately, we can obtain the projected images. and ;

[0017] Step 2.4 In the bipartite graph Up, search from the source vertex and target vertex Non-intersecting paths Meanwhile, taking into account the crack length Hydraulic resistance related to hydraulic characteristics The factors are assigned different weights. From the vertex To the top transport time It can be expressed by equation (1):

[0018]

[0019] In the formula, , representing a vertex and vertex The distance between the corresponding crack intersection lines; ,in The permeability of the fracture;

[0020] From the source vertex and target vertex path transport time and path The top sides The sum can be expressed by equation (2):

[0021]

[0022] Based on the principle of minimizing weighted values, the path with n non-intersecting edges is selected as the weighted n-shortest path, also known as the "dominant path", where n is the upper limit of the number of non-intersecting dominant paths in the fracture network model.

[0023] Step 3. After generating the Sub-DFN sub-model and the Full-DFN original model of the fractured rock mass, generate triangular mesh data files and tetrahedral mesh data files for the associated fracture network and rock matrix. Specifically, this includes: converting fracture geometric parameters into an input file of a specific format and inputting it into mesh generation software to perform triangular mesh generation on the fracture surface and matrix boundary surface; to ensure that the matrix boundary surface triangular mesh is obtained simultaneously during fracture surface triangular mesh generation, the vertex and center coordinates of the matrix boundary surface must be added to the input file for matrix boundary surface triangular mesh generation; during triangular mesh generation, the mesh can be selectively coarsened or refined using a preset ratio of maximum to minimum mesh size to improve computational efficiency; finally, output the triangular mesh data files.

[0024] After generating the triangular mesh, the rock matrix is ​​then tetrahedral meshed. To ensure that the triangular mesh shape of the fracture surface and matrix boundary remains unchanged during tetrahedral meshing of the rock matrix, the input file should include the node list and face list of the triangular mesh elements of the fracture surface and matrix boundary. After importing the input file into the tetrahedral meshing software, the triangular meshes of the fracture surface and matrix boundary are fixed and constrained, and modification of them is prohibited during the tetrahedral meshing process. Next, mesh quality parameters are set to balance mesh quality and computational efficiency. Finally, tetrahedral meshing is performed, and the tetrahedral mesh data file is output.

[0025] Step 4. Convert the data files of the triangular mesh elements of the fracture surface and matrix boundary surface and the tetrahedral mesh elements of the rock matrix to obtain the fracture-matrix rock mass sub-model Sub-DFM and the fracture-matrix rock mass full-model Full-DFM. The models can be further converted into visualization files or used for seepage and mass transfer simulation and analysis.

[0026] Furthermore, the three-dimensional geometric feature parameters of the fracture network in step 1 are:

[0027] The length, width, and height of the simulation area are determined based on the actual situation of the simulated object;

[0028] The geometric parameters of the fracture network, determined based on the actual situation of the simulated object, include the number, shape, density, and length.

[0029] This invention also provides an application of a method for identifying and extracting dominant paths in rock mass seepage mass transfer simulation that integrates fracture-matrix influences in the field of rock engineering.

[0030] The beneficial effects of this invention are as follows: By introducing a dominant path identification method based on bipartite graph theory and integrating the identified dominant fracture channels with the rock matrix system for integrated modeling, this invention not only fully considers molecular diffusion and matrix adsorption between fractures and the surrounding rock matrix, achieving a higher accuracy simulation of solute transport behavior in the fracture-matrix system, but also significantly reduces the number of fractures and simplifies the fracture network structure through dominant path extraction. While ensuring the characteristics of the main control channels for flow and mass transfer, it reduces computational costs and improves simulation efficiency and engineering practicality. Especially in long-term migration scenarios such as nuclear waste disposal, this invention can effectively predict nuclide migration behavior and has significant engineering application value. Attached Figure Description

[0031] Figure 1 This is a flowchart of the present invention;

[0032] Figure 2 A schematic diagram of the original Full-DFN model of the fractured rock mass;

[0033] Figure 3 This is a bipartite diagram of the fracture surface and its intersection line;

[0034] Figure 4 Sub-DFN bipartite diagram of fractured rock mass sub-model;

[0035] Figure 5 A schematic diagram of the Sub-DFN sub-model of fractured rock mass;

[0036] Figure 6 This is a schematic diagram of the Sub-DFM mesh for the fractured-matrix rock mass sub-model.

[0037] Figure 7 The figure shows the simulation results of solute migration;

[0038] Figure 8 To simulate the solute breakthrough curve;

[0039] Figure 9 This is a simulation of the breakdown time distribution. Detailed Implementation

[0040] To make the above-mentioned objectives, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0041] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0042] Example 1

[0043] This invention provides a method for identifying and extracting dominant pathways in rock mass seepage and mass transfer simulation that integrates the effects of fractures and matrix, such as... Figure 1 As shown, the method includes the following steps:

[0044] Step 1: First, open the dfnWorks open-source software in the Ubuntu system and add variables to the input file to set the length, width and height of the simulation region, the number of fracture network groups, and the number, shape, density and length of fracture surfaces.

[0045] Secondly, based on the dfnGen module of the dfnWorks open-source software, the Monte Carlo simulation method is used to generate, as shown in the figure. Figure 2 The random three-dimensional discrete fracture network model shown in the figure, after verifying that the fracture density and number of fractures meet the preset requirements, outputs the original fractured rock mass model Full-DFN;

[0046] Step 2: Based on bipartite graph theory, a structured model of the fracture network is performed. This structured modeling abstracts the fracture surface and fracture intersection line as two vertex sets of a bipartite graph. By establishing a mapping relationship between the "fracture surface and intersection line," a mathematical expression of the complex fracture geometric topology is achieved, such as... Figure 3 As shown; based on this bipartite graph structure, the connectivity and flow accessibility between fractures can be systematically described, thus providing a unified graph theory framework for dominant path identification; furthermore, combined with the edge-non-intersecting path algorithm, weighted calculation of fracture length, aperture, permeability, and other multi-parameter constraints is performed to systematically identify and extract dominant paths in the seepage or mass transfer process. The number of dominant paths can be set according to the simulation accuracy and time requirements, such as 1, 5, 10, up to an upper limit of n. Then, based on the extracted dominant paths, a simplified fractured rock mass sub-model Sub-DFN is constructed; wherein, the structured modeling of the fracture network based on bipartite graph theory includes the following sub-steps:

[0047] First, define the gap network as a tuple. ,in Represents a collection of fracture surfaces. Represents the set of fracture intersections;

[0048] Secondly, establish a fracture network To bipartite graph mapping ;in, The vertex in the middle corresponds to the fracture surface, that is ; The vertex in the diagram corresponds to the intersection of the cracks, that is... If and only if the vertex The corresponding fracture surface is the vertex. When one of the cracks involved in the intersection line exists, there is an edge. ;

[0049] Then, the bipartite graph At the fracture surface Intersection of fracture lines By projecting the images onto the surface separately, we can obtain the projected images. and To characterize the inflow and outflow boundary information in the fracture network model, in the vertex set... Add source vertex and target vertex ;

[0050] Next, in the bipartite graph Up, search from the source vertex and target vertex Non-intersecting paths Meanwhile, taking into account the crack length Hydraulic resistance related to hydraulic characteristics The factors are assigned different weights. From the vertex To the top transport time It can be expressed by equation (1):

[0051]

[0052] In the formula, , representing a vertex and vertex The distance between the corresponding crack intersection lines; ,in The permeability of the fracture;

[0053] From the source vertex and target vertex path transport time and path The top sides The sum can be expressed by equation (2):

[0054]

[0055] Based on the principle of minimizing weighted values, the path with n non-intersecting edges is selected as the weighted n-shortest path, also known as the "dominant path", where n is the upper limit of the number of non-intersecting dominant paths in the fracture network model.

[0056] Extracting different numbers of dominant paths to construct a sub-DFN model of fractured rock mass, such as... Figure 4 , 5 The figures shown are a bipartite diagram and a schematic diagram of the Sub-DFN submodel of fractured rock mass containing 5 dominant paths.

[0057] Step 3: First, using a Python script, the fracture geometry parameters (including fracture size and fracture normal vector) of the generated Sub-DFN and Full-DFN fractured rock mass sub-models are automatically converted into input files of a specific format. Then, the Delaunay open-source triangulation toolbox DFNsMeshGenerator3D is called to simultaneously perform triangulation on the 3D fracture network and matrix boundary surfaces. To ensure that six matrix boundary surface triangulations are obtained simultaneously during fracture surface triangulation, the vertex and center coordinates of six matrix boundary surfaces need to be added to the input file, and these six matrix boundary surfaces are used as fracture surfaces for triangulation. When generating the triangular mesh, the mesh can be selectively coarsened or refined by setting the ratio of the maximum to the minimum mesh size to improve computational efficiency. Considering the important role of the intersection of the fracture surface and the matrix boundary in seepage and mass transfer, the mesh can be refined in this region to make the calculation of physical processes occurring at the fracture intersection more accurate, such as changes in pressure gradient and velocity. At the same time, the mesh is coarsened in regions far from the fracture interaction to save computational resources. Finally, the triangular mesh data file is output.

[0058] After generating the triangular mesh, the rock matrix is ​​then meshed using tetrahedrons. The main challenge lies in maintaining the triangular mesh of the fracture surface while simultaneously meshing the surrounding rock matrix using tetrahedrons. This invention utilizes a Python script to extract triangular mesh elements from the fracture surface and matrix boundary surface. A set of disjoint 3D piecewise linear composites (PLCs) is constructed in the input file, including a node list and a face list of the triangular mesh elements. The Delaunay tetrahedron generator TetGen is then called to mesh the rock matrix using tetrahedrons. Through these settings, it is ensured that the tetrahedron elements adapt to and remain consistent with the triangular mesh elements of the fracture surface and matrix boundary surface. Finally, a tetrahedron mesh data file is output.

[0059] Step 4: Use a Python script to convert the data files of triangular mesh elements for fracture surfaces and matrix boundaries, and tetrahedral mesh elements for the rock mass matrix, to obtain the Sub-DFM and Full-DFM models of fracture-matrix rock mass. For example... Figure 6 This is a schematic diagram of the mesh for the fracture-matrix sub-DFM model. The diagram shows that the fractures within the three-dimensional cubic computational domain are discretized into triangular meshes, while the surrounding rock matrix is ​​discretized into tetrahedral meshes. Furthermore, the fracture-matrix model can be converted into a visualization file or a finite element mesh for seepage and mass transfer simulations. For example, a .vtk format file can be generated and visualized using Paraview, or a .msh format file can be generated and imported into OpenGeosys (OGS) for seepage and solute migration simulations. Figure 7 The figure shown is a simulation result of solute migration, which shows the concentration distribution of the sub-DFM model of fracture-matrix rock mass when the outlet solute concentration of the full-DFM model of fracture-matrix rock mass reaches 50% at t=0.5PV, which includes 5 dominant paths.

[0060] Example 2

[0061] To evaluate the influence of the rock matrix on solute transport and the accuracy and computational efficiency of the dominant path extraction algorithm of this invention in the simulation of solute migration in fractured rock masses, models considering and not considering matrix effects were established based on different numbers of dominant paths extracted. These included: a fractured rock mass sub-model (Sub-DFN) without considering matrix effects, a fractured rock mass original Full-DFN without considering matrix effects, a fractured-matrix rock mass sub-model (Sub-DFM) considering matrix effects, and a fractured-matrix rock mass full-model (Full-DFM) considering matrix effects. Numerical simulations of solute migration were then compared and analyzed.

[0062] We constructed five sets of models to simulate and compare seepage and mass transfer:

[0063] According to the method of this invention, a geometric model of a fracture network is first randomly generated based on the Monte Carlo method. The fracture network model is a cube with dimensions of 15m × 15m × 15m, composed of randomly generated disk-shaped fractures with equal fracture radii. The fracture permeability follows a log-normal distribution, wherein the mean parameter... The variance parameter σ² = 0.5 represents the original fractured rock mass model, Full-DFN. Subsequently, using the bipartite graph theory-based structured modeling and dominant path identification algorithm proposed in this invention, different numbers of dominant paths are extracted to construct fractured rock mass sub-models, Sub-DFN, containing 1, 5, 10, and n dominant paths, respectively. Next, the fracture surfaces and matrix boundaries of the three-dimensional fracture network are divided into triangular and tetrahedral meshes, resulting in fracture-matrix rock mass sub-models, Sub-DFM, and the full fracture-matrix rock mass model, Full-DFM, which consider matrix diffusion and adsorption and contain 1, 5, 10, and n dominant paths. Finally, numerical simulations of seepage and solute migration are performed.

[0064] like Figure 8The simulation results of one set of models are shown, displaying the normalized solute breakthrough curves of the sub-network and full-network rock mass models corresponding to the DFM model considering matrix effects and the DFN model not considering matrix effects under different numbers of dominant paths. The results show that the solute breakthrough curve of the DFM model considering matrix effects exhibits more significant "early arrival" and "tailing" phenomena, that is, the solute arrives at the outlet earlier, but the concentration decays slowly in the later stage, which is consistent with the physical mechanisms such as molecular diffusion and adsorption retention in the actual fracture-matrix system. In contrast, the solute breakthrough curve of the DFN model not considering matrix effects is relatively steep, and the "tailing" is not obvious, making it difficult to accurately characterize the long-term migration behavior of solutes, thus verifying the necessity of considering matrix effects in modeling.

[0065] also, Figure 8 By comparing the solute breakthrough curves of the Sub-DFM (Sub-DFM) and Full-DFM (Full-DFM) models of fractured-matrix rock masses, it can be seen that as the number of dominant paths increases, the curves of the sub-network models gradually approach those of the original full-network model. Even when only one dominant path is extracted, the breakthrough curve of the Sub-DFM model still reflects the basic characteristics of Full-DFM: "early arrival" and "tailing," and maintains a consistent overall trend with Full-DFM. When the number of dominant paths increases to 5 and 10, the agreement between the breakthrough curves of Sub-DFM and Full-DFM significantly improves. In particular, when n dominant paths are extracted, the breakthrough curves of Sub-DFM and Full-DFM are highly consistent. This proves that the method of this invention, after simplifying the fracture network structure, can still effectively retain the key channels controlling solute migration and reflect the characteristics of solute migration, thereby ensuring the accuracy and reliability of the simulation results.

[0066] As the number of extracted dominant paths increases, the accuracy of the Sub-DFM model approaches that of the Full-DFM model, but its computational complexity is significantly lower. This indicates that the simplified fracture-matrix rock mass sub-model Sub-DFM of this invention fully achieves the core objective of significantly reducing computational costs while maintaining simulation accuracy, providing a feasible path for solving computational challenges in large-scale engineering applications. Simultaneously, this also demonstrates that the network rock mass model considering matrix effects in this invention can effectively retain the key channels controlling solute migration even after simplification, reflecting the core characteristics of solute migration. Therefore, the network rock mass model considering matrix effects in this invention has strong practicality and engineering significance.

[0067] Figure 9The figure shows the breakdown time distribution results of five fracture-matrix models for seepage and mass transfer simulations. The 45° reference line in the figure represents the breakdown time that is completely consistent with that of Full-DFM. The results show that when only one dominant path is included, the breakdown time of Sub-DFM deviates significantly from that of Full-DFM, with most points located above the reference line and exhibiting high dispersion. As the number of dominant paths increases, the data points gradually move closer to the reference line, and the breakdown time of Sub-DFM gradually shortens and approaches the result of Full-DFM. When the number of dominant paths reaches 10, the breakdown time of Sub-DFM is in high agreement with that of Full-DFM, further demonstrating that the method of this invention effectively achieves model simplification and computational efficiency improvement while ensuring simulation accuracy.

[0068] Furthermore, with the same number of dominant paths, the breakdown times of the DFM and DFN models showed significant differences. The circular data points representing the breakdown time of the DFM model were generally shifted to the left, indicating an earlier breakdown time; while the diamond data points representing the breakdown time of the DFN model were generally shifted to the right, indicating a relatively delayed breakdown time. This further reveals that the influence of the DFM model considering matrix effects on the solute transport process is necessary and important.

[0069] The above analysis shows that the number of dominant paths is a key factor affecting the model's simulation accuracy and stability. However, considering the matrix effect can guarantee its prediction accuracy to a certain extent. Significant differences exist in the simulation results between the DFM model considering the matrix effect and the DFN model not considering it. In practical fracture network modeling and risk assessment, constructing a three-dimensional fracture network rock mass model that simultaneously considers the influence of dominant paths and the rock matrix helps improve the accuracy of solute transport simulation in fracture networks and enhances overall simulation efficiency, which has great practical significance.

[0070] Those skilled in the art can readily make various changes and modifications based on the provided textual description, drawings, and claims, without departing from the spirit and scope of the invention as defined by the claims. Any modifications or equivalent variations made to the above embodiments based on the technical concept and essence of the invention fall within the protection scope defined by the claims of this invention.

Claims

1. A method for identifying and extracting dominant pathways in rock mass seepage and mass transfer simulation that integrates fracture-matrix influences, characterized in that, Includes the following steps: Step 1. Determine the three-dimensional geometric characteristic parameters of the fracture network. Use the Monte Carlo simulation method to generate a random three-dimensional discrete fracture network model. After verifying that the fracture density and number of fractures meet the preset requirements, output the original fracture rock mass model Full-DFN. Step 2. Based on bipartite graph theory, a structured model of the fracture network is constructed; then, combined with the edge-non-intersecting path algorithm, multi-parameter constraints are calculated with weights to identify and extract the dominant paths in the seepage or mass transfer process; and based on the extracted dominant paths, a simplified fractured rock mass sub-model Sub-DFN is constructed. Step 3. After generating the Sub-DFN sub-model of fractured rock mass and the Full-DFN original model of fractured rock mass, generate triangular mesh data files and tetrahedral mesh data files that associate the fracture network and the rock matrix; Step 4. Convert the data files of triangular mesh elements of fracture surfaces and matrix boundaries, and tetrahedral mesh elements of rock matrix to obtain the Sub-DFM and Full-DFM models of fracture-matrix rock mass. Then, convert the models into visualization files or use them for seepage and mass transfer simulation and analysis.

2. The method for identifying and extracting dominant pathways in rock mass seepage and mass transfer simulation based on the influence of fractures and matrix, as described in claim 1, is characterized in that... The three-dimensional geometric feature parameters of the fracture network in step 1 are: The length, width, and height of the simulation area are determined based on the actual situation of the simulated object; The geometric parameters of the fracture network, determined based on the actual situation of the simulated object, include the number, shape, density, and length.

3. The method for identifying and extracting dominant pathways in rock mass seepage and mass transfer simulation based on the influence of fractures and matrix, as described in claim 1, is characterized in that... Step 2, which involves structurally modeling the fracture network based on bipartite graph theory, includes the following sub-steps: Step 2.1 Define the fracture network as a tuple ,in Represents a collection of fracture surfaces. Represents the set of fracture intersection lines; Step 2.2 Establishing a fracture network To bipartite graph mapping ;in, The vertex in the middle corresponds to the fracture surface, that is ; The vertex in the diagram corresponds to the intersection of the crack, i.e. If and only if the vertex The corresponding fracture surface is the vertex. When one of the cracks involved in the intersection line exists, there is an edge. ; Step 2.3 involves characterizing the inflow and outflow boundary information in the fracture network model, within the vertex set... Add source vertex and target vertex ; bipartite graph At the fracture surface Intersection of fracture lines By projecting the images onto the surface separately, we can obtain the projected images. and ; Step 2.4 In the bipartite graph Up, search from the source vertex and target vertex Non-intersecting paths Meanwhile, taking into account the crack length Hydraulic resistance related to hydraulic characteristics The factors are assigned different weights. From the vertex To the top transport time It can be expressed by equation (1): In the formula, , representing a vertex and vertex The distance between the corresponding fracture intersection lines; ,in The permeability of the fracture; From the source vertex and target vertex path transport time and path The top sides The sum is expressed by equation (2): Based on the principle of minimizing weighted values, the path with n non-intersecting edges is selected as the weighted n-shortest path, also known as the "dominant path", where n is the upper limit of the number of non-intersecting dominant paths in the fracture network model.

4. The method for identifying and extracting dominant pathways in rock mass seepage and mass transfer simulation based on the influence of fractures and matrix, as described in claim 1, is characterized in that... In step 2, the structured modeling involves abstracting the fracture surface and the fracture intersection line as two sets of vertices in a bipartite graph. By establishing a mapping relationship between the "fracture surface and the intersection line", a mathematical expression of the complex fracture geometric topology can be achieved.

5. The method for identifying and extracting dominant pathways in rock mass seepage and mass transfer simulation based on the influence of fractures and matrix, as described in claim 1, is characterized in that... In step 2, the number of dominant paths is set according to the simulation accuracy and time requirements.

6. The method for identifying and extracting dominant pathways in rock mass seepage and mass transfer simulation based on the influence of fractures and matrix, as described in claim 1, is characterized in that... In step 2, the parameters involved in the multi-parameter constraint include, but are not limited to, fracture length, aperture, and permeability.

7. The method for identifying and extracting dominant pathways in rock mass seepage and mass transfer simulation based on the influence of fractures and matrix, as described in claim 1, is characterized in that... Step 3 includes the following steps: S3.1 Convert the fracture geometry parameters into an input file of a specific format and input it into the mesh generation software to perform triangular mesh generation on the fracture surface and the matrix boundary surface. To ensure that the triangular mesh of the matrix boundary surface is obtained simultaneously when performing triangular mesh generation on the fracture surface, the vertex and center coordinates of the matrix boundary surface must be added to the input file for triangular mesh generation on the matrix boundary surface. During triangular mesh generation, the mesh is selectively coarsened or refined by using a preset ratio of the maximum to the minimum mesh size to improve computational efficiency. Finally, the triangular mesh data file is output. After generating the triangular mesh data file in step S3.2, tetrahedral meshing is performed on the rock matrix. To ensure that the triangular mesh shape of the fracture surface and matrix boundary remains unchanged during tetrahedral meshing of the rock matrix, the input file should include the node list and face list of the triangular mesh elements of the fracture surface and matrix boundary. After importing the input file into the tetrahedral meshing software, the triangular meshes of the fracture surface and matrix boundary are fixed and constrained, and modification of them is prohibited during the tetrahedral meshing process. Then, mesh quality parameters are set to balance mesh quality and computational efficiency. Finally, tetrahedral meshing is performed, and the tetrahedral mesh data file is output.

8. The method for identifying and extracting dominant pathways in rock mass seepage and mass transfer simulation incorporating fracture-matrix influence, as described in any one of claims 1-7, is characterized in that... Applications in the field of rock engineering.

Images