Three-dimensional numerical model InSAR deformation zoning buffer mapping method for long and large tunnels

Abstract

This application belongs to the technical fields of InSAR deformation monitoring, numerical simulation of underground engineering, spatial data mapping, 3D numerical model displacement boundary construction, and tunnel engineering deformation data processing. Specifically, it discloses a method for InSAR deformation partitioning and buffering mapping of a 3D numerical model for long tunnels. The method includes: acquiring InSAR deformation data and node data to be mapped; performing coordinate transformation; dividing the tunnel axis into multiple main mapping segments and setting buffer zones between adjacent segments; determining the segment to which the node belongs based on its axial mileage; selecting candidate deformation points from the segment and buffer zone; constructing distance attenuation weights based on spatial relationships and data quality; calculating candidate displacement values; fusing the candidate displacement values ​​of two adjacent segments to obtain the final displacement component for nodes located in the buffer zone; and outputting the correspondence between node numbers and the final displacement components. This application can continuously and stably map InSAR deformation to nodes in a 3D numerical model of long tunnels.

Description

Technical Field

[0001] This application belongs to the technical fields of InSAR deformation monitoring, underground engineering numerical simulation, spatial data mapping, three-dimensional numerical model displacement boundary construction and tunnel engineering deformation data processing. More specifically, it relates to an InSAR deformation partitioning buffer mapping method for three-dimensional numerical models of long tunnels. Background Technology

[0002] Synthetic Aperture Radar Interferometry (InSAR) can acquire information on large-scale surface deformation and has been widely used in fields such as surface subsidence, landslide deformation, mining area subsidence, underground engineering disturbance monitoring, and regional tectonic deformation analysis. For long tunnel projects (tunnels with a total length of over 3000 m), InSAR can provide continuous or quasi-continuous surface deformation information along the tunnel route and the areas on both sides, providing an important data source for the construction of displacement boundaries in three-dimensional numerical models, engineering disturbance analysis, and regional deformation evaluation.

[0003] Deformation data obtained from InSAR inversion can be represented as line-of-sight deformation from radar, or it can be obtained through fusion of ascending and descending orbit data, multi-view observations, GNSS constraints, ground monitoring data, or deformation decomposition methods to obtain horizontal deformation, vertical deformation, or deformation components in the east, north, and vertical directions. For three-dimensional numerical models, model nodes typically need to have displacement components in one or more directions applied in the model coordinate system. Therefore, it is necessary to convert the InSAR regional deformation data into nodal displacement data that matches the nodes of the three-dimensional numerical model.

[0004] However, InSAR deformation data is typically spatially discrete points, raster points, or regular grid data, while 3D numerical models are discretized using methods such as finite element, finite difference, or discrete element methods. Their basic computational objects are model nodes and elements. Significant differences exist between InSAR deformation points and 3D numerical model nodes in terms of spatial resolution, coordinate system, data organization, and spatial distribution. For 3D numerical models of long tunnels, if traditional global interpolation methods are used to calculate the distance between all InSAR deformation points and all model nodes to be mapped and then interpolate, problems arise such as high computational cost, significant interference from distant deformation points, and the impact of low-quality or anomalous InSAR points on the stability of the results. If interpolation is only performed along the longitudinal direction of the tunnel without setting buffers or continuous fusion mechanisms, sudden changes in node displacement can easily occur at the boundaries of adjacent sections. Furthermore, existing methods struggle to simultaneously adapt to the coordinate organization of curved tunnel axes and the mapping of multi-directional deformation components, and the output data is difficult to directly use for loading displacement boundaries in numerical models.

[0005] Therefore, how to continuously and stably map InSAR regional deformation data to the nodes of the three-dimensional numerical model of long tunnels is an urgent problem to be solved. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the purpose of this application is to provide a method for InSAR deformation partitioning buffer mapping of a three-dimensional numerical model of a long tunnel, which can continuously and stably map InSAR regional deformation data to nodes of the three-dimensional numerical model of a long tunnel.

[0007] To achieve the above objectives, in a first aspect, this application provides an InSAR deformation partitioning buffer mapping method for a three-dimensional numerical model of a long tunnel, comprising the following steps: S10, acquire InSAR deformation data of the tunnel area and node data to be mapped of the three-dimensional numerical model. The InSAR deformation data includes InSAR deformation point coordinates, deformation components in at least one direction and data quality information. The node data to be mapped includes node number and node coordinates. S20, Establish a local coordinate system for the tunnel based on the tunnel axis, and uniformly transform the InSAR deformation points and the nodes to be mapped to the local coordinate system for the tunnel; S30, the three-dimensional numerical model is divided into multiple main mapping sections along the tunnel axis, and a buffer zone is set between adjacent main mapping sections; S40, for each node to be mapped, determine its main mapping segment based on its axial mileage, select InSAR deformation points as candidate deformation points from the corresponding main mapping segment and its buffer, construct distance attenuation weights based on the spatial relationship between the node to be mapped and the candidate deformation points and the data quality information of the candidate deformation points, and calculate the candidate displacement value of the node to be mapped in the corresponding main mapping segment based on the distance attenuation weights. S50, for a node to be mapped located in the buffer, the displacement candidate values ​​corresponding to two adjacent main mapping segments are used respectively, and the displacement candidate values ​​are fused according to the relative position of the node to be mapped in the buffer to obtain the final displacement component of the node to be mapped; for a node to be mapped not located in the buffer, its corresponding displacement candidate value is used as the final displacement component. S60 outputs the correspondence between the node number to be mapped and the final displacement component, forming the node displacement mapping data of the three-dimensional numerical model.

[0008] As a further preferred embodiment, the deformation components include at least one of the following: radar line-of-sight deformation component, eastward deformation component, northward deformation component, vertical deformation component, horizontal deformation component, deformation component along the tunnel axis, and deformation component perpendicular to the tunnel axis.

[0009] As a further preferred embodiment, when the InSAR deformation data includes eastward deformation components, northward deformation components, and vertical deformation components, candidate deformation point screening, distance attenuation weight calculation, and buffer fusion are performed on the eastward deformation components, northward deformation components, and vertical deformation components respectively to obtain the eastward displacement components, northward displacement components, and vertical displacement components of the node to be mapped.

[0010] As a further preferred embodiment, the tunnel local coordinate system includes axial mileage coordinates along the tunnel axis, horizontal coordinates perpendicular to the tunnel axis, and vertical coordinates; for curved tunnels, the axial mileage coordinates are determined based on the projection position of the point to be converted on the tunnel axis, and the horizontal coordinates are determined based on the horizontal and vertical distances from the point to be converted to the tunnel axis.

[0011] As a further preferred embodiment, the length of the main mapping section is determined based on at least one of the following: tunnel length, three-dimensional numerical model mesh size, three-dimensional numerical model node density, InSAR deformation point density, local deformation gradient, and tunnel axis curvature; the length of the main mapping section is 100 meters to 500 meters.

[0012] As a further preferred embodiment, the buffer length is determined based on at least one of the following: the average spacing of InSAR deformation points, the average node spacing of the three-dimensional numerical model, the local deformation gradient, and the curvature of the tunnel axis; the buffer length is 5% to 50% of the length of the main mapping segment.

[0013] As a further preferred option, when screening candidate deformation points, the candidate deformation points must meet at least one of the following conditions: the candidate deformation point is located within the main mapping segment and its buffer of the node to be mapped; the distance between the candidate deformation point and the node to be mapped is less than the preset search radius; the data quality index of the candidate deformation point is greater than the preset threshold; the deformation value of the candidate deformation point is not determined to be an outlier; when the number of candidate deformation points is less than the preset minimum number, the search radius is expanded, the buffer range is expanded, or InSAR deformation points in adjacent main mapping segments that meet the quality requirements are called to supplement them.

[0014] As a further preferred embodiment, the distance attenuation weight is constructed according to the following formula:

[0015] in, For the first InSAR deformation point pairs of nodes The weights; For InSAR deformation point quality weights; Weighted distance; To prevent stable terms with zero distance; The distance decay exponent; The weighted distanced Calculate using the following formula:

[0016] or

[0017] in, , These are the axial mileage coordinates of the InSAR deformation point and the model node, respectively; , These are the lateral coordinates of the InSAR deformation point and the model node, respectively. This is the horizontal distance weighting coefficient; , These are the vertical coordinates of the InSAR deformation point and the model node, respectively; Weighting coefficient for vertical distance.

[0018] As a further preferred approach, for nodes to be mapped located within the buffer, fusion is performed as follows: First, the fusion coefficient is calculated based on the normalized position of the node to be mapped in the buffer. ,in , These are the axial coordinates of the start and end points of the buffer zone, respectively. Here are the axial coordinates of the nodes. The value range is 0 to 1; then according to Calculate the final displacement vector, where For the first The candidate node displacements calculated from each main mapping segment For the first Candidate node displacements calculated in the main mapping section.

[0019] Secondly, this application provides an InSAR regional deformation partitioning buffer mapping system for three-dimensional numerical models of long tunnels, used to implement the method described in any one of the above, comprising: The data acquisition module is used to acquire InSAR regional deformation data and 3D numerical model node data to be mapped; The coordinate transformation module is used to establish a local coordinate system for the tunnel based on the tunnel axis, and to uniformly transform the InSAR deformation points and nodes to be mapped to the local coordinate system for the tunnel. The partitioning module is used to divide the three-dimensional numerical model into multiple main mapping sections along the tunnel axis; The buffer setting module is used to set buffers between adjacent main mapping segments; The candidate point filtering module is used to filter InSAR deformation points in the corresponding main mapping segment and its buffer based on the main mapping segment to which the node to be mapped belongs. The displacement calculation module is used to calculate the candidate displacement value of the node to be mapped based on the distance attenuation weight between the node to be mapped and the candidate deformation point. The buffer fusion module is used to fuse candidate displacements of adjacent main mapping segments for nodes to be mapped located in the buffer. The displacement component conversion module is used to convert the final displacement components into nodal displacement components in the coordinate system of the three-dimensional numerical model. The quality evaluation module is used to evaluate the quality of the mapping results for each node to be mapped. The parameter adjustment module is used to adjust the parameters affecting the mapping results when the quality evaluation module determines that the node displacement mapping results do not meet the preset quality requirements, and to feed back the adjusted parameters to the corresponding processing module so that the subsequent mapping process can be re-executed. The data output module is used to output the correspondence between the node number to be mapped and the final displacement component.

[0020] Compared with existing technologies, this application has the following advantages: This application establishes a local coordinate system for the tunnel, divides the main mapping segment along the tunnel axis, sets a buffer zone, filters local candidate InSAR deformation points, calculates candidate node displacements based on distance attenuation weights, and fuses candidate displacements of adjacent segments for buffer nodes, thereby obtaining continuous, stable, and directly usable node displacement data for loading the displacement boundary of a three-dimensional numerical model. Specific advantages are reflected in the following aspects: (1) Improve the mapping efficiency of long tunnel models This application divides the main mapping section along the tunnel axis, so that each node to be mapped only calls the InSAR deformation points in its own section and buffer to participate in the calculation, avoiding the calculation of the global distance between all InSAR deformation points and all model nodes, and significantly reducing the amount of calculation.

[0021] (2) Reduce interference from long-distance deformation points By filtering local candidate points, the node to be mapped is only affected by InSAR deformation points within a certain mileage and spatial range around it, thus avoiding unreasonable influence of distant deformation points on the local node displacement calculation.

[0022] (3) Improve the displacement continuity at the boundary of the zone By setting buffers between adjacent main mapping sections and using the adjacent section candidate displacement fusion method for buffer nodes, the sudden displacement at the boundary caused by simple partition mapping can be avoided, and the displacement of nodes in the long tunnel model can be continuously transitioned along the axial direction.

[0023] (4) Improve the stability of mapping results This application can combine InSAR deformation point coherence, confidence level and outlier removal mechanism to perform quality control on candidate deformation points, and reduce the impact of low-quality InSAR points or anomalous deformation points on nodal displacement results.

[0024] (5) Applicable to deformation data in multiple directions This application is applicable not only to horizontal deformation mapping, but also to the mapping of three-dimensional deformation components consisting of radar line-of-sight deformation, vertical deformation, eastward deformation, northward deformation, and vertical deformation.

[0025] (6) Applicable to curved or long-distance tunnels This application establishes a local coordinate system based on the tunnel axis, and organizes InSAR deformation points and model nodes with axial mileage and lateral distance, which is applicable to straight tunnels, curved tunnels and long-distance underground engineering models.

[0026] (7) The output results are easy to call up by numerical models. The final output of this application shows the correspondence between node numbers and displacement components, which can be directly used as displacement boundary loading data for three-dimensional numerical models, reducing the workload of data conversion between InSAR deformation data and numerical simulation software. Attached Figure Description

[0027] Figure 1 This is the overall technical flow diagram provided in this application; Figure 2 This is a schematic diagram showing the spatial relationship between InSAR deformation points and the three-dimensional numerical model of a long tunnel provided in this application; Figure 3 This is a schematic diagram of the local coordinate system of the tunnel provided in this application; Figure 4 This is a schematic diagram of the main mapping segment and buffer zone division provided in this application; Figure 5 This is a schematic diagram of the candidate InSAR deformation point screening provided in this application; Figure 6 This is a schematic diagram of buffer node displacement fusion provided in this application; Figure 7 This is a schematic diagram of the regional deformation component conversion provided in this application; Figure 8 This is a schematic diagram of the node displacement mapping data output structure provided in this application; Figure 9 This is a schematic diagram of the system module structure provided in this application. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0029] To address the problems of low computational efficiency, significant interference from distant deformation points, discontinuous displacement at partition boundaries, significant impact from local anomalies, insufficient processing of multi-directional deformation components, and difficulty in directly using node displacement data for loading numerical models in existing InSAR deformation field-to-3D numerical model node mapping processes, this application provides an InSAR regional deformation partition buffer mapping method for long tunnel 3D numerical models. This method involves establishing a local tunnel coordinate system, dividing the main mapping segment along the tunnel axis, setting a buffer zone, filtering local candidate InSAR deformation points, calculating candidate node displacements based on distance attenuation weights, and fusing candidate displacements from adjacent segments for buffer nodes. This yields continuous, stable node displacement data that can be directly used for loading displacement boundaries in 3D numerical models.

[0030] like Figure 1 As shown, the InSAR regional deformation partitioning buffer mapping method for three-dimensional numerical models of long tunnels provided in this application includes the following steps: S1, acquire InSAR regional deformation data. Acquire InSAR regional deformation data covering the long tunnel engineering area, such as... Figure 2 As shown. InSAR regional deformation data includes the coordinates of InSAR deformation points and deformation components in at least one direction.

[0031] The deformation components may include at least one of the following: radar line-of-sight (LOS) deformation component; eastward (E) deformation component; northward (N) deformation component; vertical (U) deformation component; horizontal deformation component; deformation component along the tunnel axis; deformation component perpendicular to the tunnel axis; and deformation components in the first, second, or third direction in the three-dimensional numerical model coordinate system. Among these, the horizontal deformation component is a preferred deformation component for InSAR regional deformation data, but this application is not limited to horizontal deformation mapping. When the InSAR regional deformation data includes vertical deformation, radar line-of-sight deformation, or deformation components in the eastward, northward, and vertical directions, corresponding nodal displacement components can also be generated according to the partitioned buffer mapping method of this application.

[0032] InSAR regional deformation data may include: InSAR deformation point number; InSAR deformation point planar coordinates; InSAR deformation point elevation (optional); deformation components in one or more directions; coherence; confidence level; standard deviation; data validity indicator; time information; observation trajectory information.

[0033] When InSAR deformation data includes deformation components in the east, north, and vertical directions, it can be denoted as... ,in , , For respectively the first The eastward, northward, and vertical deformation components of each InSAR deformation point.

[0034] S2, Obtain the node data to be mapped from the 3D numerical model. Extract the node data to be mapped from the 3D numerical model of the long tunnel.

[0035] The node data to be mapped includes: node number; node 3D coordinates; boundary type to which the node belongs; element to which the node belongs; whether the node needs to be loaded with displacement; model region information where the node is located; and displacement direction information to be loaded by the node.

[0036] The nodes to be mapped include, but are not limited to: top surface nodes of the model; lateral boundary nodes of the model; end boundary nodes of the model; specified loading boundary nodes; and other model nodes that need to receive InSAR regional deformation data.

[0037] In this application, the node to be mapped refers to a three-dimensional numerical model node that needs to be assigned or have displacement components generated by InSAR regional deformation data. This node can be a model surface node or a boundary node specified according to engineering needs.

[0038] S3, Establish the local coordinate system of the tunnel. Establish the local coordinate system of the tunnel based on the tunnel axis, such as... Figure 3 As shown. The tunnel local coordinate system includes: axial mileage coordinates along the tunnel axis. Horizontal coordinates perpendicular to the tunnel axis Vertical coordinates .

[0039] For any InSAR deformation point Global coordinates are In the local coordinate system, it is denoted as For any model node Global coordinates are In the local coordinate system, it is denoted as .

[0040] A local coordinate system can be established for a straight tunnel based on the tunnel's starting point, ending point, and axial direction vector.

[0041] For curved tunnels, a local coordinate system can be established based on the design axis, centerline polyline, spline curve, or mileage markers. For any node to be mapped or InSAR deformation point, first determine its nearest projection point on the tunnel axis, and then use the axial mileage corresponding to that projection point as the axial coordinate. The horizontal and vertical distances from this point to the tunnel axis are used as the lateral coordinates. Vertical coordinates Take the height value of the node or deformation point directly.

[0042] S4, Divide the main mapping section along the tunnel axis. Divide the three-dimensional numerical model into multiple main mapping sections along the tunnel axis ( Figure 4 Each main mapping segment corresponds to a tunnel mileage range. Each primary mapping segment can be represented as: , and These are the start and end mileages of the main mapping segment. The length of the main mapping segment is... .

[0043] The length of the main mapping section can be determined based on at least one of the following factors: total tunnel length; 3D numerical model mesh size; 3D numerical model node density; InSAR deformation point spatial resolution; InSAR deformation point density; local deformation gradient; tunnel axis curvature; computational efficiency requirements.

[0044] In one preferred embodiment, the length of the main mapping section is 100 m to 500 m. In another preferred embodiment, for a 13 km long tunnel, the length of the main mapping section is 200 m.

[0045] S5, Set up buffers. Buffers are set at both ends of each main mapping segment to create overlapping mapping regions between adjacent main mapping segments. This provides sufficient candidate InSAR deformation points for the nodes to be mapped, enabling a continuous transition of displacement between adjacent segments, such as... Figure 4 As shown. (The following is a list of numbers / items / etc.) Each main mapping segment The extended mapping segment is ,and , These are the buffer lengths at the left and right ends of the main mapping segment, respectively, and can be set according to actual engineering conditions.

[0046] The buffer length can be adaptively determined based on the following factors: average spacing between InSAR deformation points; average node spacing of the 3D numerical model; local deformation gradient; tunnel axis curvature; node density; and data quality.

[0047] In a preferred embodiment, the buffer length can be 5% to 50% of the main mapping segment length. For example, when the main mapping segment length is 200 m, the buffer length can be 20 m to 100 m, preferably 50 m.

[0048] The buffer serves several purposes: expanding the InSAR deformation point search range for each main mapping segment; reducing abrupt changes in the number of candidate points at partition boundaries; providing overlapping areas for fusion of candidate displacements in adjacent segments; ensuring continuous transition of model node displacements along the tunnel axis; and reducing boundary displacement jumps caused by simple partition mapping.

[0049] The above scope is not a limitation on the scope of protection of this application and can be adjusted according to model length, node density, InSAR point spacing and deformation gradient.

[0050] S6, filter local candidate InSAR deformation points. For any node to be mapped. According to its axial mileage coordinates Determine the main mapping segment and buffer range to which it belongs.

[0051] like Figure 5 As shown, a candidate deformation point can satisfy at least one of the following conditions: the candidate deformation point is located within the main mapping segment and its buffer of the node to be mapped; the distance between the candidate deformation point and the node to be mapped is less than the preset search radius; the data quality index of the candidate deformation point is greater than the preset threshold; the coherence of the candidate deformation point is greater than the preset threshold; the deformation value of the candidate deformation point is not determined to be a local outlier; and the number of candidate deformation points meets the minimum requirement.

[0052] When the number of candidate points is less than the preset minimum number, the search radius can be expanded, the buffer range can be expanded, or InSAR deformation points that meet the quality requirements in adjacent main mapping segments can be called to supplement them.

[0053] S7, Construct distance decay weights. For the candidate deformation point set... Each InSAR deformation point Calculate its relationship with the node to be mapped. The distance.

[0054] Two-dimensional weighted distance (applicable to nodes on the top surface):

[0055] in, , These are the axial mileage coordinates of the InSAR deformation point and the model node, respectively; , These are the lateral coordinates of the InSAR deformation point and the model node, respectively. This is the horizontal distance weighting coefficient.

[0056] 3D weighted distance (applicable to nodes not on the top surface or where vertical influence needs to be considered):

[0057] in, , These are the vertical coordinates of the InSAR deformation point and the model node, respectively; Weighting coefficient for vertical distance.

[0058] Two-dimensional distance is suitable when the InSAR deformation point is located on the surface plane and the node to be mapped is located on the upper surface node; when the node to be mapped or the mapping object contains non-upper surface nodes, three-dimensional weighted distance can be used.

[0059] Candidate point weights are calculated based on distance:

[0060] in, For the first InSAR deformation point pairs of nodes The weights; For InSAR deformation point quality weights; Weighted distance; To prevent stable terms with zero distance; The distance attenuation index is preferably 1 to 4.

[0061] By using this distance decay weight, the candidate point displacements can be weighted and averaged in subsequent steps to generate candidate node displacement values.

[0062] S8, Calculate candidate displacement values ​​for the node to be mapped. Based on the deformation components and distance attenuation weights of the candidate InSAR deformation points, calculate the displacement values ​​for the node to be mapped. Candidate displacement values ​​in the corresponding direction.

[0063] If the InSAR deformation data is a single deformation component Then the candidate displacements of the node are:

[0064] If InSAR deformation data includes deformation components in the east, north, and vertical directions... Then the candidate displacements of the nodes are calculated in components:

[0065]

[0066]

[0067] Or it can be represented in vector form. ,in, , .

[0068] If the tunnel's local coordinate system is used, the displacement components along the tunnel's axial and lateral directions can also be calculated:

[0069]

[0070] S9, Buffer Node Continuous Fusion. For a node to be mapped located within the buffer of an adjacent main mapping segment, two candidate displacement values ​​are calculated using the candidate deformation point sets of the two adjacent main mapping segments, such as... Figure 6 As shown.

[0071] Assuming nodes Located in the The first main mapping segment and the first The buffers between the main mapping segments then yield the following results: , . For the first The candidate node displacements calculated from each main mapping segment For the first Candidate node displacements calculated in the main mapping section.

[0072] The fusion coefficient is determined based on the relative position of the nodes within the buffer. ,in, , These are the axial coordinates of the start and end points of the buffer zone, respectively. Here are the axial coordinates of the nodes. The value range is from 0 to 1.

[0073] The final displacement vectors of the nodes are fused using linear weighting:

[0074] in, , ,and .

[0075] If the candidate displacement contains eastward, northward, and vertical components, then the components are fused separately:

[0076]

[0077]

[0078] Besides linear fusion, a smooth transition function can also be used. For example:

[0079] in, This represents the normalized position of the node within the buffer. .

[0080] By using the above fusion method, the displacement abrupt change caused by partition mapping at the junction of adjacent sections can be reduced, and the node displacement can be continuously transitioned in the tunnel axis.

[0081] S10, Deformation component transformation to model coordinate system. When InSAR regional deformation data contains eastward, northward, and vertical components, it is necessary to transform the candidate nodal displacement components into nodal displacement components in the model coordinate system based on the relationship between the 3D numerical model coordinate system and the geographic coordinate system or the tunnel local coordinate system. Figure 7 As shown.

[0082] If the 3D numerical model coordinate system is consistent with the geographic coordinate system, then the following direct correspondences can be established: the eastward E component corresponds to the model's X-direction displacement; the northward N component corresponds to the model's Y-direction displacement; and the vertical U component corresponds to the model's Z-direction displacement. When the InSAR regional deformation data contains eastward, northward, and vertical components, it is necessary to convert the candidate nodal displacement components into nodal displacement components in the model coordinate system based on the relationship between the 3D numerical model coordinate system and the geographic coordinate system or the tunnel's local coordinate system.

[0083] If the model coordinate system is inconsistent with the geographic coordinate system, a coordinate transformation matrix is ​​used for conversion:

[0084] in: , , Let be the displacement component of the j-th node in the model coordinate system; , , These are the eastward, northward, and vertical displacement components of the j-th node in the geographic coordinate system. This is the coordinate transformation matrix from the geographic coordinate system to the model coordinate system.

[0085] If the nodal displacement is expressed in the tunnel local coordinate system as It can be based on the unit vector along the tunnel axis. and the horizontal unit vector is The axial and lateral displacement components are converted into displacements in the X and Y directions of the model:

[0086]

[0087] Vertical component It can be directly used as the vertical displacement of the model, or converted according to the vertical axis direction of the model.

[0088] S11, Generate node displacement mapping data. Output a 3D numerical model node displacement mapping data table. The node displacement mapping data table should include at least: node number; node coordinates; boundary type to which the node belongs; main mapping segment number to which the node belongs; whether the node is located in the buffer zone; final displacement components of the node; number of InSAR candidate points participating in the mapping; weights and / or mapping confidence; data quality identifier; and displacement component type. A schematic diagram of the node displacement mapping data table structure is shown below. Figure 8 As shown.

[0089] Output formats can include: text files; CSV files; Excel files; boundary condition files readable by numerical simulation software; FLAC3D command files; ABAQUS input files; ANSYS load files; and COMSOL interpolation function files.

[0090] Based on the same inventive concept, this application also provides an InSAR regional deformation partitioning buffer mapping system for three-dimensional numerical models of long tunnels. Figure 9 (This is used to execute the method of this application, converting InSAR regional deformation data into displacement data of nodes to be mapped in a three-dimensional numerical model.)

[0091] Data acquisition module: Acquires InSAR regional deformation data and 3D numerical model node data to be mapped. InSAR data includes deformation point number, coordinates, deformation components in at least one direction, and data quality indicators; model node data includes number, 3D coordinates, boundary type, element to which it belongs, and whether displacement information needs to be loaded.

[0092] Coordinate Transformation Module: Establishes a local coordinate system based on the tunnel axis, transforming InSAR deformation points and model nodes to the tunnel's local coordinate system. For straight tunnels, a coordinate system can be established using the start point, end point, and axis direction; for curved tunnels, a coordinate system can be established using the design axis or spline curve.

[0093] Partitioning module: Divide the model into multiple main mapping segments along the tunnel axis. The segment length can be determined based on factors such as tunnel length, node density, InSAR point density, and deformation gradient.

[0094] Buffer setting module: Sets buffers between adjacent main mapping segments to form overlapping mapping regions. The buffer length can be fixed or adaptively determined based on node spacing, deformation gradient, or tunnel curvature.

[0095] Candidate point filtering module: Filters candidate InSAR deformation points based on the main mapping segment and buffer to which the node belongs. It can filter based on conditions such as range, search radius, data quality, outliers and minimum number. If necessary, it can call candidate points from adjacent segments to supplement them.

[0096] Displacement calculation module: Calculates candidate node displacement values ​​using the distance relationship between candidate InSAR deformation points and nodes, along with distance attenuation weights, while simultaneously considering data quality. For multi-directional deformation components, candidate displacement values ​​are calculated separately for each direction.

[0097] Buffer fusion module: Fusion of candidate displacements of adjacent main mapping segments for nodes in the buffer. The fusion method can be weighted by a linear or smooth function to obtain the final displacement of the node; non-buffer nodes directly use the calculation results of their respective segments.

[0098] Displacement component conversion module: Converts InSAR deformation components or displacement components in the local coordinate system of the tunnel into nodal displacement components in the coordinate system of the three-dimensional numerical model, supporting straight lines, curved tunnels and local coordinate system conversion.

[0099] Quality evaluation module: Evaluates the quality of the mapping results for each node, including the number of candidate points, average coherence, spatial distribution uniformity, displacement jumps, and whether it is within the buffer zone, and outputs the mapping confidence score. Nodes that pass the evaluation proceed to the data output module; otherwise, they proceed to the parameter adjustment module.

[0100] Parameter adjustment module: When the mapping result does not meet the quality requirements, the main mapping segment length, buffer length, candidate point search radius, quality threshold, distance decay parameter, fusion weight or coordinate transformation parameter are adjusted and fed back to the corresponding module for recalculation.

[0101] Data output module: Outputs 3D numerical model node displacement mapping data, including node number, coordinates, boundary type, segment number, whether it is a buffer, final displacement components, number of candidate points and mapping confidence, and can generate input file formats for different numerical simulation software.

[0102] The present application will now be described in conjunction with specific embodiments.

[0103] Example 1: Fixed-zone buffer mapping of horizontal deformation in a long tunnel using InSAR This embodiment uses the InSAR horizontal deformation component as an example to illustrate the partitioned buffer mapping process of this application, and does not constitute a limitation on the type of deformation component processed in this application. This embodiment uses a three-dimensional numerical model of a long tunnel as an example. The tunnel is 13 km long, and the lateral width of the three-dimensional numerical model is 1400 m. The InSAR regional deformation data covers an area of ​​1500 m × 13000 m. In this embodiment, the horizontal deformation component obtained by InSAR inversion is used for mapping. The InSAR data includes the plane coordinates of the deformation points, the horizontal displacement component, and coherence information. Boundary nodes that require displacement are extracted from the three-dimensional numerical model, forming a set of nodes to be mapped.

[0104] S1. Establish a local coordinate system for the tunnel based on the tunnel design axis. Convert the planar coordinates of the InSAR deformation points into tunnel axial mileage coordinates. s and horizontal coordinates n And convert the coordinates of the nodes to be mapped in the 3D numerical model into ( s , n , z (Coordinate form)

[0105] S2, the main mapping section is divided along the tunnel axis into 200-m lengths. For a 13 km tunnel, it can be divided into multiple main mapping sections. A 50-m buffer zone is set at both ends of each main mapping section. For the... k Each main mapping segment has a range of 1. Its extended mapping range is .

[0106] For any node to be mapped, based on its mileage coordinates Determine the primary mapping segment. Then, filter InSAR deformation points within the corresponding extended mapping range, while removing points with coherence below a set threshold and points with localized abnormal deformation.

[0107] S3, for the retained candidate InSAR deformation points, calculate the weighted distance between them and the node to be mapped. , This is the horizontal distance weighting coefficient related to the region where the node is located.

[0108] S4, Construct distance decay weights ,in, This represents the quality weight corresponding to the coherence or confidence level of InSAR deformation points. As a stable term, This is the distance decay index.

[0109] S5, calculate the horizontal displacement of the nodes based on the weights: , For nodes located within the buffer zones of two adjacent main mapping segments, two candidate displacement values ​​are calculated using candidate InSAR points from the preceding and following segments, respectively: , .

[0110] S6, Calculate the fusion coefficient based on the node's position within the buffer. .

[0111] S7, the horizontal displacement of the node is obtained as follows: , .

[0112] S8 outputs a nodal displacement mapping table. This table includes the node number, node coordinates, segment number, whether it is located in the buffer zone, the final horizontal displacement component, the number of candidate InSAR points, and the mapping confidence level. This output can be directly converted into a displacement boundary input file for 3D numerical simulation software.

[0113] Example 2: Adaptive Partitioning and Buffering Mapping of Long Curved Tunnels in InSAR 3D Region Deformation This embodiment uses a long tunnel with a curved axis as an example to illustrate how this application can be used in curved tunnels, non-uniform mesh three-dimensional numerical models, and three-dimensional region deformation data.

[0114] The tunnel is approximately 10 km long, and its axis consists of multiple straight sections, transition curves, and circular curves. The 3D numerical model uses a denser grid near the tunnel chambers and a sparser grid in areas farther from the chambers, resulting in a non-uniform distribution of model nodes along both the tunnel's axial and lateral directions. InSAR regional deformation data covers the tunnel route and a certain area on both sides, presented as regular grids or discrete deformation points. This InSAR regional deformation data includes eastward deformation components. Northward deformation component and vertical deformation component .

[0115] S1. Acquire InSAR 3D regional deformation data covering the tunnel engineering area, including the planar coordinates of deformation points, eastward deformation components, northward deformation components, vertical deformation components, and coherence indices. Extract the nodes to be mapped from the 3D numerical model. The nodes to be mapped include the top surface nodes, lateral boundary nodes, and specified boundary nodes of the model. For each node, extract its node number, 3D coordinates, and the boundary type it belongs to.

[0116] S2. Establish a local coordinate system for the curved tunnel based on the tunnel design axis. For any InSAR deformation point or model node, first calculate the nearest projection point from that point to the tunnel axis. Use the tunnel mileage corresponding to that projection point as the axial mileage coordinate of that point, and the horizontal and vertical distances from that point to the axis as the lateral coordinates. For curved sections, the axial direction of the local coordinate system is taken as the tangential direction of the tunnel axis at the projected mileage of that point, and the lateral direction is taken as the horizontal normal direction perpendicular to the tangential direction.

[0117] S3. The main mapping segment is adaptively divided based on node density, InSAR deformation point density, tunnel axis curvature, and local deformation gradient. When the model node density is high, the InSAR deformation gradient is large, or the tunnel axis curvature is large in a local area of ​​the tunnel, a shorter main mapping segment is used; when the model node density is low, the InSAR deformation change is gradual, or the tunnel axis is approximately straight, a longer main mapping segment is used.

[0118] For example, the length of the main mapping segment can vary between 100 m and 400 m: In areas with dense caverns, near faults, or with large deformation gradients, the length of the main mapping section is taken as 100 m to 150 m. In areas where deformation is relatively gentle and the tunnel axis is approximately straight, the length of the main mapping section is taken as 250 m to 400 m; In the normal section, the length of the main mapping section is approximately 200 m. S4. Set an adaptive buffer based on the data resolution. The buffer length is determined based on the average spacing of InSAR deformation points, the average spacing of model nodes, the local deformation gradient, and the curvature of the tunnel axis. In areas with high InSAR point density and high model node density, the buffer length can be appropriately reduced; in areas with low InSAR point density, high tunnel curvature, or large differences in candidate displacements between adjacent segments, the buffer length can be appropriately increased.

[0119] For example, the buffer length can be 10% to 40% of the main mapping segment length. For a region with a main mapping segment length of 100 m, the buffer length can be 20 m to 40 m; for a region with a main mapping segment length of 400 m, the buffer length can be 40 m to 100 m.

[0120] S5. For each node to be mapped, determine its primary mapping segment based on its axial mileage. Within this primary mapping segment and its buffer zone, select candidate InSAR deformation points. The candidate point selection considers the following conditions: The candidate point is located within the extended mapping segment; The axial and lateral distances between candidate points and nodes meet the set range; The coherence of candidate points is not lower than a preset threshold; The deformation values ​​at the candidate points were not identified as local outliers. The number of candidate points meets the minimum requirement.

[0121] If the number of valid InSAR deformation points near a node is insufficient, the search radius should be expanded or InSAR points in adjacent segments that meet the quality requirements should be used to supplement the search.

[0122] S6. For each node to be mapped, calculate the candidate value of the node displacement based on the distance between the candidate InSAR deformation point and the node, the InSAR data quality, and the local coordinate direction.

[0123] In this embodiment, the following distance expression can be used:

[0124] in, and These are the axial mileage coordinates of the candidate InSAR deformation point and the node to be mapped, respectively. and These are the lateral coordinates of the candidate InSAR deformation point and the node to be mapped, respectively. This is the horizontal distance weighting coefficient related to the region where the node is located.

[0125] When a node is located in an area with a large curvature of the tunnel axis, the weighting coefficient of the lateral distance can be appropriately increased to reduce the influence of laterally distant deformation points on the node displacement calculation.

[0126] S7, Distance decay weight ,in, This represents the quality weight corresponding to the coherence or confidence level of InSAR deformation points. As a stable term, This is the distance decay index.

[0127] S8, for the deformation components in the eastward, northward, and vertical directions, calculate the candidate displacements of the nodes respectively:

[0128]

[0129]

[0130] S9. For nodes located within the buffer of adjacent main mapping segments, two candidate displacement vectors are calculated using data from the preceding and following main mapping segments, respectively. Then, a weighted fusion is performed based on the relative positions of the nodes in the buffer.

[0131] Linear fusion can be expressed as ,in, and These are the candidate displacement vectors obtained from two adjacent main mapping segments.

[0132] S10, in curved tunnels or regions with large deformation gradients, a smooth transition function can also be used. ,in, This represents the normalized position of the node within the buffer. This smoothing function makes the weight changes at the start and end points of the buffer more gradual, thereby further reducing abrupt displacement changes at the boundaries of adjacent segments.

[0133] S11 transforms the eastward, northward, and vertical nodal displacement components into nodal displacement components in the 3D numerical model coordinate system. When the 3D numerical model coordinate system is consistent with the geographic coordinate system, the eastward component corresponds to the model's X-direction displacement, the northward component corresponds to the model's Y-direction displacement, and the vertical component corresponds to the model's Z-direction displacement. When the 3D numerical model coordinate system is inconsistent with the geographic coordinate system, the eastward, northward, and vertical components are transformed into nodal displacement components in the model coordinate system using a coordinate transformation matrix. , , Three-directional displacement components.

[0134] S12 outputs mapping quality metrics for each node to be mapped, including the number of candidate InSAR points, average coherence, average distance from the node to the candidate points, whether it is located in the buffer zone, the difference in candidate displacement between adjacent segments, and mapping confidence.

[0135] S13, output the node displacement mapping table, which includes: node number; global coordinates of the node; local coordinates of the node; the number of the main mapping segment to which it belongs; whether it is located in the buffer; and the coordinates of the node in the model coordinate system. , , Shift component; number of candidate InSAR points; mapping confidence; data quality identifier.

[0136] It should be noted that this application is applicable not only to horizontal deformation mapping, but also to the mapping of InSAR three-dimensional regional deformation data to nodes of three-dimensional numerical models of long tunnels; it is applicable not only to long straight tunnels and fixed partition lengths, but also to long curved tunnels, non-uniform mesh models, adaptive partitions, and adaptive buffer settings.

[0137] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for InSAR deformation partitioning and buffering mapping of a three-dimensional numerical model of a long tunnel, characterized in that, Includes the following steps: S10, acquire InSAR deformation data of the tunnel area and node data to be mapped of the three-dimensional numerical model. The InSAR deformation data includes InSAR deformation point coordinates, deformation components in at least one direction and data quality information. The node data to be mapped includes node number and node coordinates. S20, Establish a local coordinate system for the tunnel based on the tunnel axis, and uniformly transform the InSAR deformation points and the nodes to be mapped to the local coordinate system for the tunnel; S30, the three-dimensional numerical model is divided into multiple main mapping sections along the tunnel axis, and a buffer zone is set between adjacent main mapping sections; S40, for each node to be mapped, determine its main mapping segment based on its axial mileage, select InSAR deformation points as candidate deformation points from the corresponding main mapping segment and its buffer, construct distance attenuation weights based on the spatial relationship between the node to be mapped and the candidate deformation points and the data quality information of the candidate deformation points, and calculate the candidate displacement value of the node to be mapped in the corresponding main mapping segment based on the distance attenuation weights. S50, for a node to be mapped located in the buffer, the displacement candidate values ​​corresponding to two adjacent main mapping segments are used respectively, and the displacement candidate values ​​are fused according to the relative position of the node to be mapped in the buffer to obtain the final displacement component of the node to be mapped; for a node to be mapped not located in the buffer, its corresponding displacement candidate value is used as the final displacement component. S60 outputs the correspondence between the node number to be mapped and the final displacement component, forming the node displacement mapping data of the three-dimensional numerical model.

2. The InSAR deformation partitioning and buffering mapping method for three-dimensional numerical models of long tunnels as described in claim 1, characterized in that, The deformation components include at least one of the following: radar line-of-sight deformation component, eastward deformation component, northward deformation component, vertical deformation component, horizontal deformation component, deformation component along the tunnel axis, and deformation component perpendicular to the tunnel axis.

3. The InSAR deformation partitioning and buffering mapping method for three-dimensional numerical models of long tunnels as described in claim 1, characterized in that, When the InSAR deformation data includes eastward deformation components, northward deformation components, and vertical deformation components, candidate deformation point screening, distance attenuation weight calculation, and buffer fusion are performed on the eastward deformation components, northward deformation components, and vertical deformation components respectively to obtain the eastward displacement components, northward displacement components, and vertical displacement components of the node to be mapped.

4. The InSAR deformation partitioning and buffering mapping method for three-dimensional numerical models of long tunnels as described in claim 1, characterized in that, The local coordinate system of the tunnel includes axial mileage coordinates along the tunnel axis, horizontal coordinates perpendicular to the tunnel axis, and vertical coordinates. For curved tunnels, the axial mileage coordinates are determined based on the projection position of the point to be converted on the tunnel axis, and the horizontal coordinates are determined based on the horizontal and vertical distances from the point to be converted to the tunnel axis.

5. The InSAR deformation partitioning and buffering mapping method for three-dimensional numerical models of long tunnels as described in claim 1, characterized in that, The length of the main mapping section is determined based on at least one of the following: tunnel length, 3D numerical model mesh size, 3D numerical model node density, InSAR deformation point density, local deformation gradient, and tunnel axis curvature; the length of the main mapping section is 100 meters to 500 meters.

6. The InSAR deformation partitioning buffer mapping method for three-dimensional numerical models of long tunnels as described in claim 1, characterized in that, The buffer length is determined based on at least one of the following: the average spacing of InSAR deformation points, the average node spacing of the three-dimensional numerical model, the local deformation gradient, and the curvature of the tunnel axis; the buffer length is 5% to 50% of the length of the main mapping segment.

7. The InSAR deformation partitioning and buffering mapping method for three-dimensional numerical models of long tunnels as described in claim 1, characterized in that, When screening candidate deformation points, the candidate deformation points must meet at least one of the following conditions: the candidate deformation point is located within the main mapping segment and its buffer of the node to be mapped; The distance between the candidate deformation point and the node to be mapped is less than the preset search radius; The data quality index of the candidate deformation points is greater than the preset threshold; The deformation values ​​of the candidate deformation points were not identified as outliers. When the number of candidate deformation points is less than the preset minimum number, the search radius is expanded, the buffer range is expanded, or InSAR deformation points that meet the quality requirements in adjacent main mapping segments are called to supplement them.

8. The InSAR deformation partitioning buffer mapping method for three-dimensional numerical models of long tunnels as described in claim 1, characterized in that, The distance attenuation weight is constructed according to the following formula: in, For the first InSAR deformation point pairs of nodes The weights; For InSAR deformation point quality weights; Weighted distance; To prevent stable terms with zero distance; The distance decay exponent; The weighted distance d Calculate using the following formula: or in, , These are the axial mileage coordinates of the InSAR deformation point and the model node, respectively; , These are the lateral coordinates of the InSAR deformation point and the model node, respectively. This is the horizontal distance weighting coefficient; , These are the vertical coordinates of the InSAR deformation point and the model node, respectively; Weighting coefficient for vertical distance.

9. The InSAR deformation partitioning and buffering mapping method for three-dimensional numerical models of long tunnels as described in claim 1, characterized in that, For nodes to be mapped that are located in the buffer, fusion is performed as follows: First, the fusion coefficient is calculated based on the normalized position of the node to be mapped in the buffer. ,in , These are the axial coordinates of the start and end points of the buffer zone, respectively. Here are the axial coordinates of the nodes. The value range is 0 to 1; then according to Calculate the final displacement vector, where For the first The candidate node displacements calculated from each main mapping segment For the first Candidate node displacements calculated in the main mapping section.

10. An InSAR regional deformation partitioning buffer mapping system for three-dimensional numerical models of long tunnels, characterized in that, To implement the method of any one of claims 1 to 9, comprising: The data acquisition module is used to acquire InSAR regional deformation data and 3D numerical model node data to be mapped; The coordinate transformation module is used to establish a local coordinate system for the tunnel based on the tunnel axis, and to uniformly transform the InSAR deformation points and nodes to be mapped to the local coordinate system for the tunnel. The partitioning module is used to divide the three-dimensional numerical model into multiple main mapping sections along the tunnel axis; The buffer setting module is used to set buffers between adjacent main mapping segments; The candidate point filtering module is used to filter InSAR deformation points in the corresponding main mapping segment and its buffer based on the main mapping segment to which the node to be mapped belongs. The displacement calculation module is used to calculate the candidate displacement value of the node to be mapped based on the distance attenuation weight between the node to be mapped and the candidate deformation point. The buffer fusion module is used to fuse candidate displacements of adjacent main mapping segments for nodes to be mapped located in the buffer. The displacement component conversion module is used to convert the final displacement components into nodal displacement components in the coordinate system of the three-dimensional numerical model. The quality evaluation module is used to evaluate the quality of the mapping results for each node to be mapped. The parameter adjustment module is used to adjust the parameters affecting the mapping results when the quality evaluation module determines that the node displacement mapping results do not meet the preset quality requirements, and to feed back the adjusted parameters to the corresponding processing module so that the subsequent mapping process can be re-executed. The data output module is used to output the correspondence between the node number to be mapped and the final displacement component.

Images