A method and apparatus for three-dimensional modeling of ore bodies

Abstract

The application discloses a kind of ore body three-dimensional modeling method and device, and the application relates to three-dimensional modeling technical field.The application includes the following steps: obtaining the geometric characteristic parameters of the ore body to be monitored, constructing a three-dimensional boundary model, and dividing the surface into several equal-sized grids;Collecting point cloud data of the ore body terrain surface and mapping it into the grid, analyzing the point cloud data by different modeling methods, generating multiple three-dimensional ore body structure models and optimizing the model set;Coupled analysis of its structure complexity, select information comparison grid;Calculate the cumulative difference between the point cloud surface features and the features of each model, and select the model with the smallest cumulative difference as the target ore body three-dimensional model;Determine the construction impact range as the monitoring area by combining real-time mining location information, obtain the vibration information and regional point cloud data of the monitoring area, dynamically update the local structure of the target ore body model, realize real-time monitoring and dynamic management of the ore body, and improve the safety and efficiency of mining.

Description

Technical Field

[0001] This invention relates to the field of three-dimensional modeling technology, specifically to a method and apparatus for three-dimensional modeling of ore bodies. Background Technology

[0002] In the process of ore body analysis and mining, obtaining the geometric structural characteristics of the ore body is a key step to ensure efficient mining and resource utilization. Traditional ore body analysis methods mainly rely on drilling and geophysical exploration. Although these methods can obtain the basic geometric parameters of the ore body, they often fail to fully reflect the complex structure and actual distribution of the ore body, leading to certain limitations in subsequent modeling and mining decisions.

[0003] In existing technologies, traditional methods primarily rely on borehole information, geophysical data, and geological modeling tools, which provide the foundation for modeling orebody geometry and making mining decisions. However, with the increasing complexity of orebody geometry and the dynamic changes in mining demands, existing traditional technologies have certain limitations and shortcomings. For example, they cannot effectively reflect the structural complexity of the orebody surface, thus failing to provide a basis for comparing information from different areas, resulting in low accuracy of the model and significant errors in subsequent analyses. Furthermore, during the mining phase, existing technologies lack solutions that incorporate real-time location information. By monitoring the impact range of mining operations, the 3D model of the orebody can be dynamically updated to ensure that its structure reflects changes in the mining environment in real time.

[0004] In the prior art, CN117422828A discloses a method and system for 3D modeling of ore body segments. This method involves: calculating the centroid coordinates of the outlines of adjacent ore body segments; projecting and aligning the centroids of the adjacent ore body segments; scaling the outlines using the centroid coordinates of the adjacent ore body segments as the center; densifying the scaled outlines; restoring the coordinates of the densified outlines; and synchronously advancing the perimeter to connect the structural surfaces. This prior art achieves ore body segment connection through geometric centroid projection alignment and synchronous perimeter advancement, solving common connection misalignment and intersection problems. It can achieve smooth connection of arbitrary outlines between parallel profiles, thus quickly constructing a 3D ore body model that meets objective expectations. However, this scheme cannot detect and maintain the model's accuracy in complex terrain areas, and it does not address the dynamic update process during mining, thus reducing the real-time performance and effectiveness of the model.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a method and apparatus for three-dimensional modeling of ore bodies to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A method for three-dimensional modeling of ore bodies, comprising the following steps:

[0009] Obtain the geometric feature parameters of the ore body to be monitored, construct a three-dimensional boundary model of the ore body based on the geometric parameters, and divide the surface area of ​​the three-dimensional boundary model of the ore body into several grids of the same size;

[0010] Point cloud data of the topographic surface of the ore body to be monitored is collected, and the point cloud distribution of the topographic surface is mapped to the corresponding grid. Based on different surface reconstruction algorithms and geometric optimization algorithms, the mapped point cloud data is analyzed to obtain several three-dimensional ore body structure models and form a set of preferred ore body three-dimensional models.

[0011] The terrain structure parameters of the grid are determined based on the point cloud distribution within the grid. The terrain structure parameters are coupled and analyzed to determine the structural complexity of the grid. Information comparison grids are selected based on the structural complexity.

[0012] The surface structure features of the information comparison grid are determined based on point cloud data, and the cumulative difference between the grid and the corresponding surface structure features in each three-dimensional ore body structure model is calculated. The three-dimensional ore body structure model corresponding to the minimum cumulative difference is taken as the target ore body three-dimensional model.

[0013] Using the real-time location information of the ore body to be monitored, the impact range of mining operations is determined. The impact range of mining operations is used as the monitoring area. Vibration information of the monitoring area is obtained, and the vibration update area of ​​the target ore body's three-dimensional model is determined. Point cloud data of the vibration update area is collected again. Based on the point cloud data of the vibration update area, the structure of the vibration update area corresponding to the target ore body's three-dimensional model is dynamically updated to complete the establishment of the ore body's three-dimensional model.

[0014] Furthermore, the geometric feature parameters of the ore body specifically include the size information and boundary morphology information of the ore body to be monitored. The size information includes the length, width and height of the ore body, and the boundary morphology information includes the top and bottom boundaries and boundary outlines of the ore body to be monitored.

[0015] The method for constructing a three-dimensional boundary model of the ore body is as follows: select three-dimensional modeling software, input the geometric feature parameters into the three-dimensional modeling software, and construct a three-dimensional boundary model of the ore body;

[0016] The surface area of ​​the ore body's three-dimensional boundary model is divided into grids of the same size. The specific steps of determining the grid division scheme include: reading the boundary data of different view faces of the ore body; determining the division range based on the minimum bounding box of the different view faces of the ore body's three-dimensional model; setting the area size of the grid; determining the number of grid nodes in each direction based on the division range of the minimum bounding box of the different view faces and the grid size; and completing the grid division of the surface area based on the number of grid nodes.

[0017] Furthermore, the surface of the ore body is scanned multiple times using a point cloud acquisition device with pre-set equipment parameters to obtain point cloud data of the surface of the ore body to be monitored. The equipment parameters include resolution, scanning range and angle. The point cloud data of the surface of the ore body to be monitored obtained from each scan is denoised. The point cloud data of the terrain obtained after denoising is registered. The point clouds obtained from different scans are registered to form a unified point cloud dataset.

[0018] The logic for constructing the preferred ore body 3D model set is as follows: Multiple surface reconstruction algorithms are used to analyze the mapped point cloud data one by one, generating multiple preliminary 3D surface models. Different geometric optimization methods are then used to optimize each preliminary 3D surface model, resulting in several 3D ore body structure models, forming the preferred ore body 3D model set. The surface reconstruction algorithms include Delaunay triangulation, Alpha shape algorithm, least squares surface fitting, Poisson surface reconstruction, and MarchingCubes algorithm. The geometric optimization methods include smoothing optimization, vertex simplification optimization, resampling optimization, and boundary reconstruction optimization.

[0019] Furthermore, the topographic structure parameters within each grid on the ore body surface are characterized by the point cloud distribution of different grids. These topographic structure parameters include surface curvature, undulation roughness, and point cloud density. The surface curvature is characterized by point cloud data coordinates, based on the following formula:

[0020] ;

[0021] In the formula, Let be the surface curvature of the i-th mesh. Let be the structural curvature of the p-th point within the i-th grid, where p is the index of the point within the grid. , This represents the total number of points within the grid.

[0022] The specific formula used to calculate the surface roughness is as follows:

[0023] ;

[0024] In the formula, Let p be the vertical coordinate of the p-th point. Let be the roughness of the i-th grid. It is the average value of the vertical coordinates of all points in the i-th grid.

[0025] Furthermore, the formula used to characterize the structural complexity within different grids on the ore body surface is as follows:

[0026] ;

[0027] In the formula, The structural complexity of the i-th grid is... This represents the normalized value of the surface curvature of the i-th mesh. This represents the normalized value of the roughness of the i-th mesh undulation. This is the normalized value of the point cloud density of the i-th grid.

[0028] Information comparison grids are selected based on structural complexity. The specific logic for selecting these grids is as follows: a structural complexity threshold is set, and the structural complexity is compared to this threshold. Based on the comparison results, the grids are then selected.

[0029] When the structural complexity When it is determined that the terrain change of the grid does not require monitoring, the grid is not recorded as an information comparison grid.

[0030] Structural complexity When the terrain of a grid is deemed to be complex, it is designated as an information comparison grid. For structurally complex thresholds.

[0031] Furthermore, the surface structure features include the average rate of change of the normal and the average vertical coordinate value of the target area. The formula used to calculate the rate of change of the normal based on point cloud data is:

[0032] ;

[0033] In the formula, Let be the average normal vector of the q-th target region within the j-th information comparison grid. Let be the average normal vector of the h-th corresponding neighborhood of the q-th target region. This represents the magnitude of the average normal vector of the q-th target region. Let be the average normal vector magnitude of the h-th corresponding neighborhood of the q-th target region. The average rate of change of the normal to the q-th target region within the grid is compared to the j-th piece of information.

[0034] Where q is the index of the randomly selected target region within the information comparison grid, j is the index of the information comparison grid, and h is the neighborhood index of the target region;

[0035] In the 3D ore body structure model, the average normal variation rate and average vertical coordinate value of the corresponding region are analyzed, their differences are calculated, and the differences of all randomly selected target regions are accumulated. The specific formula used for the calculation is as follows:

[0036] ;

[0037] In the formula, To accumulate the differences, , These represent the average rate of change of normal and the average vertical coordinate value of the region corresponding to the q-th target region within the grid compared to the j-th information in the 3D ore body structure model. Let be the average vertical coordinate value of the q-th target region within the j-th information comparison grid. The total number of information comparison grids;

[0038] The three-dimensional ore body structure model with the smallest cumulative difference is selected as the target ore body three-dimensional model.

[0039] Furthermore, the real-time location information of the ore body to be mined is analyzed to determine the impact range of the mining operations. The specific formula used to calculate the impact range of the mining operations is as follows:

[0040] ;

[0041] In the formula, Due to the distance affected by mining operations, To affect the distance reference value, For mining depth, The factor representing the influence of mining depth is... The influence coefficient of soil and rock strength. To measure the compressive strength of the soil and rock in the ore body to be monitored, This is a reference value for the compressive strength of soil and rock.

[0042] Using the current mining and construction site as the center, and the distance affected by the mining and construction as the radius... Using a radius, the impact range of mining operations is determined. Vibration monitoring units are set up in the grid within the impact range to monitor regional vibration information, including vibration frequency and vibration amplitude.

[0043] The vibration update region of the three-dimensional model of the target ore body is determined based on the following logic:

[0044] If the vibration frequency And vibration amplitude When this happens, it is determined that the grid does not need to be updated;

[0045] Conversely, the grid is dynamically updated to the vibration update region of the target ore body's 3D model;

[0046] By using the surface structure features of the vibration update area, the structure of the target ore body's 3D model is dynamically updated. Specifically, the point cloud data of the vibration update area is collected again, and the structure of the vibration update area is dynamically updated using the model construction method corresponding to the target ore body's 3D model, thereby achieving local updates to the structure of the target ore body's 3D model.

[0047] The present invention also provides a three-dimensional ore body modeling device, which is used to perform the above-described three-dimensional ore body modeling method, comprising:

[0048] The terrain division and analysis module is used to obtain the geometric feature parameters of the ore body to be monitored, construct a three-dimensional boundary model of the ore body based on the geometric parameters, and divide the surface area of ​​the three-dimensional boundary model of the ore body into several grids of the same size.

[0049] The surface feature mapping module is used to collect point cloud data of the topographic surface of the ore body to be monitored, and map the point cloud distribution of the topographic surface to the corresponding grid. Based on different surface reconstruction algorithms and geometric optimization algorithms, the mapped point cloud data is analyzed to obtain several three-dimensional ore body structure models, forming a set of preferred ore body three-dimensional models.

[0050] The fine analysis and comparison module is used to determine the terrain structure parameters of the grid based on the point cloud distribution within the grid, coupled analysis of the terrain structure parameters to determine the structural complexity of the grid, and filter out information comparison grids based on the structural complexity.

[0051] The 3D model building module is used to determine the surface structure features of the information comparison grid based on point cloud data, and calculate the cumulative difference between the grid and the corresponding surface structure features in each 3D ore body structure model. The 3D ore body structure model corresponding to the minimum cumulative difference is taken as the target ore body 3D model.

[0052] The dynamic update and correction module is used to determine the impact range of mining operations based on the real-time location information of the ore body to be monitored, take the impact range of mining operations as the monitoring area, acquire vibration information of the monitoring area, determine the vibration update area of ​​the target ore body 3D model, collect point cloud data of the vibration update area again, and dynamically update the vibration update area structure of the target ore body 3D model based on the point cloud data of the vibration update area to complete the establishment of the ore body 3D model.

[0053] Compared with the prior art, the beneficial effects of the present invention are:

[0054] First, by acquiring the geometric feature parameters of the ore body to be monitored, a three-dimensional boundary model is constructed, which can comprehensively reflect the morphology and structural characteristics of the ore body. By combining multiple partitioning schemes and minimizing the area of ​​non-mesh regions, the mesh partitioning scheme is effectively optimized, ensuring the accuracy of the model in detail processing and enabling a more realistic representation of the complex geometry of the ore body.

[0055] Furthermore, this scheme obtains the structural complexity within different grids on the ore body surface by coupling surface structure parameters. This feature allows for the selection of representative grids during model building, forming an information comparison region. This information comparison region ensures the accuracy of the model, guaranteeing the applicability and effectiveness of the selected model in actual mining, while also saving computational resources.

[0056] Finally, the solution includes the acquisition of real-time location information and its dynamic updating of the model. By monitoring vibration information within the area, the structural features of the target ore body's 3D model are adjusted in a timely manner, achieving the goal of real-time monitoring and dynamic management of the ore body, thus improving mining safety and efficiency. Attached Figure Description

[0057] Figure 1 This is a schematic diagram of the overall method flow of the present invention;

[0058] Figure 2 Mapping diagrams of structural complexity for different grids;

[0059] Figure 3 A bar chart comparing the rate of change of normals in the model and the point cloud mesh region;

[0060] Figure 4 This is a schematic diagram of the overall device structure of the present invention. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0062] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0063] Example:

[0064] Please see Figure 1 The present invention provides a technical solution:

[0065] A method for three-dimensional modeling of ore bodies, comprising the following steps:

[0066] Step 1: Obtain the geometric feature parameters of the ore body to be monitored, construct a three-dimensional boundary model of the ore body based on the geometric parameters, and divide the surface area of ​​the three-dimensional boundary model of the ore body into several grids of the same size.

[0067] The geometric feature parameters of the ore body specifically include the size information and boundary morphology information of the ore body to be monitored. The size information includes the length, width and height of the ore body, and the boundary morphology information includes the top and bottom boundaries and the boundary outline of the ore body to be monitored.

[0068] The dimensional information of an ore body is mainly used to describe its spatial extent and size, including its length, width, and height. These parameters can be obtained through methods such as borehole surveying. By laying boreholes, the depth, thickness, and boundary crossing points of the ore body can be obtained, directly acquiring the dimensional information of the underground ore body. This method is particularly suitable for measuring the thickness (height) of the ore body, but borehole layout is limited by cost and geological complexity.

[0069] Geological mapping, through geological mapping of the earth's surface, determines the surface distribution range of the ore body and is suitable for obtaining surface size information of the ore body; 3D scanning and remote sensing technology, using technologies such as laser scanning (LiDAR) and UAV remote sensing, obtains three-dimensional topographic data of the ore body surface, and performs laser scanning or UAV image acquisition on the ore body surface area. It has high resolution and can quickly obtain the surface range of the ore body.

[0070] The boundary morphology information of an ore body refers to its surface shape, boundary contour, and specific distribution characteristics of the top and bottom surfaces. This information can be obtained through the following methods: Borehole data is used not only to obtain the ore body's dimensions but also to infer its boundary morphology information. The starting and ending points of borehole penetrations into the ore body, such as the intersection of the top and bottom surfaces, are extracted. Data from multiple borehole points are combined to form the boundary point cloud data of the ore body. Open-pit ore body surface scanning and analysis methods utilize LiDAR or UAV technology to directly scan the ore body surface, generating the actual morphology of the ore body's top surface.

[0071] The method for constructing a three-dimensional boundary model of the ore body is as follows: select three-dimensional modeling software, input the geometric feature parameters into the three-dimensional modeling software, and construct a three-dimensional boundary model of the ore body;

[0072] Constructing a 3D boundary model of an ore body requires specialized modeling software. The following are commonly used 3D geological modeling software: Leapfrog, specifically designed for geological modeling, supports the rapid generation of 3D ore body models; Surpac, providing powerful 3D visualization and modeling capabilities, supporting mine design, resource estimation, etc.; [other software names omitted] can directly process borehole data and geological profiles to generate wireframe or mesh models of ore bodies; Micromine and Datamine are also included.

[0073] Import the geometric feature parameters into the selected software, verify the imported data and ensure its spatial consistency, extract the depth points of the top and bottom surfaces of the ore body into point cloud data, extract the contour points of the ore body boundary from the surface measurement data, fit the boundary contour points of the ore body into a smooth contour line, combine the top and bottom surfaces with the boundary contour line to construct the wireframe model of the ore body, and form the three-dimensional boundary model of the ore body.

[0074] The surface area of ​​the ore body's three-dimensional boundary model is divided into grids of the same size. The specific steps of determining the grid division scheme include: reading the boundary data of different view faces of the ore body; determining the division range based on the minimum bounding box of the different view faces of the ore body's three-dimensional model; setting the area size of the grid; determining the number of grid nodes in each direction based on the division range of the minimum bounding box of the different view faces and the grid size; and completing the grid division of the surface area based on the number of grid nodes.

[0075] Step 2: Collect point cloud data of the topographic surface of the ore body to be monitored, and map the point cloud distribution of the topographic surface to the corresponding grid. Analyze the mapped point cloud data based on different surface reconstruction algorithms and geometric optimization algorithms to obtain several three-dimensional ore body structure models and form a set of preferred three-dimensional ore body models.

[0076] The surface of the ore body is scanned multiple times using a point cloud acquisition device with pre-set parameters to obtain point cloud data of the surface of the ore body to be monitored. The device parameters include resolution, scanning range and angle. The point cloud data of the surface of the ore body to be monitored obtained from each scan is denoised. The point cloud data of the terrain after denoising is registered. The point clouds obtained from different scans are registered to form a unified point cloud dataset.

[0077] The logic for constructing the preferred ore body 3D model set is as follows: Multiple surface reconstruction algorithms are used to analyze the mapped point cloud data one by one, generating multiple preliminary 3D surface models. Different geometric optimization methods are then used to optimize each preliminary 3D surface model, resulting in several 3D ore body structure models, forming the preferred ore body 3D model set. The surface reconstruction algorithms include Delaunay triangulation, Alpha shape algorithm, least squares surface fitting, Poisson surface reconstruction, and MarchingCubes algorithm. The geometric optimization methods include smoothing optimization, vertex simplification optimization, resampling optimization, and boundary reconstruction optimization.

[0078] Multiple surface reconstruction algorithms were applied to the preprocessed point cloud data to generate several preliminary 3D surface models of the ore body. The choice of surface reconstruction algorithm affects the applicability and accuracy of the model. Appropriate algorithms were selected based on the characteristics of the ore body: the Delaunay triangulation algorithm constructs a triangular mesh based on the point cloud data, connecting all points into triangular patches to generate the surface; the Poisso surface reconstruction algorithm generates a smooth 3D surface based on the point cloud normal vector information using the Poisson equation; the Alpha shape algorithm generates the convex hull of the point cloud by adjusting parameters, identifies the boundaries of the point cloud, and reconstructs the ore body surface; the least squares surface fitting method fits the optimal smooth surface using the least squares method based on the point cloud data, minimizing the error between the point cloud and the surface; the Poisson surface reconstruction method generates a smooth surface based on the point cloud data and normal vectors using the Poisson equation; and the MarchingCubes algorithm discretizes the point cloud data into a 3D mesh, generating isosurfaces by processing each mesh unit one by one, ultimately constructing the 3D surface.

[0079] To improve the accuracy and rationality of 3D ore body structure models, it is usually necessary to optimize the initially generated 3D surface model. This mainly includes the following steps: smoothing, which eliminates noise and overly sharp curves on the model surface to make the model smoother, using Gaussian smoothing and Laplacian smoothing algorithms to adjust areas with large surface curvature changes; vertex simplification optimization, which reduces the number of surface polygons, lowers model complexity, and improves computational efficiency; resampling optimization, which adjusts the distribution of mesh vertices and homogenizes vertex density without destroying the geometry, using a uniform sampling algorithm to redistribute vertices on the model surface; and boundary reconstruction optimization, which repairs the model boundaries to make them clearer and more reasonable.

[0080] Through a combination of different surface reconstructions and geometric optimizations, a complete three-dimensional ore body structure model is finally generated.

[0081] Step 3: Determine the terrain structure parameters of the grid based on the point cloud distribution within the grid, perform coupled analysis on the terrain structure parameters to determine the structural complexity of the grid, and select information comparison grids based on the structural complexity.

[0082] The topographic structure parameters within each grid on the ore body surface are characterized by the point cloud distribution of different grids. These parameters include surface curvature, undulation roughness, and point cloud density. The surface curvature is characterized by point cloud data coordinates, based on the following formula:

[0083] ;

[0084] In the formula, Let be the surface curvature of the i-th mesh. Let be the structural curvature of the p-th point within the i-th grid, where p is the index of the point within the grid. , This represents the total number of points within the grid.

[0085] It should be noted that the structural curvature at point p within the i-th grid is... This requires calculating the curvature using the three-dimensional coordinates of the point cloud data. Surface curvature describes the degree of bending of a local surface in the point cloud, reflecting the geometric characteristics of the local topography on the ore body surface. Specific steps include selecting a local neighborhood: determining the k-nearest neighbor or radius search to determine the k-nearest neighbor. The local surface is fitted using the neighborhood points of the i-th point: a quadratic surface fitting method is used to describe the i-th point. The local geometry of each point is used to calculate the principal curvature based on the derivative of the fitted surface, and the Gaussian curvature is obtained as the structural curvature of the p-th point in the i-th grid.

[0086] The specific formula used to calculate the surface roughness is as follows:

[0087] ;

[0088] In the formula, Let p be the vertical coordinate of the p-th point. Let be the roughness of the i-th grid. It is the average value of the vertical coordinates of all points in the i-th grid.

[0089] The formula used to characterize the structural complexity within different grids on the ore body surface is:

[0090] ;

[0091] In the formula, The structural complexity of the i-th grid is... This represents the normalized value of the surface curvature of the i-th mesh. This represents the normalized value of the roughness of the i-th mesh undulation. This is the normalized value of the point cloud density of the i-th grid.

[0092] It should be noted that the structural complexity of the i-th grid... Used to characterize the structural complexity within different grids on the ore body surface, comprehensively reflecting the geometric and distribution characteristics of local grids on the ore body surface, among which The larger the value, the more complex the structure within the i-th grid.

[0093] in is the normalized curvature value of the surface structure of the i-th mesh, representing the degree of curvature of the local geometry of the mesh surface. The greater the curvature, the more complex the mesh surface, for example, with more protrusions, depressions, and wrinkles, and drastic changes in surface morphology. and Proportional to the square root of curvature The operation prevents high curvature from contributing too much to the structural complexity index. The effect of curvature on complexity is non-linear, and the square root operation is more in line with the actual law that surface complexity gradually saturates as curvature increases.

[0094] This is the normalized value of the undulation roughness of the i-th grid. Roughness describes the variation in height difference between surface points, reflecting the severity of surface fluctuations. The effect of undulation roughness on complexity is non-linear: when the roughness is small, its variation contributes significantly to structural complexity; while when the roughness is high, its influence gradually saturates, and a logarithmic function is used to determine its impact. To better simulate this relationship.

[0095] It is the normalized value of the point cloud density of the i-th grid. The point cloud density describes the distribution density of data points within the grid and indirectly reflects the complexity of the local structure of the surface. When the same point cloud acquisition device with set device parameters is used to acquire surface structure, when the surface has more details, such as wrinkles, cracks, pits or high surface roughness, the device needs to use a higher point density to capture these complex features. Therefore, the higher the point cloud density, the more drastic the geometric changes of the surface structure and the higher the complexity of the local area.

[0096] Information comparison grids are selected based on structural complexity. The specific logic for selecting these grids is as follows: a structural complexity threshold is set, and the structural complexity is compared to this threshold. Based on the comparison results, the grids are then selected.

[0097] When the structural complexity When it is determined that the terrain change of the grid does not require monitoring, the grid is not recorded as an information comparison grid.

[0098] Structural complexity When the terrain of a grid is deemed to be complex, it is designated as an information comparison grid. The structural complexity threshold. Where the structural complexity threshold... The specific settings should be based on expert experience.

[0099] Step 4: Based on the point cloud data, determine the surface structure features of the information comparison grid, and calculate the cumulative difference between the grid and the corresponding surface structure features in each three-dimensional ore body structure model. The three-dimensional ore body structure model corresponding to the minimum cumulative difference is taken as the target ore body three-dimensional model.

[0100] The surface structure features include the region's average normal change rate and the region's average vertical coordinate value. The formula used to calculate the normal change rate based on point cloud data is:

[0101] ;

[0102] In the formula, Let be the average normal vector of the q-th target region within the j-th information comparison grid. Let be the average normal vector of the h-th corresponding neighborhood of the q-th target region. This represents the magnitude of the average normal vector of the q-th target region. Let be the average normal vector magnitude of the h-th corresponding neighborhood of the q-th target region. The average rate of change of the normal to the q-th target region within the grid is compared to the j-th piece of information.

[0103] The average normal vector of the target region is obtained by weighted averaging of the normal vectors of the small patches that make up the target region. The corresponding neighborhood of the target region is specifically set as a fixed radius neighborhood. With the center of the target region as the center, a fixed radius is set, and all regions within the radius except the target region are divided into several corresponding neighborhoods with equal area.

[0104] Where q is the index of the randomly selected target region within the information comparison grid, j is the index of the information comparison grid, and h is the index of the corresponding neighborhood.

[0105] In the 3D ore body structure model, the rate of change of the normal and the vertical coordinate values ​​of the corresponding regions are analyzed, their differences are calculated, and the differences of all randomly selected target regions are accumulated. The specific formula used for the calculation is as follows:

[0106] ;

[0107] In the formula, To accumulate the differences, , These represent the average rate of change of normal and the average vertical coordinate value of the region corresponding to the q-th target region within the grid compared to the j-th information in the 3D ore body structure model. Let be the average vertical coordinate value of the q-th target region within the j-th information comparison grid. This represents the total number of information comparison grids.

[0108] It should be noted that the cumulative difference This is used to represent the cumulative difference in the rate of change of the normal and the difference in the vertical coordinates of a selected target area in a three-dimensional ore body structure model. The larger the value, the greater the structural difference between the three-dimensional ore body structure model and the actual ore body to be monitored.

[0109] The rate of change of the normal reflects the degree of change in the local shape of the 3D ore body surface, directly describing the geometric complexity of the surface in that region. The accuracy of the model's geometric properties can be assessed by calculating the difference between the rates of change of the normals in the point cloud data and the model. The vertical coordinate value directly reflects the consistency between the ore body model and the point cloud data in the height direction; the height difference... The calculation aims to evaluate the spatial deviation between the model and the actual point cloud data. By accumulating the differences in the rate of change of the normal and the differences in the vertical coordinates of all target areas, the overall deviation of the entire ore body model from the point cloud data in terms of geometry and location can be obtained.

[0110] The three-dimensional ore body structure model with the smallest cumulative difference is selected as the target ore body three-dimensional model.

[0111] Step 5: Using the real-time location information of the ore body to be monitored, determine the impact range of the mining operation, take the impact range of the mining operation as the monitoring area, obtain the vibration information of the monitoring area, determine the vibration update area of ​​the target ore body 3D model, collect the point cloud data of the vibration update area again, and dynamically update the vibration update area structure corresponding to the target ore body 3D model based on the point cloud data of the vibration update area to complete the establishment of the ore body 3D model.

[0112] To analyze the real-time location information of the ore body to be mined, the impact range of mining operations is determined. The specific formula used to calculate the impact range of mining operations is as follows:

[0113] ;

[0114] In the formula, Due to the distance affected by mining operations, To affect the distance reference value, For mining depth, The factor representing the influence of mining depth is... The influence coefficient of soil and rock strength. To measure the compressive strength of the soil and rock in the ore body to be monitored, This is a reference value for the compressive strength of the soil and rock, which can be obtained from the reference data corresponding to the soil and rock of the ore body to be monitored.

[0115] It should be noted that, This is a standard influence range, usually given based on empirical data or model simulation results. It represents the influence range of mining operations under certain standard conditions, such as a specific mining depth and soil strength. The specific range is determined based on expert experience and the actual ore body. Through analysis of... The scope of influence can be dynamically adjusted based on actual mining conditions.

[0116] Mining depth It is one of the key factors affecting the scope of the construction impact. Generally, the deeper the mining, the greater the disturbance to the surrounding rock and soil, and the wider the impact range, expressed in square root form. The greater the depth, the smaller the incremental impact per unit depth, preventing over-correction. The contribution of depth to the area of ​​influence has been adjusted. The specific value is related to geological conditions, construction methods, etc., and is set based on expert experience.

[0117] The compressive strength of soil and rock is a key factor affecting the resistance of the ore body and surrounding rock mass to deformation or failure. When the soil and rock strength is high, the rock mass is more stable and the range of influence will be smaller. When the soil and rock strength is low, the rock mass is more likely to deform or fail and the range of influence will be larger. The contribution of changes in soil and rock strength to the affected area was adjusted. The specific value is closely related to the properties of the rock mass and the geological environment, and is set in combination with expert experience.

[0118] Using the current mining and construction site as the center, and the distance affected by the mining and construction as the radius... Using a radius, the impact range of mining operations is determined. Vibration monitoring units are set up in the grid within the impact range to monitor regional vibration information, including vibration frequency and vibration amplitude. The specific size of the grid within the impact range can be set according to the size of the impact range to improve vibration monitoring capabilities and save computing resources.

[0119] The vibration update region of the three-dimensional model of the target ore body is determined based on the following logic:

[0120] If the vibration frequency And vibration amplitude When this happens, it is determined that the grid does not need to be updated;

[0121] Conversely, the grid is dynamically updated to the vibration update region of the target ore body's 3D model;

[0122] By using the surface structure features of the vibration update area, the structure of the target ore body's 3D model is dynamically updated. Specifically, the point cloud data of the vibration update area is collected again, and the structure of the vibration update area is dynamically updated using the model construction method corresponding to the target ore body's 3D model, thereby achieving local updates to the structure of the target ore body's 3D model.

[0123] Please see Figure 4 The present invention also provides a three-dimensional ore body modeling device, which is used to perform the above-described three-dimensional ore body modeling method, comprising:

[0124] The terrain division and analysis module is used to obtain the geometric feature parameters of the ore body to be monitored, construct a three-dimensional boundary model of the ore body based on the geometric parameters, and divide the surface area of ​​the three-dimensional boundary model of the ore body into several grids of the same size.

[0125] The surface feature mapping module is used to collect point cloud data of the topographic surface of the ore body to be monitored, and map the point cloud distribution of the topographic surface to the corresponding grid. Based on different surface reconstruction algorithms and geometric optimization algorithms, the mapped point cloud data is analyzed to obtain several three-dimensional ore body structure models, forming a set of preferred ore body three-dimensional models.

[0126] The fine analysis and comparison module is used to determine the terrain structure parameters of the grid based on the point cloud distribution within the grid, coupled analysis of the terrain structure parameters to determine the structural complexity of the grid, and filter out information comparison grids based on the structural complexity.

[0127] The 3D model building module is used to determine the surface structure features of the information comparison grid based on point cloud data, and calculate the cumulative difference between the grid and the corresponding surface structure features in each 3D ore body structure model. The 3D ore body structure model corresponding to the minimum cumulative difference is taken as the target ore body 3D model.

[0128] The dynamic update and correction module is used to determine the impact range of mining operations based on the real-time location information of the ore body to be monitored, take the impact range of mining operations as the monitoring area, acquire vibration information of the monitoring area, determine the vibration update area of ​​the target ore body 3D model, collect point cloud data of the vibration update area again, and dynamically update the vibration update area structure of the target ore body 3D model based on the point cloud data of the vibration update area to complete the establishment of the ore body 3D model.

[0129] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0130] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0131] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0132] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for three-dimensional modeling of ore bodies, characterized in that, The specific steps include: Obtain the geometric feature parameters of the ore body to be monitored, construct a three-dimensional boundary model of the ore body based on the geometric parameters, and divide the surface area of ​​the three-dimensional boundary model of the ore body into several grids of the same size; Point cloud data of the topographic surface of the ore body to be monitored is collected, and the point cloud distribution of the topographic surface is mapped to the corresponding grid. Based on different surface reconstruction algorithms and geometric optimization algorithms, the mapped point cloud data is analyzed to obtain several three-dimensional ore body structure models and form a set of preferred three-dimensional ore body models. The terrain structure parameters of the grid are determined based on the point cloud distribution within the grid. The terrain structure parameters are coupled and analyzed to determine the structural complexity of the grid. Information comparison grids are selected based on the structural complexity. The surface structure features of the information comparison grid are determined based on point cloud data, and the cumulative difference between the grid and the corresponding surface structure features in each three-dimensional ore body structure model is calculated. The three-dimensional ore body structure model corresponding to the minimum cumulative difference is taken as the target ore body three-dimensional model. Using the real-time location information of the ore body to be monitored, the impact range of mining construction is determined. The impact range of mining construction is used as the monitoring area. Vibration information of the monitoring area is obtained, the vibration update area of ​​the target ore body three-dimensional model is determined, the point cloud data of the vibration update area is collected again, and the vibration update area structure corresponding to the target ore body three-dimensional model is dynamically updated based on the point cloud data of the vibration update area to complete the establishment of the ore body three-dimensional model. The formula used to characterize the structural complexity within different grids on the ore body surface is: In the formula, The structural complexity of the i-th grid is... This represents the normalized value of the surface curvature of the i-th mesh. This represents the normalized value of the roughness of the i-th mesh undulation. This is the normalized value of the point cloud density of the i-th grid. Information comparison grids are selected based on structural complexity. The specific logic for selecting these grids is as follows: a structural complexity threshold is set, and the structural complexity is compared to this threshold. Based on the comparison results, the grids are then selected. When the structural complexity When it is determined that the terrain change of the grid does not require monitoring, the grid is not recorded as an information comparison grid. Structural complexity When the terrain of a grid is deemed to be complex, it is designated as an information comparison grid. For structurally complex thresholds.

2. The method for three-dimensional modeling of ore bodies according to claim 1, characterized in that: The geometric feature parameters of the ore body specifically include the size information and boundary morphology information of the ore body to be monitored. The size information includes the length, width and height of the ore body, and the boundary morphology information includes the top and bottom boundaries and the boundary outline of the ore body to be monitored. The method for constructing a three-dimensional boundary model of the ore body is as follows: select three-dimensional modeling software, input the geometric feature parameters into the three-dimensional modeling software, and construct a three-dimensional boundary model of the ore body; The surface area of ​​the ore body's three-dimensional boundary model is divided into grids of the same size. The specific steps of determining the grid division scheme include: reading the boundary data of different view faces of the ore body; determining the division range based on the minimum bounding box of the different view faces of the ore body's three-dimensional model; setting the area size of the grid; determining the number of grid nodes in each direction based on the division range of the minimum bounding box of the different view faces and the grid size; and completing the grid division of the surface area based on the number of grid nodes.

3. The method for three-dimensional modeling of ore bodies according to claim 2, characterized in that: The surface of the ore body is scanned multiple times using a point cloud acquisition device with pre-set parameters to obtain point cloud data of the surface of the ore body to be monitored. The device parameters include resolution, scanning range and angle. The point cloud data of the surface of the ore body to be monitored obtained from each scan is denoised. The point cloud data of the terrain after denoising is registered. The point clouds obtained from different scans are registered to form a unified point cloud dataset. The logic for constructing the preferred ore body 3D model set is as follows: Multiple surface reconstruction algorithms are used to analyze the mapped point cloud data one by one, generating multiple preliminary 3D surface models. Different geometric optimization methods are then used to optimize each preliminary 3D surface model, resulting in several 3D ore body structure models, forming the preferred ore body 3D model set. The surface reconstruction algorithms include Delaunay triangulation, Alpha shape algorithm, least squares surface fitting, Poisson surface reconstruction, and MarchingCubes algorithm. The geometric optimization methods include smoothing optimization, vertex simplification optimization, resampling optimization, and boundary reconstruction optimization.

4. The method for three-dimensional modeling of ore bodies according to claim 1, characterized in that: The topographic structure parameters within each grid on the ore body surface are characterized by the point cloud distribution of different grids. These parameters include surface curvature, undulation roughness, and point cloud density. The surface curvature is characterized by point cloud data coordinates, based on the following formula: In the formula, Let be the surface curvature of the i-th mesh. Let be the structural curvature of the p-th point within the i-th grid, where p is the index of the point within the grid. , This represents the total number of points within the grid. The specific formula used to calculate the surface roughness is as follows: In the formula, Let p be the vertical coordinate of the p-th point. Let be the roughness of the i-th grid. It is the average value of the vertical coordinates of all points in the i-th grid.

5. The method for three-dimensional modeling of ore bodies according to claim 1, characterized in that: The surface structure features include the average rate of change of the normal and the average vertical coordinate value of the target area. The formula used to calculate the rate of change of the normal based on point cloud data is: In the formula, Let be the average normal vector of the q-th target region within the j-th information comparison grid. Let be the average normal vector of the h-th corresponding neighborhood of the q-th target region. This represents the magnitude of the average normal vector of the q-th target region. Let be the average normal vector magnitude of the h-th corresponding neighborhood of the q-th target region. The average rate of change of the normal to the q-th target region within the grid is used to compare the j-th piece of information. Where q is the index of the randomly selected target region within the information comparison grid, j is the index of the information comparison grid, and h is the neighborhood index of the target region; In the 3D ore body structure model, the average normal variation rate and average vertical coordinate value of the corresponding region are analyzed, their differences are calculated, and the differences of all randomly selected target regions are accumulated. The specific formula used for the calculation is as follows: In the formula, To accumulate the differences, , These represent the average rate of change of normal and the average vertical coordinate value of the region corresponding to the q-th target region within the grid compared to the j-th information in the 3D ore body structure model. Let be the average vertical coordinate value of the q-th target region within the j-th information comparison grid. The total number of information comparison grids; The three-dimensional ore body structure model with the smallest cumulative difference is selected as the target ore body three-dimensional model.

6. A coal mine monitoring method based on video images according to claim 5, characterized in that: To analyze the real-time location information of the ore body to be mined, the impact range of mining operations is determined. The specific formula used to calculate the impact range of mining operations is as follows: In the formula, Due to the distance affected by mining operations, To affect the distance reference value, For mining depth, The factor representing the influence of mining depth is... The influence coefficient of soil and rock strength. To measure the compressive strength of the soil and rock in the ore body to be monitored, This is a reference value for the compressive strength of soil and rock. Using the current mining and construction site as the center, and the distance affected by the mining and construction as the radius... Using a radius, the impact range of mining operations is determined. Vibration monitoring units are set up in the grid within the impact range to monitor regional vibration information, including vibration frequency and vibration amplitude. The vibration update region of the three-dimensional model of the target ore body is determined based on the following logic: If the vibration frequency And vibration amplitude When this happens, it is determined that the grid does not need to be updated; Conversely, the grid is dynamically updated to the vibration update region of the target ore body's 3D model; By using the surface structure features of the vibration update area, the structure of the target ore body's 3D model is dynamically updated. Specifically, the point cloud data of the vibration update area is collected again, and the structure of the vibration update area is dynamically updated using the model construction method corresponding to the target ore body's 3D model, thereby achieving local updates to the structure of the target ore body's 3D model.

7. A three-dimensional modeling device for ore bodies, characterized in that: The aforementioned ore body three-dimensional modeling device is used to execute the ore body three-dimensional modeling method according to any one of claims 1-6, comprising: The terrain division and analysis module is used to obtain the geometric feature parameters of the ore body to be monitored, construct a three-dimensional boundary model of the ore body based on the geometric parameters, and divide the surface area of ​​the three-dimensional boundary model of the ore body into several grids of the same size. The surface feature mapping module is used to collect point cloud data of the topographic surface of the ore body to be monitored, and map the point cloud distribution of the topographic surface to the corresponding grid. Based on different surface reconstruction algorithms and geometric optimization algorithms, the mapped point cloud data is analyzed to obtain several three-dimensional ore body structure models, forming a set of preferred ore body three-dimensional models. The fine analysis and comparison module is used to determine the terrain structure parameters of the grid based on the point cloud distribution within the grid, coupled analysis of the terrain structure parameters to determine the structural complexity of the grid, and filter out information comparison grids based on the structural complexity. The 3D model building module is used to determine the surface structure features of the information comparison grid based on point cloud data, and calculate the cumulative difference between the grid and the corresponding surface structure features in each 3D ore body structure model. The 3D ore body structure model corresponding to the minimum cumulative difference is taken as the target ore body 3D model. The dynamic update and correction module is used to determine the impact range of mining operations based on the real-time location information of the ore body to be monitored, take the impact range of mining operations as the monitoring area, acquire vibration information of the monitoring area, determine the vibration update area of ​​the target ore body 3D model, collect point cloud data of the vibration update area again, and dynamically update the vibration update area structure of the target ore body 3D model based on the point cloud data of the vibration update area to complete the establishment of the ore body 3D model.

Images