A method for constructing a mine three-dimensional model

Abstract

The application discloses a kind of construction methods of mine three-dimensional model, it is related to geological exploration technical field.The method includes: obtaining the exploration comprehensive data of the mine to be modeled;The roadway of the mine to be modeled is scanned, and a plurality of point cloud data are obtained;Point cloud data are sequentially subjected to noise detection and denoising, and obtain the point cloud data after denoising;The point cloud data after denoising are divided into a plurality of grids in three-dimensional space, and each grid is marked by grid feature vector, and grid data are obtained;Based on grid data and exploration comprehensive data, modeling is carried out by SKUA-GOCAD, and the mine three-dimensional model of the mine to be modeled is obtained;When the mine to be modeled is newly drilled or roadway is exposed, the mine three-dimensional model is updated in real time by underground small geological anomaly body and tunneling point cloud data;Underground small geological anomaly body is detected by geophysical prospecting means including geological radar.The method improves the modeling accuracy of mine three-dimensional model.

Description

Technical Field

[0001] This invention relates to the field of geological exploration technology, and in particular to a method for constructing a three-dimensional model of a mine. Background Technology

[0002] Three-dimensional modeling of mines is the core foundation of intelligent coal mine construction. By integrating multi-source data such as drilling and geophysical exploration, a high-precision three-dimensional model of the mine can be constructed, which can intuitively present the morphology of the ore body, grade distribution and fault structure, significantly improving the accuracy of resource reserve estimation. Its dynamic update capability and modeling accuracy directly affect mine safety production and resource development decisions.

[0003] Traditional tunnel surveying methods, such as Real-Time Kinematic (RTK), cross-section methods, and total station surveying, rely on manual splicing of profiles or interpolation during the modeling process. The model is a static result, and it needs to be modified or even rebuilt when new boreholes or tunnels are exposed. The model is difficult to update, resulting in low modeling accuracy of the 3D mine model. Summary of the Invention

[0004] Therefore, it is necessary to provide a method for constructing a 3D model of a mine to address the aforementioned technical problems. This method improves the modeling accuracy of the 3D mine model.

[0005] The present invention adopts the following technical solution:

[0006] This invention provides a method for constructing a three-dimensional model of a mine, comprising:

[0007] Obtain comprehensive exploration data for the mine to be modeled; the comprehensive exploration data includes stratigraphic information, roadway information, geological anomaly information, borehole information, and well information;

[0008] The roadways of the mine to be modeled are scanned to obtain multiple point cloud data; each point cloud data includes three-dimensional coordinates and attribute information; the attribute information includes roof, floor and sidewalls;

[0009] The point cloud data is subjected to noise detection and denoising in sequence to obtain the denoised point cloud data; the noise includes people in the alley, screws on the inner wall of the alley, and light tubes on the inner wall of the alley.

[0010] The denoised point cloud data is divided into multiple grids in three-dimensional space, and each grid is labeled using a grid feature vector to obtain grid data. The grid feature vector includes three-dimensional coordinates, reflectivity, and variance. Variance is used to represent the density of points in a grid.

[0011] Based on grid data and comprehensive exploration data, a 3D model of the mine to be modeled is obtained by using SKUA-GOCAD.

[0012] When new boreholes or roadways are exposed in the mine to be modeled, the 3D model of the mine is updated in real time using small geological anomalies underground and point cloud data after excavation; the small geological anomalies underground are detected by geophysical means including ground-penetrating radar.

[0013] Preferably, the point cloud data is subjected to noise detection and denoising sequentially to obtain denoised point cloud data, specifically including:

[0014] Noise detection is performed on point cloud data using the deep learning model VoxelNet.

[0015] The detected noise points are separated from the point cloud data to generate denoised point cloud data.

[0016] Preferably, based on grid data and comprehensive exploration data, SKUA-GOCAD is used for modeling to obtain a 3D model of the mine to be modeled, specifically including:

[0017] Attribute models are constructed based on attribute information in grid data;

[0018] Three-dimensional models of strata are constructed based on stratigraphic information, three-dimensional models of tunnels are constructed based on tunnel information, three-dimensional models of geological anomalies are constructed based on geological anomaly information, three-dimensional models of boreholes are constructed based on borehole information, and three-dimensional models of boreholes are constructed based on well information.

[0019] The three-dimensional models of the stratigraphy, tunnels, geological anomalies, boreholes, and wells are determined as the basic three-dimensional models.

[0020] By integrating the correlation between grid data and exploration aggregate data, a stratigraphic body is generated;

[0021] The attribute model, the basic 3D model, and the formation volume are defined as the 3D model of the mine.

[0022] Preferably, constructing an attribute model based on attribute information in the grid data specifically includes:

[0023] Correct the grid data;

[0024] Convert the corrected grid data to .las format;

[0025] Import the converted .las format data into SKUA-GOCAD to build an attribute model.

[0026] Preferably, the grid data is corrected, specifically including:

[0027] In the process of modeling in SKUA-GOCAD, discrete smoothing interpolation is used to establish the stratigraphic interface and fault interface of the tunnel.

[0028] The grid data is corrected by stratigraphic sections and fault interfaces to simulate the spatial distribution, location, and contact relationships of structures; the structures are different types of interfaces; the structures include strata and geological anomalies.

[0029] Preferably, the 3D model of the mine is updated in real time using small geological anomalies underground and point cloud data after tunneling, specifically including:

[0030] During tunnel excavation, small geological anomalies are obtained underground by drilling ground radar, and the three-dimensional model of the mine is updated in real time based on the small geological anomalies.

[0031] After the tunneling process is completed, the newly generated tunnel is scanned using three-dimensional laser scanning technology to obtain the first .las file;

[0032] Based on the first .las file, a .txt file is generated using the VoxelNet deep learning algorithm, and the generated .txt algorithm is then converted into a second .las file;

[0033] Import the second .las file into SKUA-GOCAD to update the 3D model of the mine in real time.

[0034] This invention provides a device for constructing a three-dimensional model of a mine, comprising:

[0035] The acquisition module is used to acquire comprehensive exploration data of the mine to be modeled; the comprehensive exploration data includes stratigraphic information, roadway information, geological anomaly information, borehole information, and well information;

[0036] The scanning module is used to scan the roadways of the mine to be modeled, obtaining multiple point cloud data; each point cloud data includes three-dimensional coordinates and attribute information; the attribute information includes the roof, floor, and sidewalls;

[0037] The noise reduction module is used to perform noise detection and noise reduction on the point cloud data in sequence to obtain the noise-reduced point cloud data; the noise includes people in the alley, screws on the inner wall of the alley, and light tubes on the inner wall of the alley.

[0038] The partitioning and labeling module is used to divide the denoised point cloud data into multiple grids in three-dimensional space, and label each grid using a grid feature vector to obtain grid data; the grid feature vector includes three-dimensional coordinates, reflectivity, and variance; the variance is used to represent the density of points in a grid;

[0039] The modeling module is used to create a 3D model of the mine to be modeled using SKUA-GOCAD based on grid data and comprehensive exploration data.

[0040] The update module is used to update the 3D model of the mine in real time when new boreholes or roadways are exposed in the mine to be modeled, using underground small geological anomalies and point cloud data after tunneling; the underground small geological anomalies are detected by geophysical means including ground-penetrating radar.

[0041] The present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for constructing a three-dimensional model of a mine.

[0042] The present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-mentioned method for constructing a three-dimensional model of a mine.

[0043] The above-mentioned at least one technical solution adopted in this invention can achieve the following beneficial effects:

[0044] The mine roadways to be modeled are scanned to obtain multiple point cloud datasets. Noise detection and denoising are then performed on these point cloud datasets to obtain denoised point cloud data. Noise includes figures within the roadways, screws on the roadway walls, and light bulbs on the roadway walls. Denoising avoids noise interference, further improving modeling accuracy. The denoised point cloud data is divided into multiple grids, and each grid is labeled using a grid feature vector to obtain grid data. This gridding process transforms the point cloud data into structured data, providing multi-dimensional features for modeling. Based on the grid data and comprehensive exploration data, SKUA-GOCAD is used for modeling to obtain a 3D model of the mine. The 3D model is updated in real-time using small geological anomalies underground and post-excavation point cloud data. Utilizing the multi-source data fusion capability of SKUA-GOCAD software, the gridded point cloud data is combined with geophysical data to generate a high-precision 3D mine model. The real-time update mechanism ensures that the 3D mine model remains consistent with the actual state of the roadways over the long term. This method improves the modeling accuracy of the 3D mine model. Attached Figure Description

[0045] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0046] Figure 1 A schematic diagram of a method for constructing a three-dimensional model of a mine, provided by the present invention;

[0047] Figure 2 The present invention provides a flowchart of VoxelNet data denoising process.

[0048] Figure 3A flowchart for three-dimensional mine modeling provided by this invention;

[0049] Figure 4 A schematic diagram of a device for constructing a three-dimensional model of a mine provided by the present invention;

[0050] Figure 5 A schematic diagram of a computer device for constructing a three-dimensional model of a mine, as provided by the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0052] The server mentioned in this invention can be a server set up on a business platform, or a device such as a desktop computer or laptop computer capable of executing the solution of this invention. For ease of explanation, the following description will only focus on the server as the executing entity.

[0053] Traditional 3D mine modeling relies on manual cross-section stitching or interpolation, which is inefficient, accuracy depends on experience, and model updates are difficult. The models are essentially static; adding new boreholes or tunnels requires modification or even reconstruction. Modeling complex mining areas with intersecting faults, folds, and ore body morphologies is challenging (e.g., using AutoCAD for 3D mine modeling). Traditional surveying methods (cross-section method, total station surveying) have limited accuracy and low data acquisition efficiency, making them unsuitable for real-time monitoring and dynamic modeling.

[0054] The SKUA-GOCAD modeling process uses Discrete Smooth Interpolation.

[0055] The 3D laser scanning (DSI) method is used to establish stratigraphic and fault interfaces, correct modeling data, and simulate the spatial distribution, location, and contact relationships of structures. It possesses the capability for 3D modeling and dynamic updating of complex geological bodies and tunnels. 3D laser scanning can efficiently acquire point clouds of the entire tunnel cross-section with millimeter-level accuracy. Traditional measurement methods are inefficient and inaccurate; total stations or cross-sectional measurements can only obtain a limited number of points, resulting in low modeling accuracy, incomplete information, high workload, slow speed, and significant difficulty. Traditional 3D modeling methods are lagging and static, often remaining at the geometric model stage, lacking attribute information, and unable to quickly perform 3D geological modeling of mines.

[0056] To address the aforementioned challenges, a rapid 3D geological modeling method for mines is proposed. Its purpose is to quickly acquire the spatial coordinates of tunnels using 3D laser scanning, and then, combined with SKUA-GOCAD, rapidly construct realistic and detailed 3D tunnel models. SKUA-GOCAD accepts various input and output data formats, such as DXF, Acview shape, XYZ, Column-based File, CPS3, LAS, SEG-Y, and VIP file formats, allowing for the application of extensive prior research findings to 3D modeling. Simultaneously, it supports EPOS database management for regional data, facilitating data sharing with other platforms and enabling rapid updates to the 3D model with new data, ensuring consistency between the model and the actual site conditions. This provides reliable 3D spatial data support for mine safety monitoring, production scheduling, disaster early warning, support design, and ventilation system optimization.

[0057] A realistic 3D model of the tunnel was established to provide a scientific basis for safety risk monitoring, support optimization, and ventilation and transportation design. High-density 3D point clouds generated by 3D laser scanning were combined with SKUA-GOCAD to achieve dynamic modeling, breaking through the limitations of traditional static models. Attributes such as lithology, grade, physical properties, and stress fields can be superimposed on the geometry, realizing for the first time a complete technical chain from data acquisition to real-time modeling to dynamic updates.

[0058] This invention proposes a combined method of 3D laser scanning and SKUA-GOCAD to replace the traditional "manual measurement + static modeling" process.

[0059] The technical solutions provided by the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0060] Figure 1 This is a schematic diagram of a method for constructing a three-dimensional model of a mine according to the present invention, which specifically includes the following steps:

[0061] S101: Obtain comprehensive exploration data for the mine to be modeled; comprehensive exploration data includes stratigraphic information, roadway information, geological anomaly information, borehole information, and well information.

[0062] Comprehensive exploration data of the mine to be modeled is obtained through drilling or 3D laser scanning technology.

[0063] S102: Scan the roadways of the mine to be modeled to obtain multiple point cloud data; each point cloud data includes three-dimensional coordinates and attribute information; the attribute information includes roof, floor and sidewalls.

[0064] The mine to be modeled was subjected to 3D laser scanning to obtain multiple point cloud data.

[0065] S103: Perform noise detection and denoising on the point cloud data in sequence to obtain the denoised point cloud data; the noise includes people in the alley, screws on the inner wall of the alley, and light tubes on the inner wall of the alley.

[0066] In an exemplary embodiment, noise detection and denoising are performed sequentially on the point cloud data to obtain denoised point cloud data. Specifically, this includes: performing noise detection on the point cloud data using a deep learning model called End-to-End Learning for Point Cloud Based 3D Object Detection (VoxelNet); separating the detected noise from the point cloud data to generate denoised point cloud data.

[0067] Specifically, the present invention provides, as follows Figure 2 The flowchart shown illustrates the VoxelNet data denoising process. It involves selecting drilling locations, borehole data, seismic data volumes, fault and stratigraphic data, and organizing the data coordinates. A 3D laser scanner is used to comprehensively scan the mine roadway, acquiring high-density point cloud data. Each point cloud data point contains 3D coordinates (X, Y, Z) and other attribute information. The scanning process obtains full-section data of the entire roadway, including information on the roof, floor, and sidewalls, accurately recording details such as the roadway's morphology, deformation, and cracks.

[0068] The VoxelNet deep learning model was used to detect and denoise 3D laser scanning point cloud data. Noise points were defined as people in the alleyway and screws and pipes attached to the inner wall. After the VoxelNet network detected the noise points, the data was then separated.

[0069] VoxelNet is a deep learning model / algorithm. First, a dataset is created using the point cloud annotation tool SemanticSegmentation Editor, annotating information such as people and pipes, before being fed into VoxelNet for training.

[0070] First, semantic segmentation and annotation are performed on the point cloud data (scattered points obtained from scanning) using annotation tools, such as distinguishing categories like "person" and "pipe." VoxelNet is used to divide the 3D space containing the point cloud into many small cubes (grids) so that the disordered points are assigned to corresponding voxels. PointNet is then used to extract features from the grid data. Each voxel contains many points; PointNet extracts features from the disordered points and represents each voxel with a feature vector. Each voxel is no longer a collection of points but a feature vector. These feature vectors are concatenated according to the original 3D positions of the voxels to obtain a regularly arranged 3D feature grid. 3D convolution is used to extract ordered 3D Voxel data. The generated data is in .txt file format, which is then uniformly converted to .las files.

[0071] S104: Divide the denoised point cloud data into multiple grids in three-dimensional space, and label each grid with a grid feature vector to obtain grid data; the grid feature vector includes three-dimensional coordinates, reflectivity and variance; variance is used to represent the density of points in a grid.

[0072] Specifically, Voxel in VoxelNet is used to divide a certain space in 3D space into a grid, and PointNet network is used to extract features from the grid data. The extracted features are used to represent the grid and placed back into 3D space. The unordered point cloud data is transformed into high-dimensional feature data, coordinate data is extracted, and a .las file is generated. Multi-source data are integrated into a unified coordinate system, and the file format is unified as .las.

[0073] S105: Based on grid data and comprehensive exploration data, a 3D model of the mine to be modeled is obtained by using SKUA-GOCAD.

[0074] In an exemplary embodiment, a 3D model of the mine to be modeled is obtained by modeling using SKUA-GOCAD based on grid data and exploration integrated data. Specifically, this includes: constructing an attribute model based on attribute information in the grid data; constructing a 3D stratigraphic model based on stratigraphic information, a 3D roadway model based on roadway information, a 3D geological anomaly model based on geological anomaly information, a 3D borehole model based on borehole information, and a 3D well model based on borehole information; determining the stratigraphic model, roadway model, geological anomaly model, borehole model, and borehole model as the basic 3D model; generating a stratigraphic body by integrating the correlation between the grid data and exploration integrated data; and determining the attribute model, basic 3D model, and stratigraphic body as the mine 3D model.

[0075] In one exemplary embodiment, constructing an attribute model based on attribute information in the grid data specifically includes: correcting the grid data; converting the corrected grid data into .las format data; and importing the converted .las format data into SKUA-GOCAD to establish the attribute model.

[0076] In an exemplary embodiment, the grid data is corrected by: using discrete smooth interpolation to establish the stratigraphic and fault interfaces of the tunnel during the modeling process in SKUA-GOCAD; correcting the grid data through stratigraphic sections and fault interfaces to simulate the spatial distribution, location, and contact relationships of the structure; the structure consists of different types of interfaces; the structure includes strata and geological anomalies.

[0077] Specifically, generating stratigraphic volumes is a built-in function of skua-gocad. After processing and interpreting all the data through the Workflow platform, it is necessary to fully consider the correlation between all the data, mainly the mutual tangency between ground planes, and then begin to materialize the strata to generate stratigraphic volumes. The basic principle of generating stratigraphic volumes is that the area enclosed by the top and bottom interfaces and boundaries constitutes a stratigraphic entity, i.e., a geological body. The specific operation process is as follows: Enter the Structure & Stratigraphy workbench → Create a stratigraphic column → Define faults → Create 2D horizontal model boundary (select model boundary) → Building the Fault Network → Building the Geologic Grid → Create a geological grid, set the grid size, thickness, number, etc. → Preview Geologic Grid → Build Geologic Grid. The stratigraphic volume construction is complete.

[0078] Specifically, the present invention provides, as follows Figure 3The flowchart shown illustrates the 3D mine modeling process. Processed data is imported into SKUA-GOCAD. The modeling process uses Discrete Smooth Interpolation (DSI) to establish stratigraphic and fault interfaces, correcting the modeling data and simulating the spatial distribution, location, and contact relationships of structures. Based on stratigraphic, roadway, geological anomaly, borehole, and well information, 3D models of the stratigraphic, roadway, geological anomaly, borehole, and well are established. Attribute data is included in the well curve data and is directly imported into SKUA-GOCAD using the .las format exported from ArcGIS to create attribute models. After processing and interpreting all the above data through the Workflow platform, the relationships between all data points need to be fully considered, primarily the intersecting relationships between stratigraphic layers. Then, the stratigraphic bodies are solidified to generate the stratigraphic volume.

[0079] S106: When new boreholes or roadways are exposed in the mine to be modeled, the 3D model of the mine is updated in real time using underground small geological anomalies and point cloud data after excavation; the underground small geological anomalies are detected by geophysical means including ground-penetrating radar.

[0080] In an exemplary embodiment, the 3D model of the mine is updated in real time using underground small geological anomalies and point cloud data after tunneling. Specifically, this includes: during tunnel excavation, underground small geological anomalies are acquired using borehole ground-penetrating radar, and the 3D model of the mine is updated in real time using these anomalies; after the tunneling process is completed, the newly generated tunnel is scanned using 3D laser scanning technology to obtain a first .las file; based on the first .las file, a .txt file is generated using the VoxelNet deep learning algorithm, and the generated .txt file is converted into a second .las file; the second .las file is imported into SKUA-GOCAD for real-time updating of the 3D model of the mine.

[0081] Verification of small geological anomalies in underground mines is often done using borehole ground-penetrating radar. Large-scale geophysical exploration in the early stages is not suitable for three-dimensional modeling of small geological anomalies in underground mines. Based on the verification by borehole ground-penetrating radar, the structure of small geological anomalies in underground mines ahead of the tunnel face is updated in real time. After the tunneling is completed, the scale of the underground roadway is updated in real time using three-dimensional laser scanning technology.

[0082] The 3D laser scanning + SKUA-GOCAD technology uses a 3D laser scanner to accurately acquire the 3D coordinates of the tunnel with millimeter-level precision. At the same time, geological information can be acquired in real time and dynamically updated and modeled in real time with SKUA-GOCAD. The whole process greatly improves the modeling speed, basically achieving instant modeling. It greatly reduces the time for data processing and model updates in traditional methods, and also reduces human error and the time and cost of later modifications.

[0083] When applying the method for constructing a three-dimensional mine model provided by this invention, it is not necessary to follow the... Figure 1 The steps shown are executed in sequence. The specific execution order of each step can be determined as needed, and this invention does not impose any restrictions on it.

[0084] The above describes a method for constructing a three-dimensional mine model according to one or more embodiments of the present invention. Based on the same idea, the present invention also provides a corresponding apparatus for constructing a three-dimensional mine model, such as... Figure 4 As shown.

[0085] Figure 4 A schematic diagram of a device for constructing a three-dimensional model of a mine provided by the present invention includes:

[0086] The acquisition module 401 is used to acquire the comprehensive exploration data of the mine to be modeled; the comprehensive exploration data includes stratigraphic information, roadway information, geological anomaly information, borehole information and well information.

[0087] The scanning module 402 is used to scan the roadways of the mine to be modeled to obtain multiple point cloud data; each point cloud data includes three-dimensional coordinates and attribute information; the attribute information includes the roof, floor and sidewalls.

[0088] The noise reduction module 403 is used to perform noise detection and noise reduction on the point cloud data in sequence to obtain the noise-reduced point cloud data; the noise includes people in the alley, screws on the inner wall of the alley, and light tubes on the inner wall of the alley.

[0089] The partitioning and labeling module 404 is used to divide the denoised point cloud data into multiple grids in three-dimensional space, and label each grid through a grid feature vector to obtain grid data; the grid feature vector includes three-dimensional coordinates, reflectivity and variance; the variance is used to represent the density of points in a grid.

[0090] Modeling module 405 is used to perform modeling based on grid data and exploration integrated data using SKUA-GOCAD to obtain a 3D model of the mine to be modeled.

[0091] The update module 406 is used to update the 3D model of the mine in real time when new boreholes or roadways are exposed in the mine to be modeled, using underground small geological anomalies and point cloud data after tunneling; the underground small geological anomalies are detected by geophysical exploration methods including ground-penetrating radar.

[0092] Specific limitations regarding the device for constructing the 3D mine model can be found in the limitations on the construction method of the 3D mine model mentioned above, and will not be repeated here. Each module in the aforementioned device for constructing the 3D mine model can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0093] The present invention also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 The provided method for constructing a 3D model of a mine.

[0094] The present invention also provides Figure 5 The schematic diagram of the computer device shown is as follows: Figure 5 As shown, at the hardware level, this computer device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for business operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then executes it to achieve the above. Figure 1 The provided method for constructing a 3D model of a mine.

[0095] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0096] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this invention.

Claims

1. A method for constructing a three-dimensional model of a mine, characterized in that, include: Obtain comprehensive exploration data for the mine to be modeled; The comprehensive exploration data includes stratigraphic information, tunnel information, geological anomaly information, borehole information, and well information; The roadways of the mine to be modeled are scanned to obtain multiple point cloud data; each point cloud data includes three-dimensional coordinates and attribute information; the attribute information includes roof, floor and sidewalls; The point cloud data is subjected to noise detection and denoising in sequence to obtain denoised point cloud data; the noise includes people in the alley, screws on the inner wall of the alley, and light tubes on the inner wall of the alley. The denoised point cloud data is divided into multiple grids in three-dimensional space, and each grid is labeled using a grid feature vector to obtain grid data; the grid feature vector includes three-dimensional coordinates, reflectivity, and variance. Variance is used to represent the density of points in a grid; Based on the grid data and the comprehensive exploration data, a three-dimensional model of the mine to be modeled is obtained by using SKUA-GOCAD. When new boreholes or roadways are exposed in the mine to be modeled, the 3D model of the mine is updated in real time using small underground geological anomalies and point cloud data after excavation; the small underground geological anomalies are detected by geophysical means including ground-penetrating radar. During the tunnel excavation process, small geological anomalies are obtained underground using borehole ground-penetrating radar, and the mine's 3D model is updated in real time based on these anomalies. After the excavation process is completed, the newly generated tunnel is scanned using 3D laser scanning technology to obtain the first .las file. Based on the first .las file, a .txt file is generated using the VoxelNet deep learning algorithm, and the generated .txt algorithm is converted into a second .las file; Import the second .las file into SKUA-GOCAD to update the 3D model of the mine in real time.

2. The method as described in claim 1, characterized in that, The step of sequentially performing noise detection and denoising on the point cloud data to obtain denoised point cloud data specifically includes: Noise detection was performed on the point cloud data using the deep learning model VoxelNet. The detected noise points are separated from the point cloud data to generate denoised point cloud data.

3. The method as described in claim 1, characterized in that, Based on the grid data and the comprehensive exploration data, a 3D model of the mine to be modeled is obtained using SKUA-GOCAD, specifically including: An attribute model is constructed based on the attribute information in the grid data; A three-dimensional model of the stratigraphy is constructed based on the stratigraphic information; a three-dimensional model of the tunnel is constructed based on the tunnel information; a three-dimensional model of the geological anomaly is constructed based on the geological anomaly information; a three-dimensional model of the borehole is constructed based on the borehole information; and a three-dimensional model of the well is constructed based on the well drilling information. The three-dimensional models of the strata, the tunnels, the geological anomalies, the boreholes, and the wells are determined as the basic three-dimensional models. By combining the correlation between the grid data and the exploration aggregate data, a stratigraphic body is generated; The attribute model, the basic three-dimensional model, and the formation volume are determined as the three-dimensional model of the mine.

4. The method as described in claim 3, characterized in that, The construction of the attribute model based on the attribute information in the grid data specifically includes: The grid data is corrected; Convert the corrected grid data to .las format; Import the converted .las format data into SKUA-GOCAD to build the attribute model.

5. The method as described in claim 4, characterized in that, The correction of the grid data specifically includes: In the process of modeling in SKUA-GOCAD, discrete smoothing interpolation is used to establish the stratigraphic interface and fault interface of the tunnel. The grid data is corrected by stratigraphic sections and fault interfaces to simulate the spatial distribution, location, and contact relationships of structures; the structures are different types of interfaces; the structures include strata and geological anomalies.

6. A device for constructing a three-dimensional model of a mine, characterized in that, include: The acquisition module is used to acquire comprehensive exploration data of the mine to be modeled; The comprehensive exploration data includes stratigraphic information, tunnel information, geological anomaly information, borehole information, and well information; The scanning module is used to scan the roadways of the mine to be modeled, obtaining multiple point cloud data; each point cloud data includes three-dimensional coordinates and attribute information; the attribute information includes the roof, floor, and sidewalls; The noise reduction module is used to sequentially detect and denoise the point cloud data to obtain denoised point cloud data; the noise includes people in the alley, screws on the inner wall of the alley, and light tubes on the inner wall of the alley. The partitioning and labeling module is used to divide the denoised point cloud data into multiple grids in three-dimensional space, and label each grid using a grid feature vector to obtain grid data; the grid feature vector includes three-dimensional coordinates, reflectivity, and variance; Variance is used to represent the density of points in a grid; The modeling module is used to perform modeling based on the grid data and the exploration integrated data using SKUA-GOCAD to obtain a three-dimensional model of the mine to be modeled. The update module is used to update the 3D model of the mine in real time when new boreholes or roadways are exposed in the mine to be modeled, using underground small geological anomalies and point cloud data after tunneling; the underground small geological anomalies are detected by geophysical means including ground-penetrating radar. During the tunnel excavation process, small geological anomalies are obtained underground using borehole ground-penetrating radar, and the mine's 3D model is updated in real time based on these anomalies. After the excavation process is completed, the newly generated tunnel is scanned using 3D laser scanning technology to obtain the first .las file. Based on the first .las file, a .txt file is generated using the VoxelNet deep learning algorithm, and the generated .txt algorithm is converted into a second .las file; Import the second .las file into SKUA-GOCAD to update the 3D model of the mine in real time.

7. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method as described in any one of claims 1 to 5.

8. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method as described in any one of claims 1 to 5.

Images