A method for constructing an integrated on - ground and underground geological dynamic grid structure

Abstract

The present invention discloses a method for constructing an integrated above-ground and underground geological dynamic grid structure, belonging to the field of geological data grid structure construction. The method comprises collecting geological spatial data; presetting a grid size and dividing the geological spatial data based on the grid size to obtain an initial grid model; clustering the geological spatial data using the DBSCAN clustering algorithm and removing noise data points to obtain a plurality of cluster regions; dynamically adjusting the grid size in each cluster region in the initial grid model based on the data point density of each cluster region to obtain a primary refined grid model; gradient-refining the cluster region junctions in the primary refined grid model to obtain a secondary refined grid model; and using a convolution operation to correct the data values of data points in the transition zone between the ground and the underground in the secondary refined grid model to obtain a final above-ground and underground integrated grid model. The present invention solves the problem that traditional grid structures cannot adapt to the spatial data distribution of different regions.

Description

Technical Field

[0001] The present invention belongs to the field of geological data grid structure construction, and in particular relates to a method for constructing an above-ground and underground integrated geological dynamic grid structure. Background Art

[0002] Existing 3D geological modeling technologies have limitations when it comes to integrating surface and underground geological information. Traditional 3D modeling methods often model surface and underground geological information separately, resulting in discrepancies in spatial location and accuracy. This separate modeling approach not only reduces the overall consistency and reliability of the model but also limits its effectiveness in practical applications. This is particularly true when considering the interactions and relationships between surface and underground facilities, making it difficult to provide a comprehensive geological view and support decision-making.

[0003] One existing technical solution is 3D geological model visualization based on borehole data. This technology uses data fusion and interpolation modeling based on geological borehole data and geophysical exploration techniques to construct a 3D visualization model of the subsurface strata. This technology utilizes the HT framework of Tupu Software for web-based 3D visualization. However, its drawbacks include large amounts of generated data, dramatic changes in the underlying layer, and significant errors in the regional model and cropping surfaces.

[0004] One existing technical solution is a DEM value-added theoretical framework and construction method oriented towards the essence of geomorphology. This approach introduces a value-added DEM, dynamically linking spatial and temporal dimensions, and transcends the limitation of traditional DEMs, which only reflect surface elevation. Disadvantages include the high requirements of spatial data structures in terms of data storage efficiency, spatial analysis complexity, and visualization, as well as the diverse and complex requirements for digital landform modeling and analysis in the geomorphology field. Summary of the Invention

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method for constructing an integrated above-ground and underground geological dynamic grid structure, which solves the problem that the traditional grid structure cannot adapt to the spatial data distribution of different regions.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: a method for constructing an integrated above-ground and underground geological dynamic grid structure, comprising:

[0007] Collect geospatial data;

[0008] Preset the grid size and divide the geological spatial data based on the grid size to obtain the initial grid model;

[0009] The DBSCAN clustering algorithm is used to cluster the geological spatial data and remove noise data points to obtain several cluster areas;

[0010] Based on the data point density of each cluster area, the grid size in each cluster area in the initial grid model is dynamically adjusted to obtain the initial refined grid model;

[0011] Based on the gradient refinement of the cluster area junction in the primary refined grid model, a secondary refined grid model is obtained;

[0012] The convolution operation is used to correct the data values of the data points in the transition zone between the ground and underground in the secondary refined grid model to obtain the final ground and underground integrated grid model.

[0013] The beneficial effects of the present invention are: adaptively dividing the grid according to the density distribution of spatial data, treating dense and sparse areas differently, and avoiding the waste of computing resources or insufficient accuracy caused by unified grid granularity. Dynamically improving spatial resolution, realizing adaptive grid construction with high resolution in dense areas and low resolution in sparse areas, improving modeling efficiency and spatial expression capabilities, and improving the accuracy and rationality of spatial data modeling. By gradient-refining the cluster boundary area, the problem of sudden changes at the boundaries of different cluster areas is avoided, so that the model has better continuity and reality in the regional transition part, and the spatial consistency of the model is enhanced. The calculation content includes the mixed input and processing of point cloud, raster, and vector data, which is easy to integrate with multiple data types and is suitable for various scenarios such as smart cities, underground pipeline networks, and mine modeling.

[0014] Furthermore, the expression of the data point density of each cluster area is:

[0015]

[0016] in, For the The density of data points in each cluster area; For the The number of data points in a cluster area; For the The convex hull volume of the cluster area.

[0017] The beneficial effect of the above further solution is that by calculating the ratio of the number of data points to the area of the region, a comparable spatial density index is obtained, avoiding the errors caused by evaluating only the number of points or area separately, so that the internal density of the region can be quantitatively measured.

[0018] Furthermore, the expression for dynamically adjusting the grid size in each cluster area in the initial grid model based on the data point density of each cluster area is:

[0019]

[0020] in, For the The grid size after the clustering area is adjusted; is the preset grid size; For the The density of data points in each cluster area; is the weighting coefficient given based on geographical characteristics; The correlation coefficient to control the influence of density on grid size; A control parameter for the logarithmic mapping used to balance the smoothness of grid adjustments.

[0021] The beneficial effect of this further approach is that by dynamically calculating the grid size for each region, the grid granularity can be flexibly adjusted according to the data point density in different regions, achieving adaptive grid division based on data distribution. By comprehensively considering factors such as the geography, function, and frequency of use of different regions, the model's ability to identify spatial heterogeneity and regional complexity is improved, and its adaptability to spatial heterogeneity is enhanced.

[0022] Furthermore, the cluster area junctions in the gradient-based refined primary grid model are refined to obtain a secondary refined grid model, specifically:

[0023] A1. Set the neighborhood radius to determine the boundary point of the current cluster region and the boundary point of the current adjacent cluster region, and determine the neighborhood of the boundary point of the current cluster region and the neighborhood of the boundary point of the current adjacent cluster region according to the neighborhood radius;

[0024] A2. Integrate the neighborhood of the boundary point of the current cluster region and the neighborhood of the boundary point of the current adjacent cluster region to obtain the boundary region;

[0025] A3. Calculate the gradient value of each data point in the boundary area;

[0026] A4. Extract data points whose gradient values are greater than the gradient threshold as data points to be refined;

[0027] A5. For the data points to be refined, the grid size of the data points is refined based on the gradient;

[0028] A6. Return to step A3 until the gradient values of all data points are less than the gradient threshold, then proceed to step A7.

[0029] A7. Return to A1 and refine the intersection point between the current cluster region and the next adjacent cluster region until the intersection points between the current cluster region and all adjacent cluster regions are refined, and then proceed to step A8.

[0030] A8. Return to A1 and refine the intersection points between the next cluster area and the adjacent cluster areas until the intersection points between all cluster areas and all adjacent cluster areas are refined to obtain a secondary refined grid model.

[0031] The beneficial effects of the above-mentioned further scheme are as follows: by calculating the gradient value of the data point, the mutation points in the junction area are dynamically identified, and then the local grid refinement is performed, which can effectively alleviate the grid jump phenomenon at the cluster boundary and enhance the expression continuity and accuracy at the junction of the cluster area. The introduction of gradient threshold and multi-round iteration mechanism can capture and optimize the areas with drastic changes in spatial attributes at a finer granularity, making the model more responsive and adaptable to heterogeneous changes such as geology and landforms. The boundary area after gradient refinement has higher consistency and smoothness, providing more reasonable input for subsequent operations such as convolution correction and spatial interpolation, improving the overall realistic expression effect and post-processing stability.

[0032] Furthermore, the expression of the gradient value of the data point in A3 is:

[0033]

[0034] in, For data points The gradient value of is the symbol of partial derivative; is the attribute value of the data point; is the horizontal coordinate of the data point; is the vertical coordinate of the data point; is the vertical coordinate of the data point.

[0035] The benefits of this further approach include: quantitatively reflecting the degree of change in data points across different directions, accurately measuring the changing trends of spatial attributes, and providing a scientific basis for boundary refinement. It also effectively improves the stability and accuracy of gradient values in situations with uneven data density or outliers, providing a foundational metric for multi-scale refinement.

[0036] Furthermore, the expression of the grid size based on the gradient-refined data points is:

[0037]

[0038]

[0039] in, For the The grid size of the data points to be refined is based on the gradient refinement; For the The grid size of the data points to be refined; For the Data points to be refined The gradient value at ; For the The horizontal coordinate of the data point to be refined; For the The vertical coordinate of the data point to be refined; For the The vertical coordinates of the data points to be refined; is a positive number; is a regulating factor used to control the intensity of the gradient's influence on the grid density; It is a harmonic function that integrates local actual values and regional gradient change trends; For data points Updated attribute values after fusing local actual values and regional gradient change trends; The weight coefficient to control the proportion of original value and gradient trend; Extract function for attribute value of data point; For data points Attribute value of is the mean gradient of the boundary area.

[0040] The beneficial effects of this further approach include: by introducing a harmonic function to fuse the original spatial attribute values with regional gradient changes, dynamic perception and response to regions with different spatial change rates are achieved, thereby rationally adjusting the grid granularity and improving the model's adaptability to geological structural differences. The setting of parameters such as weight coefficients and adjustment factors makes the method highly adjustable, improving the spatial resolution of boundary areas and optimizing the spatial continuity of the transition zone between aboveground and underground.

[0041] Furthermore, the convolution operation is used to correct the data values of the data points in the transition zone between the ground and underground in the secondary refined grid model to obtain the final ground and underground integrated grid model, specifically:

[0042] Identify the transition zone between the ground and underground in the secondary refined grid model; the transition zone is the overlapping part of the underground cluster area and the ground cluster area in the secondary refined grid model;

[0043] Calculate the local density of each data point in the transition zone;

[0044] Based on the local density of each data point in the transition zone, the convolution kernel size of the convolution operation of each data point in the transition zone is determined respectively;

[0045] For each data point in the transition zone, the convolution operation is iterated until the maximum number of iterations is reached or the difference between two adjacent iteration results is less than the preset threshold. The iterative output result of the convolution operation on each data point in the transition zone is used as the modified data value of each data point in the transition zone to obtain the final above-ground and underground integrated grid model.

[0046] The above-mentioned further solution has the following beneficial effects: it effectively eliminates data mutations, jumps, and faults, achieves a natural connection and smooth transition in the model across spatial transition zones, and improves data continuity and smoothness in the transition zone between the surface and underground. Dynamically adjusting the convolution kernel size based on local density allows for more detailed local corrections in data-dense areas, while avoiding overfitting in sparse areas, thereby more accurately preserving local geological features and structural contours. This prevents excessive iterations and information loss, enhancing numerical stability.

[0047] Furthermore, the expression of the local density of each data point in the transition zone is:

[0048]

[0049] in, The transition zone The local density of data points; The transition zone The number of data points in the neighborhood of a data point; The transition zone The neighborhood volume of a data point; is the neighborhood radius of the data points in the transition zone.

[0050] The beneficial effect of the above further solution is that the calculation of local density realizes the perception of the local spatial distribution characteristics of each data point and enhances the spatial response capability of the model.

[0051] Furthermore, the expression of the convolution kernel size of the convolution operation of each data point in the transition zone is:

[0052]

[0053] in, The transition zone The convolution kernel size of the convolution operation of the data point; is a constant used to control the scale of the convolution kernel; The transition zone The local density of data points; It is a parameter used to control the influence of density on the size of the convolution kernel.

[0054] The beneficial effect of this further approach is that the convolution kernel size of each data point is adjusted based on local density, resulting in smaller kernels in high-density areas to preserve detail and larger kernels in low-density areas to enhance smoothness. By parameterizing the influence of density on the convolution kernel, the convolution scale is controlled to ensure continuity and stability, thereby improving the flexibility and adaptability of the entire transition zone modeling.

[0055] Furthermore, the expression of the convolution operation is:

[0056]

[0057] in, For the The output result of the iterative convolution operation indicates that the data point The value at is changed to ; is the convolution kernel function used to weight neighborhood points according to distance; For the The output result of the iterative convolution operation; The transition zone The number of data points in the neighborhood of a data point; The transition zone data points and the transition zone The Euclidean distance of the data points; Parameters that control weight decay; For the The horizontal coordinate of each data point; For the The vertical coordinate of the data point; For the The vertical coordinates of the data points; For the The horizontal coordinate of each data point; For the The vertical coordinate of the data point; For the The vertical coordinates of the data points; The transition zone The convolution kernel size for the convolution operation of data points.

[0058] The beneficial effects of the above further scheme are: the convolution operation based on distance weighting ensures smooth transition of data, effectively realizes the spatial weight attenuation mechanism between data points, and improves the expression continuity of transition zone data. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0060] The specific embodiments of the present invention are described below to facilitate understanding of the present invention by those skilled in the art. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations utilizing the concepts of the present invention are protected.

[0061] like Figure 1 As shown, in one embodiment of the present invention, a method for constructing an integrated above-ground and underground geological dynamic grid structure includes:

[0062] Collect geospatial data;

[0063] Preset the grid size and divide the geological spatial data based on the grid size to obtain the initial grid model;

[0064] The DBSCAN clustering algorithm is used to cluster the geological spatial data and remove noise data points to obtain several cluster areas;

[0065] Based on the data point density of each cluster area, the grid size in each cluster area in the initial grid model is dynamically adjusted to obtain the initial refined grid model;

[0066] Based on the gradient refinement of the cluster area junction in the primary refined grid model, a secondary refined grid model is obtained;

[0067] The convolution operation is used to correct the data values of the data points in the transition zone between the ground and underground in the secondary refined grid model to obtain the final ground and underground integrated grid model.

[0068] In this embodiment, the geological spatial data includes:

[0069] ① Three-dimensional spatial coordinate data: The spatial position coordinates (X, Y, Z) of each geological point, used to express the spatial position of the geological body on the surface or underground.

[0070] ② Geological attribute data: attribute values corresponding to each point or cell, such as lithology type (such as sandstone, shale, etc.), density, resistivity, porosity and other physical parameters.

[0071] ③ Stratum number or layer identification: whether it is above ground / underground identification.

[0072] ④ Topography and landform data: digital elevation model (DEM), contour lines, surface slope and other information.

[0073] ⑤ Underground structure data: data describing the structure of underground rock formations, such as drilling records, seismic exploration results, and geological profiles.

[0074] ⑥Data source information: data collection method (such as lidar, remote sensing, drilling, geological measurement), collection accuracy, timestamp, coordinate system information (such as WGS84, CGCS2000, etc.).

[0075] When constructing a grid structure, the above data are used to obtain the coordinates of the data points, and the stratum number or layer identification is used to identify whether the data points are above ground or underground. In different application scenarios, the geological attribute data, topographic and landform data, and underground structure data are selected as the attribute values of the data points, and the data source information is used to identify the relevant information of the attribute values.

[0076] For example, when constructing a density grid model, the attribute value of each data point is density; when constructing a resistivity grid model, the attribute value of each data point is resistivity; when constructing a surface slope grid model, the attribute value of each data point is surface slope, and so on;

[0077] At the same time, terrain and landform data are also hidden constraints for building raster models, which are used to limit the location of data points.

[0078] The resulting raster model has application value in the following aspects:

[0079] In terms of three-dimensional geological modeling, by integrating surface and underground spatial information, the final integrated ground and underground grid model can be used to construct a more detailed three-dimensional geological structure model, providing a data basis for geological surveys, mineral resource assessments, etc.

[0080] In the application of remote sensing and spatial analysis, during the processing of remote sensing images and lidar data, the final integrated ground and underground raster model can be used as a spatial reference framework for data fusion and feature extraction, thereby improving the efficiency and accuracy of subsequent analysis.

[0081] In this embodiment, in the integrated spatial data of the ground and underground, the data density in different areas varies greatly, and the use of traditional unified grid accuracy often cannot effectively reflect the spatial characteristics of the data.

[0082] The DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm adaptively adjusts the grid divisions based on the density distribution of the data. Clustering based on the density distribution of data points can identify dense and sparse areas and remove noise points, effectively handling irregular cluster patterns in spatial data. It automatically adapts to the data characteristics of different regions, avoiding the limitations of pre-set grid sizes found in traditional methods.

[0083] The expression of the data point density of each cluster area is:

[0084]

[0085] in, For the The density of data points in each cluster area; For the The number of data points in a cluster area; For the The convex hull volume of the cluster area.

[0086] The expression for dynamically adjusting the grid size in each cluster area in the initial grid model based on the data point density of each cluster area is:

[0087]

[0088] in, For the The grid size after the clustering area is adjusted; is the preset grid size; For the The density of data points in each cluster area; is the weighting coefficient given based on geographical characteristics; The correlation coefficient to control the influence of density on grid size; A control parameter for the logarithmic mapping used to balance the smoothness of grid adjustments.

[0089] In this embodiment, during the process of adaptive grid precision, a mathematical relationship is established between the density information in the clustering result and the grid size adjustment, and the clustering result (ie, density information) is mapped to the grid precision.

[0090] In this example, a higher density indicates a greater concentration of points within the area, representing denser data and, therefore, higher grid accuracy. To better represent integrated above- and underground spatial data, an adaptive grid model is designed based on clustering results and the characteristics of geographic information. This guidance effectively improves the rationality and accuracy of grid division, dynamically adapting to the complexity of diverse geographic environments and providing more accurate data analysis results.

[0091] Clustering results can be used to identify distinct geological clusters. Semantic annotation is performed based on the spatial distribution of clusters, introducing additional geographic features into the formula. Leveraging geographic information (such as topography, building distribution, and underground pipeline locations), different regions can have different density mapping rules, which can affect the grid size.

[0092] This adaptive grid precision adjustment method based on cluster density, guided by various geographic information, can improve the accuracy and efficiency of spatial data processing. Complex areas with high data density use finer grids, while open areas or relatively uniform underground areas use larger grids.

[0093] The gradient-based refinement of the cluster region junctions in the primary refined grid model to obtain a secondary refined grid model is as follows:

[0094] A1. Set the neighborhood radius to determine the boundary point of the current cluster region and the boundary point of the current adjacent cluster region, and determine the neighborhood of the boundary point of the current cluster region and the neighborhood of the boundary point of the current adjacent cluster region according to the neighborhood radius;

[0095] A2. Integrate the neighborhood of the boundary point of the current cluster region and the neighborhood of the boundary point of the current adjacent cluster region to obtain the boundary region;

[0096] A3. Calculate the gradient value of each data point in the boundary area;

[0097] A4. Extract data points whose gradient values are greater than the gradient threshold as data points to be refined;

[0098] A5. For the data points to be refined, the grid size of the data points is refined based on the gradient;

[0099] A6. Return to step A3 until the gradient values of all data points are less than the gradient threshold, then proceed to step A7.

[0100] A7. Return to A1 and refine the intersection point between the current cluster region and the next adjacent cluster region until the intersection points between the current cluster region and all adjacent cluster regions are refined, and then proceed to step A8.

[0101] A8. Return to A1 and refine the intersection points between the next cluster area and the adjacent cluster areas until the intersection points between all cluster areas and all adjacent cluster areas are refined to obtain a secondary refined grid model.

[0102] In this embodiment, the boundary area is usually where the data changes more dramatically, and requires fine gradient calculation and refinement to improve accuracy. The boundary area refers to the boundary area between two cluster areas, which usually has large data changes. The steps to identify the boundary area are as follows:

[0103] Cluster boundary definition: For every two clusters, we take the boundary points from the clustering results and calculate the neighborhood information of each cluster boundary point to determine which points are on the cluster boundary. We then use these boundary points for further refinement.

[0104] Gradient change detection: In the boundary area, by calculating the gradient value of each point on the boundary , to determine which areas have the largest changes. The gradient changes more dramatically at the junction, which is usually where refinement is needed.

[0105] In the boundary area, refinement is performed based on the gradient information to ensure that the boundary area has sufficient accuracy to accurately depict the transition between the two cluster areas.

[0106] Gradient Refinement: When the gradient is greater than a threshold, the grid is refined in the boundary area. The refinement process can add grid points through interpolation methods such as cubic spline interpolation or bilinear interpolation to make the data smoother and increase the resolution.

[0107] Since the boundary areas between cluster regions usually have complex spatial structures, the refinement process can be performed iteratively to further improve the accuracy and eliminate unnecessary computational burden.

[0108] First iteration: For the boundary areas between each cluster area, preliminary refinement is performed based on the gradient value to increase the grid density and process areas with large changes.

[0109] Second Iteration: After the first iteration, the gradients of the boundary region are recalculated and the region is further refined based on the new gradient values. If necessary, the precision can be increased or decreased. For regions with less gradient change, the precision can be reduced to save computational effort.

[0110] Multiple rounds of iteration: After each round of iterative refinement, check the accuracy change of the refined area. If the result after refinement meets the expectations, you can stop the iteration; otherwise, you can continue to iterate until the accuracy requirements are met.

[0111] The expression of the gradient value of the data point in A3 is:

[0112]

[0113] in, For data points The gradient value of is the symbol of partial derivative; is the attribute value of the data point; is the horizontal coordinate of the data point; is the vertical coordinate of the data point; is the vertical coordinate of the data point.

[0114] The expression of the grid size based on the gradient refinement data point is:

[0115]

[0116]

[0117] in, For the The grid size of the data points to be refined is based on the gradient refinement; For the The grid size of the data points to be refined; For the Data points to be refined The gradient value at ; For the The horizontal coordinate of the data point to be refined; For the The vertical coordinate of the data point to be refined; For the The vertical coordinates of the data points to be refined; is a positive number; is a regulating factor used to control the intensity of the gradient's influence on the grid density; It is a harmonic function that integrates local actual values and regional gradient change trends; For data points Updated attribute values after fusing local actual values and regional gradient change trends; The weight coefficient to control the proportion of original value and gradient trend; Extract function for attribute value of data point; For data points Attribute value of is the mean gradient of the boundary area.

[0118] In this embodiment, A very small positive number used to avoid division by 0.

[0119] The convolution operation is used to correct the data values of the data points in the transition zone between the ground and underground in the secondary refined grid model to obtain the final ground and underground integrated grid model, specifically:

[0120] Identify the transition zone between the ground and underground in the secondary refined grid model; the transition zone is the overlapping part of the underground cluster area and the ground cluster area in the secondary refined grid model;

[0121] Calculate the local density of each data point in the transition zone;

[0122] Based on the local density of each data point in the transition zone, the convolution kernel size of the convolution operation of each data point in the transition zone is determined respectively;

[0123] For each data point in the transition zone, the convolution operation is iterated until the maximum number of iterations is reached or the difference between two adjacent iteration results is less than the preset threshold. The iterative output result of the convolution operation on each data point in the transition zone is used as the modified data value of each data point in the transition zone to obtain the final above-ground and underground integrated grid model.

[0124] In this embodiment, the integrated spatial data of the ground and underground usually includes the overlapping parts of the ground and underground, and the data complexity is relatively high. In order to adapt to the updated grid size and refine the accuracy of the data, the data can be locally modified through convolution operations. Flexibly adapt to the data characteristics of different regions and avoid excessive noise interference. Improve the spatial resolution and accuracy of the data, especially in complex areas such as underground structures and ground overlapping areas. By analyzing the spatial position and height information of different clusters in the DBSCAN results, the overlapping areas of the ground and underground are identified and the areas that need to be convolution modified are marked. This area is usually the transition zone between the two clusters after DBSCAN clustering. When processing the overlapping areas of the ground and underground, the size of the convolution kernel is dynamically adjusted according to the local density of the point cloud data, and the noise is removed and the data is refined through the convolution operation. For the overlapping parts, the size and step size of the convolution kernel should be dynamically optimized according to the different ground and underground structures. Through the convolution operation, noise points are removed, valid data is retained, and data accuracy is improved.

[0125] The local density of each point in the clustered transition zone needs to be calculated to adjust the convolution kernel size. For denser areas (e.g., where there is a lot of overlap between the ground and underground points), the convolution kernel size should be smaller. For less dense areas (e.g., at the boundary between the ground and underground), the convolution kernel size should be larger.

[0126] Let the convolution kernel size and local density The purpose of the convolution operation is to smooth the data and remove noise through local weighted averaging. After the convolution kernel size is determined, the convolution operation can be used for local smoothing.

[0127] The expression of the local density of each data point in the transition zone is:

[0128]

[0129] in, The transition zone The local density of data points; The transition zone The number of data points in the neighborhood of a data point; The transition zone The neighborhood volume of a data point; is the neighborhood radius of the data points in the transition zone.

[0130] The expression of the convolution kernel size of the convolution operation of each data point in the transition band is:

[0131]

[0132] in, The transition zone The convolution kernel size of the convolution operation of the data point; is a constant used to control the scale of the convolution kernel; The transition zone The local density of data points; It is a parameter used to control the influence of density on the size of the convolution kernel.

[0133] The expression of the convolution operation is:

[0134]

[0135] in, For the The output result of the iterative convolution operation indicates that the data point The value at is changed to ; is the convolution kernel function used to weight neighborhood points according to distance; For the The output result of the iterative convolution operation; The transition zone The number of data points in the neighborhood of a data point; The transition zone data points and the transition zone The Euclidean distance of the data points; Parameters that control weight decay; For the The horizontal coordinate of each data point; For the The vertical coordinate of the data point; For the The vertical coordinates of the data points; For the The horizontal coordinate of each data point; For the The vertical coordinate of the data point; For the The vertical coordinates of the data points; The transition zone The convolution kernel size for the convolution operation of data points.

Claims

1. A method for constructing an integrated above-ground and underground geological dynamic grid structure, characterized in that: include: Collect geospatial data; Preset the grid size and divide the geological spatial data based on the grid size to obtain the initial grid model; The DBSCAN clustering algorithm is used to cluster the geological spatial data and remove noise data points to obtain several cluster areas; Based on the data point density of each cluster area, the grid size in each cluster area in the initial grid model is dynamically adjusted to obtain the initial refined grid model; Based on the gradient refinement of the cluster region junction in the primary refined grid model, a secondary refined grid model is obtained; for the data points to be refined at the cluster region junction, the grid size of the data points is refined based on the gradient; the expression of the grid size based on the gradient refined data points is: in, For the The grid size of the data points to be refined is based on the gradient refinement; For the The grid size of the data points to be refined; For the Data points to be refined The gradient value at ; For the The horizontal coordinate of the data point to be refined; For the The vertical coordinate of the data point to be refined; For the The vertical coordinates of the data points to be refined; is a positive number; is a regulating factor used to control the intensity of the gradient's influence on the grid density; It is a harmonic function that integrates local actual values and regional gradient change trends; For data points Updated attribute values after fusing local actual values and regional gradient change trends; The weight coefficient to control the proportion of original value and gradient trend; Extract function for attribute value of data point; For data points Attribute value of is the mean gradient of the boundary area; The convolution operation is used to correct the data values of the data points in the transition zone between the ground and underground in the secondary refined grid model to obtain the final ground and underground integrated grid model.

2. The method for constructing an integrated above-ground and underground geological dynamic grid structure according to claim 1, characterized in that: The expression of the data point density of each cluster area is: in, For the The density of data points in each cluster area; For the The number of data points in a cluster area; For the The convex hull volume of the cluster area.

3. The method for constructing an integrated above-ground and underground geological dynamic grid structure according to claim 1, characterized in that: The expression for dynamically adjusting the grid size in each cluster area in the initial grid model based on the data point density of each cluster area is: in, For the The grid size after the clustering area is adjusted; is the preset grid size; For the The density of data points in each cluster area; is the weighting coefficient given based on geographical characteristics; The correlation coefficient to control the influence of density on grid size; A control parameter for the logarithmic mapping used to balance the smoothness of grid adjustments.

4. The method for constructing an integrated above-ground and underground geological dynamic grid structure according to claim 1, characterized in that: The gradient-based refinement of the cluster region junctions in the primary refined grid model to obtain a secondary refined grid model is as follows: A1. Set the neighborhood radius to determine the boundary point of the current cluster region and the boundary point of the current adjacent cluster region, and determine the neighborhood of the boundary point of the current cluster region and the neighborhood of the boundary point of the current adjacent cluster region according to the neighborhood radius; A2. Integrate the neighborhood of the boundary point of the current cluster region and the neighborhood of the boundary point of the current adjacent cluster region to obtain the boundary region; A3. Calculate the gradient value of each data point in the boundary area; A4. Extract data points whose gradient values are greater than the gradient threshold as data points to be refined; A5. For the data points to be refined, the grid size of the data points is refined based on the gradient; A6. Return to step A3 until the gradient values of all data points are less than the gradient threshold, then proceed to step A7. A7. Return to A1 and refine the intersection point between the current cluster region and the next adjacent cluster region until the intersection points between the current cluster region and all adjacent cluster regions are refined, and then proceed to step A8. A8. Return to A1 and refine the intersection points between the next cluster area and the adjacent cluster areas until the intersection points between all cluster areas and all adjacent cluster areas are refined to obtain a secondary refined grid model.

5. The method for constructing an integrated above-ground and underground geological dynamic grid structure according to claim 4, characterized in that: The expression of the gradient value of the data point in A3 is: in, For data points The gradient value of is the symbol of partial derivative; is the attribute value of the data point; is the horizontal coordinate of the data point; is the vertical coordinate of the data point; is the vertical coordinate of the data point.

6. The method for constructing an integrated above-ground and underground geological dynamic grid structure according to claim 1, characterized in that: The convolution operation is used to correct the data values of the data points in the transition zone between the ground and underground in the secondary refined grid model to obtain the final ground and underground integrated grid model, specifically: Identify the transition zone between the ground and underground in the secondary refined grid model; the transition zone is the overlapping part of the underground cluster area and the ground cluster area in the secondary refined grid model; Calculate the local density of each data point in the transition zone; Based on the local density of each data point in the transition zone, the convolution kernel size of the convolution operation of each data point in the transition zone is determined respectively; For each data point in the transition zone, the convolution operation is iterated until the maximum number of iterations is reached or the difference between two adjacent iteration results is less than the preset threshold. The iterative output result of the convolution operation on each data point in the transition zone is used as the modified data value of each data point in the transition zone to obtain the final above-ground and underground integrated grid model.

7. The method for constructing an integrated above-ground and underground geological dynamic grid structure according to claim 6, characterized in that: The expression of the local density of each data point in the transition zone is: in, The transition zone The local density of data points; The transition zone The number of data points in the neighborhood of a data point; The transition zone The neighborhood volume of a data point; is the neighborhood radius of the data points in the transition zone.

8. The method for constructing an integrated above-ground and underground geological dynamic grid structure according to claim 6, characterized in that: The expression of the convolution kernel size of the convolution operation of each data point in the transition band is: in, The transition zone The convolution kernel size of the convolution operation of the data point; is a constant used to control the scale of the convolution kernel; The transition zone The local density of data points; It is a parameter used to control the influence of density on the size of the convolution kernel.

9. The method for constructing an integrated above-ground and underground geological dynamic grid structure according to claim 6, characterized in that: The expression of the convolution operation is: in, For the The output result of the iterative convolution operation indicates that the data point The value at is changed to ; is the convolution kernel function used to weight neighborhood points according to distance; For the The output result of the iterative convolution operation; The transition zone The number of data points in the neighborhood of a data point; The transition zone data points and the transition zone The Euclidean distance of the data points; Parameters that control weight decay; For the The horizontal coordinate of each data point; For the The vertical coordinate of the data point; For the The vertical coordinates of the data points; For the The horizontal coordinate of each data point; For the The vertical coordinate of the data point; For the The vertical coordinates of the data points; The transition zone The convolution kernel size for the convolution operation of data points.

Images