Geological 3D modeling method and system based on multi-data fusion

Abstract

The present invention relates to the field of three-dimensional object recognition technology, specifically to a geological three-dimensional modeling method and system based on multi-data fusion, comprising the following steps: dividing density segments according to the node spacing ratio, obtaining an adjustment coefficient for the elevation and horizontal difference of the segment, offsetting the elevation of the nodes in the high-density area and correcting the angle of the low-density area, reconstructing a multi-layer profile, extracting the segment length and angle coupling characteristics, comparing the coupling difference, and generating three-dimensional splicing information in combination with the original coordinates. In the present invention, the density segments are divided according to the node spacing ratio, and the adjustment coefficient is formed by combining the elevation and horizontal difference to achieve the node offset in the high-density area and the angle correction in the low-density area, enhance the morphological details, form a coupling feature with the length on both sides of the node and the turning angle, and extract the difference in the multi-profile comparison. The segment length, elevation offset and original coordinates are combined to complete the three-dimensional splicing to make the structure more coherent. The processing with density segmentation, node fine-tuning and coupling difference as the core effectively releases the data space association and improves the restoration and interpretation accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional object recognition, and in particular to a geological three-dimensional modeling method and system based on multi-data fusion. Background Art

[0002] The field of three-dimensional object recognition technology includes the technical content of using computers to collect data, analyze information and reconstruct spatial structures of three-dimensional objects or scenes in the real world. Its core lies in the recognition and representation of the geometric shape, topological relationship and spatial distribution characteristics of three-dimensional objects by processing the acquired images, point clouds, optical scanning data, geological profile data and other forms of spatial data. This technical field mainly covers multi-sensor data acquisition and alignment, three-dimensional coordinate data extraction and organization, spatial geometric feature analysis and three-dimensional graphics reconstruction, and is widely used in scenarios such as geological exploration, engineering surveying, urban planning and mineral resource development. In the process of development of this field, a relatively systematic technical system has been formed, including data acquisition technology, three-dimensional data analysis methods and geometric modeling tools, which can obtain and express high-precision three-dimensional information of complex natural environments or artificial structures.

[0003] Among them, the geological 3D modeling method based on multi-data fusion refers to a technical solution that, in response to the modeling needs of the 3D structure of geological bodies, introduces geological data from different sources, including borehole columnar data, geological profile mapping data, geophysical exploration results, and surface elevation measurement results, and adopts a method of unified spatial coordinate conversion, data interpolation fitting, and topological relationship constraints to jointly process different types of data, ultimately forming a 3D model that can be used to describe the internal spatial structure and stratigraphic distribution of geological bodies. The patent subject mainly focuses on how to integrate heterogeneous geological data under a unified geological coordinate system, how to use the spatial correlation between data to construct the continuity of point, line, surface, and body elements, and how to use multi-dimensional geometric splicing to restore the 3D structure of geological bodies.

[0004] Existing 3D object recognition and geological 3D modeling technologies rely heavily on coordinate unification, interpolation, and topological constraints after multi-sensor data acquisition to reconstruct spatial structures. These methods lack proportional analysis of the node spacing distribution within geological profiles and segmented processing of differential density. This results in an inability to dynamically adjust node elevation and angle responses in data-dense or data-sparse areas, leading to localized spatial distortion. Existing processes often rely on global fitting and unidirectional interpolation to construct 3D models. These processes lack fine-grained dynamic comparison of microscopic coupling features such as turning angles and segment lengths between different profiles, leading to geometric shifts and misaligned geologic features at the intersections of multiple profiles. For example, in mineral exploration, inconsistencies between layers across different survey lines often lead to misjudgments of geological body strike and thickness, impacting the accuracy of reserve calculations and engineering designs. This reliance on a single spatial unification constraint and lack of multi-layer coupling correction between nodes often fails to fully capture the detailed variations and true extension of complex geological bodies in 3D space. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a geological three-dimensional modeling method and system based on multi-data fusion.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a geological three-dimensional modeling method based on multi-data fusion, comprising the following steps:

[0007] S1: Obtain node coordinate sequences based on geological profile mapping results, extract horizontal spacing information between nodes, calculate average characteristics based on all node spacing, form node spacing ratio sequences, divide the profile into differential density segments based on the ratio characteristics, and obtain profile segment density labels;

[0008] S2: using the subdivided segments determined by the profile segment density labels, searching for the elevation and horizontal coordinates of the corresponding position nodes in the geophysical exploration profile, combining the elevation differences and spatial position differences between the nodes, to form a subdivided segment adjustment coefficient sequence;

[0009] S3: Directionally offset the node elevations in the high-density segment according to the segment adjustment coefficient sequence, determine the offset trend based on the node elevation correspondence information in the geological borehole columnar data, and correct the node angle variation in the low-density segment. After integration, a multi-layer profile reconstruction primitive is formed.

[0010] S4: For the nodes in the multi-layer profile reconstruction primitive, extract the length relationship of the line segments on both sides of the node and the node turning angle information to form a coupling feature set, and compare it with the similar coupling features of adjacent profile nodes to obtain the multi-layer profile coupling difference features.

[0011] As a further solution of the present invention, the profile segment density label includes a node spacing ratio sequence, a segment difference density feature, and a segment density identifier; the subdivided segment adjustment coefficient sequence includes an elevation difference coefficient, a spatial position difference coefficient, and a node adjustment sequence; the multi-layer profile reconstruction primitive includes a high-density segment node offset result, a low-density segment angle correction result, and a profile reconstruction node set; the multi-layer profile coupling difference feature includes a line segment length relationship difference, a turning angle difference, and a coupling feature difference between profiles.

[0012] As a further solution of the present invention, the specific steps of S1 are:

[0013] S101: Obtaining a node coordinate sequence in geological profile mapping results, determining the horizontal spacing between nodes based on the plane coordinate values ​​of adjacent nodes through the spatial difference relationship between the corresponding coordinates, and obtaining a node horizontal spacing sequence;

[0014] S102: According to the node horizontal spacing sequence, the node spacing information in the sequence is compared with the overall spacing characteristics of the sequence, and a node spacing ratio sequence is obtained according to the proportional relationship between the node spacing and the overall characteristics;

[0015] S103: Call the node spacing ratio sequence, identify the position exceeding the ratio difference threshold according to the difference between adjacent ratio values, divide the profile into continuous segments with different density, and mark the segments with corresponding density attributes to obtain the profile segment density label.

[0016] As a further solution of the present invention, the specific steps of S2 are:

[0017] S201: Obtaining subdivided segments determined by the profile segment density labels, calling corresponding position nodes on the geophysical exploration profile based on the subdivided segments, extracting elevation values ​​and horizontal coordinate values ​​recorded by the nodes, and generating node spatial distribution information;

[0018] S202: Based on the node spatial distribution information, for the elevation values ​​and horizontal coordinate values ​​between adjacent nodes, by comparing the elevation changes with the spatial position changes, forming associated data for describing elevation fluctuations and position shifts, and obtaining a sequence of position change feature values ​​between nodes;

[0019] S203: Calling the position change characteristic value sequence between the nodes, using the corresponding elevation difference and position difference for each set of characteristic values, determining the corresponding relationship and collecting the adjustment index set within the subdivided segment to obtain the subdivided segment adjustment coefficient sequence.

[0020] As a further solution of the present invention, the specific steps of S3 are:

[0021] S301: Based on the subdivided segment adjustment coefficient sequence, the elevation value of each node in the segment is called and combined with the corresponding coefficient, and the offset amplitude of the node elevation value is sequentially deduced within the corresponding segment. The offset amplitudes are orderly sorted according to the node positions within the same segment to obtain a node elevation offset amplitude sequence;

[0022] S302: Based on the node elevation offset amplitude sequence and in combination with the node elevation information in the geological borehole columnar data, the node offset amplitudes are sequentially compared with the columnar elevations, and the node offset trend within the segment is determined based on the positive and negative relationship of the differences, thereby obtaining a node offset trend sequence.

[0023] S303: Call the node offset trend sequence, retrieve the node angle change value in the low-density segment in turn and correct the node angle, then integrate the corrected low-density segment node angle with the node offset amplitude and offset trend of the high-density segment in turn to obtain a multi-layer profile reconstruction primitive.

[0024] As a further solution of the present invention, the specific steps of S4 are:

[0025] S401: Obtaining a node in the multi-layer cross-section reconstruction primitive, determining the proportional relationship between the lengths of the line segments on the left and right sides of the node based on the length information of the line segments on both sides of the node and the coordinates of the node's location, extracting a length relationship attribute using the combined features of the node and the line segments on both sides, and generating a node length relationship value;

[0026] S402: Based on the node length relationship value, call the angle information formed between the adjacent line segments on both sides of the node to determine the correspondence between the length relationship value and the angle value, form a joint description based on the geometric characteristics of the node, and obtain the node coupling feature value;

[0027] S403: For the node coupling characteristic value, call the coupling characteristic values ​​of the corresponding nodes of the adjacent sections, determine the degree of value difference through the corresponding relationship between the nodes, form the coupling characteristic change expression of the node position on the multi-layer section, and obtain the multi-layer section coupling difference feature.

[0028] As a further embodiment of the present invention, the coupling characteristic values ​​of the corresponding nodes of adjacent sections refer to the coupling characteristic values ​​extracted from the nodes on different sections in the multi-layer section data, specifically the node coupling characteristics formed by coupling the length relationship value of the line segments on both sides of each node with the angle value;

[0029] The coupling feature change expression is based on the obtained coupling feature values ​​of the corresponding nodes of adjacent sections. By comparing the coupling feature value differences between the corresponding nodes one by one, the data expression describing the change of node features between sections is obtained.

[0030] As a further embodiment of the present invention, the method further comprises:

[0031] S5: Based on the multi-layer profile coupling difference feature, combining the segment length relationship of the node and the elevation offset information in the previous step, as well as the original coordinates of the node in the geological profile mapping results and the geophysical exploration profile, to form geological three-dimensional splicing information;

[0032] The geological three-dimensional splicing information includes node segment length relationships, elevation offset information, and node original coordinates.

[0033] As a further solution of the present invention, the specific steps of S5 are:

[0034] S501: Based on the multi-layer profile coupling difference feature, the original coordinates and elevation offset information of the nodes in the geological profile mapping results and the geophysical exploration profile are called, and the corresponding segment length connection is extracted according to the position relationship of the nodes in the layer profile to obtain the node segment length difference feature value set;

[0035] S502: calling the node segment length difference feature value set, combining the node elevation offset information, and performing correspondence sorting between the segment length difference feature values ​​and the elevation offset values ​​for the nodes on the same layer to obtain a node offset relationship sequence;

[0036] S503: Based on the node offset relationship sequence and in combination with the original coordinate values ​​of the nodes, the spatial correspondence of the nodes in the vertical and horizontal directions is integrated to generate geological three-dimensional splicing information.

[0037] The geological 3D modeling system based on multi-data fusion includes:

[0038] The node ratio density module obtains the node coordinate sequence in geological profile mapping, calculates the horizontal spacing between nodes, forms a node spacing ratio sequence based on the proportional relationship between the node spacing and the average value, divides the profile difference density segment according to the ratio sequence, and obtains the profile segment density label;

[0039] The section adjustment coefficient module determines the subdivided section position based on the profile section density label, obtains the elevation and horizontal coordinates of the nodes in the geophysical exploration profile, calculates the elevation difference and horizontal difference between the nodes to form a ratio sequence, and obtains the subdivided section adjustment coefficient sequence;

[0040] The node elevation offset module extracts the elevations of high-density segment nodes and their corresponding elevations in the geological borehole columnar data based on the segment adjustment coefficient sequence, calculates the difference to form an offset sequence, operates the offset sequence with the segment adjustment coefficient sequence, and performs ratio screening based on the angle change amplitude of the low-density segment nodes to obtain multi-layer profile reconstruction primitives.

[0041] The multi-layer coupling difference module extracts the length relationship of the line segments on both sides of the node and the node turning angle based on the multi-layer profile reconstruction primitive to form a coupling feature set, and performs difference calculation with the feature set of the corresponding node of the adjacent profile to obtain the multi-layer profile coupling difference feature;

[0042] The three-dimensional splicing information module calls the multi-layer section coupling difference feature, combines the node segment length relationship, the multi-layer section reconstruction primitives and the original coordinates of the nodes in the difference section, performs cumulative calculation, and generates geological three-dimensional splicing information.

[0043] Compared with the prior art, the advantages and positive effects of the present invention are:

[0044] In the present invention, by dividing the node spacing ratio into density segments and combining the elevation and horizontal coordinate differences to form an adjustment coefficient, the node direction offset in high-density areas and the angle correction in low-density areas are accurately achieved, the expression of spatial morphological details is enhanced, the lengths on both sides of the nodes and the turning angles form a coupling feature, and the difference is extracted in the multi-section comparison. The segment length, elevation offset and original coordinates are combined to achieve three-dimensional splicing, making the geological body structure distribution more coherent. This type of continuous processing with density segmentation, node fine-tuning and coupling difference as the core effectively releases the spatial correlation between different geological data and significantly improves the accuracy of structural restoration and interpretation. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Schematic diagram of the steps of the present invention;

[0046] Figure 2 This is a schematic flow chart of step S1 of the present invention;

[0047] Figure 3 This is a schematic flow chart of step S2 of the present invention;

[0048] Figure 4 This is a schematic flow chart of step S3 of the present invention;

[0049] Figure 5 This is a schematic flow chart of step S4 of the present invention;

[0050] Figure 6 This is a schematic flow chart of step S5 of the present invention;

[0051] Figure 7 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0053] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0054] See also Figure 1 , a geological 3D modeling method based on multi-data fusion includes the following steps:

[0055] S1: Obtain node coordinate sequences based on geological profile mapping results, extract horizontal spacing information between nodes, combine the average characteristics calculated from all node spacings to form a node spacing ratio sequence, divide the profile into differential density segments based on the ratio characteristics, and obtain profile segment density labels;

[0056] S2: Using the subdivided segments determined by the profile segment density labels, retrieve the elevation and horizontal coordinates of the corresponding position nodes in the geophysical exploration profile, and combine the elevation differences and spatial position differences between the nodes to form a subdivided segment adjustment coefficient sequence;

[0057] S3: Based on the subdivided segment adjustment coefficient sequence, the node elevations in the high-density segment are directional shifted, and the shift trend is determined by combining the node elevation correspondence information in the geological drill hole columnar data. The node angle variation amplitude is corrected in the low-density segment, and the multi-layer profile reconstruction primitive is formed after integration processing;

[0058] S4: For the nodes in the multi-layer profile reconstruction primitive, the length relationship of the line segments on both sides of the node and the node turning angle information are extracted to form a coupling feature set, which is then compared with the similar coupling features of adjacent profile nodes to obtain the multi-layer profile coupling difference features;

[0059] S5: Based on the multi-layer profile coupling difference characteristics, combined with the node segment length relationship and the elevation offset information in the previous step, as well as the original coordinates of the nodes in the geological profile mapping results and geophysical exploration profiles, the geological three-dimensional splicing information is formed.

[0060] The profile segment density label includes the node spacing ratio sequence, segment difference density characteristics, and segment density identification. The subdivided segment adjustment coefficient sequence includes the elevation difference coefficient, the spatial position difference coefficient, and the node adjustment sequence. The multi-layer profile reconstruction primitives include the high-density segment node offset results, the low-density segment angle correction results, and the profile reconstruction node set. The multi-layer profile coupling difference characteristics include the line segment length relationship difference, the turning angle difference, and the coupling feature difference between profiles. The geological three-dimensional splicing information includes the node segment length relationship, the elevation offset information, and the original coordinates of the node.

[0061] See also Figure 2 , the specific steps of S1 are:

[0062] S101: Obtaining a node coordinate sequence in geological profile mapping results, determining the horizontal spacing between nodes based on the plane coordinate values ​​of adjacent nodes through the spatial difference relationship between the corresponding coordinates, and obtaining a node horizontal spacing sequence;

[0063] When obtaining the node coordinate sequence in the geological profile mapping results, first extract the node data from the completed profile mapping file, which is usually a DWG or DXF format drawing. Output the node coordinate list through CAD tools or directly convert it into a CSV file. Read the X and Y coordinate values ​​of each node in sequence, reading them one by one according to the node number from small to large. Then, starting from the first node, gradually calculate the horizontal distance between the current node and the next node backward. Specifically, take the coordinates of node 1 as 100.0 and 50.0, and the coordinates of node 2 as 103.0 and 54.0 as an example. First calculate the X coordinate difference as 3.0 and the Y coordinate difference as 4.0. The horizontal distance can be understood as the geometric relationship between 3.0 and 4.0. The length value of 5.0 is obtained, and the same operation is continued for all nodes to obtain the horizontal distances of all adjacent nodes from node 1 to node 2, node 2 to node 3, etc., to form a complete node horizontal spacing sequence. If there are some values ​​in the sequence that are significantly larger or smaller, a simpler method can be used to judge first, such as using the sum of absolute differences for preliminary comparison. If a single distance is found to be greater than twice the average level, it is recorded as an abnormality. Subsequently, the corresponding node is located in the CAD software to check whether there are any measurement point number errors or repeated annotations. The whole process repeatedly traverses the node list to ensure that the horizontal spacing between all nodes can be correctly filed into the spacing sequence, and finally obtain reliable node horizontal spacing sequence data.

[0064] S102: Based on the node level spacing sequence, the node spacing information in the sequence is compared with the overall spacing characteristics of the sequence, and a node spacing ratio sequence is obtained based on the proportional relationship between the node spacing and the overall characteristics;

[0065] According to the sequence of node horizontal spacing, it is necessary to first perform overall characteristic statistics on all spacings, including calculating the average value, overall distribution range and fluctuation of a set of horizontal spacings. The average value can be understood as the sum of all node horizontal spacings divided by the number of node spacings. For example, if the spacings are 5.0, 4.8, 5.3, 7.5, 4.6 and 5.1 respectively, the total average value is the sum of these values ​​divided by 6, which is about 5.4. Then, each node spacing is compared with the average value to calculate the ratio of each node spacing to the average value. For example, the ratio of 5.0 to 5.4 is about 0.93, and the ratio of 7.5 to 5.4 is about 1.3. 9. The node spacing ratio sequence is gradually obtained as 0.93, 0.89, 0.98, 1.39, 0.86, and 0.95. These ratios are then differentiated by thresholds. For example, ratios between 0.8 and 1.2 are considered normal, those exceeding 1.2 are considered significantly high, and those less than 0.8 are considered significantly low. If the ratio 1.39 falls within the significantly high range, it is marked as a high-spacing node in the sequence, and the ratio 0.86 falls within the normal range and is still marked as normal. By traversing the ratios of all node spacing to the average value, the node spacing ratio sequence is identified and a data basis is provided for subsequent further analysis.

[0066] S103: Calling the node spacing ratio sequence, identifying the locations exceeding the ratio difference threshold according to the differences between adjacent ratio values, dividing the profile into continuous segments with different density, and labeling the segments with corresponding density attributes to obtain the profile segment density labels;

[0067] After calling the node spacing ratio sequence, the difference between each pair of adjacent ratios is analyzed step by step. The ratio difference value is calculated by subtracting the previous ratio from the latter ratio and then taking the absolute value. For example, the difference between 0.89 and 0.93 is about 0.04, and the difference between 1.39 and 0.98 is about 0.41. The threshold value is set to 0.3 in the ratio difference value sequence to distinguish normal changes from significant jumps. If 0.41 exceeds 0.3, it is marked as a significant difference point. Then, the entire sequence is divided into several continuous segments based on the significant difference points. For example, the first segment from the beginning to the third node contains only the ratios of 0.93, 0.89, and 0.98. The fourth segment The points form a segment with a ratio of 1.39. The fifth and sixth nodes form the next segment with ratios of 0.86 and 0.95. The density attribute of each segment is then judged based on the average ratio value. If the average ratio of the segment is between 0.8 and 1.2, it is marked as medium density. If it is greater than 1.2, it is marked as high density. If it is less than 0.8, it is marked as low density. For example, if the average ratio of the second segment is 1.39, it is marked as high density. Other segments with an average ratio of around 0.9 are marked as medium density. This completes the process of converting the node spacing ratio sequence into a segment list with density attribute annotations. Finally, the segment numbers and their corresponding density attributes can be listed in the geological profile or data report as important results of the profile analysis.

[0068] See also Figure 3 , the specific steps of S2 are:

[0069] S201: Obtaining subdivided segments determined by profile segment density labels, calling corresponding position nodes on the geophysical exploration profile based on the subdivided segments, extracting elevation values ​​and horizontal coordinate values ​​recorded by the nodes, and generating node spatial distribution information;

[0070] Once the spatial distribution of nodes is established, each pair of adjacent nodes within a segment is processed one by one. The elevation values ​​are sequentially subtracted to determine the elevation change between the adjacent points. For example, if nodes X is between 10m and 20m, and H is between 121.3m and 120.9m, the elevation difference is -0.4m, indicating a 0.4m drop in elevation from 10m to 20m. The horizontal distance between the two points, ΔX (here, 10m), is also recorded. Continuing to process all nodes within the segment sequentially, a series of elevation differences and corresponding horizontal distances is generated, such as (-0.4m, 10m), (0.3m, 10m), (-0.2m, 10m), and so on. To reflect this relationship, the elevation change per unit horizontal distance, or slope, can be calculated. This value is the elevation difference divided by the horizontal distance. For example, -0.4 divided by 10 gives -0.04, representing a 4cm drop in elevation for every 1m horizontal distance. To facilitate subsequent processing of different situations based on the degree of slope change, a benchmark value can be set. For example, set to ±0.02, which means that when the absolute value of the slope is greater than 0.02, it is considered a significant change, and when it is less than 0.02, it is considered a moderate change. Therefore, the above -0.04 is judged to be a significant decrease because it is less than -0.02. If a certain segment is 0.03 in the subsequent calculation, it is considered a significant increase. This information and the numerical values ​​are summarized one by one into a sequence to form a complete list of characteristic values ​​of position changes between nodes. Arranged in sequence, it can be read as (from 10m to 20m: elevation difference -0.4m, horizontal distance 10m, slope -0.04). (From 20m to 30m: elevation difference 0.3m, horizontal distance 10m, slope 0.03), and so on, until the end of the subdivided segment.

[0071] S202: Based on the spatial distribution information of the nodes, the elevation values ​​and horizontal coordinate values ​​between adjacent nodes are compared with the elevation changes and spatial position changes to form associated data for describing elevation fluctuations and position shifts, thereby obtaining a sequence of characteristic values ​​of position changes between nodes.

[0072] The specific calculation formula by comparing the elevation change and the spatial position change is:

[0073] ;

[0074] Calculate the composite change ratio between nodes , which is used to subsequently obtain the characteristic value sequence of position changes between nodes;

[0075] in, Representative With the The dimensionless composite change ratio between adjacent nodes, Representative The elevation value of each node (unit: m), Representative The elevation value of each node (unit: m), Representative The horizontal coordinate value of each node (unit: m), Representative The horizontal coordinate value of each node (unit: m), Representative The other horizontal coordinate value of the node (unit: m), Representative The other horizontal coordinate value of the node (unit: m), Represents the average value of all node elevation values ​​(unit: m), Represents the average absolute deviation of all node elevation values ​​in the node set relative to the average elevation value. Represents the total number of nodes, Represents the selected benchmark length for elevation normalization (unit: m);

[0076] The meaning and acquisition methods of the parameters are as follows:

[0077] Indicates the The elevation value of each node (in meters) is obtained through the GNSS elevation measurement data of the total station;

[0078] Indicates the The elevation value of each node (in meters) is also obtained through the measured elevation value;

[0079] It represents the average value of all node elevation values, which is calculated by the arithmetic mean of all node elevation sampling values;

[0080] Indicates the With the The horizontal x-coordinate value of each node (in meters) is derived from the coordinate acquisition results of the survey coordinate locator or geographic information system;

[0081] Indicates the With the The horizontal y coordinate value of each node (in meters), the data collection method is the same as above Indicates the total number of nodes used for calculation, which is directly obtained by the total number of spatial node layouts;

[0082] Represents the normalized benchmark length, which is selected based on the average value of the maximum horizontal spacing between nodes in the area and is calculated as follows: ,in is the number of all adjacent point pairs, is the reference point coordinate;

[0083] The actual sampling data of a monitoring area are as follows:

[0084] meters (measured by laser scanner);

[0085] meters (measured by laser scanner);

[0086] In the node collection , whose elevation values ​​are 113.6, 116.2, 114.9, 112.8, and 117.5 meters respectively;

[0087] Node coordinates: , , , , all coordinate values ​​are in meters;

[0088] Normalized reference length Meters, calculated based on the average of the maximum horizontal distances between adjacent nodes

[0089] Calculate the intermediate values ​​of parameters: rice;

[0090] ;

[0091] The mean absolute deviation is: rice;

[0092] After normalization: ;

[0093] Substitute into the numerator: , after normalization: ;

[0094] Horizontal distance part: ;

[0095] Overall molecule: ;

[0096] Substituting into the denominator: , , the product is: , the absolute value is still 2.511;

[0097] Denominator as a whole: ;

[0098] Substitute into the main formula: ;

[0099] The results show that the composite change ratio between nodes is 0.824. The larger the value, the more significant the difference between the elevation change and the horizontal distance. This result is used to determine the degree of mutation in the sequence of characteristic values ​​of position changes between nodes. Combining multiple node sequences can further identify areas sensitive to elevation disturbances.

[0100] This formula forms a composite change value of the coupled elevation and horizontal position by taking the square root of the sum of the squares of the normalized elevation difference and the horizontal Euclidean distance in the numerator and then subtracting the average elevation deviation in the node set. In the calculation, the normalized elevation difference is divided by the reference length to eliminate the problem of direct addition of different physical dimensions. The square and square root structure is used to calculate the straight-line distance between two points in the plane coordinate system. The absolute value is used to ensure that the quantified offset is non-negative and represents the overall degree of deviation. The denominator is obtained by multiplying the normalized elevation difference with the horizontal coordinate difference to obtain the joint change amplitude, and then adding one to avoid the risk of zero division and maintain numerical smoothness. The absolute value is taken to ensure that the direction does not affect the scale determination. The overall operation logic quantifies the comprehensive change intensity between nodes in the form of a superposition of elevation and horizontal coordinates. This allows local elevation fluctuations and horizontal displacement to be simultaneously incorporated into the numerator calculation of the change factor. The coupling suppression effect of the denominator balances extreme change values, ultimately forming a dimensionless ratio reflecting the coupled offset characteristics of terrain elevation and horizontal position between node pairs.

[0101] The specific significance of calculating the composite change ratio between nodes lies in constructing a quantitative indicator that can comprehensively measure the landform or site change status between two points by simultaneously integrating the elevation fluctuation amplitude and horizontal position movement distance between nodes. This ratio not only reflects the elevation difference between adjacent nodes in the vertical direction, but also combines their spatial spacing in the plane, thereby effectively describing the overall offset intensity caused by terrain fluctuation, settlement, uplift or horizontal displacement within the local area. The size of this ratio can be used to identify potential abnormal change sections in surface morphology or engineering structures, providing a basis for subsequent sorting of the degree of difference between nodes and locating locations where deformation risks may exist.

[0102] S203: Calling the sequence of position change eigenvalues ​​between nodes, for each set of eigenvalues, using the corresponding elevation difference and position difference, determining the corresponding relationship and collecting the adjustment index set within the subdivided segment to obtain the subdivided segment adjustment coefficient sequence;

[0103] Using the sequence of characteristic values ​​of inter-node position change obtained from the point-by-point calculations above, we can extract the slope and horizontal distance values ​​for each segment, group by group, and combine them with a pre-selected adjustment sensitivity parameter to form a sequence of adjustment coefficients. For example, a sensitivity coefficient of 2.5 can be chosen. By multiplying the slope value by the sensitivity coefficient and adding it to 1, we can obtain the adjustment coefficient for each segment. For a slope of -0.04, for example, the calculation is 1 plus 2.5 times -0.04, resulting in 0.9, indicating a 90% adjustment ratio for this segment. For a slope of 0.03, the adjustment coefficient is 1.075, meaning a 107.5% adjustment ratio. To facilitate batch processing, a spreadsheet can be populated with slope data in one column and a formula in another column: 1 plus 2.5 times the slope value for the corresponding row. This automatically calculates the adjustment coefficient for each row. After repeated calculations, a sequence of adjustment coefficients is formed within the entire subdivision, such as (0.9, 1.075, 0.95, 1.02...). This sequence can be directly used to adjust the earthwork volume, elevation markup, or other engineering data of the corresponding section in the design drawings or calculations, ensuring that the subdivision changes are reasonably distributed according to the actual measured elevation fluctuations.

[0104] See also Figure 4 , the specific steps of S3 are:

[0105] S301: Based on the subdivided segment adjustment coefficient sequence, the elevation value of each node in the segment is called and combined with the corresponding coefficient, and the offset amplitude of the node elevation value is deduced in the corresponding segment in turn. The offset amplitudes are sorted in order according to the node position in the same segment to obtain the node elevation offset amplitude sequence;

[0106] First, the entire line needs to be divided into multiple small sections of equal length. For example, a line length of 500m can be divided into small sections of 10m each, totaling 50 sections. These small sections are then graded according to geological weathering conditions or lithological differences. For example, the fully weathered section is assigned an adjustment coefficient of 0.6, the strongly weathered section is 0.8, the weakly weathered section is 1.0, and the unweathered section is 1.2. They are arranged in sequence. Then, the original elevation value of each node is called according to its location. For example, the first node is 105.3m and the second node is 104.8m. The adjustment coefficient of the section where the node is located is directly multiplied by the node elevation. For example, the first node is in the fully weathered section. The corresponding adjusted value is 63.2m. The second node is in the strongly weathered section, and the corresponding value is 83.8m. Continue to process all nodes and calculate in sequence to obtain a set of node elevation offset amplitude sequences. Through spreadsheet tools or script batch operations, the node elevations and section adjustment coefficients can be input in batches to automatically generate offset amplitude values. These offset amplitude values ​​are then sorted in order according to the mileage position of the nodes, ensuring that they are arranged in order from the starting point to the end point in the list. For example, the starting point 0m corresponds to 63.2m, the mileage 20m corresponds to 83.8m, and the mileage 30m corresponds to 104.2m, and step by step to the end point of the line, completing the entire node elevation offset amplitude sequence arrangement.

[0107] S302: Based on the node elevation offset amplitude sequence and the node elevation information in the geological borehole columnar data, the node offset amplitudes are sequentially compared with the columnar elevations, and the node offset trend within the segment is determined based on the positive and negative relationship of the differences, thereby obtaining a node offset trend sequence.

[0108] According to the sequence of node elevation offset amplitudes that have been sorted, it is necessary to further compare the columnar elevation values ​​of the corresponding nodes in the borehole columnar geological data one by one. For example, at the first node position, the offset amplitude is 63.2m, and the position in the borehole columnar data is 60.2m, with a difference of 3m. In this case, if the difference is positive, it is considered that the offset trend is downward, and if it is negative, it is considered that the offset trend is upward. The offset direction can be represented by +1 for downward and -1 for upward. According to this point-by-point comparison of the difference, for example, the offset amplitude of the second node is 83.8m, the columnar data is 81.0m, the difference is 2.8m, and the direction is +1. Section 5 The point offset amplitude is 91.5m, the columnar data is 95.0m, the difference is −3.5m, and the direction is −1. This method continuously determines the offset trend from the first node to the last node, forming a complete node offset trend sequence. For example, the sequence obtained is [+1, +1, −1, +1, −1, ...]. This process can be calculated in a spreadsheet using a column to subtract the offset amplitude from the columnar elevation and directly mark the positive and negative values ​​to represent the offset direction using conditional formatting. Alternatively, a simple script can be used to traverse the two columns of data and automatically record the direction mark. Ultimately, the offset trend of each node in the corresponding segment is clearly defined in the sequence.

[0109] S303: Calling the node offset trend sequence, sequentially retrieving node angle change values ​​in low-density sections and correcting the node angles, and then sequentially integrating the corrected low-density section node angles with the node offset amplitudes and offset trends of high-density sections to obtain multi-layer profile reconstruction primitives;

[0110] The specific calculation formula for sequentially retrieving the node angle change value and correcting the node angle in the low-density section is:

[0111] ;

[0112] Calculate the node correction angle value , then the corrected low-density segment node angles are sequentially integrated with the node offset amplitude and offset trend of the high-density segment to obtain the multi-layer profile reconstruction primitives;

[0113] in, Representative The corrected angle value of the low-density segment node, Representative Node and The angle change value calculated between the associated nodes, Representative Node and The weight factor of each node, Representative The initial angle offset value of each node, Represents the average value of the initial angle offset values ​​of all nodes in the current low-density segment, Representative The standard deviation of the angle change of each node, Representatives and The total number of nodes associated with a node;

[0114] Angle correction value In the calculation of , the node is located in the low-density segment, and the total number of adjacent nodes is , the cross-section is sampled by a 3D laser scanning system to obtain the following angle change values ​​and initial angle offset values:

[0115] The angle changes between node 4 and its five associated nodes are as follows, obtained by taking the arc cosine difference of the direction vector changes of the nodes within five consecutive data frames. The angle units are all degrees:

[0116] ;

[0117] Weighting Factor Calculated by the Euclidean distance between nodes and the signal-to-noise ratio measurement, according to the quantitative standard:

[0118] The weight is based on the normalized value of the signal-to-noise ratio, and the quantification method is the signal-to-noise ratio of the current comparison node. Perform Sigmoid normalization transformation on the ratio of the average signal-to-noise ratio of all associated nodes;

[0119] The resulting normalized weights are:

[0120] ;

[0121] Compute the weighted sum this way:

[0122] ,

[0123] Original angle offset value of node 4 The initial angle offset values ​​of the five nodes in the low-density section are as follows:

[0124] 3.1, 2.7, 3.4, 2.8, 3.0, average:

[0125] ;

[0126] therefore:

[0127] ;

[0128] Standard deviation The calculation method is based on the Average square root of the squares of the differences between the values ​​and the mean:

[0129] average value: ;

[0130] Standard Deviation: ;

[0131] The denominator is calculated as: ;

[0132] Finally, substitute all of them into the main formula: ;

[0133] The result shows that the correction angle of node 4 is 1.36 degrees, indicating that the original angle value has a certain degree of discreteness. The correction value can be used to reconstruct and integrate the direction vectors of multi-layer profile nodes. This value will be brought into the subsequent node position vector fitting or contour combination logic steps as the reference input for the final profile vector field direction determination.

[0134] The operational logic of this formula is to form a weighted cumulative value by multiplying the angle change values ​​between nodes by their corresponding weights one by one and then summing them up. This value reflects the contribution strength of each adjacent node to the angle correction of the current node. The absolute value of the difference between the node's own offset and the average offset of the segment multiplied by the total number of nodes is then subtracted and the square root is taken to reflect the degree of deviation between the node and the overall structural consistency and to weaken the influence of extreme deviations in a square root manner. The denominator uses the square root of the standard deviation of the node angle change divided by the number of adjacent nodes, plus one. The purpose is to increase the denominator when the fluctuation is large to suppress the abnormal correction amplitude and ensure that the angle correction tends to be smooth under the influence of multiple adjacent nodes. The overall formula ensures that the output angle correction value is non-negative through the absolute value, forming a set of angle correction logic framework based on the joint action of local weighting, deviation adjustment and variance smoothing.

[0135] The node correction angle value represents the numerical adjustment result of the original angle of the node by comprehensively considering the angle change relationship between the node and its adjacent nodes, the importance of each adjacent node, and the offset degree of the node itself relative to the same section as a whole when analyzing the multi-node profile structure. This value is directly used to replace the original angle information of the node to reflect the optimization direction of the node based on the consistency of the local structural trend and the overall offset, ensuring that the subsequent generation of multi-layer profile contours, reconstruction of node direction vectors, and calculation of node spatial positions can be based on angle input data that is more consistent with the actual structural continuity and local smoothness characteristics.

[0136] See also Figure 5 , the specific steps of S4 are:

[0137] S401: Obtain a node in a multi-layer cross-section reconstruction primitive, determine the proportional relationship between the lengths of the line segments on the left and right sides of the node based on the length information of the line segments on both sides of the node and the coordinates of the node location, extract the length relationship attribute using the combined features of the node and the line segments on both sides, and generate a node length relationship value;

[0138] To obtain the nodes in the multi-layer cross-section reconstruction primitive, first directly read the cross-section coordinate information of each layer through 3D scanning or CAD cross-section data files, traverse each cross-section contour line from beginning to end, and collect the coordinates of each node in a list. For example, for a typical part's multi-layer cross-section file, there are usually about 150 nodes per layer. After all the node data are accumulated, a complete node point set is formed. Then, each node is selected in turn, and its position in the current cross-section is first identified. The previous node on the left and the next node on the right of the node are recorded. The length of the line segments on both sides is obtained in turn through the coordinate difference between the nodes. If the distance between the two points on the left side of a node is 2.3mm and the right side is 4.6mm, then the ratio of the length of the left and right sides of the node can be directly expressed as 2.3 to 4.6, that is, the ratio value is approximately 0.5. Then list the left and right ratio values ​​of all nodes one by one to form a sequence, continue to check this sequence, and limit the ratio value to a reasonable range of 0.4 to 2.5. If the ratio is less than 0.4 or greater than 2.5, the corresponding node will be placed in a separate sequence and marked. After that, the nodes in the normal range are combined with the node's own coordinates and the length information on both sides to generate a length relationship attribute combination. For example, for a node with a ratio value of 0.5, the recordable length relationship attribute is the total length of 2.3 plus 4.6 to get 6.9, the difference is 2.3 minus 4.6 to get -2.3, and then attached with a ratio value of 0.5. After all nodes execute this process, a complete node length relationship attribute sequence is formed, and the ratio value is extracted separately from it to form a numerical sequence of node length relationship for subsequent calculations.

[0139] S402: Based on the node length relationship value, call the angle information formed between the adjacent line segments of the node to determine the correspondence between the length relationship value and the angle value, form a joint description based on the geometric characteristics of the node, and obtain the node coupling feature value;

[0140] According to the value of the node length relationship, the angle information formed by the adjacent line segments on both sides of the node is further called. First, the position of a node is taken out from the node list. Then, the two line segments on its left and right are viewed with the node as the center. Their directions are directly compared in text form through the three-dimensional coordinate points. For example, if the left line segment extends 1 in the X direction and 2 in the Y direction, and the right line segment extends 2 in the X direction and the Y direction remains basically unchanged, then the angle is between 60 and 70 degrees through geometric understanding. In this way, the angle value of each node is determined by its coordinate position change one by one, and then the length ratio value of the node is paired with the corresponding angle value. For example, if the ratio value is 0.5 and the angle is 63.5 degrees, it can be recorded as (0. 5,63.5°), continue to traverse all nodes and form a corresponding sequence of length ratio values ​​and angles to provide data for further analysis. For the specific mechanical blade cavity detection situation, the reasonable angle range can be set as 30° to 150°. For angle values ​​less than 30° or greater than 150°, they can be directly classified as too small or too large. The geometric properties of the node are then comprehensively described as a combination of length ratio and angle. For example, for a node with a ratio value of 0.5 and an angle of 63.5°, its coupling eigenvalue can be literally understood as approximately equal to the ratio value multiplied by the angle size, which is approximately 31.7. This process is then repeated to obtain the comprehensive coupling values ​​of all nodes to form a multi-node coupling characteristic sequence.

[0141] S403: For the node coupling characteristic value, call the coupling characteristic values ​​of the corresponding nodes of the adjacent sections, determine the degree of value difference through the corresponding relationship between the nodes, form the coupling characteristic change expression of the node position on the multi-layer section, and obtain the multi-layer section coupling difference feature;

[0142] For the coupling eigenvalues ​​of all nodes, a comparison is started on the multi-layer profile. For example, the current profile layer and the next layer are matched by coordinate matching or shortest distance matching. The corresponding nodes in the previous and next layers are found in turn, and their coupling eigenvalues ​​are taken out for comparison. The difference in coupling values ​​is calculated. For example, if the coupling value of the current layer node is 31.7 and the corresponding value of the next layer is 29.8, the difference is 1.9. In this way, a set of coupling difference sequences is obtained by processing all corresponding nodes. During the workpiece inspection process, a threshold of 2.5 can be set for screening. If the difference of the node is greater than 2.5, it is identified as a node with a large difference. The proportion of these nodes in all nodes is further counted. For example, the statistical results show that 90% of the nodes have a difference between 1.5 and 2.5, which can be understood as within the normal range. The remaining approximately 10% of the nodes may be concentrated in a larger difference segment. Their spatial position and profile layer number need to be additionally recorded for later structural evaluation or correction, and finally a node coupling difference characteristic representation on the multi-layer profile is formed.

[0143] See also Figure 6, the specific steps of S5 are:

[0144] S501: Based on the multi-layer profile coupling difference feature, the original coordinates and elevation offset information of the nodes in the geological profile surveying results and the geophysical exploration profile are called, and the corresponding segment length connection is extracted according to the position relationship of the nodes in the layer profile to obtain the node segment length difference feature value set;

[0145] Based on the demarcation point data extracted from the multi-layer geological profile mapping results, the node number and corresponding coordinate information, including X, Y coordinates and Z elevation, are read point by point. At the same time, the elevation of the corresponding node and the offset in the X and Y directions are obtained from the geophysical profile results. The corresponding index between the nodes is established. For adjacent nodes in the same layer, the spatial distance between the two points is first calculated based on the original mapping coordinates. For example, the distance from node 1 to node 2 is about 11.36m. Then, the distance between the same two points is calculated based on the geophysical coordinates to be about 12.68m. By subtracting the two, the segment length difference of 1.32m is obtained. This calculation method is repeated. For all nodes, a set of difference values ​​such as 1.32m, 0.95m and 1.10m is obtained. This set of data is partitioned according to the absolute value size, and the threshold is set to 1m. All values ​​greater than 1m are marked as prominent change segments, represented by 1, and less than or equal to 1m are relatively stable segments, represented by 0. A set of sequences similar to [1,0,1] is obtained, which is used to determine the significance of the spatial offset of the nodes in the subsequent steps. Subsequently, these nodes are marked with different colors in GIS in a graphical way, which can clearly distinguish which sections have obvious extension or contraction on the profile map, and finally form a set of segment length difference feature values ​​for subsequent analysis.

[0146] S502: calling the node segment length difference feature value set, combining the node elevation offset information, and performing correspondence sorting between the segment length difference feature values ​​and the elevation offset values ​​for the nodes on the same layer to obtain a node offset relationship sequence;

[0147] After obtaining the node segment length difference feature value set, this data is paired with the node elevation offset information. For example, the segment length difference of nodes 1, 2, and 3 is 1.32m, 0.95m, and 1.10m, and the elevation offset is 1m. They are respectively combined into node 1 (1.32, 1), node 2 (0.95, 1), and node 3 (1.10, 1). Then, all segment length differences are uniformly converted according to the maximum value. Taking the maximum value of 1.32m as the benchmark, node 2 is about 72%, node 3 is about 83%, and the elevation offset is 100% in this case. Then, the segment length is proportionally converted by assigning weights. The difference accounts for 60%, and the elevation offset accounts for 40%, thus obtaining a comprehensive node offset value, which is approximately 83% for node 2 and approximately 90% for node 3. Based on this comprehensive value, the offset significance is divided according to the interval, with 85% to 100% being recorded as a higher segment, 60% to 85% as a medium segment, and below 60% as a lower segment. Therefore, nodes 1, 2, and 3 are all judged to be higher segments. This offset relationship sequence can be formed into a data table in spreadsheet software, and can also be drawn into a line graph in drawing software to facilitate viewing of the overall fluctuation trend. Finally, this node offset relationship sequence is stored and prepared for entering the three-dimensional splicing link.

[0148] S503: Based on the node offset relationship sequence and the original coordinate values ​​of the nodes, the vertical and horizontal spatial correspondences of the nodes are integrated to generate geological three-dimensional splicing information;

[0149] According to the obtained node offset relationship sequence, the spatial coordinates of the nodes are adjusted one by one according to this ratio. First, the elevation is superimposed in the longitudinal direction according to the offset ratio. For example, node 1 is offset by about 96%. The benchmark value for longitudinal superposition is 2m. Therefore, the elevation of node 1 changes from 50m to about 52m, node 2 increases from 48m to about 50m, and node 3 increases from 47m to about 49m. At the same time, fine-tuning is also performed in the plane direction according to the same offset ratio. The benchmark 1m is used to distribute half in the X and Y directions. Node 1 increases by about 0.68m in the X direction and about 0.68m in the Y direction, so the new coordinates are 100.68m and 200.68m. The new X and Y coordinates of all nodes are calculated in this way. Finally, the updated X, Y, and Z coordinates of all nodes are compiled and output to generate a geological 3D splicing information, which can be used to import into point cloud software to view the distribution status of nodes in three-dimensional space.

[0150] See also Figure 2 , a geological 3D modeling system based on multi-data fusion, including:

[0151] The node ratio density module obtains the node coordinate sequence in geological profile mapping, calculates the horizontal spacing between nodes, forms a node spacing ratio sequence based on the proportional relationship between the node spacing and the average value, divides the profile difference density segment according to the ratio sequence, and obtains the profile segment density label;

[0152] The section adjustment coefficient module determines the subdivision section location based on the profile section density label, obtains the elevation and horizontal coordinates of the nodes in the geophysical exploration profile, calculates the elevation difference and horizontal difference between the nodes to form a ratio sequence, and obtains the subdivision section adjustment coefficient sequence;

[0153] The node elevation offset module extracts the elevations of high-density segment nodes and their corresponding elevations in the geological borehole columnar data based on the segment adjustment coefficient sequence, calculates the difference to form an offset sequence, and then operates on the offset sequence and the segment adjustment coefficient sequence. The offset sequence is then screened by combining the angle change amplitude of the low-density segment nodes to obtain a multi-layer profile reconstruction primitive.

[0154] The multi-layer coupling difference module extracts the length relationship of the line segments on both sides of the node and the node turning angle based on the multi-layer profile reconstruction primitive to form a coupling feature set. The difference calculation is performed with the feature set of the corresponding nodes of the adjacent profile to obtain the multi-layer profile coupling difference feature;

[0155] The 3D splicing information module calls the multi-layer profile coupling difference feature, combines the node segment length relationship, multi-layer profile reconstruction primitives and the original coordinates of the nodes in the difference profile, performs cumulative calculations, and generates geological 3D splicing information.

[0156] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A geological 3D modeling method based on multi-data fusion, characterized in that: The following steps are involved: S1: Obtain node coordinate sequences based on geological profile mapping results, extract horizontal spacing information between nodes, calculate average characteristics based on all node spacing, form node spacing ratio sequences, divide the profile into differential density segments based on the ratio characteristics, and obtain profile segment density labels; S2: using the subdivided segments determined by the profile segment density labels, searching for the elevation and horizontal coordinates of the corresponding position nodes in the geophysical exploration profile, combining the elevation differences and spatial position differences between the nodes, to form a subdivided segment adjustment coefficient sequence; S3: Directionally offset the node elevations in the high-density segment according to the segment adjustment coefficient sequence, determine the offset trend based on the node elevation correspondence information in the geological borehole columnar data, and correct the node angle variation in the low-density segment. After integration, a multi-layer profile reconstruction primitive is formed. S4: for the nodes in the multi-layer profile reconstruction primitive, extract the length relationship of the line segments on both sides of the node and the node turning angle information to form a coupling feature set, and compare it with the similar coupling features of adjacent profile nodes to obtain the multi-layer profile coupling difference feature; The method further comprises: S5: Based on the multi-layer profile coupling difference feature, combining the segment length relationship of the node and the elevation offset information in the previous step, as well as the original coordinates of the node in the geological profile mapping results and the geophysical exploration profile, to form geological three-dimensional splicing information; The geological three-dimensional splicing information includes node segment length relationships, elevation offset information, and node original coordinates.

2. The geological three-dimensional modeling method based on multi-data fusion according to claim 1, characterized in that: The profile segment density label includes a node spacing ratio sequence, a segment difference density feature, and a segment density identifier; the subdivided segment adjustment coefficient sequence includes an elevation difference coefficient, a spatial position difference coefficient, and a node adjustment sequence; the multi-layer profile reconstruction primitive includes a high-density segment node offset result, a low-density segment angle correction result, and a profile reconstruction node set; the multi-layer profile coupling difference feature includes a line segment length relationship difference, a turning angle difference, and a coupling feature difference between profiles.

3. The geological three-dimensional modeling method based on multi-data fusion according to claim 1, characterized in that: The specific steps of S1 are: S101: Obtaining a node coordinate sequence in geological profile mapping results, determining the horizontal spacing between nodes based on the plane coordinate values ​​of adjacent nodes through the spatial difference relationship between the corresponding coordinates, and obtaining a node horizontal spacing sequence; S102: According to the node horizontal spacing sequence, the node spacing information in the sequence is compared with the overall spacing characteristics of the sequence, and a node spacing ratio sequence is obtained according to the proportional relationship between the node spacing and the overall characteristics; S103: Call the node spacing ratio sequence, identify the position exceeding the ratio difference threshold according to the difference between adjacent ratio values, divide the profile into continuous segments with different density, and mark the segments with corresponding density attributes to obtain the profile segment density label.

4. The geological three-dimensional modeling method based on multi-data fusion according to claim 3 is characterized in that: The specific steps of S2 are: S201: Obtaining subdivided segments determined by the profile segment density labels, calling corresponding position nodes on the geophysical exploration profile based on the subdivided segments, extracting elevation values ​​and horizontal coordinate values ​​recorded by the nodes, and generating node spatial distribution information; S202: Based on the node spatial distribution information, for the elevation values ​​and horizontal coordinate values ​​between adjacent nodes, by comparing the elevation changes with the spatial position changes, forming associated data for describing elevation fluctuations and position shifts, and obtaining a sequence of position change feature values ​​between nodes; S203: Calling the position change characteristic value sequence between the nodes, using the corresponding elevation difference and position difference for each set of characteristic values, determining the corresponding relationship and collecting the adjustment index set within the subdivided segment to obtain the subdivided segment adjustment coefficient sequence.

5. The geological three-dimensional modeling method based on multi-data fusion according to claim 4 is characterized in that: The specific steps of S3 are: S301: Based on the subdivided segment adjustment coefficient sequence, the elevation value of each node in the segment is called and combined with the corresponding coefficient, and the offset amplitude of the node elevation value is sequentially deduced within the corresponding segment. The offset amplitudes are orderly sorted according to the node positions within the same segment to obtain a node elevation offset amplitude sequence; S302: Based on the node elevation offset amplitude sequence and in combination with the node elevation information in the geological borehole columnar data, the node offset amplitudes are sequentially compared with the columnar elevations, and the node offset trend within the segment is determined based on the positive and negative relationship of the differences, thereby obtaining a node offset trend sequence. S303: Call the node offset trend sequence, retrieve the node angle change value in the low-density segment in turn and correct the node angle, then integrate the corrected low-density segment node angle with the node offset amplitude and offset trend of the high-density segment in turn to obtain a multi-layer profile reconstruction primitive.

6. The geological three-dimensional modeling method based on multi-data fusion according to claim 5, characterized in that: The specific steps of S4 are: S401: Obtaining a node in the multi-layer cross-section reconstruction primitive, determining the proportional relationship between the lengths of the line segments on the left and right sides of the node based on the length information of the line segments on both sides of the node and the coordinates of the node's location, extracting a length relationship attribute using the combined features of the node and the line segments on both sides, and generating a node length relationship value; S402: Based on the node length relationship value, call the angle information formed between the adjacent line segments on both sides of the node to determine the correspondence between the length relationship value and the angle value, form a joint description based on the geometric characteristics of the node, and obtain the node coupling feature value; S403: For the node coupling characteristic value, call the coupling characteristic values ​​of the corresponding nodes of the adjacent sections, determine the degree of value difference through the corresponding relationship between the nodes, form the coupling characteristic change expression of the node position on the multi-layer section, and obtain the multi-layer section coupling difference feature.

7. The geological three-dimensional modeling method based on multi-data fusion according to claim 6, characterized in that: The coupling characteristic values ​​of the corresponding nodes of adjacent sections refer to the coupling characteristic values ​​extracted from the nodes on different sections in the multi-layer section data, specifically the node coupling characteristics formed by coupling the length relationship value of the line segments on both sides of each node with the angle value; The coupling feature change expression is based on the obtained coupling feature values ​​of the corresponding nodes of adjacent sections. By comparing the coupling feature value differences between the corresponding nodes one by one, the data expression describing the change of node features between sections is obtained.

8. The geological three-dimensional modeling method based on multi-data fusion according to claim 1, characterized in that: The specific steps of S5 are: S501: Based on the multi-layer profile coupling difference feature, the original coordinates and elevation offset information of the nodes in the geological profile mapping results and the geophysical exploration profile are called, and the corresponding segment length connection is extracted according to the position relationship of the nodes in the layer profile to obtain the node segment length difference feature value set; S502: calling the node segment length difference feature value set, combining the node elevation offset information, and performing correspondence sorting between the segment length difference feature values ​​and the elevation offset values ​​for the nodes on the same layer to obtain a node offset relationship sequence; S503: Based on the node offset relationship sequence and in combination with the original coordinate values ​​of the nodes, the spatial correspondence of the nodes in the vertical and horizontal directions is integrated to generate geological three-dimensional splicing information.

9. The geological 3D modeling system based on multi-data fusion is characterized by: The geological three-dimensional modeling method based on multi-data fusion according to any one of claims 1 to 8, wherein the system comprises: The node ratio density module obtains the node coordinate sequence in geological profile mapping, calculates the horizontal spacing between nodes, forms a node spacing ratio sequence based on the proportional relationship between the node spacing and the average value, divides the profile difference density segment according to the ratio sequence, and obtains the profile segment density label; The section adjustment coefficient module determines the subdivided section position based on the profile section density label, obtains the elevation and horizontal coordinates of the nodes in the geophysical exploration profile, calculates the elevation difference and horizontal difference between the nodes to form a ratio sequence, and obtains the subdivided section adjustment coefficient sequence; The node elevation offset module extracts the elevations of high-density segment nodes and their corresponding elevations in the geological borehole columnar data based on the segment adjustment coefficient sequence, calculates the difference to form an offset sequence, operates the offset sequence with the segment adjustment coefficient sequence, and performs ratio screening based on the angle change amplitude of the low-density segment nodes to obtain multi-layer profile reconstruction primitives. The multi-layer coupling difference module extracts the length relationship of the line segments on both sides of the node and the node turning angle based on the multi-layer profile reconstruction primitive to form a coupling feature set, and performs difference calculation with the feature set of the corresponding node of the adjacent profile to obtain the multi-layer profile coupling difference feature; The three-dimensional splicing information module calls the multi-layer section coupling difference feature, combines the node segment length relationship, the multi-layer section reconstruction primitives and the original coordinates of the nodes in the difference section, performs cumulative calculation, and generates geological three-dimensional splicing information.

Images