A model construction and non-Darcy seepage simulation method based on real rough fractures

CN118095122BActive Publication Date: 2024-11-12SHANDONG UNIV OF SCI & TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410195905.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-02-22
Publication Date
2024-11-12
Estimated Expiration
2044-02-22

AI Technical Summary

Technical Problem

[0008]本发明的主要目的在于提供一种基于真实粗糙裂隙的模型构建与非达西渗流模拟方法,以解决现有技术中建立的渗流模型不具有普适性、模型求解效率低、缺乏对实际粗糙裂隙进行建模的问题

Benefits of technology

[0092] Based on the real rough crack surface characteristics, a rough crack model is generated that is similar to the morphological characteristics and properties of natural cracks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118095122B_ABST
    Figure CN118095122B_ABST
Patent Text Reader

Abstract

The present invention provides a model construction and non-Darcy seepage simulation method based on real rough fractures, which belongs to the technical field of fracture seepage simulation, and specifically comprises the following steps: collecting rocks of different lithologies in the target layer to prepare the upper and lower walls of the rough fractures. Obtaining a dataset of morphological characteristics of the upper and lower walls of the rough fractures. Analyzing the distribution characteristics of the fracture opening. Analyzing the fracture roughness. Establishing a set to determine whether the generated fracture model meets the requirements on the fracture opening and roughness characteristics. Generate a rough fracture dataset with similar characteristics and properties to the actual rough fractures. Establishing a numerical model of rough fractures and carrying out nonlinear seepage solution of rough fractures. Establishing nonlinear seepage indicators of fractured rock masses for different rock types and fracture roughness in the study area. The technical solution of the present invention overcomes the problems that the seepage model established in the prior art is not universal, the model solution efficiency is low, and there is a lack of modeling of actual rough fractures.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of fracture seepage simulation, and in particular to a model construction and non-Darcy seepage simulation method based on real rough fractures. Background Art

[0002] At present, the construction of underground projects such as tunnels, mines, water conservancy and hydropower in my country has entered an unprecedented period of prosperity. Underground projects are inevitably built in fractured rock areas with complex geological conditions. Accidents such as sudden (gushing) water and engineering collapse induced by the instability of fractured rock mass seepage can easily cause a large number of casualties, serious economic losses and adverse social impacts. Therefore, the study of fractured rock mass seepage characteristics is the focus of the research on underground engineering disaster prevention and control.

[0003] A single rough fracture is the basic component of a rock fracture network. The construction of a single rough fracture model and the solution of the seepage process are the basis for a deep understanding of the overall seepage characteristics of a natural fractured rock network. To this end, domestic and foreign experts and scholars have conducted extensive research on single rough fracture modeling and single rough fracture seepage.

[0004] At present, there are two main methods for generating rough fracture surfaces. One is the rough fracture modeling method based on fractal theory. Scholars such as Xie Heping, Sun Hongquan, and Liu Baoguo have conducted in-depth research on rough fracture joint surfaces based on fractal theory. They not only established the relationship between fractal dimension and roughness coefficient, but also proposed a rough fracture modeling method based on fractal dimension. The second is the Fourier transform modeling method based on spatial frequency distribution. This method analyzes the geometric morphology characteristics of the fracture surface and proposes a construction model of the rough fracture surface using the probability density function and power spectrum density function of the rough fracture surface. Foreign scholars such as Brown, Glover, and Ogilvie have elaborated on this method in detail and proposed a rough fracture mathematical model.

[0005] In terms of seepage, scholars have analyzed the flow rate, pressure gradient and other fluid mechanics elements when the fluid flows through the rough fractures to reveal the seepage characteristics in the rough fractures. With the increase of the Reynolds number, the seepage process in the rough fractures shows nonlinear characteristics. Therefore, based on the results of indoor tests, researchers have successively proposed nonlinear rough fracture seepage models such as the Forchheimer equation, the Izbash equation, the hypercube law and the local cubic law. However, due to the limitation of the number of rough fracture surfaces made indoors, the study of rough fracture seepage has problems such as a small number of experiments and poor repeatability of experiments, which limits the use conditions of the description model and has certain deviations from actual engineering. Therefore, on the basis of establishing a reasonable rough fracture surface model (increasing the number and repeatability of experiments), conducting large-scale analysis and research on rough fracture characteristics and nonlinear seepage characteristics is the basis for building a more accurate rough fracture seepage model.

[0006] In summary, there are four problems in the current research: (1) Rough fracture surfaces have the characteristics of large morphological variability and a wide range of fracture openings. If the parameters are changed without considering the actual situation of the study area and the research is not targeted, the results will be different. (2) Seepage tests are carried out based on a single or a small number of rough fracture models. The number of samples is small and the established seepage model is not universal. (3) Existing software cannot realize the continuous calculation of rough fracture modeling-roughness evaluation-nonlinear seepage solution. The efficiency of a single solution is low and it is impossible to carry out large-scale automatic solution calculation to provide a scientific basis for actual engineering. (4) When there are too many fracture models, commercial software often cannot directly model them. There are many difficulties in assigning fracture attributes and setting boundary conditions, and necessary secondary development is required. Self-developed simulation programs have fast calculation speed and strong convergence, but they often have problems such as low development level and small application scope.

[0007] Therefore, a model construction and non-Darcy seepage simulation method based on real rough fractures is needed. Summary of the invention

[0008] The main purpose of the present invention is to provide a model construction and non-Darcy seepage simulation method based on real rough fractures, so as to solve the problems that the seepage model established in the prior art is not universal, the model solving efficiency is low, and there is a lack of modeling of actual rough fractures.

[0009] To achieve the above object, the present invention provides a model construction and non-Darcy seepage simulation method based on real rough fractures, which specifically includes the following steps:

[0010] S1, collect rocks of different lithology in the target layer, classify and label them; prepare rock test blocks into cuboids according to rock types; analyze the mechanical causes of the fracture surface, use axial splitting and shear test methods to prepare rough fractures, and prepare the upper and lower walls of rough fractures.

[0011] S2, use three-dimensional laser scanning technology to scan the upper and lower walls of the rough cracks to obtain the upper and lower wall morphology feature data set of the rough cracks.

[0012] S3, preprocess the upper and lower wall morphology feature datasets of the rough cracks to establish the upper and lower wall height datasets Z of the cracks top With Z bot .

[0013] S4, analyzing the crack opening distribution characteristics according to the crack upper and lower wall height data sets obtained in step S3.

[0014] S5, analyzing the matching degree between the upper and lower walls of the rough fracture according to the data sets obtained in step S3 and step S4.

[0015] S6, performing crack roughness analysis based on the height data set obtained in S3.

[0016] S7, based on steps S4 to S6, analyze and establish a rough fracture set Rockname of different rock types in the study area to determine whether the generated fracture model meets the requirements on fracture opening and roughness characteristics.

[0017] S8, based on the analysis of steps S4 to S6, a rough fracture data set with characteristics and properties similar to those of actual rough fractures is generated.

[0018] S9, based on the MATLAB with COMSO platform, a rough fracture numerical model is established and the nonlinear seepage solution of rough fracture is carried out.

[0019] S10, looping steps S8 to S9, obtaining the seepage characteristics of different rock types in different roughness fractures in the study area, and establishing nonlinear seepage indicators of fractured rock mass for different rock types and fracture roughness in the study area.

[0020] Furthermore, step S3 is specifically as follows:

[0021] Using the data interpolation function of Surfer software, by setting the number of nodes M and N fixed in the x and y directions, the upper and lower walls of the crack are divided into M×N square grids, and the upper and lower wall height datasets Z of the crack are established. top With Z bot :

[0022]

[0023]

[0024] In the formula, Represents the height value of the grid node (M, N) on the upper wall of the crack, Represents the height value of the grid node (M, N) at the lower wall of the crack.

[0025] Furthermore, step S4 specifically includes the following steps:

[0026] S4.1, based on Z top With Z bot , subtract the z values ​​of the upper and lower fracture surface position coordinates to obtain the fracture aperture dataset Aperture under the natural state:

[0027]

[0028] In the formula, Ap M,N Represents the crack opening between the upper and lower walls of the crack at the grid node (M,N) of the crack.

[0029] S4.2, analyze the distribution characteristics of the fracture aperture data in the fracture aperture data set Aperture, draw a fracture aperture frequency histogram, analyze the probability density function that the fracture aperture distribution conforms to, select the probability density function for description, and determine the distribution mean Mean and standard deviation SD.

[0030] Furthermore, step S5 specifically includes the following steps:

[0031] S5.1, import the upper and lower wall height datasets and the crack aperture dataset into Matlab software, and use the fast Fourier transform function in Matlab to calculate the power spectral density PSD of the upper and lower wall height datasets and the crack aperture dataset; the calculation parameters required by the fast Fourier transform function include: the wave number k in the Fourier component, the sampling frequency v, and the wavelength λ, where k = 1 / λ, v = 2π / λ, and the wavelength λ is equal to the length of the crack section, which is 100 mm here.

[0032] S5.2, based on the obtained PSD curves of each data set, the correlation coefficient FSDR is calculated using formula (4), and the relationship curve between PSDR and wavelength λ is plotted in the form of double logarithmic coordinates to analyze the degree of crack matching;

[0033]

[0034] Where, FSD 裂隙开度 is the power spectral density curve of the crack aperture dataset, PSD 上壁面 is the power spectral density curve of the crack upper wall, PSD 下壁面 is the power spectrum density curve of the crack lower wall.

[0035] S5.3, according to the relationship curve between the correlation coefficient PSDR and the wavelength λ, determine the minimum matching coefficient MinMF, the maximum matching coefficient MaxMF, the mismatch length ML and the transition length TL between the upper and lower walls of the crack.

[0036] Furthermore, step S6 specifically includes the following steps:

[0037] S6.1, based on the data set of the upper and lower wall heights of the fracture in step S3, use Z 2s Descriptive method for crack roughness evaluation, Z 2s It is expressed as:

[0038]

[0039] Where △m and △n are the stage spacings in the x and y directions respectively; m,n Represents the height value.

[0040] When the upper and lower walls of the crack are different, use the upper and lower crack surfaces The average value is expressed as:

[0041]

[0042] In the formula The average Z of the crack 2s Roughness, z is the upper and lower wall of the crack 2s Roughness.

[0043] S6.2, the tortuosity Rs characterization method is used to characterize the roughness of the crack surface from the perspective of height distribution characteristics. Rs is the actual area of ​​the crack surface S real The ratio of the fracture surface projection area S to the fracture surface projection area S' can be obtained by measurement, and the actual fracture surface area S real It is expressed by formula (7):

[0044]

[0045] When the upper and lower surfaces of the crack are different, the average value of the upper and lower crack walls is used. Indicates the roughness of the crack surface, the formula is:

[0046]

[0047] Where Rs top With Rs. bot are the Rs values ​​on the crack surface and the lower surface, S real-top represents the actual area of ​​the crack upper wall, S real-bot represents the actual area of ​​the lower wall of the crack, S top represents the projection area of ​​the crack wall, S bot Represents the projection area of ​​the lower wall of the crack.

[0048] Further, step S7 is specifically as follows:

[0049]

[0050] In the formula, Ap-Mean max is the maximum value of the mean value of the crack opening distribution of different rough crack models, Ap-Mean min is the minimum value of the mean value of the crack aperture distribution of different rough crack models, Ap-SD max is the maximum value of the standard deviation of the crack opening distribution of different rough crack models, Ap-SD min is the minimum value of the standard deviation of the fracture aperture distribution of different rough fracture models, Frac-Z2s max is the crack roughness index Z of different rough crack models 2sMaximum, Frac-Z2s min is the crack roughness index Z of different rough crack models 2s Minimum value, Frac-Rs max Frac-Rs is the maximum value of the crack roughness index Rs of different rough crack models, min is the minimum value of the crack roughness index Rs for different rough crack models.

[0051] Further, step S8 is specifically as follows:

[0052] S8.1, the rough fracture is modeled by the inverse Fourier transform modeling method based on spatial frequency distribution, wherein the parameters required by the inverse Fourier transform modeling method include: sequence set F(m, n) and the number of nodes M and N, wherein the sequence set F(m, n) is specifically expressed as:

[0053] F(m,n)={Rad cos(Phase), Rad sin(Phase)};

[0054] Wherein, Rad and Phase are specific parameters in the sequence set F(m, n) respectively.

[0055] S8.2, repeat step S8.1, clarify the two calculation parameters Rad and Phase in the inverse Fourier transform function; use the inverse Fourier function function IFFT2D(F(m, n), M, N) in Matlab to automatically generate the upper and lower wall height data sets of the rough fracture.

[0056] S8.3, repeat steps S2 to S6 with the generated rough fracture upper and lower wall height data sets, calculate the fracture opening distribution characteristics and roughness characteristics, and compare them with the rough fracture set Rockname constructed in step S7. When the fracture parameters meet the standard fracture opening distribution and roughness characteristic set standards, it is determined to be a valid rough fracture model and the rough fracture set data file is saved; when it does not meet the standard set requirements, it is determined to be an invalid rough fracture model, the set data is cleared, and step S8 is repeated.

[0057] Furthermore, step S8.1 specifically includes:

[0058] S8.1.1, input the matching coefficient, mismatch length, and transition length parameters between the upper and lower walls of the rough fracture.

[0059] S8.1.2, let i = 1 ~ M, j = 1 ~ ~ N, and iterate the steps S8.1.2 ~ S8.1.4 to solve the inverse Fourier function parameter set F (m, n) within the rough crack area.

[0060] S8.1.3, in the iterative calculation, the parameter Rad is the attenuation function, that is, the power spectrum of the rough crack wall, which gradually decays according to the distribution of the function A(m, n), expressed as:

[0061]

[0062] In the formula, is the spectrum index, indicating the decay rate, D is the fractal dimension of rough cracks.

[0063] S8.1.4, the fracture wall parameter phase is based on the matching degree between the upper and lower walls of the rough fracture, and is calculated by comparing the node position length A(m, n) = (m 2 +n 2 ) 1 / 2 Determine the size of the mismatch length ML:

[0064] When the node position length A(m, n)<ML / 2, the rough crack wall parameter phase is equal to phase=2πRan1.

[0065] When the node position length A(m, m)≥ML / 2, the upper and lower wall parameters of the rough fracture are phase 上 and phase 下 The random numbers Ran1, Ran2 and Ran3 are used for calculation, which are:

[0066] phase 上 =2πRan1;

[0067] phase 下 =2πRan3;

[0068] The specific judgment of the relationship between Ran1 and Ran3 is as follows:

[0069] Ran3=γRan1+(1-γ)Ran2

[0070]

[0071] In the formula, k c is the wave number of the mismatch wavelength, k c =2π / λ c ,λ c is the mismatch wavelength, and k is the wave number in the Fourier component.

[0072] Further, step S9 specifically includes:

[0073] S9.1, based on the MATLAB with COMSO platform, load the rough fracture set data file generated in step S8, and use the parametric scanning function to realize the construction of the upper and lower rough fracture walls; then, build a cuboid, use the Boozer algorithm to find the union of the cuboid and the upper and lower rough fracture walls, and build a rough fracture numerical model.

[0074] S9.2, using the MATLAB with COMSO platform, the control equation of the software platform is called, in which the Navier-Stokes equation is used as the control equation of fluid flow. When the fluid density is constant and the fluid is incompressible, the NS equation is expressed as:

[0075]

[0076] In the formula, u is the fluid velocity, p is the fluid pressure, ρ is the fluid density, μ is the fluid dynamic viscosity, and F is the volume force, that is, the external force acting on the fluid. is the Laplace operator.

[0077] The continuity equation representing mass conservation is called to form a coupled equation group with formula (12). The continuity equation is expressed as:

[0078] At the same time, the turbulent Standradk-ε model is used to simulate the nonlinear seepage process of fluid in rough fractures, where the turbulent kinetic energy k T The constraint equations for turbulence dissipation ε are:

[0079]

[0080]

[0081] Among them, P k is the turbulent kinetic energy generation term, expressed as:

[0082]

[0083] Where, turbulent viscosity μ T =ρC μ k 2 / ε, the turbulence model parameters take the default parameters: C ε1 =1.44; C ε2 =1.44; C μ =0.09.

[0084] S9.3, numerically set boundary conditions for rough fractures, taking the x direction as the fluid flow direction, setting pressure boundary conditions at the model inlet and outlet boundaries, and taking the y direction wall as an impermeable boundary.

[0085] S9.4, use the COMSO simulation software to divide the model grid, call the solver to solve the flow field and pressure field; after the solution, calculate the seepage velocity at the inlet of the rough fracture model, and obtain the steady-state volume flow rate Q under different fracture opening distribution conditions by integrating the flow velocity at the inlet, specifically:

[0086] Q = Au (17);

[0087] In the formula: A = ∫dh, A is the cross-sectional area perpendicular to the direction of fluid flow, and u is the velocity of the fluid passing through the cross-sectional area.

[0088] S9.5, using the cubic law, using the model inlet pressure P and volume flow rate Q, calculate the equivalent fracture opening at different volume flow rates as a reference indicator for nonlinear seepage in the study area. The specific calculation formula is:

[0089]

[0090] Where he is the equivalent crack opening, and W is the crack model width.

[0091] The present invention has the following beneficial effects:

[0092] Based on the real rough fracture surface characteristics, a rough fracture model with similar morphological characteristics and properties to natural fractures is generated.

[0093] Realize the construction of digital model of rough fracture surface and the solution of non-Darcy seepage process.

[0094] It solves the problem of automatically carrying out large-scale (quantity) rough fracture modeling-roughness evaluation-non-Darcy seepage process solution at one time without the need for a single test or single operation, which is helpful in understanding the seepage characteristics of the study area and providing a scientific basis for engineering design and scheme optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] In order to more clearly illustrate the specific implementation of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the specific implementation or the prior art description. Obviously, the drawings described below are some implementations of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0096] Figure 1 The relationship between the correlation coefficient PSDR and the wavelength λ is shown. DETAILED DESCRIPTION

[0097] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0098] A model construction and non-Darcy seepage simulation method based on real rough fractures specifically includes the following steps:

[0099] S1, conduct field surveys in the target area, collect rocks of different lithology in the target layer, and classify and label them; prepare rock test blocks into 100mm×100mm×100mm cuboids according to rock types; analyze the mechanical causes of the crack surface, use axial splitting and shear test methods to prepare rough cracks, and prepare the upper and lower walls of the rough cracks.

[0100] S2, use three-dimensional laser scanning technology to scan the upper and lower walls of the rough cracks to obtain the upper and lower wall morphology feature data set of the rough cracks.

[0101] S3, preprocess the upper and lower wall morphology feature datasets of the rough cracks to establish the upper and lower wall height datasets Z of the cracks top With Z bot .

[0102] S4, analyzing the crack opening distribution characteristics according to the crack upper and lower wall height data sets obtained in step S3.

[0103] S5, analyzing the matching degree between the upper and lower walls of the rough fracture according to the data sets obtained in step S3 and step S4.

[0104] S6, performing crack roughness analysis based on the height data set obtained in S3.

[0105] S7, based on steps S4 to S6, analyze and establish a rough fracture set Rockname of different rock types in the study area to determine whether the generated fracture model meets the requirements on fracture opening and roughness characteristics.

[0106] S8, based on the analysis of steps S4 to S6, a rough fracture data set with characteristics and properties similar to those of actual rough fractures is generated.

[0107] S9, based on the MATLAB with COMSO platform, a rough fracture numerical model is established and the nonlinear seepage solution of rough fracture is carried out.

[0108] S10, looping steps S8 to S9, obtaining the seepage characteristics of different rock types in different roughness fractures in the study area, and establishing nonlinear seepage indicators of fractured rock mass for different rock types and fracture roughness in the study area.

[0109] Specifically, step S3 is as follows:

[0110] According to the morphology and typical feature points of the upper and lower walls of the crack, the data of the upper and lower crack surfaces are spliced ​​to simulate the closed state of the crack. Using the data interpolation function of the Surfer software, by setting the number of nodes M and N fixed in the x and y directions, the upper and lower walls of the crack are divided into M×N square grids, and the upper and lower wall height data sets Z of the crack are established. top With Z bot :

[0111]

[0112]

[0113] In the formula, Represents the height value of the grid node (M, N) on the upper wall of the crack, Represents the height value of the grid node (M, N) at the lower wall of the crack.

[0114] Specifically, step S4 includes the following steps:

[0115] S4.1, based on Z top With Z bot , subtract the z values ​​of the upper and lower fracture surface position coordinates to obtain the fracture aperture dataset Aperture under the natural state:

[0116]

[0117] In the formula, Ap M,N Represents the crack opening between the upper and lower walls of the crack at the grid node (M,N) of the crack.

[0118] S4.2, analyze the distribution characteristics of the fracture aperture data in the fracture aperture data set Aperture, draw a fracture aperture frequency histogram, analyze the probability density function that the fracture aperture distribution conforms to, select the probability density function for description, and determine the distribution mean Mean and standard deviation SD.

[0119] Specifically, step S5 includes the following steps:

[0120] S5.1, import the upper and lower wall height data sets and the crack aperture data set into Matlab software, and use the fast Fourier transform function in Matlab to calculate the power spectral density PSD of the upper and lower wall height data sets and the crack aperture data set; the calculation parameters required by the fast Fourier transform function include: the wave number k in the Fourier component, the sampling frequency v, and the wavelength λ, where k = 1 / λ, v = 2π / λ, and the wavelength λ is equal to the length of the crack section, which is 100 mm in this embodiment.

[0121] S5.2, based on the obtained PSD curves of each data set, the correlation coefficient PSDR is calculated using formula (4), and the relationship curve between PSDR and wavelength λ is plotted in the form of double logarithmic coordinates to analyze the degree of crack matching;

[0122]

[0123] In the formula, PSD 裂隙开度 is the power spectral density curve of the crack aperture dataset, PSD 上壁面 is the power spectral density curve of the crack upper wall, PSD 下壁面 is the power spectrum density curve of the crack lower wall.

[0124] S5.3, according to the relationship curve between the correlation coefficient PSDR and the wavelength λ, determine the minimum matching coefficient MinMF, the maximum matching coefficient MaxMF, the mismatch length ML and the transition length TL between the upper and lower walls of the crack. Figure 1 As shown, the minimum PSDR corresponding to the wavelength in the curve is the minimum matching coefficient, the PSDR corresponding to the wavelength is the maximum matching coefficient, the position between the maximum wavelength where the minimum matching occurs and the minimum wavelength where the maximum matching occurs in the curve is the transition length (TL), and the equidistant position between the maximum wavelength where the minimum matching occurs and the minimum wavelength where the maximum matching occurs is the mismatch length (ML).

[0125] Specifically, step S6 includes the following steps:

[0126] S6.1, using Z 2s The description method characterizes the roughness of the fracture surface from the perspective of the three-dimensional morphological characteristics of the fracture surface. Based on the upper and lower wall height data sets in step S3, the Z 2s Descriptive method for crack roughness evaluation, Z 2s It is expressed as:

[0127] Where Δm and Δn are the stage spacings in the x and y directions respectively; z m,n Represents the height value.

[0128] When the upper and lower walls of the crack are different, use the upper and lower crack surfaces The average value is expressed as:

[0129]

[0130] In the formula The average Z of the crack 2s Roughness, Z is the upper and lower wall of the crack 2s Roughness.

[0131] S6.2, the tortuosity Rs characterization method is used to characterize the roughness of the crack surface from the perspective of height distribution characteristics. Rs is the actual area of ​​the crack surface S real The ratio of the crack surface projection area S to the crack surface projection area S' can be obtained by measurement. In this embodiment, the crack projection area is 100mm×100mm, and the actual crack surface area S real It is expressed by formula (7):

[0132]

[0133] When the upper and lower surfaces of the crack are different, the average value of the upper and lower crack walls is used. Indicates the roughness of the crack surface, the formula is:

[0134]

[0135] Where Rs top With Rs. bot are the Rs values ​​on the crack surface and the lower surface, S real-top represents the actual area of ​​the crack upper wall, S real-bot represents the actual area of ​​the lower wall of the crack, S top represents the projection area of ​​the crack wall, S bot Represents the projection area of ​​the lower wall of the crack.

[0136] Specifically, step S7 is as follows:

[0137]

[0138] In the formula, Ap-Mean max is the maximum value of the mean value of the crack opening distribution of different rough crack models, Ap-Mean min is the minimum value of the mean value of the crack aperture distribution of different rough crack models, Ap-SD max is the maximum value of the standard deviation of the crack opening distribution of different rough crack models, Ap-SD min is the minimum value of the standard deviation of the fracture aperture distribution of different rough fracture models, Frac-Z2s max is the crack roughness index Z of different rough crack models 2s Maximum, Frac-Z2smin is the crack roughness index Z of different rough crack models 2s Minimum value, Frac-Rs max Frac-Rs is the maximum value of the crack roughness index Rs of different rough crack models, min is the minimum value of the crack roughness index Rs for different rough crack models.

[0139] Specifically, step S8 is as follows:

[0140] S8.1, the rough fracture is modeled by the inverse Fourier transform modeling method based on spatial frequency distribution, wherein the parameters required by the inverse Fourier transform modeling method include: sequence set F(m, n) and the number of nodes M and N, wherein the sequence set F(m, n) is specifically expressed as:

[0141] F(m,n)={Rad cos(Phase), Rad sin(Phase)};

[0142] Wherein, Rad and Phase are specific parameters in the sequence set F(m, n) respectively.

[0143] S8.2, repeat step S8.1, clarify the two calculation parameters Rad and Phase in the inverse Fourier transform function; use the inverse Fourier function function IFFT2D(F(m, n), M, N) in Matlab to automatically generate the upper and lower wall height data sets of the rough fracture.

[0144] S8.3, repeat steps S2 to S6 with the generated rough fracture upper and lower wall height data sets, calculate the fracture opening distribution characteristics and roughness characteristics, and compare them with the rough fracture set Rockname constructed in step S7. When the fracture parameters meet the standard fracture opening distribution and roughness characteristic set standards, it is determined to be a valid rough fracture model, and the rough fracture set data file is saved in .txt format; when it does not meet the standard set requirements, it is determined to be an invalid rough fracture model, the set data is cleared, and step S8 is repeated.

[0145] Specifically, step S8.1 specifically includes:

[0146] S8.1.1, input the matching coefficient, mismatch length, and transition length parameters between the upper and lower walls of the rough fracture.

[0147] S8.1.2, let i = 1 ~ M, j = 1 ~ ~ N, and iterate the steps S8.1.2 ~ S8.1.4 to solve the inverse Fourier function parameter set F (m, n) within the rough crack area.

[0148] S8.1.3, in the iterative calculation, the parameter Rad is the attenuation function, that is, the power spectrum of the rough crack wall, which gradually decays according to the distribution of the function A(m, n), expressed as:

[0149]

[0150] In the formula, is the spectrum index, indicating the decay rate, , D is the fractal dimension of rough cracks.

[0151] S8.1.4, the fracture wall parameter phase is based on the matching degree between the upper and lower walls of the rough fracture, and is calculated by comparing the node position length A(m, n) = (m 2 +n 2 ) 1 / 2 Determine the size of the mismatch length ML:

[0152] When the node position length A(m, n)<ML / 2, the rough crack wall parameter phase is equal to phase=2πRan1.

[0153] When the node position length A(m, m)≥ML / 2, the upper and lower wall parameters of the rough fracture are phase 上 and phase 下 The random numbers Ran1, Ran2 and Ran3 are used for calculation, which are:

[0154] phase 上 =2πRan1;

[0155] phase 下 =2πRan3;

[0156] Among them, Ran1 and Ran3 have the following relationship: combining the matching coefficient and transition length parameter, the two random numbers are linearly fused using linear weights, so that the linear weighting is zero at half of the mismatch length ML, and then gradually increases to the maximum wavelength of the crack to reach the maximum matching coefficient MaxMF of the crack surface. The specific judgment of the relationship between Ran1 and Ran3 is as follows:

[0157] Ran3=γRan1+(1-γ)Ran2

[0158]

[0159] In the formula, k c is the wave number of the mismatch wavelength, k c =2π / λ c ,λ c is the mismatch wavelength, and k is the wave number in the Fourier component.

[0160] Specifically, step S9 specifically includes:

[0161] S9.1, based on the MATLAB with COMSO platform, load the rough fracture set data file generated in step S8, and use the parametric scanning function to realize the construction of the upper and lower rough fracture walls; then, build a 100mm×100mm×100mm cuboid, use the Boozer algorithm to find the union of the cuboid and the upper and lower rough fracture walls, and establish a rough fracture numerical model.

[0162] S9.2, using the MATLAB with COMSO platform, the control equation of the software platform is called, in which the Navier-Stokes equation is used as the control equation of fluid flow. When the fluid density is constant and the fluid is incompressible, the NS equation is expressed as:

[0163]

[0164] In the formula, u is the fluid velocity, p is the fluid pressure, ρ is the fluid density, μ is the fluid dynamic viscosity, and F is the volume force, that is, the external force acting on the fluid. is the Laplace operator.

[0165] The continuity equation representing mass conservation is called to form a coupled equation group with formula (12). The continuity equation is expressed as:

[0166]

[0167] At the same time, the turbulent Standradk-ε model is used to simulate the nonlinear seepage process of fluid in rough fractures, where the turbulent kinetic energy k T The constraint equations for turbulence dissipation ε are:

[0168]

[0169]

[0170] Among them, P k is the turbulent kinetic energy generation term, expressed as:

[0171]

[0172] Where, turbulent viscosity μ T =ρC μ k 2 / ε, the turbulence model parameters take the default parameters: C ε1 =1.44; C ε2 =1.44; C μ =0.09.

[0173] S9.3, numerically set boundary conditions for rough fractures, taking the x direction as the fluid flow direction, setting pressure boundary conditions at the model inlet and outlet boundaries, and taking the y direction wall as an impermeable boundary.

[0174] S9.4, use the COMSO simulation software to divide the model grid, call the solver to solve the flow field and pressure field; after the solution, calculate the seepage velocity at the inlet of the rough fracture model, and obtain the steady-state volume flow rate Q under different fracture opening distribution conditions by integrating the flow velocity at the inlet, specifically:

[0175] Q = Au (17);

[0176] In the formula: A = ∫dh, A is the cross-sectional area perpendicular to the direction of fluid flow, and u is the velocity of the fluid passing through the cross-sectional area.

[0177] S9.5, using the cubic law, using the model inlet pressure P and volume flow rate Q, calculate the equivalent fracture opening at different volume flow rates as a reference indicator for nonlinear seepage in the study area. The specific calculation formula is:

[0178]

[0179] Where he is the equivalent crack opening, and W is the crack model width.

[0180] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by technicians in this technical field within the essential scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A model construction and non-Darcy seepage simulation method based on real rough fractures, It is characterized in that The specific steps include: S1, collect rocks of different lithology in the target layer, classify and label them; prepare rock test blocks into cuboids according to rock types; analyze the mechanical causes of the fracture surface, prepare rough fractures using axial splitting and shear test methods, and prepare the upper and lower walls of rough fractures; S2, using three-dimensional laser scanning technology to scan the upper and lower walls of the rough cracks to obtain the upper and lower wall morphology feature data set of the rough cracks; S3, preprocess the upper and lower wall morphology feature datasets of the rough cracks to establish the upper and lower wall height datasets Z of the cracks top With Z bot ; S4, analyzing the distribution characteristics of the crack opening according to the crack upper and lower wall height data sets obtained in step S3; S5, analyzing the matching degree between the upper and lower walls of the rough fracture according to the data sets obtained in step S3 and step S4; S6, performing crack roughness analysis based on the height data set obtained in S3; S7, based on steps S4 to S6, analyze and establish a rough fracture set Rockname of different rock types in the study area, to determine whether the generated fracture model meets the requirements on fracture opening and roughness characteristics; S8, based on the analysis of steps S4 to S6, generating a rough fracture data set having characteristics and properties similar to those of actual rough fractures; S9, based on the MATLAB with COMSO platform, a rough fracture numerical model was established to solve the nonlinear seepage of rough fractures; S10, looping steps S8 to S9, obtaining seepage characteristics of different rock types in different roughness fractures in the study area, and establishing nonlinear seepage indicators of fractured rock mass for different rock types and fracture roughness in the study area; Step S9 specifically includes: S9.1, based on the MATLAB with COMSO platform, load the rough fracture set data file generated in step S8, and use the parametric scanning function to realize the construction of the upper and lower rough fracture walls; then, build a cuboid, and use the Boozer algorithm to find the union of the cuboid and the upper and lower rough fracture walls to establish a rough fracture numerical model; S9.2, using the MATLAB with COMSO platform, the control equation of the software platform is called, in which the Navier-Stokes equation is used as the control equation of fluid flow. When the fluid density is constant and the fluid is incompressible, the NS equation is expressed as: In the formula, u is the fluid velocity, p is the fluid pressure, ρ is the fluid density, μ is the fluid dynamic viscosity, and F is the volume force, that is, the external force acting on the fluid. is the Laplace operator; The continuity equation representing mass conservation is called to form a coupled equation group with formula (12). The continuity equation is expressed as: At the same time, the turbulent Standradk-ε model is used to simulate the nonlinear seepage process of fluid in rough fractures, where the turbulent kinetic energy k T The constraint equations for turbulence dissipation ε are: Among them, P k is the turbulent kinetic energy generation term, expressed as: Where, turbulent viscosity μ T =ρC μ k 2 / ε, the turbulence model parameters take the default parameters: C ε1 =1.44; C ε2 =1.44; C μ =0.09; S9.3, numerically set boundary conditions for rough fractures, taking the x direction as the fluid flow direction, setting pressure boundary conditions at the model inlet and outlet boundaries, and taking the y direction wall as an impermeable boundary; S9.4, use the COMSO simulation software to divide the model grid, call the solver to solve the flow field and pressure field; after the solution, calculate the seepage velocity at the inlet of the rough fracture model, and obtain the steady-state volume flow rate Q under different fracture opening distribution conditions by integrating the flow velocity at the inlet, specifically: Q = Au (17); Where: A = ∫dh, A is the cross-sectional area perpendicular to the direction of fluid flow, u is the velocity of the fluid passing through the cross-sectional area; S9.5, using the cubic law, using the model inlet pressure P and volume flow rate Q, calculate the equivalent fracture opening at different volume flow rates as a reference indicator for nonlinear seepage in the study area. The specific calculation formula is: Where he is the equivalent crack opening, and W is the crack model width.

2. A model construction and non-Darcy seepage simulation method based on real rough fractures according to claim 1, It is characterized in that Step S3 is specifically as follows: Using the data interpolation function of Surfer software, by setting the number of nodes M and N fixed in the x and y directions, the upper and lower walls of the crack are divided into M×N square grids, and the upper and lower wall height datasets Z of the crack are established. top With Z bot : In the formula, Represents the height value of the grid node (M,N) on the upper wall of the crack, Represents the height value of the grid node (M,N) at the lower wall of the crack.

3. A model construction and non-Darcy seepage simulation method based on real rough fractures according to claim 2, It is characterized in that Step S4 specifically includes the following steps: S4.1, based on Z top With Z bot , subtract the z values ​​of the upper and lower fracture surface position coordinates to obtain the fracture aperture dataset Aperture under the natural state: In the formula, Ap M,N Represents the crack opening between the upper and lower walls of the crack at the grid node (M, N) of the crack; S4.2, analyze the distribution characteristics of the fracture aperture data in the fracture aperture data set Aperture, draw a fracture aperture frequency histogram, analyze the probability density function that the fracture aperture distribution conforms to, select the probability density function for description, and determine the distribution mean Mean and standard deviation SD.

4. A model construction and non-Darcy seepage simulation method based on real rough fractures according to claim 3, It is characterized in that Step S5 specifically includes the following steps: S5.1, import the upper and lower wall height datasets and the crack aperture dataset into Matlab software, and use the fast Fourier transform function in Matlab to calculate the power spectral density PSD of the upper and lower wall height datasets and the crack aperture dataset; the calculation parameters required by the fast Fourier transform function include: the wave number k in the Fourier component, the sampling frequency v, and the wavelength λ, where k = 1 / λ, v = 2π / λ, and the wavelength λ is equal to the length of the crack section; S5.2, based on the obtained PSD curves of each data set, the correlation coefficient PSDR is calculated using formula (4), and the relationship curve between PSDR and wavelength λ is plotted in the form of double logarithmic coordinates to analyze the degree of crack matching; In the formula, PSD 裂隙开度 is the power spectral density curve of the crack aperture dataset, PSD 上壁面 is the power spectral density curve of the crack upper wall, PSD 下壁面 is the power spectrum density curve of the crack lower wall; S5.3, according to the relationship curve between the correlation coefficient PSDR and the wavelength λ, determine the minimum matching coefficient MinMF, the maximum matching coefficient MaxMF, the mismatch length ML and the transition length TL between the upper and lower walls of the crack.

5. A model construction and non-Darcy seepage simulation method based on real rough fractures according to claim 4, It is characterized in that Step S6 specifically includes the following steps: S6.1, based on the data set of the upper and lower wall heights of the fracture in step S3, use Z 2s Descriptive method for crack roughness evaluation, Z 2s It is expressed as: Where △m and △n are the stage spacings in the x and y directions respectively; m,n Represents the height value; When the upper and lower walls of the crack are different, use the upper and lower crack surfaces The average value is expressed as: In the formula The average Z of the crack 2s Roughness, Z is the upper and lower wall of the crack 2s Roughness; S6.2, the tortuosity Rs characterization method is used to characterize the roughness of the crack surface from the perspective of height distribution characteristics. Rs is the actual area of ​​the crack surface S real The ratio of the fracture surface projection area s to the fracture surface projection area s' can be obtained by measurement, and the actual fracture surface area s real It is expressed by formula (7): When the upper and lower surfaces of the crack are different, the average value of the upper and lower crack walls is used. Indicates the roughness of the crack surface, the formula is: Where Rs top With Rs. bot are the Rs values ​​on the crack surface and the lower surface, S real-top represents the actual area of ​​the crack upper wall, S real-bot represents the actual area of ​​the lower wall of the crack, S top represents the projection area of ​​the crack wall, S bot Represents the projection area of ​​the lower wall of the crack.

6. A model construction and non-Darcy seepage simulation method based on real rough fractures according to claim 5, It is characterized in that Step S7 is specifically as follows: Rockname={Ap-Mean max ,Ap-Mean min ,Ap-SD max ,Ap-SD min ,Frac-Z2s max ,Frac-Z2s min ,Frac-Rs max ,Frac-Rs min } (9); In the formula, Ap-Mean max is the maximum value of the mean value of the crack opening distribution of different rough crack models, Ap-Mean min is the minimum value of the mean value of the crack aperture distribution of different rough crack models, Ap-SD max is the maximum value of the standard deviation of the crack opening distribution of different rough crack models, Ap-SD min is the minimum value of the standard deviation of the fracture aperture distribution of different rough fracture models, Frac-Z2s max is the crack roughness index Z of different rough crack models 2s Maximum, Frac-Z2s min is the crack roughness index Z of different rough crack models 2s Minimum value, Frac-Rs max Frac-Rs is the maximum value of the crack roughness index Rs of different rough crack models, min is the minimum value of the crack roughness index Rs for different rough crack models.

7. A model construction and non-Darcy seepage simulation method based on real rough fractures according to claim 6, It is characterized in that Step S8 is specifically as follows: S8.1, the rough fracture is modeled by the inverse Fourier transform modeling method based on spatial frequency distribution, wherein the parameters required by the inverse Fourier transform modeling method include: sequence set F(m, n) and the number of nodes M and N, wherein the sequence set F(m, n) is specifically expressed as: F(m,n)={Rad cOS(Phase), Rad sin(Phase)}; In the formula, Rad and Phase are specific parameters in the sequence set F(m,n); S8.2, repeat step S8.1, clarify the two calculation parameters Rad and Phase in the inverse Fourier transform function; use the inverse Fourier function functionIFFT2D(F(m, n), M, N) in Matlab to automatically generate the upper and lower wall height data sets of the rough fracture; S8.3, repeat steps S2 to S6 with the generated rough fracture upper and lower wall height data sets, calculate the fracture opening distribution characteristics and roughness characteristics, and compare them with the rough fracture set Rockname constructed in step S7. When the fracture parameters meet the standard fracture opening distribution and roughness characteristic set standards, it is determined to be a valid rough fracture model and the rough fracture set data file is saved; when it does not meet the standard set requirements, it is determined to be an invalid rough fracture model, the set data is cleared, and step S8 is repeated.

8. A model construction and non-Darcy seepage simulation method based on real rough fractures according to claim 7, It is characterized in that Step S8.1 specifically includes: S8.1.1, input the matching coefficient, mismatch length and transition length parameters between the upper and lower walls of the rough fracture; S8.1.2, let i = 1 ~ M, j = 1 ~ N, and iterate the execution of steps S8.1.2 ~ S8.1.4 to solve the inverse Fourier function parameter set F (m, n) within the rough fracture region; S8.1.3, in the iterative calculation, the parameter Rad is the attenuation function, that is, the power spectrum of the rough crack wall, which gradually decays according to the distribution of the function A(m, n), expressed as: In the formula, is the spectrum index, indicating the decay rate, D is the fractal dimension of rough cracks; S8.1.4, the fracture wall parameter phase is based on the matching degree between the upper and lower walls of the rough fracture, and is calculated by comparing the node position length A(m, n) = (m 2 +n 2 ) 1 / 2 Determine the size of the mismatch length ML: When the node position length A(m, m) < ML / 2, the rough fracture wall parameter phase is equal to phase = 2πRan 1 ; When the node position length A(m, m)≥ML / 2, the upper and lower wall parameters of the rough fracture are phase 上 and phase 下 Using random number Ran 1 Ran 2 and Ran 3 The calculations are: phase 上 =2πRan 1 ; phase 下 =2πRan 3 ; Ran 1 With Ran 3 The specific judgment of the relationship is as follows: Ran 3 =yRan 1 +(1-γ)Ran 2 In the formula, k c is the wave number of the mismatch wavelength, k c =2π / λ c ,λ c is the mismatch wavelength, and k is the wave number in the Fourier component.