CN118332902A - Three-dimensional numerical simulation method for evolution of seepage field of fractured rock mass tunnel - Google Patents
Three-dimensional numerical simulation method for evolution of seepage field of fractured rock mass tunnel Download PDFInfo
- Publication number
- CN118332902A CN118332902A CN202410375015.XA CN202410375015A CN118332902A CN 118332902 A CN118332902 A CN 118332902A CN 202410375015 A CN202410375015 A CN 202410375015A CN 118332902 A CN118332902 A CN 118332902A
- Authority
- CN
- China
- Prior art keywords
- fracture
- seepage
- rock mass
- model
- fractured rock
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000011435 rock Substances 0.000 title claims abstract description 122
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000004088 simulation Methods 0.000 title claims abstract description 50
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 61
- 239000003673 groundwater Substances 0.000 claims abstract description 38
- 230000035699 permeability Effects 0.000 claims abstract description 23
- 238000004364 calculation method Methods 0.000 claims abstract description 10
- 238000004422 calculation algorithm Methods 0.000 claims description 30
- 239000011159 matrix material Substances 0.000 claims description 28
- 238000009826 distribution Methods 0.000 claims description 19
- 238000013528 artificial neural network Methods 0.000 claims description 14
- 238000013527 convolutional neural network Methods 0.000 claims description 14
- 125000004122 cyclic group Chemical group 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 12
- 238000004891 communication Methods 0.000 claims description 10
- 238000007639 printing Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 239000013598 vector Substances 0.000 claims description 9
- 238000010586 diagram Methods 0.000 claims description 7
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 5
- 238000002474 experimental method Methods 0.000 claims description 5
- 239000000945 filler Substances 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 5
- 230000033001 locomotion Effects 0.000 claims description 5
- 230000000704 physical effect Effects 0.000 claims description 5
- 238000013179 statistical model Methods 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 4
- 230000001133 acceleration Effects 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 206010017076 Fracture Diseases 0.000 description 188
- 208000010392 Bone Fractures Diseases 0.000 description 172
- 238000004458 analytical method Methods 0.000 description 6
- 238000012795 verification Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000002265 prevention Effects 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 238000003012 network analysis Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000013515 script Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011960 computer-aided design Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 230000003204 osmotic effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000004457 water analysis Methods 0.000 description 1
Abstract
The invention discloses a three-dimensional numerical simulation method for the evolution of a seepage field of a fractured rock mass tunnel, which belongs to the field of computation of seepage of fractured rock mass and comprises the following steps: s1, calibrating fracture data acquired by an unmanned aerial vehicle; s2, reconstructing a crack trace and a crack structural surface; s3, identifying the extension path and width change of the crack; s4, constructing a fracture structure surface network model; s5, establishing a discrete fracture network model of the fractured rock mass; s6, analyzing discrete fracture network connectivity of the fractured rock mass; s7, simulating and calculating the groundwater seepage value of the fractured rock mass; s8, acquiring an actual physical model permeation parameter; s9, comparing the simulated permeability parameters with the actual physical model permeability parameters, if the results are inconsistent, adjusting and optimizing a discrete fracture network model of the fractured rock mass, and carrying out iterative calculation until the results are consistent. The invention can accurately describe the underground fracture network, and improve the numerical simulation precision of the seepage field, thereby effectively providing guarantee for preventing and controlling the tunnel burst water.
Description
Technical Field
The invention relates to the technical field of fracture rock mass seepage calculation, in particular to a three-dimensional numerical simulation method for the evolution of a fracture rock mass tunnel seepage field.
Background
Rock mass is widely distributed in nature and occupies about 2/3 of the land surface area. The rock mass is composed of rock blocks and structural surfaces. The structural surfaces cut surrounding rocks to form rock blocks, and the structural surfaces which are staggered form a fracture network. The fracture network of the rock mass is an important factor in determining the mechanical, hydraulic and engineering stability of the rock mass. Under the rock mass fracture scale, the seepage flow of the rock mass is very small and can be ignored, and a channel formed by a fracture network is a main channel for underground water circulation. Especially in tunnel construction, a fracture network is critical for tunnel gushing water analysis, and in a part of a typical tunnel, fracture communication water guide is a main reason for tunnel gushing water.
Accurately characterizing the seepage characteristics of a three-dimensional rock fracture is an important means for analyzing groundwater migration and geothermal resource development. Unlike homogeneous or characterized voxels, the fracture network is non-homogeneous, and the mechanics, hydraulics and thermodynamics in fractured rock mass are all affected by their heterogeneous nature. The osmotic coefficient, the water storage coefficient and the heat conductivity parameters of the whole geologic body are equivalent to single values, so that certain limitations exist, and the non-uniformity of the fracture scale and the groundwater seepage rule under a non-uniform medium cannot be truly reflected. Accurately characterizing the morphology and hydrodynamic properties of the fracture is an important method to solve this problem, but directly observing the widely distributed fracture inside the rock mass is a difficult problem to be solved, so simulating the fracture inside the rock mass by mathematical methods becomes a viable way.
The existing fracture network simulation method is characterized in that the seepage channel of the fracture is analyzed by analyzing the communication condition of the fracture, and then relevant seepage analysis is carried out. However, many current implementations are two-dimensional simulation, so that it is difficult to reveal seepage characteristics under a three-dimensional view angle. Moreover, the engineering and scientific research circles simulate the whole seepage characteristics through multiple physical field simulation software more, and the seepage characteristics of the engineering and scientific research circles are not explored from the fracture scale and an effective simulation method is established.
In the field of geological engineering and resource development, rock fracture network analysis is critical for understanding and predicting groundwater flow, rock mass stability and heat transfer behavior. Conventional fracture analysis methods often rely on direct rock sample observations or simplified mathematical models, which often fail to accurately capture detailed characteristics of complex fracture networks. In recent years, despite significant developments in computer aided design and digital petrophysical technology, the prior art still faces the problems of difficult data acquisition, insufficient model accuracy and low computational efficiency.
The Chinese patent literature with publication number of CN117574653A and publication date of 2024, 02 month and 20 discloses a mesh subdivision method for groundwater seepage prevention numerical simulation of a complex fractured rock mass, firstly, a complex fractured rock mass geometric model is constructed, and a two-dimensional fixed point of the model is obtained; then, carrying out discrete processing on the line patterns in the model to obtain one-dimensional fixed points and one-dimensional constraints of the model; taking both a two-dimensional fixed point and a one-dimensional fixed point of the model as nodes, calculating the distance between the nodes, combining two nodes with the distance between the nodes smaller than the minimum feature refusal distance into one node, and correcting the connection mode of the combined nodes and the one-dimensional constraint of the model; finally, uniformly arranging nodes in the model, deleting part of the nodes according to the position probability of the nodes, generating triangular grids by the reserved nodes, adjusting the positions of the nodes, and generating the triangular grids by considering one-dimensional constraint of the model; finally, the quality of the triangular mesh is evaluated, and the mesh is optimized by reducing and increasing nodes.
The mesh subdivision method for the complex fracture rock mass groundwater seepage prevention numerical simulation disclosed by the patent document solves the problem of dispersion of a complex fracture network model, saves the calculation cost, and can simulate the discontinuity and the non-uniformity of groundwater seepage. However, the underground fracture network can not be accurately described, the numerical simulation precision of the seepage field is low, and the prevention and control of the tunnel gushing water can not be effectively guaranteed.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides the three-dimensional numerical simulation method for the tunnel seepage field evolution of the fractured rock mass, which can accurately describe an underground fracture network and improve the numerical simulation precision of the seepage field, thereby effectively providing guarantee for preventing and controlling the water burst of the tunnel.
The invention is realized by the following technical scheme:
the three-dimensional numerical simulation method for the evolution of the seepage field of the fractured rock mass tunnel is characterized by comprising the following steps of:
s1, acquiring crack data of a rock outcrop area in the field through an unmanned aerial vehicle, and calibrating the crack data acquired by the unmanned aerial vehicle;
S2, denoising pretreatment is carried out on the fracture data, three-dimensional point cloud data in the fracture data are converted into a continuous three-dimensional surface model, and fracture trace lines and fracture structural surfaces are reconstructed to obtain a three-dimensional reconstruction model;
S3, training a convolutional neural network model, using the convolutional neural network model to a three-dimensional reconstruction model, identifying geometric features of the fissures, training a cyclic neural network, and processing spatial distribution of the fissures through the cyclic neural network to identify extension paths and width changes of the fissures;
s4, constructing a fracture structure surface network model;
s5, establishing a discrete fracture network model of the fractured rock mass;
S6, analyzing discrete fracture network connectivity of the fractured rock mass through a graph theory algorithm;
s7, simulating and calculating the groundwater seepage value of the fractured rock mass to obtain simulated seepage parameters;
s8, printing the standard sample model into a physical model through 3D, and obtaining the permeation parameters of the actual physical model;
s9, comparing the simulated permeability parameters with the actual physical model permeability parameters, if the results are inconsistent, adjusting and optimizing a discrete fracture network model of the fractured rock mass, and carrying out iterative calculation until the results are consistent.
In the step S1, field acquisition refers to that an unmanned aerial vehicle flies and scans the outcrop area of the rock around a tunnel work area in a full coverage mode, three-dimensional point cloud data generated by high-resolution images and laser scanning are acquired, and crack data comprising the spatial distribution, geometric characteristics and physical properties of cracks are acquired.
In the step S1, the calibration of the fracture data collected by the unmanned aerial vehicle means that the occurrence, trace length, opening degree, roughness coefficient, connectivity, filler, density, group number and longitude and latitude of rock mass fractures around a tunnel work area are manually obtained, and the calibration of the fracture data is performed.
In the step S2, the denoising pretreatment of the fracture data refers to removing noise through a Gaussian filtering algorithm of a cross-platform computer vision library, and removing non-target objects through a random sample consistency algorithm of the cross-platform computer vision library.
In S3, the geometric features of the fracture include the occurrence, opening, trace length, density, and roughness coefficient.
The step S4 specifically comprises the following steps:
S41, converting the occurrence into spherical coordinates through the method 1;
1 (1)
Wherein,Is the tendency of the fracture structure surface,Is the inclination angle of the fracture structure,Is a horizontal angle, and is provided with a plurality of grooves,Is a vertical angle;
S42, converting the spherical coordinates into normal vectors of the fracture structural surface through a model 2;
2, 2
Wherein,、AndThe directions of the fracture structure surfaces are cosine in a rectangular coordinate system;
S43, calculating the trace length of the crack by using the crack structural surface trend of the crack as the direction of the disk and the opening degree of the crack as the thickness of the disk, determining the size of the disk, superposing a disk model, and establishing a crack structural surface network model;
3
Wherein,For the length of the track to be the same,Is the radius of the disc.
The step S5 specifically comprises the following steps:
s51, dividing the fracture shape into a plurality of groups, and acquiring an average value of normal vectors of each group of fractures as a basic parameter of fracture simulation;
S52, checking the statistical distribution of each group of cracks, and determining a statistical model as a parameter of Monte Carlo simulation;
S53, simulating the occurrence, trace length, opening degree and position of the fracture through Monte Carlo, establishing a single fracture through a disc model, and overlapping the single fracture to establish a discrete fracture network model of the fractured rock mass.
The step S6 specifically comprises the following steps:
s61, representing cracks in the fractured rock mass by using a cylindrical model, wherein the cracks are mutually communicated to form a water guide channel;
S62, performing orthogonal slicing and continuous slicing on a discrete fracture network model of the fractured rock mass to obtain a two-dimensional fracture trace distribution map of each position;
S63, numbering the cracks to obtain a crack diagram;
s64, marking the intersecting state of the cracks as a two-dimensional adjacent matrix for representing the communication relation of the cracks;
S65, judging whether two three-dimensional fracture discs are intersected or not so as to judge whether the fractures are communicated or not, and marking the fractures as a three-dimensional adjacent matrix;
s66, traversing an adjacency matrix of a discrete fracture network model of the fractured rock mass based on a depth-first algorithm in a graph theory path finding algorithm, and finding all communicated fractures and formed communication paths.
The step S7 specifically comprises the following steps:
S71, dividing seepage of a fractured rock mass into seepage of a rock matrix and seepage of a fracture network, wherein the seepage of the rock matrix adopts a porous medium seepage model, and the seepage of the fracture network adopts a cubic theorem seepage model;
s72, obtaining the groundwater seepage flow of each crack according to the cube law;
4. The method is to
Wherein,For the seepage flow of the underground water,Is the density of the groundwater, and is used for the treatment of the ground water,The acceleration of the gravity is that,Is the opening degree of the fracture opening,Is the motion viscosity coefficient of the underground water,Is the water head difference of the underground water,Is hydraulic conductivity;
S73, calculating the groundwater head of the two-dimensional fracture through the method 5;
5. The method is to
Wherein,Is an underground water head;
S74, establishing a plurality of linear equation sets through the formula 5, solving the linear equation sets to obtain a water head value of each intersection point, and substituting the water head value into the formula 4 to obtain the groundwater seepage flow of each crack;
S75, optimizing a communicated fracture network by adopting a maximum flow minimum cutting theorem in a network flow optimization algorithm;
s76, establishing a seepage equation of the rock matrix through a formula 6, establishing a seepage equation of a fracture network through a formula 7, and calculating groundwater seepage values to obtain simulation seepage parameters;
6. The method is to
Wherein,In order to make the water content be the specific water content,In order for the saturation to be effective,In order to be a specific reserve quantity,Is the water potential of the underground water,In order to be able to take time,For the groundwater flow speed in the rock matrix,Is the source and sink flow;
7. The method of the invention
Wherein,Is the opening degree of the crack,Is the specific water volume in the fracture,Is the storage ratio of the fracture,Is the water potential in the crack, and the water potential in the crack,Is the permeability of a single fracture, and the permeability of the single fracture is the same,Is the water exchange quantity between the rock matrix and the fracture.
The step S8 specifically comprises the following steps:
S81, intercepting a fracture network in a three-dimensional discrete fracture network through a standard cylinder to obtain a standard sample model;
S82, printing the standard sample model into a physical model through 3D;
s83, obtaining the actual physical model permeability parameters through geophysical well logging prediction or seepage experiments.
The beneficial effects of the invention are mainly shown in the following aspects:
1. compared with the prior art, the method can accurately describe the underground fracture network, improve the numerical simulation precision of the seepage field, and effectively provide guarantee for preventing and controlling the tunnel gushing water.
2. The invention can remarkably improve the collection and processing efficiency of rock mass cracks, simplify the whole numerical simulation flow and realize the automatic processing of most flows, greatly simplify the manual operation, improve the modeling efficiency, improve the numerical simulation precision of the rock crack network, and have important practical application value in tunnel seepage field evolution and gushing water disaster analysis.
3. According to the invention, the data acquired by the unmanned aerial vehicle and the laser radar are calibrated by manually acquiring the crack data around the tunnel work area on site, so that the accuracy of the acquired data is improved.
4. According to the invention, the convolutional neural network and the cyclic neural network are written through the deep learning framework to automatically identify and extract the fracture characteristics, so that the artificial extraction is reduced, and the improvement of the productivity and the fracture treatment efficiency are facilitated.
5. According to the invention, the two-dimensional fracture trace diagram of the discrete fracture network is obtained in a slicing mode, so that the seepage field simulation of the discrete fracture network can be carried out at any position.
6. According to the invention, the seepage field evolution simulation of the discrete fracture network and the movement of the groundwater are carried out through the finite element and the ion tracking algorithm, so that the high-precision simulation of the groundwater seepage can be realized.
7. According to the invention, through the practical physical model seepage experiment verification, compared with the analysis solution verification which only uses a mathematical formula, the method is more practical, and is favorable for iteratively improving the seepage model precision of the fracture network model.
8. According to the invention, only one programming environment is needed, a plurality of software switches are not needed, and most processes realize automatic operation through scripts, so that the numerical simulation process is greatly simplified.
Drawings
The invention will be further specifically described with reference to the drawings and detailed description below:
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a graph of the results of orthogonal slicing of a discrete fracture network model according to the present invention;
FIG. 3 is a two-dimensional trace plot of a discrete fracture network of the present invention;
FIG. 4 is a numbered result diagram of a two-dimensional trace of the present invention;
FIG. 5 is a graph of the water direction of the fissures in the two-dimensional trace of the present invention;
FIG. 6 is a graph of a discrete fracture network model of the present invention;
The marks in the figure: 1. the first, 2, second, 3, third, 4, fourth, 5, fifth, 6, sixth, 7, seventh, 8, eighth, 9, ninth, 10, tenth, ①, first, ②, second, ③, third, ④, fourth, ⑤, fifth, ⑥, sixth, ⑦, seventh, ⑧, eighth.
Detailed Description
Example 1
Referring to fig. 1-6, the three-dimensional numerical simulation method for the evolution of the seepage field of the fractured rock mass tunnel comprises the following steps:
s1, acquiring crack data of a rock outcrop area in the field through an unmanned aerial vehicle, and calibrating the crack data acquired by the unmanned aerial vehicle;
S2, denoising pretreatment is carried out on the fracture data, three-dimensional point cloud data in the fracture data are converted into a continuous three-dimensional surface model, and fracture trace lines and fracture structural surfaces are reconstructed to obtain a three-dimensional reconstruction model;
S3, training a convolutional neural network model, using the convolutional neural network model to a three-dimensional reconstruction model, identifying geometric features of the fissures, training a cyclic neural network, and processing spatial distribution of the fissures through the cyclic neural network to identify extension paths and width changes of the fissures;
s4, constructing a fracture structure surface network model;
s5, establishing a discrete fracture network model of the fractured rock mass;
S6, analyzing discrete fracture network connectivity of the fractured rock mass through a graph theory algorithm;
s7, simulating and calculating the groundwater seepage value of the fractured rock mass to obtain simulated seepage parameters;
s8, printing the standard sample model into a physical model through 3D, and obtaining the permeation parameters of the actual physical model;
s9, comparing the simulated permeability parameters with the actual physical model permeability parameters, if the results are inconsistent, adjusting and optimizing a discrete fracture network model of the fractured rock mass, and carrying out iterative calculation until the results are consistent.
Compared with the prior art, the embodiment can accurately describe the underground fracture network, improve the numerical simulation precision of the seepage field, and further can effectively provide guarantee for preventing and controlling the tunnel gushing water.
Example 2
Referring to fig. 1-6, the three-dimensional numerical simulation method for the evolution of the seepage field of the fractured rock mass tunnel comprises the following steps:
s1, acquiring crack data of a rock outcrop area in the field through an unmanned aerial vehicle, and calibrating the crack data acquired by the unmanned aerial vehicle;
S2, denoising pretreatment is carried out on the fracture data, three-dimensional point cloud data in the fracture data are converted into a continuous three-dimensional surface model, and fracture trace lines and fracture structural surfaces are reconstructed to obtain a three-dimensional reconstruction model;
S3, training a convolutional neural network model, using the convolutional neural network model to a three-dimensional reconstruction model, identifying geometric features of the fissures, training a cyclic neural network, and processing spatial distribution of the fissures through the cyclic neural network to identify extension paths and width changes of the fissures;
s4, constructing a fracture structure surface network model;
s5, establishing a discrete fracture network model of the fractured rock mass;
S6, analyzing discrete fracture network connectivity of the fractured rock mass through a graph theory algorithm;
s7, simulating and calculating the groundwater seepage value of the fractured rock mass to obtain simulated seepage parameters;
s8, printing the standard sample model into a physical model through 3D, and obtaining the permeation parameters of the actual physical model;
s9, comparing the simulated permeability parameters with the actual physical model permeability parameters, if the results are inconsistent, adjusting and optimizing a discrete fracture network model of the fractured rock mass, and carrying out iterative calculation until the results are consistent.
Preferably, in the step S1, the field acquisition refers to that the unmanned aerial vehicle flies and scans the outcrop area around the tunnel work area in a full coverage manner, acquires the three-dimensional point cloud data generated by the high-resolution image and the laser scanning, and acquires the fracture data, including the spatial distribution, the geometric characteristics and the physical properties of the fracture.
In the step S1, the calibration of the fracture data collected by the unmanned aerial vehicle means that the occurrence, trace length, opening degree, roughness coefficient, connectivity, filler, density, group number and longitude and latitude of rock mass fractures around a tunnel work area are manually obtained, and the calibration of the fracture data is performed.
In the step S2, the denoising pretreatment of the fracture data refers to removing noise through a Gaussian filtering algorithm of a cross-platform computer vision library, and removing non-target objects through a random sample consistency algorithm of the cross-platform computer vision library.
In S3, the geometric features of the fracture include the occurrence, opening, trace length, density, and roughness coefficient.
The embodiment is a preferred implementation mode, can remarkably improve the collection and treatment efficiency of rock mass cracks, simplify the whole numerical simulation flow and realize the automatic treatment of most flows, greatly simplify manual operation, improve modeling efficiency, improve the accuracy of numerical simulation of a rock crack network, and has important practical application value in tunnel seepage field evolution and gushing water disaster analysis.
Example 3
Referring to fig. 1-6, the three-dimensional numerical simulation method for the evolution of the seepage field of the fractured rock mass tunnel comprises the following steps:
s1, acquiring crack data of a rock outcrop area in the field through an unmanned aerial vehicle, and calibrating the crack data acquired by the unmanned aerial vehicle;
S2, denoising pretreatment is carried out on the fracture data, three-dimensional point cloud data in the fracture data are converted into a continuous three-dimensional surface model, and fracture trace lines and fracture structural surfaces are reconstructed to obtain a three-dimensional reconstruction model;
S3, training a convolutional neural network model, using the convolutional neural network model to a three-dimensional reconstruction model, identifying geometric features of the fissures, training a cyclic neural network, and processing spatial distribution of the fissures through the cyclic neural network to identify extension paths and width changes of the fissures;
s4, constructing a fracture structure surface network model;
s5, establishing a discrete fracture network model of the fractured rock mass;
S6, analyzing discrete fracture network connectivity of the fractured rock mass through a graph theory algorithm;
s7, simulating and calculating the groundwater seepage value of the fractured rock mass to obtain simulated seepage parameters;
s8, printing the standard sample model into a physical model through 3D, and obtaining the permeation parameters of the actual physical model;
s9, comparing the simulated permeability parameters with the actual physical model permeability parameters, if the results are inconsistent, adjusting and optimizing a discrete fracture network model of the fractured rock mass, and carrying out iterative calculation until the results are consistent.
In the step S1, field acquisition refers to that an unmanned aerial vehicle flies and scans the outcrop area of the rock around a tunnel work area in a full coverage mode, three-dimensional point cloud data generated by high-resolution images and laser scanning are acquired, and crack data comprising the spatial distribution, geometric characteristics and physical properties of cracks are acquired.
In the step S1, the calibration of the fracture data collected by the unmanned aerial vehicle means that the occurrence, trace length, opening degree, roughness coefficient, connectivity, filler, density, group number and longitude and latitude of rock mass fractures around a tunnel work area are manually obtained, and the calibration of the fracture data is performed.
In the step S2, the denoising pretreatment of the fracture data refers to removing noise through a Gaussian filtering algorithm of a cross-platform computer vision library, and removing non-target objects through a random sample consistency algorithm of the cross-platform computer vision library.
In S3, the geometric features of the fracture include the occurrence, opening, trace length, density, and roughness coefficient.
Further preferably, the step S4 specifically includes:
S41, converting the occurrence into spherical coordinates through the method 1;
1 (1)
Wherein,Is the tendency of the fracture structure surface,Is the inclination angle of the fracture structure,Is a horizontal angle, and is provided with a plurality of grooves,Is a vertical angle;
S42, converting the spherical coordinates into normal vectors of the fracture structural surface through a model 2;
2, 2
Wherein,、AndThe directions of the fracture structure surfaces are cosine in a rectangular coordinate system;
S43, calculating the trace length of the crack by using the crack structural surface trend of the crack as the direction of the disk and the opening degree of the crack as the thickness of the disk, determining the size of the disk, superposing a disk model, and establishing a crack structural surface network model;
3
Wherein,For the length of the track to be the same,Is the radius of the disc.
The step S5 specifically comprises the following steps:
s51, dividing the fracture shape into a plurality of groups, and acquiring an average value of normal vectors of each group of fractures as a basic parameter of fracture simulation;
S52, checking the statistical distribution of each group of cracks, and determining a statistical model as a parameter of Monte Carlo simulation;
S53, simulating the occurrence, trace length, opening degree and position of the fracture through Monte Carlo, establishing a single fracture through a disc model, and overlapping the single fracture to establish a discrete fracture network model of the fractured rock mass.
The step S6 specifically comprises the following steps:
s61, representing cracks in the fractured rock mass by using a cylindrical model, wherein the cracks are mutually communicated to form a water guide channel;
S62, performing orthogonal slicing and continuous slicing on a discrete fracture network model of the fractured rock mass to obtain a two-dimensional fracture trace distribution map of each position;
S63, numbering the cracks to obtain a crack diagram;
s64, marking the intersecting state of the cracks as a two-dimensional adjacent matrix for representing the communication relation of the cracks;
S65, judging whether two three-dimensional fracture discs are intersected or not so as to judge whether the fractures are communicated or not, and marking the fractures as a three-dimensional adjacent matrix;
s66, traversing an adjacency matrix of a discrete fracture network model of the fractured rock mass based on a depth-first algorithm in a graph theory path finding algorithm, and finding all communicated fractures and formed communication paths.
In this embodiment, the data collected by the unmanned aerial vehicle and the laser radar are calibrated by manually collecting the crack data around the tunnel work area on site, so that the accuracy of the collected data is improved.
The convolutional neural network and the cyclic neural network are written through the deep learning framework to automatically identify and extract the fracture characteristics, so that artificial extraction is reduced, and the improvement of productivity and the fracture treatment efficiency are facilitated.
Example 4
Referring to fig. 1-6, the three-dimensional numerical simulation method for the evolution of the seepage field of the fractured rock mass tunnel comprises the following steps:
s1, acquiring crack data of a rock outcrop area in the field through an unmanned aerial vehicle, and calibrating the crack data acquired by the unmanned aerial vehicle;
S2, denoising pretreatment is carried out on the fracture data, three-dimensional point cloud data in the fracture data are converted into a continuous three-dimensional surface model, and fracture trace lines and fracture structural surfaces are reconstructed to obtain a three-dimensional reconstruction model;
S3, training a convolutional neural network model, using the convolutional neural network model to a three-dimensional reconstruction model, identifying geometric features of the fissures, training a cyclic neural network, and processing spatial distribution of the fissures through the cyclic neural network to identify extension paths and width changes of the fissures;
s4, constructing a fracture structure surface network model;
s5, establishing a discrete fracture network model of the fractured rock mass;
S6, analyzing discrete fracture network connectivity of the fractured rock mass through a graph theory algorithm;
s7, simulating and calculating the groundwater seepage value of the fractured rock mass to obtain simulated seepage parameters;
s8, printing the standard sample model into a physical model through 3D, and obtaining the permeation parameters of the actual physical model;
s9, comparing the simulated permeability parameters with the actual physical model permeability parameters, if the results are inconsistent, adjusting and optimizing a discrete fracture network model of the fractured rock mass, and carrying out iterative calculation until the results are consistent.
In the step S1, field acquisition refers to that an unmanned aerial vehicle flies and scans the outcrop area of the rock around a tunnel work area in a full coverage mode, three-dimensional point cloud data generated by high-resolution images and laser scanning are acquired, and crack data comprising the spatial distribution, geometric characteristics and physical properties of cracks are acquired.
In the step S1, the calibration of the fracture data collected by the unmanned aerial vehicle means that the occurrence, trace length, opening degree, roughness coefficient, connectivity, filler, density, group number and longitude and latitude of rock mass fractures around a tunnel work area are manually obtained, and the calibration of the fracture data is performed.
In the step S2, the denoising pretreatment of the fracture data refers to removing noise through a Gaussian filtering algorithm of a cross-platform computer vision library, and removing non-target objects through a random sample consistency algorithm of the cross-platform computer vision library.
In S3, the geometric features of the fracture include the occurrence, opening, trace length, density, and roughness coefficient.
The step S4 specifically comprises the following steps:
S41, converting the occurrence into spherical coordinates through the method 1;
1 (1)
Wherein,Is the tendency of the fracture structure surface,Is the inclination angle of the fracture structure,Is a horizontal angle, and is provided with a plurality of grooves,Is a vertical angle;
S42, converting the spherical coordinates into normal vectors of the fracture structural surface through a model 2;
2, 2
Wherein,、AndThe directions of the fracture structure surfaces are cosine in a rectangular coordinate system;
S43, calculating the trace length of the crack by using the crack structural surface trend of the crack as the direction of the disk and the opening degree of the crack as the thickness of the disk, determining the size of the disk, superposing a disk model, and establishing a crack structural surface network model;
3
Wherein,For the length of the track to be the same,Is the radius of the disc.
The step S5 specifically comprises the following steps:
s51, dividing the fracture shape into a plurality of groups, and acquiring an average value of normal vectors of each group of fractures as a basic parameter of fracture simulation;
S52, checking the statistical distribution of each group of cracks, and determining a statistical model as a parameter of Monte Carlo simulation;
S53, simulating the occurrence, trace length, opening degree and position of the fracture through Monte Carlo, establishing a single fracture through a disc model, and overlapping the single fracture to establish a discrete fracture network model of the fractured rock mass.
The step S6 specifically comprises the following steps:
s61, representing cracks in the fractured rock mass by using a cylindrical model, wherein the cracks are mutually communicated to form a water guide channel;
S62, performing orthogonal slicing and continuous slicing on a discrete fracture network model of the fractured rock mass to obtain a two-dimensional fracture trace distribution map of each position;
S63, numbering the cracks to obtain a crack diagram;
s64, marking the intersecting state of the cracks as a two-dimensional adjacent matrix for representing the communication relation of the cracks;
S65, judging whether two three-dimensional fracture discs are intersected or not so as to judge whether the fractures are communicated or not, and marking the fractures as a three-dimensional adjacent matrix;
s66, traversing an adjacency matrix of a discrete fracture network model of the fractured rock mass based on a depth-first algorithm in a graph theory path finding algorithm, and finding all communicated fractures and formed communication paths.
Still further preferably, the step S7 specifically includes:
S71, dividing seepage of a fractured rock mass into seepage of a rock matrix and seepage of a fracture network, wherein the seepage of the rock matrix adopts a porous medium seepage model, and the seepage of the fracture network adopts a cubic theorem seepage model;
s72, obtaining the groundwater seepage flow of each crack according to the cube law;
4. The method is to
Wherein,For the seepage flow of the underground water,Is the density of the groundwater, and is used for the treatment of the ground water,The acceleration of the gravity is that,Is the opening degree of the fracture opening,Is the motion viscosity coefficient of the underground water,Is the water head difference of the underground water,Is hydraulic conductivity;
S73, calculating the groundwater head of the two-dimensional fracture through the method 5;
5. The method is to
Wherein,Is an underground water head;
S74, establishing a plurality of linear equation sets through the formula 5, solving the linear equation sets to obtain a water head value of each intersection point, and substituting the water head value into the formula 4 to obtain the groundwater seepage flow of each crack;
S75, optimizing a communicated fracture network by adopting a maximum flow minimum cutting theorem in a network flow optimization algorithm;
s76, establishing a seepage equation of the rock matrix through a formula 6, establishing a seepage equation of a fracture network through a formula 7, and calculating groundwater seepage values to obtain simulation seepage parameters;
6. The method is to
Wherein,In order to make the water content be the specific water content,In order for the saturation to be effective,In order to be a specific reserve quantity,Is the water potential of the underground water,In order to be able to take time,For the groundwater flow speed in the rock matrix,Is the source and sink flow;
7. The method of the invention
Wherein,Is the opening degree of the crack,Is the specific water volume in the fracture,Is the storage ratio of the fracture,Is the water potential in the crack, and the water potential in the crack,Is the permeability of a single fracture, and the permeability of the single fracture is the same,Is the water exchange quantity between the rock matrix and the fracture.
The step S8 specifically comprises the following steps:
S81, intercepting a fracture network in a three-dimensional discrete fracture network through a standard cylinder to obtain a standard sample model;
S82, printing the standard sample model into a physical model through 3D;
s83, obtaining the actual physical model permeability parameters through geophysical well logging prediction or seepage experiments.
The embodiment is an optimal implementation mode, and a two-dimensional fracture trace diagram of the discrete fracture network is obtained in a slicing mode, so that seepage field simulation of the discrete fracture network can be conducted at any position.
The seepage field evolution simulation of the discrete fracture network and the movement of the groundwater are carried out through the finite element and ion tracking algorithm, so that the high-precision simulation of the groundwater seepage can be realized.
And compared with the analysis solution verification by only using a mathematical formula, the seepage experiment verification by the actual physical model is more practical, and is favorable for iteratively improving the seepage model precision of the fracture network model. Only one programming environment is needed, a plurality of software switches are not needed, and most processes realize automatic operation through scripts, so that the numerical simulation process is greatly simplified.
The adjacency matrix of the present invention is shown below:
Wherein 0 means no intersection and 1 means intersection.
The graph theory algorithm is realized through a complex network analysis library.
The invention obtains the average value of each group of fracture normal vectors as the basic parameter of fracture simulation through a machine learning library.
The invention determines a statistical model and is realized by a scientific calculation library as a parameter of Monte Carlo simulation.
Claims (10)
1. The three-dimensional numerical simulation method for the evolution of the seepage field of the fractured rock mass tunnel is characterized by comprising the following steps of:
s1, acquiring crack data of a rock outcrop area in the field through an unmanned aerial vehicle, and calibrating the crack data acquired by the unmanned aerial vehicle;
S2, denoising pretreatment is carried out on the fracture data, three-dimensional point cloud data in the fracture data are converted into a continuous three-dimensional surface model, and fracture trace lines and fracture structural surfaces are reconstructed to obtain a three-dimensional reconstruction model;
S3, training a convolutional neural network model, using the convolutional neural network model to a three-dimensional reconstruction model, identifying geometric features of the fissures, training a cyclic neural network, and processing spatial distribution of the fissures through the cyclic neural network to identify extension paths and width changes of the fissures;
s4, constructing a fracture structure surface network model;
s5, establishing a discrete fracture network model of the fractured rock mass;
S6, analyzing discrete fracture network connectivity of the fractured rock mass through a graph theory algorithm;
s7, simulating and calculating the groundwater seepage value of the fractured rock mass to obtain simulated seepage parameters;
s8, printing the standard sample model into a physical model through 3D, and obtaining the permeation parameters of the actual physical model;
s9, comparing the simulated permeability parameters with the actual physical model permeability parameters, if the results are inconsistent, adjusting and optimizing a discrete fracture network model of the fractured rock mass, and carrying out iterative calculation until the results are consistent.
2. The method for three-dimensional numerical simulation of the evolution of the seepage field of the fractured rock mass tunnel according to claim 1, wherein the method comprises the following steps of: in the step S1, field acquisition refers to that an unmanned aerial vehicle flies and scans the outcrop area of the rock around a tunnel work area in a full coverage mode, three-dimensional point cloud data generated by high-resolution images and laser scanning are acquired, and crack data comprising the spatial distribution, geometric characteristics and physical properties of cracks are acquired.
3. The method for three-dimensional numerical simulation of the evolution of the seepage field of the fractured rock mass tunnel according to claim 1, wherein the method comprises the following steps of: in the step S1, the calibration of the fracture data collected by the unmanned aerial vehicle means that the occurrence, trace length, opening degree, roughness coefficient, connectivity, filler, density, group number and longitude and latitude of rock mass fractures around a tunnel work area are manually obtained, and the calibration of the fracture data is performed.
4. The method for three-dimensional numerical simulation of the evolution of the seepage field of the fractured rock mass tunnel according to claim 1, wherein the method comprises the following steps of: in the step S2, the denoising pretreatment of the fracture data refers to removing noise through a Gaussian filtering algorithm of a cross-platform computer vision library, and removing non-target objects through a random sample consistency algorithm of the cross-platform computer vision library.
5. The method for three-dimensional numerical simulation of the evolution of the seepage field of the fractured rock mass tunnel according to claim 1, wherein the method comprises the following steps of: in S3, the geometric features of the fracture include the occurrence, opening, trace length, density, and roughness coefficient.
6. The method for three-dimensional numerical simulation of the evolution of the seepage field of the fractured rock mass tunnel according to claim 1, wherein the method comprises the following steps of: the step S4 specifically comprises the following steps:
S41, converting the occurrence into spherical coordinates through the method 1;
1 (1)
Wherein,Is the tendency of the fracture structure surface,Is the inclination angle of the fracture structure,Is a horizontal angle, and is provided with a plurality of grooves,Is a vertical angle;
S42, converting the spherical coordinates into normal vectors of the fracture structural surface through a model 2;
2, 2
Wherein,、AndThe directions of the fracture structure surfaces are cosine in a rectangular coordinate system;
S43, calculating the trace length of the crack by using the crack structural surface trend of the crack as the direction of the disk and the opening degree of the crack as the thickness of the disk, determining the size of the disk, superposing a disk model, and establishing a crack structural surface network model;
3
Wherein,For the length of the track to be the same,Is the radius of the disc.
7. The method for three-dimensional numerical simulation of the evolution of the seepage field of the fractured rock mass tunnel according to claim 1, wherein the method comprises the following steps of: the step S5 specifically comprises the following steps:
s51, dividing the fracture shape into a plurality of groups, and acquiring an average value of normal vectors of each group of fractures as a basic parameter of fracture simulation;
S52, checking the statistical distribution of each group of cracks, and determining a statistical model as a parameter of Monte Carlo simulation;
S53, simulating the occurrence, trace length, opening degree and position of the fracture through Monte Carlo, establishing a single fracture through a disc model, and overlapping the single fracture to establish a discrete fracture network model of the fractured rock mass.
8. The method for three-dimensional numerical simulation of the evolution of the seepage field of the fractured rock mass tunnel according to claim 1, wherein the method comprises the following steps of: the step S6 specifically comprises the following steps:
s61, representing cracks in the fractured rock mass by using a cylindrical model, wherein the cracks are mutually communicated to form a water guide channel;
S62, performing orthogonal slicing and continuous slicing on a discrete fracture network model of the fractured rock mass to obtain a two-dimensional fracture trace distribution map of each position;
S63, numbering the cracks to obtain a crack diagram;
s64, marking the intersecting state of the cracks as a two-dimensional adjacent matrix for representing the communication relation of the cracks;
S65, judging whether two three-dimensional fracture discs are intersected or not so as to judge whether the fractures are communicated or not, and marking the fractures as a three-dimensional adjacent matrix;
s66, traversing an adjacency matrix of a discrete fracture network model of the fractured rock mass based on a depth-first algorithm in a graph theory path finding algorithm, and finding all communicated fractures and formed communication paths.
9. The method for three-dimensional numerical simulation of the evolution of the seepage field of the fractured rock mass tunnel according to claim 1, wherein the method comprises the following steps of: the step S7 specifically comprises the following steps:
S71, dividing seepage of a fractured rock mass into seepage of a rock matrix and seepage of a fracture network, wherein the seepage of the rock matrix adopts a porous medium seepage model, and the seepage of the fracture network adopts a cubic theorem seepage model;
s72, obtaining the groundwater seepage flow of each crack according to the cube law;
4. The method is to
Wherein,For the seepage flow of the underground water,Is the density of the groundwater, and is used for the treatment of the ground water,The acceleration of the gravity is that,Is the opening degree of the fracture opening,Is the motion viscosity coefficient of the underground water,Is the water head difference of the underground water,Is hydraulic conductivity;
S73, calculating the groundwater head of the two-dimensional fracture through the method 5;
5. The method is to
Wherein,Is an underground water head;
S74, establishing a plurality of linear equation sets through the formula 5, solving the linear equation sets to obtain a water head value of each intersection point, and substituting the water head value into the formula 4 to obtain the groundwater seepage flow of each crack;
S75, optimizing a communicated fracture network by adopting a maximum flow minimum cutting theorem in a network flow optimization algorithm;
s76, establishing a seepage equation of the rock matrix through a formula 6, establishing a seepage equation of a fracture network through a formula 7, and calculating groundwater seepage values to obtain simulation seepage parameters;
6. The method is to
Wherein,In order to make the water content be the specific water content,In order for the saturation to be effective,In order to be a specific reserve quantity,Is the water potential of the underground water,In order to be able to take time,For the groundwater flow speed in the rock matrix,Is the source and sink flow;
7. The method of the invention
Wherein,Is the opening degree of the crack,Is the specific water volume in the fracture,Is the storage ratio of the fracture,Is the water potential in the crack, and the water potential in the crack,Is the permeability of a single fracture, and the permeability of the single fracture is the same,Is the water exchange quantity between the rock matrix and the fracture.
10. The method for three-dimensional numerical simulation of the evolution of the seepage field of the fractured rock mass tunnel according to claim 1, wherein the method comprises the following steps of: the step S8 specifically comprises the following steps:
S81, intercepting a fracture network in a three-dimensional discrete fracture network through a standard cylinder to obtain a standard sample model;
S82, printing the standard sample model into a physical model through 3D;
s83, obtaining the actual physical model permeability parameters through geophysical well logging prediction or seepage experiments.
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118332902A true CN118332902A (en) | 2024-07-12 |
Family
ID=
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10656068B2 (en) | Method for providing a numerical model of a sample of rock | |
| CN111339691A (en) | Intelligent geotechnical engineering parameter three-dimensional analysis and evaluation system and method based on voxler software | |
| Bilskie et al. | Topographic accuracy assessment of bare earth lidar-derived unstructured meshes | |
| CN105279790A (en) | Fracture network 3D digital core modeling method | |
| CN102609982B (en) | Topology discovery method of space geological data based on unstructured mode | |
| Bruna et al. | A new methodology to train fracture network simulation using multiple-point statistics | |
| Koike et al. | Incorporation of fracture directions into 3D geostatistical methods for a rock fracture system | |
| CN115937467B (en) | Method and system for dividing random three-dimensional fracture in upscale model grid | |
| CN115100373A (en) | Surface and underground environment three-dimensional integrated representation method, device, medium and equipment | |
| Comunian et al. | Hierarchical simulation of aquifer heterogeneity: implications of different simulation settings on solute-transport modeling|[Simulation hiérarchique de l’hétérogénéité des aquifères: implications de différentes configurations de simulation sur la modélisation du transport de solutés] | |
| CN109376209B (en) | Contaminated site database 3D model display system | |
| CN118332902A (en) | Three-dimensional numerical simulation method for evolution of seepage field of fractured rock mass tunnel | |
| CN114898057B (en) | Geological profile three-dimensional space information extraction method and device and storage medium | |
| Salah | Geostatistical analysis of groundwater levels in the south Al Jabal Al Akhdar area using GIS | |
| CN105279305B (en) | A kind of Three Dimensional Ground laser scanner technique survey station choosing method | |
| CN113920135A (en) | Ion type rare earth ore digital core modeling method | |
| WO2021231170A1 (en) | Systems and methods for generating vertical and lateral heterogeneity indices of reservoirs | |
| CN108537883B (en) | Geological modeling method based on MapReduce framework | |
| CN117745979B (en) | Three-dimensional fracture-pore coupling network simulation generation method and system | |
| CN115469361B (en) | Clastic rock stratum three-dimensional geological modeling method | |
| CN111401451B (en) | Automatic identification method for reservoir configuration interface in reservoir geological model | |
| Wang et al. | Improved skeleton extraction method considering surface feature of natural micro fractures in unconventional shale/tight reservoirs | |
| Wang et al. | Existence analysis of hydraulic conductivity representative elementary volume in fractured rocks based on three-dimensional discrete fracture network method | |
| CN116385683B (en) | Three-dimensional small drainage basin channel fractal dimension calculation method and system | |
| CN115564908B (en) | Three-dimensional modeling system for ground surface matrix layer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication |