A method for quickly constructing a three-dimensional complex rock mass geological model

Abstract

The application relates to the technical field of three-dimensional geological modeling, in particular to a method for rapidly constructing a three-dimensional complex rock mass geological model, which comprises the following steps: obtaining an initial three-dimensional geological model of an underground complex rock mass in a to-be-detected geological area and marking the initial three-dimensional geological model as a to-be-detected initial three-dimensional geological model; screening each cube surface that may cause ambiguity problems; obtaining a three-dimensional coordinate range that may cause ambiguity problems, and obtaining false characteristic values of each hole boundary vertex set; training a binary classification model to obtain a judgment threshold of the false characteristic values of all hole boundary vertex sets of the to-be-detected initial three-dimensional geological model, screening each false hole from the holes to which all the hole boundary vertex sets belong, repairing the false holes, and obtaining a final three-dimensional geological model of the underground complex rock mass in the to-be-detected geological area. The application aims to improve the authenticity and integrity of the three-dimensional complex rock mass geological model.

Description

Technical Field

[0001] This application relates to the field of three-dimensional geological modeling technology, specifically to a method for rapidly constructing a three-dimensional complex rock mass geological model. Background Technology

[0002] A three-dimensional (3D) geological model of complex rock masses is a computer-generated model that describes the geological structure of complex underground rock masses in three dimensions. Because it can visually display the distribution of geological structures within underground rock masses, such as faults, karst caves, and fissures, 3D geological models are widely used in engineering applications such as tunnel construction, mining, and dam construction. Therefore, rapidly constructing a 3D geological model of complex rock masses can provide more reliable technical support for engineering design and construction.

[0003] Existing methods typically involve interpolating the 3D coordinates of all measuring points in the acquired geological survey data, and then using the moving cubes algorithm to mesh the interpolated 3D data to obtain a 3D geological model. However, this process ignores the ambiguity inherent in the moving cubes algorithm. When directly using the moving cubes algorithm to quickly construct a 3D complex rock mass geological model of the target geological area, false voids due to ambiguity may appear in the final 3D complex rock mass geological model. These false voids are not the actual caves, fissures, or other voids that exist within the complex underground rock mass of the target geological area. The generation of false voids reduces the realism and completeness of the final 3D complex rock mass geological model, thereby affecting the accuracy of subsequent engineering design and construction. Summary of the Invention

[0004] In view of the above, it is necessary to provide a rapid construction method for three-dimensional complex rock mass geological models, which improves the realism and completeness of the three-dimensional complex rock mass geological models compared with traditional rapid construction methods.

[0005] The method for rapidly constructing a three-dimensional complex rock mass geological model proposed in this application adopts the following technical solution:

[0006] One embodiment of this application provides a method for rapidly constructing a three-dimensional complex rock mass geological model, the method comprising the following steps:

[0007] Using all the measuring points in the geological exploration data of the complex underground rock mass in the geological area to be tested, the moving cube algorithm is used to obtain the initial three-dimensional geological model of the complex underground rock mass in the geological area to be tested, and it is recorded as the initial three-dimensional geological model to be tested.

[0008] A hole identification process is performed on the initial 3D geological model to be tested, resulting in a set of vertex boundaries for each hole in the initial 3D geological model. For any isomorphic surface formed during the acquisition of the initial 3D geological model using the moving cube algorithm and any voxel it passes through, the function value of the surface function of the isomorphic surface is compared with the function value obtained by substituting the vertex coordinates on each cube face of the voxel into the surface function. Cube faces that may have ambiguity are then filtered out. The range of ambiguity is obtained by using the 3D coordinates of the vertices of all the cube faces that may have ambiguity from all the voxels formed during the acquisition of the initial 3D geological model. Finally, the false feature value of each set of vertex boundaries for each hole is obtained by counting the number of vertices within the range of the 3D coordinates.

[0009] Based on all the measuring points in the geological exploration data of a preset number of underground complex rock masses in different geological regions collected in history, a binary classification model is trained to obtain the judgment threshold of false feature values ​​of all hole boundary vertex sets of the initial three-dimensional geological model to be tested. Then, false holes are screened from the holes to which all hole boundary vertex sets belong, and the false holes are repaired to obtain the final three-dimensional geological model of the underground complex rock mass in the geological region to be tested.

[0010] In one embodiment, the process of obtaining the set of vertices of the hole boundary is as follows:

[0011] The hole identification method of three-dimensional triangular mesh model is used to identify holes in the initial three-dimensional geological model to be tested, and the set of boundary vertices composed of all boundary edges in the initial three-dimensional geological model to be tested is obtained.

[0012] The vertices in the boundary vertex set are used as nodes of the undirected graph. An edge is established between the two vertices of the same triangular facet in the initial 3D geological model to be tested in the boundary vertex set and their corresponding two nodes in the undirected graph. An undirected graph is then created. All cycles in the undirected graph are extracted. All the vertices corresponding to all nodes in each cycle in the boundary vertex set are used to form the boundary vertex set of each hole in the initial 3D geological model to be tested.

[0013] In one embodiment, the process of screening the cube faces that would lead to ambiguity is as follows:

[0014] Substitute the coordinates of the four vertices on any cube face of any volume element into the surface function of any isosurface to obtain function values ​​f(b1), f(b2), f(b3), and f(b4), where f(b1) and f(b3) are the function values ​​of two vertices on the same diagonal, and f(b2) and f(b4) are the function values ​​of two vertices on the other diagonal. Denote any isosurface as w, and denote the function value of the surface function of isosurface w as f(w). Based on the differences between function values ​​f(b1), f(b2), f(b3), f(b4) and f(w), determine the feature vector of any cube face, which is used to evaluate whether any cube face is a cube face that will have an ambiguity problem.

[0015] In one embodiment, the feature vector is obtained by using the calculation formulas k1(b)=(f(b1)-f(w))*(f(b3)-f(w)), k2(b)=(f(b2)-f(w))*(f(b4)-f(w)), k3(b)=-(f(b1)-f(w))*(f(b2)-f(w)) to form the feature vector of any cube face.

[0016] In one embodiment, the method for evaluating whether any cube facet is a cube facet that will have an ambiguity problem is as follows: if all components in the feature vector of any cube facet are greater than 0, then the cube facet is regarded as a cube facet that will have an ambiguity problem.

[0017] In one embodiment, the process of obtaining the range of three-dimensional coordinates that would lead to ambiguity is as follows:

[0018] For the 3D coordinates of all vertices of the cube face where ambiguity occurs, obtain the minimum and maximum values ​​of the X coordinate (xmin and xmax), the minimum and maximum values ​​of the Y coordinate (ymin and ymax), and the minimum and maximum values ​​of the Z coordinate (zmin and zmax) to determine the range of 3D coordinates where ambiguity occurs.

[0019] In one embodiment, the three-dimensional coordinate range is X = [xmin, xmax], Y = [ymin, ymax], and Z = [zmin, zmax].

[0020] In one embodiment, the process of obtaining the false feature value is as follows:

[0021] From any set of vertices at the boundary of a hole, select all vertices whose three-dimensional coordinates fall within the range of the three-dimensional coordinates. Use the ratio of the number of selected vertices to the total number of vertices in the set of vertices at the boundary of a hole as the false feature value of the set of vertices at the boundary of a hole.

[0022] In one embodiment, the training of the binary classification model, used to obtain the threshold for judging false feature values ​​of the set of all hole boundary vertices of the initial three-dimensional geological model to be tested, includes:

[0023] The initial three-dimensional geological models of underground complex rock masses in a predetermined number of different geological regions collected in history are denoted as each reference initial three-dimensional geological model. The spurious feature values ​​of the set of all hole boundary vertices of all reference initial three-dimensional geological models are used to train a binary classification model.

[0024] The judgment threshold is the value of the false feature value corresponding to the segmentation plane in the binary classification model.

[0025] In one embodiment, the false hole is the hole to which the set of boundary vertices of holes with false feature values ​​greater than the judgment threshold belongs.

[0026] This application has at least the following beneficial effects:

[0027] This application first identifies the set of cavity boundary vertices, which are composed of the boundary points of each cavity in the initial three-dimensional geological model to be tested. Then, by comparing the function values ​​of the isosurface with the function values ​​of the cube face vertices, it accurately locates each cube face that may have ambiguity. This allows for the accurate screening of all cube faces that may have ambiguity. Furthermore, by using the three-dimensional coordinate range of the cube faces that may have ambiguity, it obtains the false feature values ​​of each cavity boundary vertex set. This quantifies the possibility that the cavity to which each cavity boundary vertex set belongs is a false cavity. This helps to accurately assess whether the cavities in the initial three-dimensional geological model of the complex underground rock mass of the geological area to be tested are false cavities caused by ambiguity in the moving cube algorithm. This avoids the need to repair real cavities such as karst caves and fissures that exist inside the complex underground rock mass of the geological area to be tested in the initial three-dimensional geological model.

[0028] Furthermore, by filtering out false spurious features caused by ambiguity issues and repairing them, the algorithm can quickly construct a 3D geological model of complex underground rock masses in the geological area under test using the moving cube algorithm. At the same time, it can specifically repair false spurious ... Attached Figure Description

[0029] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 A flowchart illustrating the steps of a rapid construction method for a three-dimensional complex rock mass geological model provided in this application;

[0031] Figure 2 A schematic diagram illustrating the process of obtaining the set of vertices at the boundary of a hole;

[0032] Figure 3 This is a schematic diagram illustrating the process of obtaining fake holes. Detailed Implementation

[0033] In the description of the embodiments of this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. It should be understood that, unless otherwise stated, " / " in this application means "or".

[0035] It should also be noted that the terms "first" and "second" in this application are used to distinguish similar objects, rather than to describe a specific order or sequence.

[0036] The following description, in conjunction with the accompanying drawings, details the specific scheme of the rapid construction method for a three-dimensional complex rock mass geological model provided in this application.

[0037] This application provides a method for rapidly constructing a three-dimensional complex rock mass geological model, specifically, the method described below. Please refer to [link to relevant documentation]. Figure 1 The method includes the following steps:

[0038] Step 1: Using all the measuring points in the geological exploration data of the complex underground rock mass in the geological area to be tested, the moving cube algorithm is used to obtain the initial three-dimensional geological model of the complex underground rock mass in the geological area to be tested, and this model is recorded as the initial three-dimensional geological model to be tested.

[0039] This application aims to repair false holes in the three-dimensional geological model of the complex underground rock mass in the test area constructed by the moving cube algorithm, so as to avoid misjudging false holes as actual cavities, fissures and other pores inside the complex underground rock mass, thereby realizing the rapid and accurate construction of the three-dimensional geological model of the complex underground rock mass in the test area.

[0040] Geological exploration data of the complex underground rock mass in the area to be tested is collected. A three-dimensional spatial interpolation algorithm is used to interpolate the three-dimensional coordinate data of all measuring points in the collected geological exploration data. Then, a moving cubes algorithm is used to mesh the interpolated three-dimensional data to obtain an initial three-dimensional geological model of the complex underground rock mass in the geological area to be tested, which is denoted as the initial three-dimensional geological model to be tested. The geological exploration data can be borehole data, planar geological maps, profile geological maps, or exploration and mining engineering maps; in this embodiment, borehole data is selected. The acquisition of borehole data and the moving cubes algorithm are both well-known technologies and will not be described further in this application.

[0041] In this embodiment, the three-dimensional spatial interpolation algorithm is specifically the Kriging method. The Kriging method is a well-known technology and will not be described in detail in this application. As other implementation methods, based on the ability to interpolate the three-dimensional coordinate data of the measurement point, the implementer may adopt other existing feasible technologies, such as the inverse power distance method or the Thiessen polygon method. This application does not impose any special restrictions.

[0042] Simultaneously, a voxel dataset of the initial 3D geological model to be tested is obtained. Each voxel data set comprises four parts: a set of vertex indices, a set of vertex coordinates, a set of isosurface values, and a set of vertex function values ​​for each voxel traversed by the moving cube algorithm during meshing. The isosurfaces are formed from 3D data obtained through interpolation. A voxel is a cube element obtained by voxelizing the 3D data obtained through interpolation during meshing. Each array in the vertex index set is an array of indices of the four vertices on each cube face of each voxel, sorted counter-clockwise. The vertex coordinate set is a set of 3D coordinates of the eight vertices of each voxel. The isosurface values ​​are the function values ​​of the surface functions traversing the isosurfaces of each voxel. The vertex function value set is a set of function values ​​obtained by substituting the 3D coordinates of the eight vertices of each voxel into the non-constant terms of the surface function. If each voxel traverses multiple isosurfaces, then each voxel corresponds to multiple voxel data sets. The voxel dataset can be used to evaluate whether ambiguity will occur in the triangular facet vertices corresponding to each voxel through which the isosurface passes in the initial 3D geological model to be tested.

[0043] Step 2: Identify holes in the initial 3D geological model to be tested, obtaining the set of vertices representing the boundaries of each hole in the initial 3D geological model to be tested. For any isomorphic surface formed during the acquisition of the initial 3D geological model to be tested using the moving cube algorithm and any voxel it passes through, filter out the cube faces that may have ambiguity by comparing the function value of the surface function of the isomorphic surface with the function value obtained by substituting the vertex coordinates on each cube face of the voxel into the surface function. Obtain the range of ambiguity by using the 3D coordinates of the vertices of all the cube faces that may have ambiguity from all the voxels formed during the acquisition of the initial 3D geological model to be tested. Obtain the false feature value of each set of vertex boundaries of holes by using the number of vertices in the set of vertex boundaries of holes that are within the range of the 3D coordinates.

[0044] The hole identification method of the three-dimensional triangular mesh model is used to identify holes in the initial three-dimensional geological model to be tested, and the set of boundary vertices composed of all boundary edges in the initial three-dimensional geological model to be tested is obtained. Among them, the boundary edges are the hole boundaries. The hole identification method of the three-dimensional triangular mesh model is a well-known technology and will not be described in detail in this application.

[0045] Generally, vertices belonging to the same cavity boundary in the initial 3D geological model to be tested usually form a closed loop polygon. Therefore, the vertices in the boundary vertex set are used as nodes of an undirected graph. An edge is established between the two vertices of the same triangular facet in the initial 3D geological model to be tested, and the corresponding two nodes in the undirected graph are used to construct an undirected graph. All loops in the undirected graph are extracted, and all vertices in the boundary vertex set corresponding to all nodes in each loop are used to form the boundary vertex set of each cavity in the initial 3D geological model to be tested. This set is used to represent the set of boundary vertices of each cavity in the initial 3D geological model to be tested. The construction of the undirected graph and the extraction of loops in the undirected graph are well-known techniques and will not be described in detail in this application. The schematic diagram of the process of obtaining the cavity boundary vertex set is shown below. Figure 2 As shown.

[0046] In the moving cube algorithm, for any cube face of any voxel, if the function values ​​of two vertices on one diagonal of the cube face are greater than the isosurface value of the isosurface passing through the voxel, and the function values ​​of two vertices on the other diagonal are less than the isosurface value of the isosurface passing through the voxel, then an ambiguity problem usually occurs on the cube face, which leads to the generation of holes in the initial 3D geological model to be tested constructed using the moving cube algorithm.

[0047] Based on the above analysis, taking any volumetric data element B in the volumetric data set of the initial 3D geological model to be tested as an example, for any array b in the vertex index array set of volumetric data element B, the first to fourth index values ​​in array b are denoted as b1, b2, b3, and b4 respectively. Since the index values ​​in array b are on the same cube face and are sorted counterclockwise, the volumetric element corresponding to volumetric data element B is denoted as volumetric element M, and the cube face of the vertices with index values ​​b1, b2, b3, and b4 on volumetric element M is denoted as cube face m. Then b1 and b3 are two vertices on one diagonal of cube face m, and b3 and b4 are two vertices on the other diagonal of cube face m. The index values ​​b1, b2, and b4 are obtained from the vertex function value set of volumetric data element B respectively. The function values ​​f(b1), f(b2), f(b3), and f(b4) of the vertices of b4 are used as the feature vectors of the cube face m, K(b) = (k1(b), k2(b), k3(b)) to determine whether ambiguity will occur on the cube face m. Here, k1(b) = (f(b1) - f(w)) * (f(b3) - f(w)), k2(b) = (f(b2) - f(w)) * (f(b4) - f(w)), and k3(b) = -(f(b1) - f(w)) * (f(b2) - f(w)). The isosurface that passes through the volume element M when constructing the volume element B is denoted as w, and f(w) represents the function value of the surface function of the isosurface w. When k1(b), k2(b), and k3(b) are all greater than 0, it is determined that ambiguity will occur on the cube face m.

[0048] From all arrays in the vertex index array set of all volume metadata in the volume metadata set, select the arrays corresponding to all cube faces that will have ambiguity problems, and form a problem array set.

[0049] When using the moving cube algorithm to mesh the interpolated 3D data, the vertices of the triangular facets in the initial 3D geological model to be tested are usually the vertices of the voxels obtained after voxelizing the interpolated 3D data, as well as the intersections of the isosurfaces formed by the interpolated 3D data and the edges of the voxels.

[0050] Based on the above analysis, taking any array c in the problem array set as an example, the three-dimensional coordinates of the vertices represented by the four index values ​​in array c are obtained from the vertex coordinate set of the data of the volume element where array c is located, thus obtaining the three-dimensional coordinate range of array c: X = [x1, x2], Y = [y1, y2], Z = [z1, z2]. This range is used to evaluate the position of the cube face corresponding to array c in the coordinate system of the initial three-dimensional geological model to be tested. Here, x1 and x2 represent the minimum and maximum values ​​of the X coordinate among the four obtained three-dimensional coordinates, y1 and y2 represent the minimum and maximum values ​​of the Y coordinate among the four obtained three-dimensional coordinates, and z1 and z2 represent the minimum and maximum values ​​of the Z coordinate among the four obtained three-dimensional coordinates. The X, Y, and Z coordinate ranges of all arrays in the problem array set are merged. The merged X, Y, and Z coordinate ranges are used as the three-dimensional coordinate ranges where ambiguity will occur. This is used to characterize the position of all cube faces that will have ambiguity in the coordinate system of the initial three-dimensional geological model in the data set of the initial three-dimensional geological model to be tested.

[0051] Taking any set of vertices D representing the boundary of a hole in the initial 3D geological model as an example, all vertices whose 3D coordinates fall within the range where ambiguity may occur are selected from the set of vertices D. The ratio of the number of selected vertices to the total number of vertices in the set of vertices D is used as the false feature value of the set of vertices D. This value is used to evaluate whether the hole corresponding to the set of vertices D in the initial 3D geological model is a false hole caused by the ambiguity in the moving cube algorithm. The larger the false feature value, the more likely the hole corresponding to the set of vertices D in the initial 3D geological model is to be a false hole.

[0052] Step 3: Based on all the measuring points in the geological exploration data of a preset number of underground complex rock masses in different geological regions collected in history, train a binary classification model to obtain the judgment threshold of false feature values ​​of all hole boundary vertex sets of the initial three-dimensional geological model to be tested. Then, filter out each false hole from the holes to which all hole boundary vertex sets belong, repair the false holes, and obtain the final three-dimensional geological model of the underground complex rock mass in the geological region to be tested.

[0053] Geological survey data of complex underground rock masses in different geological regions obtained from historical data collection were acquired. Using the same acquisition method as the initial 3D geological model of the complex underground rock masses in the geological region to be surveyed, initial 3D geological models of complex underground rock masses in each geological region were constructed and designated as control initial 3D geological models. A predetermined number of control initial 3D geological models containing false cavities were manually selected from all the constructed control initial 3D geological models. Using the same acquisition method as for the false feature values ​​of the set of vertices at the cavity boundaries D, the sets of vertices at the cavity boundaries of each selected control initial 3D geological model, as well as the false feature values ​​of each set of vertices at the cavity boundaries D, were obtained. The eigenvalues ​​are obtained by training a binary classification model using a support vector machine (SVM) model with all the spurious feature values. Spurious feature values ​​belonging to and not belonging to the set of vertices representing the boundary of holes caused by ambiguity in the moving cube algorithm are labeled 1 and 0, respectively. The F1-Score is used as the evaluation metric for the SVM model. The value of the spurious feature value corresponding to the segmentation plane in the trained binary classification model is used as the threshold for judging spurious feature values. This threshold is used to determine whether the set of vertices representing the boundary of holes in the initial 3D geological model is composed of spurious holes caused by ambiguity in the moving cube algorithm. The training process of the SVM model is a well-known technique and will not be described further in this application.

[0054] In this embodiment, the preset quantity is 200. The preset quantity is preset by the user and can be set by the implementer. This application does not impose any special restrictions.

[0055] For the set of all cavity boundary vertices in the initial 3D geological model to be tested, cavities belonging to the set of vertices whose false feature values ​​are greater than a judgment threshold are identified as false cavities. A triangular mesh cavity repair method based on radial basis functions is used to repair these false cavities. The resulting 3D geological model is then used as the final 3D geological model of the complex underground rock mass in the geological area to be tested. The triangular mesh cavity repair method based on radial basis functions is a known technique and will not be described in detail here. As other implementation methods, implementers may use other existing feasible methods to repair false cavities in the initial 3D geological model to be tested, and this application does not impose any special restrictions. A schematic diagram of the process for obtaining false cavities is shown below. Figure 3 As shown.

[0056] In summary, this application first identifies the set of cavity boundary vertices composed of the boundary points of each cavity in the initial three-dimensional geological model to be tested. Then, by comparing the function values ​​of the isosurface with the function values ​​of the cube face vertices, it accurately locates each cube face that may have ambiguity issues. This allows for the accurate screening of all cube faces that may have ambiguity issues. Furthermore, by using the three-dimensional coordinate range of the cube faces that may have ambiguity issues, it obtains the false feature values ​​of each cavity boundary vertex set. This quantifies the possibility that the cavity to which each cavity boundary vertex set belongs is a false cavity. This helps to accurately evaluate whether the cavities in the initial three-dimensional geological model of the complex underground rock mass of the geological area to be tested are false cavities caused by ambiguity issues in the moving cube algorithm. This avoids the need to repair real cavities such as karst caves and fissures that exist inside the complex underground rock mass of the geological area to be tested in the initial three-dimensional geological model.

[0057] Furthermore, by filtering out false spurious features caused by ambiguity issues and repairing them, the algorithm can quickly construct a 3D geological model of complex underground rock masses in the geological area under test using the moving cube algorithm. At the same time, it can specifically repair false spurious ...

[0058] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0059] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from its essential characteristics. Therefore, the embodiments described above should be considered exemplary and non-limiting in all respects.

Claims

1. A method for rapidly constructing a three-dimensional geological model of a complex rock mass, characterized in that, The method includes the following steps: Using all the measuring points in the geological exploration data of the complex underground rock mass in the geological area to be tested, the moving cube algorithm is used to obtain the initial three-dimensional geological model of the complex underground rock mass in the geological area to be tested, and it is recorded as the initial three-dimensional geological model to be tested. A hole identification process is performed on the initial 3D geological model to be tested, resulting in a set of vertex boundaries for each hole in the initial 3D geological model. For any isomorphic surface formed during the acquisition of the initial 3D geological model using the moving cube algorithm and any voxel it passes through, the function value of the surface function of the isomorphic surface is compared with the function value obtained by substituting the vertex coordinates on each cube face of the voxel into the surface function. Cube faces that may have ambiguity are then filtered out. The range of ambiguity is obtained by using the 3D coordinates of the vertices of all the cube faces that may have ambiguity from all the voxels formed during the acquisition of the initial 3D geological model. Finally, the false feature value of each set of vertex boundaries for each hole is obtained by counting the number of vertices within the range of the 3D coordinates. Based on all the measuring points in the geological exploration data of underground complex rock masses in a preset number of different geological regions collected in history, a binary classification model is trained to obtain the judgment threshold of false feature values ​​of all hole boundary vertex sets of the initial three-dimensional geological model to be tested. Then, false holes are screened from the holes to which all hole boundary vertex sets belong, and the false holes are repaired to obtain the final three-dimensional geological model of underground complex rock masses in the geological region to be tested. The process of obtaining the false feature value is as follows: select all vertices whose three-dimensional coordinates are within the range of the three-dimensional coordinates from any set of vertices at the boundary of the hole, and use the ratio of the number of selected vertices to the total number of vertices in the set of vertices at the boundary of the hole as the false feature value of the set of vertices at the boundary of the hole.

2. The method for rapidly constructing a three-dimensional complex rock mass geological model as described in claim 1, characterized in that, The process of obtaining the set of vertices of the hole boundary is as follows: The hole identification method of three-dimensional triangular mesh model is used to identify holes in the initial three-dimensional geological model to be tested, and the set of boundary vertices composed of all boundary edges in the initial three-dimensional geological model to be tested is obtained. The vertices in the boundary vertex set are used as nodes of the undirected graph. An edge is established between the two vertices of the same triangular facet in the initial 3D geological model to be tested in the boundary vertex set and their corresponding two nodes in the undirected graph. An undirected graph is then created. All cycles in the undirected graph are extracted. All the vertices corresponding to all nodes in each cycle in the boundary vertex set are used to form the boundary vertex set of each hole in the initial 3D geological model to be tested.

3. The method for rapidly constructing a three-dimensional complex rock mass geological model as described in claim 1, characterized in that, The process of filtering out the cube faces that would lead to ambiguity is as follows: Substitute the coordinates of the four vertices on any cube face of any volume element into the surface function of any isosurface to obtain function values ​​f(b1), f(b2), f(b3), and f(b4), where f(b1) and f(b3) are the function values ​​of two vertices on the same diagonal, and f(b2) and f(b4) are the function values ​​of two vertices on the other diagonal. Denote any isosurface as w, and denote the function value of the surface function of isosurface w as f(w). Based on the differences between function values ​​f(b1), f(b2), f(b3), f(b4) and f(w), determine the feature vector of any cube face, which is used to evaluate whether any cube face is a cube face that will have an ambiguity problem.

4. The method for rapidly constructing a three-dimensional complex rock mass geological model as described in claim 3, characterized in that, The method for obtaining the feature vector is as follows: using the calculation formula , , The calculated k1(b), k2(b), and k3(b) form the feature vector of any cube face.

5. The method for rapidly constructing a three-dimensional complex rock mass geological model as described in claim 3, characterized in that, The method for evaluating whether any cube facet is a cube facet that will have an ambiguity problem is as follows: if all components in the feature vector of any cube facet are greater than 0, then any cube facet is considered as a cube facet that will have an ambiguity problem.

6. The method for rapidly constructing a three-dimensional complex rock mass geological model as described in claim 1, characterized in that, The process of obtaining the range of three-dimensional coordinates that may cause ambiguity is as follows: For the 3D coordinates of all vertices of the cube face where ambiguity occurs, obtain the minimum and maximum values ​​of the X coordinate (xmin and xmax), the minimum and maximum values ​​of the Y coordinate (ymin and ymax), and the minimum and maximum values ​​of the Z coordinate (zmin and zmax) to determine the range of 3D coordinates where ambiguity occurs.

7. The method for rapidly constructing a three-dimensional complex rock mass geological model as described in claim 5, characterized in that, The range of the three-dimensional coordinates is X=[xmin,xmax], Y=[ymin,ymax], and Z=[zmin,zmax].

8. The method for rapidly constructing a three-dimensional complex rock mass geological model as described in claim 1, characterized in that, The trained binary classification model is used to obtain the threshold for judging false feature values ​​of the set of all hole boundary vertices in the initial 3D geological model to be tested, including: The initial three-dimensional geological models of underground complex rock masses in a predetermined number of different geological regions collected in history are denoted as each reference initial three-dimensional geological model. The spurious feature values ​​of the set of all hole boundary vertices of all reference initial three-dimensional geological models are used to train a binary classification model. The judgment threshold is the value of the false feature value corresponding to the segmentation plane in the binary classification model.

9. The method for rapidly constructing a three-dimensional complex rock mass geological model as described in claim 1, characterized in that, The false holes are those whose false feature values ​​are greater than the judgment threshold and belong to the set of boundary vertices of the holes.

Images