A method for simulating seepage in a multi-scale fractured medium based on coupled finite elements

By constructing a multi-scale fractured medium seepage simulation method, and combining the finite element method and the Galerkin finite element method, the problems of high computational cost and complex mesh generation in multi-scale fractured medium seepage simulation are solved, and accurate simulation of groundwater flow in complex rock masses is achieved.

CN120542189BActive Publication Date: 2025-11-04HEFEI UNIV OF TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202510687386.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-05-27
Publication Date
2025-11-04
Estimated Expiration
2045-05-27

AI Technical Summary

Technical Problem

Existing technologies struggle to accurately characterize the seepage characteristics of multi-scale fractured media when simulating groundwater flow in complex fractured rock masses, resulting in high computational costs and complex mesh generation.

Method used

A multi-scale fractured medium seepage simulation method based on coupled finite element method is adopted. By constructing a multi-scale fracture geometric model, the coupling relationship between small, medium and large scale fractures and matrix is ​​established, the mesh generation process is simplified, and the Galerkin finite element method is used for numerical discretization to establish a multi-scale discrete fracture-matrix model.

Benefits of technology

It achieves accurate simulation of the seepage process in multi-scale fractured media, reduces computational costs, and truly reflects the complex impact of fractures of different scales on the overall water flow.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120542189B_ABST
    Figure CN120542189B_ABST
Patent Text Reader

Abstract

The application relates to a seepage simulation method in a multi-scale fractured medium based on coupling finite elements, which comprises the following steps: 1. acquiring statistical distribution characteristics of related parameters of multi-scale fractures in a simulation domain, and constructing a multi-scale fracture geometric model; 2. respectively performing intersection line detection on each scale fracture in the multi-scale fracture geometric model; 3. simplifying the multi-scale fracture geometric model; 4. adopting coupling units to establish coupling relationships among the each scale fracture models; and 5. simulating a multi-scale discrete fracture-matrix model. The application is based on a coupling finite element method, and couples three scale fracture conceptual models, so that the difficulties in calculation precision, calculation cost and mesh partitioning of a single traditional model in solving multi-scale problems are avoided, and a feasible new way is provided for large-scale complex rock mass underground water flow analysis.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the field of fracture water simulation in complex fractured rock mass, and in particular, to a seepage simulation method in multi-scale fractured medium based on coupled finite elements. BACKGROUND

[0002] Fractured medium is an important channel for groundwater flow in rock mass, and widely exists in natural groundwater systems and many rock mass engineering, such as mining, groundwater resource development, nuclear waste geological disposal site, tunnel and underground cavern engineering construction and other fields. The existence of fractures significantly changes the seepage characteristics of rock mass, and affects the migration path and flow velocity of groundwater. Therefore, accurately describing the water flow process in fractured medium has great scientific significance and engineering value.

[0003] However, natural fractured rock mass usually shows complex geometric structure and scale diversity characteristics. Actual fracture network often exists multi-scale fractures, including a few large-scale fractures with clear geometric shape, and a large number of small-scale fractures with random space position. This scale difference significantly increases the difficulty of accurate characterization of fracture parameters, and also poses great challenges to numerical simulation methods. At present, common groundwater flow simulation methods mainly include equivalent porous medium (EPM) model, discrete fracture network (DFN) model and discrete fracture-matrix (DFM) model. Among them, EPM is simple to calculate, but it equates fractures to anisotropic continuous permeability field, ignores the real geometric structure of fracture network, and is difficult to accurately capture the local flow rule of groundwater. DFN model can more truly reflect the guiding effect of fracture network on water flow by accurately describing the geometric characteristics of fractures, but the calculation cost is high when the number of fractures is large and the scale span is large. The DFM model considers the spatial matching of fractures and matrix grid, which leads to the modeling of complex fracture-matrix system not being friendly, such as the generation of degenerated three-dimensional grid in the area with small intersection angle will lead to numerical solution not converging. Although the grid refinement can alleviate this influence, but the excessively refined grid will bring more serious calculation and memory pressure.

[0004] In summary, the existing technology is still limited to single scale in the seepage simulation of fracture-matrix system, and has limited ability to describe multi-scale fracture-matrix system. SUMMARY

[0005] To solve the technical problems proposed in the background art, the present application provides an efficient and accurate seepage simulation method in multi-scale fractured medium based on coupled finite elements, so as to couple the conceptual model of three-scale fractures, to avoid the difficulties of single traditional model in solving multi-scale problems in terms of calculation accuracy, calculation cost and grid partition, and to provide a new feasible way for large-scale complex rock mass groundwater flow analysis.

[0006] To achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0007] The present invention provides a method for simulating seepage in multi-scale fractured media based on coupled finite element method, characterized by the following steps:

[0008] Step 1: Obtain the statistical distribution characteristics of relevant parameters of multi-scale fractures in the simulation domain, and construct a multi-scale fracture geometric model based on the statistical distribution characteristics; the multi-scale fractures include: small-scale fractures, medium-scale fractures and large-scale fractures;

[0009] Step 2: Perform intersection line detection on the cracks at each scale in the multi-scale crack geometric model to obtain the intersection lines between cracks at each scale and the intersection lines between cracks and the boundary of the simulation domain;

[0010] Step 3: Based on the intersection lines between fractures at each scale and the intersection lines between fractures and the boundary of the simulation domain, the multi-scale fracture geometric model is simplified to obtain the conceptual model of the fracture geometric model at each scale, including: equivalent porous medium model, discrete fracture-matrix model, and discrete fracture network model.

[0011] Step 4: Establish the coupling relationship between the conceptual models of the fracture geometric models at each scale to obtain the multi-scale discrete fracture-matrix model;

[0012] Step 5: Simulate the multi-scale discrete fracture-matrix model to obtain a simulation diagram of the head distribution of the multi-scale discrete fracture-matrix model.

[0013] The present invention provides a method for simulating seepage in multi-scale fractured media based on coupled finite element method, wherein step 3 includes:

[0014] Step 301: Calculate the volume of the characterization unit of the small-scale fracture geometric model and the matrix, and determine the overall permeability coefficient of the characterization unit through numerical simulation or experiment. Then, use the overall permeability coefficient as the equivalent permeability coefficient after coupling the small-scale fracture and the matrix, and establish an equivalent porous medium model of coupling the small-scale fracture geometric model and the matrix.

[0015] Step 302: Using the boundary of the simulated domain as the outer boundary and the intersection of the large-scale fracture geometric model and the matrix as the inner boundary, the simulated domain and the large-scale fracture are spatially matched and discretized using the three-dimensional Delaunay tetrahedral mesh generation algorithm to obtain the discrete fracture-matrix model of the large-scale fracture. The matrix is ​​discretized into spatial tetrahedral matrix elements, and the large-scale fracture is discretized into triangular large-scale fracture elements, which are used as one face of the matrix elements. The nodes of the large-scale fracture elements have the same index as the nodes of the matrix elements.

[0016] Step 303, only considering the intersection line between mesoscale fractures and the intersection line between mesoscale fractures and the boundary of the simulation domain, using a two-dimensional Delaunay triangular unit grid generation algorithm to independently grid each fracture in the mesoscale fracture geometry model, generate corresponding triangular mesoscale fracture units, and form a discrete fracture network model of the mesoscale fracture.

[0017] Further, the step 4 comprises:

[0018] Step 401, replacing the permeability coefficient of the matrix in the discrete fracture-matrix model of the large-scale fracture with the overall permeability coefficient of the equivalent porous medium model to realize the coupling between the matrix, the small-scale fracture and the large-scale fracture;

[0019] Using the Galerkin finite element method to numerically discretize the matrix unit and the large-scale fracture unit respectively, thereby obtaining the algebraic form of the discrete fracture-matrix model of the large-scale fracture using formula (1):

[0020]

[0021] In formula (1), D m and D fL are the hydraulic conductivity matrix of the matrix unit node set and the hydraulic conductivity matrix of the large-scale fracture unit node set; H m and H fL are the head value vector of the matrix unit node set and the head value vector of the large-scale fracture unit node set; F m and F fL are the water quantity vector of the matrix unit node set and the water quantity vector of the large-scale fracture unit node set, represents the equivalent to, represents the total hydraulic conductivity matrix of the discrete fracture-matrix model of the large-scale fracture, represents the total head value vector of the discrete fracture-matrix model of the large-scale fracture, represents the total water quantity vector of the discrete fracture-matrix model of the large-scale fracture.

[0022] Step 402, using the Galerkin finite element method to numerically discretize the triangular unit grid of the discrete fracture network model of the mesoscale fracture, thereby combining formula (2) with the algebraic form of the discrete fracture-matrix model of the large-scale fracture:

[0023]

[0024] In formula (2): represents the total hydraulic conductivity matrix of the discrete fracture-matrix model of the large-scale fracture, D Ma transmissibility matrix of a set of nodes of a meso-scale fracture element, a total head value vector of a discrete fracture-matrix model representing macro-scale fractures, H M a head value vector of a set of nodes of a meso-scale fracture element, a total flux vector of a discrete fracture-matrix model representing macro-scale fractures, F M a flux vector of a set of nodes of a meso-scale fracture element, F E a water flow exchange term between a grid node of a meso-scale fracture element and a node of a matrix element, D is a total transmissibility matrix of a multi-scale discrete fracture-matrix model, H is a total head value vector of a multi-scale discrete fracture-matrix model, F is a total flux vector of a multi-scale discrete fracture-matrix model, represents an equivalent to;

[0025] Step 403, searching for an external matrix element of each node of a meso-scale fracture element, and taking the meso-scale fracture element node corresponding to the index of the searched matrix element as a coupling node, so as to form a 5-node coupling element together with the 4 nodes of the external matrix element;

[0026] Step 404, based on all the 5-node coupling elements, obtaining an interpolated head value of the coupling node in the matrix element by using the basis function on the matrix element node;

[0027] Step 405, based on the head value of the meso-scale fracture element node and the interpolated head value of the coupling node, calculating a head difference in the spatial position of the meso-scale fracture element node;

[0028] Step 406, based on the 5-node coupling element and the head difference, obtaining a water flow exchange term F between the meso-scale fracture element node and the matrix element node by using formula (3) E :

[0029]

[0030] In formula (3), n f is the number of meso-scale fracture element nodes; N m is a vector composed of the values of the basis function of the matrix element at the positions of the meso-scale fracture element nodes; represents a vector in which only the position of the i-th meso-scale fracture element node of the meso-scale fracture element is 1, and the rest is 0; θ represents a parameter of water flux exchange between the meso-scale fracture and the matrix; S i is a measure of an influence area related to the i-th meso-scale fracture element node; a total head value vector of a discrete fracture-matrix model representing macro-scale fractures, H is a head value vector of a meso-scale fracture element node, T represents transposition, represents a water flow exchange matrix between the mesoscale fracture element nodes and the matrix element nodes;

[0031] Step 407, the algebraic form of the water flow exchange term and the algebraic form of the discrete fracture-matrix model of the large-scale fracture are combined, so that a multi-scale discrete fracture-matrix model is obtained by using formula (4):

[0032]

[0033] In formula (4), represents a total water conductivity matrix of the multi-scale discrete fracture-matrix model, represents a water flow exchange matrix between the mesoscale fracture element nodes and the matrix element nodes, represents a water head value vector of the multi-scale discrete fracture-matrix model, represents a water quantity vector of the multi-scale discrete fracture-matrix model.

[0034] Further, the step 5 comprises:

[0035] Step 501, setting a boundary condition on the boundary of the multi-scale discrete fracture-matrix model, wherein the boundary condition is a constant water head boundary or a flow boundary;

[0036] Step 502, solving the multi-scale discrete fracture-matrix model under the boundary condition by using a direct or iterative solving method to obtain a water head numerical solution of each node in the entire simulation domain, which is used to draw a groundwater water head distribution map.

[0037] The electronic device comprises a memory and a processor, and the memory is used for storing a program supporting the processor to execute the seepage simulation method in a multi-scale fracture medium.

[0038] The computer readable storage medium stores a computer program, and the computer program is executed by the processor to execute the steps of the seepage simulation method in the multi-scale fracture medium.

[0039] Compared with the prior art, the seepage simulation method in the multi-scale fracture medium has the following beneficial effects:

[0040] 1. This invention fully considers the multi-scale characteristics of rock mass fractures, classifying fracture media into three scales: small-scale, medium-scale, and large-scale. It then employs an equivalent porous media model, a discrete fracture network model, and a discrete fracture-matrix model to generalize fractures of different scales, respectively. In this way, the invention solves the problem that traditional models cannot simultaneously and accurately characterize the dominant role of large-scale fractures in overall water flow, the local water-conducting effect of medium-scale fractures, and the subtle seepage characteristics of small-scale fractures. This allows the entire seepage process to more realistically reflect the complex influence of fractures of different scales on overall water flow.

[0041] 2. This invention employs the coupled finite element method, which establishes the coupling relationship between geometrically complex medium-sized cracks and large-scale and small-scale cracks and the matrix through coupled nodes and coupled elements. This simplifies the complex mesh generation process and avoids the high computational cost of traditional methods when dealing with multi-scale problems. Attached Figure Description

[0042] Figure 1 This is a flowchart of a method for simulating seepage in multi-scale fractured media based on coupled finite element method according to the present invention;

[0043] Figure 2 This is a schematic diagram of the multi-scale fracture geometry model and mesh generation of the present invention;

[0044] Figure 3 This is a schematic diagram illustrating the construction principle of the coupling unit of the present invention;

[0045] Figure 4 This is a schematic diagram of the coupling node head interpolation calculation according to the present invention;

[0046] Figure 5 This is a schematic diagram of the head distribution in the simulation domain and fissures of the present invention. Detailed Implementation

[0047] In this embodiment, to address the difficulties in spatial discretization and low computational efficiency caused by the multi-scale nature and geometric complexity of fractured media in natural rock masses, a multi-scale fracture simulation method combining matched and unmatched mesh partitioning is proposed. This method involves fracture parameter generation, spatial discretization, fracture-matrix mesh coupling, and hydraulic head calculation, belonging to the field of hydrogeological numerical simulation technology. (See also...) Figure 1 This method for simulating seepage in multi-scale fractured media based on coupled finite element method includes:

[0048] Step 1: Obtain the statistical distribution characteristics of relevant parameters of multi-scale fractures in the simulation domain, and construct a multi-scale fracture geometric model based on the statistical distribution characteristics; multi-scale fractures include: small-scale fractures, medium-scale fractures and large-scale fractures;

[0049] It should be noted that the simulation domain here is the range set before the actual groundwater flow is simulated, based on engineering requirements. According to the embodiments provided in this application, the statistical characteristics of fracture geometric parameters include the fracture's dip direction, angle, length, and width, etc., which are not specifically limited in this application. The setting of the simulation domain and the acquisition of the statistical characteristics of fracture geometric parameters are described in detail in the prior art and will not be elaborated further here.

[0050] Based on the statistical distribution characteristics of the multi-scale fracture-related parameters shown in Table 1, the Monte Carlo method was used to generate a specific number of geometric parameters for small-scale, medium-scale, and large-scale fractures.

[0051] Table 1. Statistical distribution characteristics of multi-scale fracture-related parameters

[0052]

[0053] like Figure 2 The geometric model and boundary conditions of large-scale and mesoscale cracks shown in part (a) are illustrated in this embodiment. The multi-scale crack geometric model accurately depicts large-scale and mesoscale cracks. Since small-scale cracks are implicitly included in the calculation process, they are not accurately depicted in the figure.

[0054] Step 2: Perform intersection line detection on the fractures at each scale in the multi-scale fracture geometric model to obtain the intersection lines between fractures at each scale and the intersection lines between the fractures and the boundary of the simulation domain:

[0055] Regarding the above steps, this embodiment uses a disk model as the crack geometry model, specifically:

[0056] Step 201: Based on the generated fracture inclination α, dip angle β, and center point P(x0,y0,z0), the equation of the space plane can be calculated:

[0057]

[0058] Step 202: Based on the generated crack radius r and center point P(x0,y0,z0), the equation of the spatial sphere can be determined:

[0059]

[0060] Step 203: By simultaneously solving the equations of the spatial plane and the spatial sphere, the expression for the cracked spatial disk model can be obtained;

[0061] Step 204: Repeat the above steps for all generated crack parameters until a disk model of all cracks is obtained;

[0062] For the above-mentioned crack disc model, it should be noted that the disc model is only a preferred solution in the present embodiment, and the intersection calculation of other crack models, such as spatial polygon, spatial ellipse and the like, is described in detail in the prior art, and will not be described in detail here.

[0063] Step 205, based on the equation of the disc model of all cracks, the intersection lines of each crack disc model with other crack disc models and the boundary of the simulation domain are calculated in a loop, and the calculated intersection line results are stored;

[0064] Step 206, based on all intersection lines in the same spatial plane, the intersection points of all intersection lines are calculated according to the spatial expression of the intersection line, and the calculated intersection point results are stored;

[0065] For the sub-step in step 2 above, it should be noted that the intersection lines between the cracks in the discrete crack network model of each scale crack and the intersection lines between the cracks and the boundary of the simulation domain are detected independently. Specifically, the intersection lines of the small-scale cracks are only detected with all small-scale cracks and the boundary of the simulation domain, the intersection lines of the medium-scale cracks are only detected with all medium-scale cracks and the boundary of the simulation domain, and the intersection lines of the large-scale cracks are only detected with all large-scale cracks and the boundary of the simulation domain, without the need to detect the intersection lines between cracks of different scales.

[0066] Step 3, based on the intersection lines between cracks of each scale and the intersection lines between cracks and the boundary of the simulation domain, the multi-scale crack geometry model is simplified to obtain a conceptual model of the crack geometry model of each scale, including: an equivalent porous medium model, a discrete crack-matrix model, and a discrete crack network model:

[0067] Step 301, calculate the volume of the small-scale crack geometry model and the matrix representation unit, and determine the overall permeability coefficient of the representation unit by numerical simulation or experiment, so as to take the overall permeability coefficient as the equivalent permeability coefficient of the small-scale crack and the matrix coupling, and establish an equivalent porous medium model of the small-scale crack geometry model and the matrix coupling;

[0068] For the above-mentioned step 301, it should be noted that the treatment of small-scale cracks is to implicitly include the water conductivity performance of the cracks into the matrix to convert into an equivalent porous medium model, that is, the small-scale cracks and the matrix are regarded as an equivalent enhanced matrix. The method for calculating the overall permeability coefficient of the small-scale crack geometry model representation unit volume and converting into an equivalent porous medium is described in detail in the prior art, and will not be described in detail here.

[0069] In the present embodiment, the permeability coefficients of the equivalent porous medium model in the X-axis, Y-axis and Z-axis three main directions are 1.38×10 -8 m / s, 1.37×10 -8 m / s and 1.24×10-8 m / s.

[0070] Step 302, the simulation domain boundary is taken as an outer boundary, and the intersection line of the large-scale fracture geometry model and the outer boundary is taken as an inner boundary, so that the simulation domain and the large-scale fracture are spatially matched and meshed by using a three-dimensional Delaunay tetrahedral mesh generation algorithm, to obtain a discrete fracture-matrix model of the large-scale fracture; wherein the matrix is discretized into matrix units of spatial tetrahedrons, and the large-scale fracture is discretized into triangular large-scale fracture units and taken as a certain face of the matrix units; and the nodes of the large-scale fracture units and the nodes of the matrix units have the same index;

[0071] For the step 302, it needs to be explained that, when the large-scale fracture matching mesh is divided, the disc model of the large-scale fracture, the intersection line of the large-scale fracture and the simulation domain boundary are taken as boundary constraints of the mesh division. Specifically, when the mesh is divided, one face of the mesh near the plane boundary is strictly matched with the boundary in geometry, one edge of the mesh near the intersection line boundary is strictly matched with the intersection line in geometry, and one node of the mesh near the intersection point is strictly matched with the intersection point in geometry. How to perform the matching mesh division and obtain high-quality mesh units is described in detail in the prior art, and will not be described in detail here.

[0072] Step 303, only considering the intersection line between the mesoscale fractures and the intersection line between the mesoscale fractures and the simulation domain boundary, using a two-dimensional Delaunay triangular unit mesh generation algorithm to independently mesh each fracture in the mesoscale fracture geometry model, to generate corresponding triangular mesoscale fracture units and form a discrete fracture network model of the mesoscale fracture.

[0073] For the independent meshing of the mesoscale fracture geometry model in the step 303, it needs to be explained that, unlike the meshing of the large-scale fracture, when the mesoscale fracture mesh is divided, only the mesoscale fracture geometry model needs to be discretized, without considering the geometric matching between the mesoscale fracture and the matrix tetrahedron. Since each mesoscale fracture is represented by a disc model on a spatial plane, the disc model can be rotated into a planar circle on a two-dimensional plane by using a normal vector of the spatial plane. On this basis, a two-dimensional Delaunay triangular unit mesh generation algorithm is used to generate triangular units of each mesoscale fracture, with the two-dimensional planar circle and the internal intersection line as boundary constraints. Then, the obtained mesh nodes are rotated again according to the normal vector to obtain triangular mesoscale fracture units in three-dimensional space.

[0074] In the specific embodiments of the application, the mesh results obtained by the step 302 and the step 303 Figure 2Fig. 2 is a schematic diagram of a discrete grid of large-scale fractures and meso-scale fractures shown in part (b) of Fig. 1, wherein a white tetrahedron unit represents a matrix unit, a blue triangular unit represents a large-scale fracture unit, and a green triangular unit represents a meso-scale fracture unit. The two groups of grids are independent of each other in space.

[0075] Step 4, coupling the conceptual models of the fracture geometries of different scales to obtain a multi-scale discrete fracture-matrix model:

[0076] Step 401, replacing the permeability coefficient of the matrix in the discrete fracture-matrix model of the large-scale fractures with the overall permeability coefficient of the equivalent porous medium model to realize the coupling between the matrix, the small-scale fractures and the large-scale fractures;

[0077] The Galerkin finite element method is used to discretize the matrix units and the large-scale fracture units, respectively, so as to obtain the algebraic form of the discrete fracture-matrix model of the large-scale fractures by using formula (1):

[0078]

[0079] In formula (1), D m and D fL are the transmissibility matrices of the node set of the matrix units and the node set of the large-scale fracture units, respectively; H m and H fL are the head value vectors of the node set of the matrix units and the node set of the large-scale fracture units, respectively; F m and F fL are the flow vectors of the node set of the matrix units and the node set of the large-scale fracture units, respectively, represents the equivalent of represents the total transmissibility matrix of the discrete fracture-matrix model of the large-scale fractures, represents the total head value vector of the discrete fracture-matrix model of the large-scale fractures, represents the total flow vector of the discrete fracture-matrix model of the large-scale fractures.

[0080] For the above step 401, it should be noted that the transmissibility performance of the small-scale fractures is implicitly combined with the matrix, and does not exhibit a specific spatial geometric structure, so the equivalent porous medium model of the small-scale fractures can be regarded as a continuous medium in the simulation region, and the permeability coefficient thereof can be directly used as the permeability coefficient of the matrix in the discrete fracture-matrix model of the large-scale fractures.

[0081] Step 402, using the Galerkin finite element method to numerically disperse the triangular element grid of the discrete fracture network model of the mesoscale fracture, so as to combine formula (2) with the algebraic form of the discrete fracture-matrix model of the large-scale fracture:

[0082]

[0083] In formula (2): D represents the total hydraulic conductivity matrix of the discrete fracture-matrix model of the large-scale fracture, D M is the hydraulic conductivity matrix of the node set of the mesoscale fracture element, H represents the total head value vector of the discrete fracture-matrix model of the large-scale fracture, H M is the head value vector of the node set of the mesoscale fracture element, F represents the total water amount vector of the discrete fracture-matrix model of the large-scale fracture, F M is the water amount vector of the node set of the mesoscale fracture element, F E is the water flow exchange term between the node set of the mesoscale fracture element and the node set of the matrix element, D is the total hydraulic conductivity matrix of the multi-scale discrete fracture-matrix model, H is the total head value vector of the multi-scale discrete fracture-matrix model, and F is the total water amount vector of the multi-scale discrete fracture-matrix model, represents the equivalence of;

[0084] For the above step 402, it should be noted that the grids of the large-scale discrete fracture-matrix model and the mesoscale discrete fracture network model are independent of each other, and the index systems of the two are also independent, so it is necessary to combine the algebraic matrices of the two according to the diagonal matrix form according to the matrix calculation principle. The two groups of grids need to be coupled according to the water flow exchange term F E in the formula, and the specific derivation of the water flow exchange term will be introduced later.

[0085] Step 403, retrieving the external matrix element of each mesoscale fracture element node, and taking the mesoscale fracture element node corresponding to the index of the retrieved matrix element as the coupling node, so as to form a 5-node coupling element together with the 4 nodes of the external matrix element;

[0086] As shown in the coupling element construction principle diagram Figure 3 , it should be noted that, for the sake of understanding, the diagram takes the coupling element construction in a two-dimensional grid as an example. The blue line segment in the figure represents the large-scale fracture, the green line segment represents the mesoscale fracture, and the red dot is the mesoscale fracture element node. As shown in the two-dimensional coupling element diagram Figure 3 (b) part, for each mesoscale fracture element node, the constructed coupling element is unique. As shown in Figure 4The coupling unit in the three-dimensional problem shown in part (c) of the above can be constructed according to the construction principle of the two-dimensional coupling unit, and can be constructed according to the spatial position relationship between the point and the tetrahedron matrix unit.

[0087] For the above step 403, in a specific embodiment, all meso-scale fracture unit nodes are traversed, for each meso-scale fracture unit node, all matrix units are further traversed, and the matrix units inside or on the boundary surface of the meso-scale fracture unit node are retrieved through the spatial position relationship between the four nodes of the matrix unit and the meso-scale fracture unit node, and are stored in the corresponding index position of the meso-scale fracture unit node.

[0088] Step 404, based on all 5-node coupling units, the interpolation head value of the coupling node in the matrix unit is obtained by using the basis function on the matrix unit node;

[0089] For the above step 404, as shown in the two-dimensional coupling node head interpolation diagram shown in part (a.b) of the above, Figure 4 In a specific embodiment of a three-dimensional problem, for each coupling unit e, the basis function of the matrix unit is used to interpolate the internal coupling node according to formula (3.1):

[0090] (3.1)

[0091] In formula (3.1): represents the interpolation head value of the coupling node of the e-th coupling unit, i represents the index of the matrix unit node of the coupling unit, represents the value of the basis function corresponding to the i-th matrix unit node of the e-th coupling unit at the (x, y, z) coordinates, represents the head value of the i-th matrix unit node of the e-th coupling unit. x, y, z are the spatial coordinates of the coupling node of the e-th coupling unit. represents the vector composed of the values of the basis function of the matrix unit node of the e-th coupling unit at the coupling node position, represents the head value vector of the matrix unit node of the e-th coupling unit.

[0092] Step 405, based on the head value of the meso-scale fracture unit node and the interpolation head value of the coupling node, the head difference of the meso-scale fracture unit node in the spatial position is calculated;

[0093] For the above step 405, in a specific embodiment, for each coupling unit e, formula (3.2) is used to calculate the head difference by combining formula (3.1):

[0094] (3.2) ​

[0095] in formula (3.2) is: an interpolated head value on the coupling node representing the e-th coupling element, a head value on the meso-scale fracture element node of the e-th coupling element, a vector consisting of values of the basis function on the coupling node position of the matrix element node of the e-th coupling element, a head value vector of the matrix element node of the e-th coupling element a head value vector of the meso-scale fracture element node of the e-th coupling element.

[0096] It should be noted that, for the above-mentioned basis function, the embodiment uses a linear basis function, and the method provided by the application is also applicable to basis functions in other finite element methods.

[0097] In step 406, based on the 5-node coupling element and the head difference, the water flow exchange term F between the meso-scale fracture element node and the matrix element node is obtained by using formula (3) E ;

[0098] The specific derivation process of the water flow exchange term F E is as follows:

[0099] The water flow exchange between the meso-scale fracture and the matrix is represented by using formula (3.3):

[0100] (3.3)

[0101] In formula (3.3): represents the water flow of the fracture flowing into the matrix, T is a cross-medium transmission parameter, h f and h m are head values of the matrix and the fracture at the same position, respectively;

[0102] For the fracture, the water flow exchange term is the water flow flowing out of the fracture to supply the surrounding matrix, and is equal in water flow, which can be represented by using formula (3.4):

[0103] (3.4)

[0104] In formula (3.4): represents the water flow of the matrix supplied by the fracture, represents the water flow of the fracture flowing into the matrix, T is a cross-medium transmission parameter, h f and h m are head values of the matrix and the fracture at the same position, respectively.

[0105] The partial differential equations of seepage in matrix and fracture are discretized and added using Galerkin finite element method, and the finite element weak form of water exchange term can be expressed by formula (3.5):

[0106] (3.5)

[0107] In formula (3.5), φ m and φ f are the base functions of finite element method at the grid nodes, T is the cross-medium transmission parameter, h f and h m are the water head values of matrix and fracture at the same position respectively, represents the mesoscale fracture domain;

[0108] The water exchange term is numerically approximated using the collocation method, and formula (3.6) is obtained:

[0109] (3.6)

[0110] In formula (3.6), n f is the number of mesoscale fracture element nodes, e represents the coupling element index, i represents the matrix element node index of the coupling element, represents the value of the base function corresponding to the i-th matrix element node of the e-th coupling element at the (x, y, z) coordinates, represents the base function of the mesoscale fracture element node of the e-th coupling element, similar to T, represents the parameter of water exchange between the fracture and the matrix, and its size represents the degree of water exchange, and a value of 0 represents no exchange, and in the embodiment, the value of θ is set to 100, represents the water head difference between the coupling node and the mesoscale fracture node of the e-th coupling element, represents the influence area of the coupling node of the e-th coupling element, and the numerical approximation of the water exchange term F E is obtained using formula (3.7);

[0111] (3.7)

[0112] In formula (3.7), φ represents the base function of the coupling node of the e-th coupling element, represents the mesoscale fracture domain;

[0113] Formula (3) can be obtained using formula (3.2) and formula (3.6):

[0114]

[0115] In formula (3), n fis the number of nodes of the meso-scale fracture element; N m is a vector composed of the values of the basis functions of the matrix element at the node positions of the meso-scale fracture element; represents a vector with value 1 only at the i th node position of the meso-scale fracture element and 0 at the other positions; θ represents a parameter of water exchange between the meso-scale fracture and the matrix; S i is a measure of the influence area related to the i th node of the meso-scale fracture element; represents a vector of the total water head values of the discrete fracture-matrix model of the macro-scale fracture, is a vector of the water head values of the nodes of the meso-scale fracture element, and T represents transposition, represents a water flow exchange matrix between the nodes of the meso-scale fracture element and the nodes of the matrix element;

[0116] For formula (3), it needs to be explained that in the process of calculating the water flow exchange matrix , it is necessary to follow the matrix assembly principle in the finite element method, that is, the index of the matrix corresponds to the index of .

[0117] Step 407, the algebraic form of the water flow exchange term and the algebraic form of the discrete fracture-matrix model of the macro-scale fracture are combined, so as to obtain the multi-scale discrete fracture-matrix model by using formula (4);

[0118] For the above step 407, it needs to be explained that in formula (3) contains both the nodes of the matrix element and the nodes of the meso-scale fracture element, so the size of this matrix is (n m +n f )×(n m +n f ), n m and n f are the number of nodes of the matrix element and the number of nodes of the meso-scale fracture element respectively, and in formula (3) can be directly moved to the left side of formula (2) for combination, so as to obtain the overall algebraic equation (4) of the multi-scale discrete fracture-matrix model as follows:

[0119]

[0120] In formula (4), represents the total water conductivity matrix of the multi-scale discrete fracture-matrix model, represents the water flow exchange matrix between the nodes of the meso-scale fracture element and the nodes of the matrix element, represents the water head value vector of the multi-scale discrete fracture-matrix model, This represents the water volume vector in a multi-scale discrete fracture-matrix model.

[0121] Step 5: Simulate the multi-scale discrete fracture-matrix model to obtain the simulated head distribution diagram of the multi-scale discrete fracture-matrix model.

[0122] Step 501: Set boundary conditions at the boundary of the multi-scale discrete fracture-matrix model, wherein the boundary conditions are constant head boundaries or flow boundaries;

[0123] Regarding step 501 above, it should be noted that the large-scale discrete fracture-matrix model and the mesoscale discrete fracture network model in the multi-scale discrete fracture-matrix model require separate boundary conditions. Specifically, the large-scale discrete fracture-matrix model requires boundary conditions on the simulation domain boundary surface, while the mesoscale discrete fracture network model requires boundary conditions on the intersection line with the simulation domain boundary surface. The methods for setting constant head and flow boundary conditions and their handling in algebraic equations are detailed in existing technologies and will not be repeated here.

[0124] like Figure 2 The purple surface in the schematic diagram of the geometric model and boundary conditions of large-scale and mesoscale fractures shown in part (a) indicates that in this embodiment, constant head conditions of interface H1 = 10 m at X = 0 and interface H2 = 0 m at X = 80 are applied to the entire domain, and the remaining interfaces are set as flow-free boundaries.

[0125] Step 502: Under the boundary conditions, solve the multi-scale discrete fracture-matrix model using direct or iterative methods to obtain the numerical solution of the hydraulic head at each node in the entire simulation domain, which is used to draw the groundwater head distribution map.

[0126] The result obtained through step 5 above is as follows: Figure 5 The hydraulic head distribution map in the overall simulation domain shown in part (a) is as follows: The hydraulic head distribution map of the mesoscale fractures in the simulation domain is shown in part (a). Figure 5 As shown in part (b) of the document.

[0127] In this embodiment, an electronic device includes a memory and a processor. The memory is used to store a program that supports the processor in executing the above-described multi-scale fractured medium seepage simulation method, and the processor is configured to execute the program stored in the memory.

[0128] In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is run by a processor, it executes the steps of the above-described multi-scale fractured medium seepage simulation method.

Claims

1. A method for simulating seepage in multi-scale fractured media based on coupled finite element method, characterized in that, Includes the following steps: Step 1: Obtain the statistical distribution characteristics of relevant parameters of multi-scale fractures in the simulation domain, and construct a multi-scale fracture geometric model based on the statistical distribution characteristics; the multi-scale fractures include: small-scale fractures, medium-scale fractures and large-scale fractures; Step 2: Perform intersection line detection on the cracks at each scale in the multi-scale crack geometric model to obtain the intersection lines between cracks at each scale and the intersection lines between cracks and the boundary of the simulation domain; Step 3: Based on the intersection lines between fractures at each scale and the intersection lines between fractures and the boundary of the simulation domain, the multi-scale fracture geometric model is simplified to obtain the conceptual model of the fracture geometric model at each scale, including: equivalent porous medium model, discrete fracture-matrix model, and discrete fracture network model. Step 301: Calculate the volume of the characterization unit of the small-scale fracture geometric model and the matrix, and determine the overall permeability coefficient of the characterization unit through numerical simulation or experiment. Then, use the overall permeability coefficient as the equivalent permeability coefficient after coupling the small-scale fracture and the matrix, and establish an equivalent porous medium model of coupling the small-scale fracture geometric model and the matrix. Step 302: Using the boundary of the simulated domain as the outer boundary and the intersection of the large-scale fracture geometric model and the matrix as the inner boundary, the simulated domain and the large-scale fracture are spatially matched and discretized using the three-dimensional Delaunay tetrahedral mesh generation algorithm to obtain the discrete fracture-matrix model of the large-scale fracture. The matrix is ​​discretized into spatial tetrahedral matrix elements, and the large-scale fracture is discretized into triangular large-scale fracture elements, which are used as one face of the matrix elements. The nodes of the large-scale fracture elements have the same index as the nodes of the matrix elements. Step 303: Considering only the intersection lines between mesoscale fractures and the intersection lines between mesoscale fractures and the boundary of the simulation domain, use the two-dimensional Delaunay triangular element mesh generation algorithm to discretize each fracture in the mesoscale fracture geometric model independently, generate the corresponding triangular mesoscale fracture elements, and form a discrete fracture network model of mesoscale fractures. Step 4: Establish the coupling relationship between the conceptual models of the fracture geometric models at each scale to obtain the multi-scale discrete fracture-matrix model; Step 5: Simulate the multi-scale discrete fracture-matrix model to obtain a simulation diagram of the head distribution of the multi-scale discrete fracture-matrix model.

2. The method for simulating seepage in multi-scale fractured media based on coupled finite element method as described in claim 1, characterized in that, Step 4 includes: Step 401: Replace the matrix permeability coefficient in the discrete fracture-matrix model of the large-scale fracture with the overall permeability coefficient of the equivalent porous medium model to achieve coupling between the matrix, small-scale fractures and large-scale fractures. The matrix element and the large-scale fracture element were numerically discretized using the Galerkin finite element method, respectively, and the algebraic form of the discrete fracture-matrix model of the large-scale fracture was obtained using equation (1): (1) In formula (1): D m and D fL These are the water conduction matrices for the matrix element node set and the large-scale fracture element node set, respectively; H m With H fL For the head vector of the matrix element node set and the head vector of the large-scale fracture element node set; F m and F fL For the water volume vectors of the matrix element node set and the large-scale fracture element node set, This means that it is equivalent to, The total water conductivity matrix of the discrete fracture-matrix model representing large-scale fractures. The vector representing the total hydraulic head value of a discrete fracture-matrix model for large-scale fractures. The total water volume vector representing a discrete fracture-matrix model of large-scale fractures; Step 402: Numerically discretize the triangular element mesh of the discrete fracture network model of the mesoscale fracture using the Galerkin finite element method, and then combine Equation (2) with the algebraic form of the discrete fracture-matrix model of the large-scale fracture: (2) In formula (2): D represents the total water conductivity matrix of the discrete fracture-matrix model of large-scale fractures. M The water conduction matrix of the mesoscale fracture element node set. H represents the total head vector of the discrete fracture-matrix model for large-scale fractures. M This is the head value vector of the mesoscale fracture element node set. F represents the total water volume vector of a discrete fracture-matrix model representing large-scale fractures. M F is the water volume vector of the mesoscale fracture element node set. E Let represent the flow exchange term between mesoscale fracture element mesh nodes and matrix element nodes, D be the total hydraulic conductivity matrix of the multiscale discrete fracture-matrix model, H be the total hydraulic head vector of the multiscale discrete fracture-matrix model, and F be the total water volume vector of the multiscale discrete fracture-matrix model. This means equivalent to; Step 403: Retrieve the external matrix element of each mesoscale fracture element node, and take the mesoscale fracture element node corresponding to the index of the retrieved matrix element as the coupling node, so that together with the 4 nodes of the external matrix element, a 5-node coupling element is formed. Step 404: Based on all 5-node coupling units, the interpolated head value of the coupling nodes within the matrix unit is obtained using the basis functions on the matrix unit nodes; Step 405: Based on the head value of the mesoscale fracture element node and the interpolated head value of the coupled node, calculate the head difference at the spatial location of the mesoscale fracture element node; Step 406: Based on the 5-node coupled unit and the head difference, the flow exchange term F between the mesoscale fracture unit node and the matrix unit node is obtained using equation (3). E : (3) In equation (3): n f N is the number of nodes in a mesoscale fracture element; m This is a vector composed of the values ​​of the basis functions of the matrix element at the nodal locations of the mesoscale fracture element; θ represents a vector whose position is 1 only at the i-th node of the mesoscale fracture element, and 0 at all other positions; θ represents the parameter of water exchange between the mesoscale fracture and the matrix; S i It is a measure of the influence region associated with the i-th mesoscale fracture element node; The vector representing the total hydraulic head value of a discrete fracture-matrix model for large-scale fractures. This is the head vector of a mesoscale fracture element node, where T denotes transpose. This represents the water flow exchange matrix between mesoscale fracture element nodes and matrix element nodes; Step 407: The algebraic form of the water flow exchange term and the algebraic form of the discrete fracture-matrix model of the large-scale fracture are combined to obtain the multi-scale discrete fracture-matrix model using equation (4): (4) In equation (4), This represents the total water conductivity matrix of the multi-scale discrete fracture-matrix model. This represents the water flow exchange matrix between mesoscale fracture element nodes and matrix element nodes. This represents the head vector of the multi-scale discrete fracture-matrix model. This represents the water volume vector in a multi-scale discrete fracture-matrix model.

3. The method for simulating seepage in multi-scale fractured media based on coupled finite element method as described in claim 1, characterized in that, Step 5 includes: Step 501: Set boundary conditions at the boundary of the multi-scale discrete fracture-matrix model, wherein the boundary conditions are constant head boundaries or flow boundaries; Step 502: Under the boundary conditions, solve the multi-scale discrete fracture-matrix model using direct or iterative methods to obtain the numerical solution of the hydraulic head at each node in the entire simulation domain, which is used to draw the groundwater head distribution map.

4. An electronic device, comprising a memory and a processor, characterized in that, The memory is used to store programs that support the processor in executing the seepage simulation method in multi-scale fractured media according to any one of claims 1-3, the processor being configured to execute the programs stored in the memory.

5. A computer-readable storage medium storing a computer program thereon, characterized in that, When the computer program is run by the processor, it performs the steps of the seepage simulation method in multi-scale fractured media as described in any one of claims 1-3.

Citation Information

Patent Citations

  • Method for underground water coupling simulation in multi-scale fissure matrix system and device thereof

    CN114065593A

  • Mine floor limestone aquifer identification method based on multiple media of cave-fracture-hole-matrix

    CN117310799A