Complex site finite element network generation method and system

Abstract

The invention relates to a complex site finite element network generation method and system. The method comprises the steps of obtaining drilling data of a complex site area; converting the drilling data of the complex site area into a rock-soil feature point set; performing maximum probability scatter interpolation processing on the rock-soil feature point set to form a site material distribution image; and taking the site material distribution image as input data, performing automatic mesh generation based on an octree algorithm, and obtaining a site finite element model for earthquake response analysis. According to the method, the drilling data is converted into the feature point set and the interpolation processing is carried out to construct the high-precision site material distribution image, and the high-precision site finite element model is constructed by carrying out grid decomposition on the site material distribution image according to the octree algorithm, so that the high-precision site finite element model can be obtained without manual intervention; the modeling efficiency and the automation degree of a complex site are remarkably improved, and the precision of model numerical analysis is ensured.

Description

Technical Field

[0001] This application relates to the field of finite element network generation technology for complex sites, and in particular to a method and system for generating finite element networks for complex sites. Background Technology

[0002] Dynamic analysis of major engineering projects requires consideration of the dynamic responses of adjacent sites. Therefore, numerical modeling of the site is crucial. Actual engineering sites often exhibit complex geological features, significant spatial uncertainties, and high drilling costs, making it challenging to build complex site models given the sparse availability of borehole data.

[0003] Existing geological modeling methods for complex sites can be broadly categorized into four types: profile connection method, multi-level digital elevation modeling method, stochastic modeling method, and generalized triangular prism modeling method. The profile connection method constructs geologically consistent profiles based on borehole data, then processes these profiles with contour lines and establishes a 3D geological model through automatic or semi-automatic triangulation. The multi-level digital elevation modeling method, based on borehole data, numbers the strata in the study area according to their sedimentary sequence, establishes the site's solid geometry through stratigraphic splicing, and uses Delaunay triangulation to mesh each soil layer. The stochastic modeling method, based on known geological data, simulates the stratigraphic distribution of unknown areas using stochastic algorithms, generating multiple possible geological geometries, and then meshes to establish the site model. The generalized triangular prism modeling method, based on stratigraphic elevation data, uses interpolation algorithms to form stratigraphic interfaces and vertically connects these interfaces using triangular prism elements to establish the site model. Despite the significant research achievements, engineers still need to spend considerable time manually analyzing borehole data during the modeling process to determine the relative positions of soil layers and form the site geometry. This entire process is time-consuming, labor-intensive, lacks automation, is prone to errors, and presents significant modeling challenges. Furthermore, generating high-quality meshes based on geometric models is also a difficult problem, often encountering issues such as frequent manual intervention, low meshing success rate, mesh distortion, or mesh node decoupling. Currently, tetrahedral mesh generation technology is relatively mature, but tetrahedral meshes have poor computational accuracy, and automated generation of high-precision hexahedral meshes is still not feasible.

[0004] In recent years, scholars both domestically and internationally have proposed various modeling and numerical analysis methods based on electronic images. Replacing traditional CAD solid models with digital images avoids complex and time-consuming manual processing during preprocessing, providing a new approach for the discretization and numerical analysis of complex models. Electronic images can be acquired in various ways; for example, the medical field can generate diagnostic images using X-rays, CT scans, and ultrasound, while the geological field can output visualized images using remote sensing technologies such as Geographic Information Systems (GIS). Based on the pixel information of digital images, mesh generation can be performed directly to establish numerical models of complex geometries. Currently, a common method is to combine digital images with quadtree / octree mesh technology to form a scaled boundary finite element model, and then perform numerical analysis of the model using the scaled boundary finite element method (SBFEM). However, the problem of dangling nodes in quadtree / octree meshes hinders their application in traditional finite element analysis.

[0005] In summary, although the finite element mesh generation method for complex sites has achieved some research results and practical applications, it still has shortcomings, mainly in the following aspects: (1) The preprocessing process for modeling requires a lot of manual intervention, the degree of automation is low, and the modeling is difficult. (2) It is difficult to generate high-quality meshes based on the geometry of complex sites, which often leads to mesh distortion or mesh node decoupling.

[0006] Therefore, there is an urgent need to provide a method for generating finite element networks for complex sites to solve the technical problems mentioned above. Summary of the Invention

[0007] This application provides a method and system for generating finite element networks for complex sites. The method analyzes and processes the acquired borehole data to accurately describe the soil structure and construct a high-precision site material distribution image. The method then uses an octree algorithm to decompose the site material distribution image into a mesh and construct a high-precision site finite element model. This method achieves the generation of a high-precision site finite element model without manual intervention, significantly improving the modeling efficiency and automation of complex sites, while ensuring the accuracy of the model's numerical analysis.

[0008] In a first aspect, this application provides a method for generating finite element networks for complex sites, comprising the following steps: acquiring borehole data of a complex site area; wherein the borehole data includes borehole coordinate data, elevation information, and material information of each soil layer interval within the borehole; converting the borehole data of the complex site area into a set of soil and rock feature points; performing maximum probability scatter interpolation on the set of soil and rock feature points to form a site material distribution image; using the site material distribution image as input data, performing automated mesh generation based on an octree algorithm to obtain a site finite element model for seismic response analysis.

[0009] Optionally, the borehole data in the complex site area is converted into a set of soil and rock feature points, including: setting a thickness threshold δ; when the soil layer thickness is ≤ δ, extracting a soil and rock feature point at the center of the soil layer; when the soil layer thickness is > δ, uniformly extracting multiple soil and rock feature points along the depth direction at intervals of δ / 2; extracting double-marked soil and rock feature points at each soil layer interface and recording the material properties of the adjacent soil layers above and below; and storing the spatial coordinates, elevation, and material property information of each soil and rock feature point as a structured set of soil and rock feature points.

[0010] Optionally, the set of soil and rock feature points is subjected to maximum probability scatter interpolation to form a site material distribution image. Specifically, this includes: assigning material numbers to different soil layer materials based on soil layer material information; creating an independent (0,1) binary indicator variable field for each soil layer material, where 1 represents that a soil and rock feature point belongs to that material and 0 represents that a soil and rock feature point does not belong to that material; performing three-dimensional spatial interpolation calculation on each indicator variable field; comparing the probability values ​​of each soil layer material at its spatial location, and taking the material type corresponding to the highest probability as the material at that location; and storing the three-dimensional soil layer material distribution result as a site material distribution image in the form of a three-dimensional array.

[0011] Optionally, the three-dimensional spatial interpolation calculation for each of the indicator variable fields includes: interpolating each of the soil and rock feature points, separating the entire interpolation process according to the corresponding soil material number to achieve multi-stage interpolation; during the multi-stage interpolation process, determining the probability of the soil material corresponding to each soil and rock feature point based on the indicator variable field to obtain the probability distribution value of each soil material in the entire set of soil and rock feature points.

[0012] Optionally, the method for generating finite element networks for complex sites further includes: when a high-precision surface model exists, spatially registering and cropping the high-precision surface model with the site material distribution image; when a high-precision surface model is lacking, if contour data exists, extracting soil and rock feature points from the contour lines to construct a constrained Delaunay triangulation and reconstructing the surface model; if no contour data exists, using borehole elevation points to construct a constrained Delaunay triangulation and reconstructing the surface model; and using the generated surface model to crop the site material distribution image and remove voxels from the area above the surface.

[0013] Optionally, the site material distribution image is used as input data, and automated mesh generation is performed based on the octree algorithm to obtain a site finite element model for seismic response analysis. This includes: presetting the maximum size Lmax and minimum size Lmin of the mesh element; calculating the local subdivision index of the mesh element based on the differences in the soil material properties; performing octree subdivision when the differences in the soil material are significant and the mesh element size is greater than the minimum size Lmin; generating pyramid-tetrahedral transition elements in the soil material interface region; and outputting the mesh element when Lmin ≤ mesh element size ≤ Lmax to obtain a non-uniform finite element mesh.

[0014] Secondly, this application provides a finite element network generation system for complex sites. The system includes: an acquisition module for acquiring borehole data of a complex site area; wherein the borehole data includes borehole coordinate data, elevation information, and material information for each soil layer interval; a processing module for converting the borehole data of the complex site area into a set of soil and rock feature points; a mapping module for performing maximum probability scatter interpolation processing on the set of soil and rock feature points to form a site material distribution image; and a network partitioning module for using the site material distribution image as input data and performing automated mesh partitioning based on an octree algorithm to obtain a site finite element model for seismic response analysis.

[0015] Thirdly, this application also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method described above when executing the computer program.

[0016] Fourthly, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method described above.

[0017] This application has at least the following advantages: The above steps mainly involve acquiring borehole data and surface elevation data of complex sites and performing scatter interpolation to obtain site material distribution images. This accurately describes the soil structure and surface morphology of the site, providing a data foundation for the subsequent construction of finite element models of complex sites. This significantly improves the modeling accuracy and automation of complex sites. The site material distribution images are then decomposed into meshes using an octree algorithm to construct the site finite element model. The mesh generation process is based on the site material distribution images, requiring no manual intervention and improving work efficiency. The octree-based element form ensures sufficiently high mesh quality, and its combination with the finite element method ensures the accuracy and efficiency of numerical analysis, thereby guaranteeing more accurate seismic analysis in the future. Attached Figure Description

[0018] Figure 1This is a diagram illustrating the application environment of a method for generating finite element networks for complex sites in one embodiment. Figure 2 This is a flowchart illustrating the steps of a method for generating a finite element network for a complex site in one embodiment. Figure 3 This is a schematic diagram showing the conversion of borehole data into soil and rock feature points in one embodiment; Figure 4 This is a schematic diagram illustrating the steps involved in constructing a site material distribution image in one embodiment; Figure 5 This is a schematic diagram showing the probability of soil and rock feature points in sequential interpolation in one embodiment; Figure 6 This is a flowchart illustrating a two-dimensional site soil layer imaging process in one embodiment; Figure 7 This is a schematic diagram showing the fitting of the ground surface in one embodiment; Figure 8 This is a schematic diagram showing the surface adjustment of a three-dimensional site material distribution image in one embodiment; Figure 9 This is a schematic diagram of a quadtree decomposition showing a two-dimensional soil layer image in one embodiment; Figure 10 This is a schematic diagram showing the finite element network generation structure of a quadtree element in one embodiment; Figure 11 This is a schematic diagram showing the finite element network generation structure of an octree element in one embodiment; Figure 12 This is a time history comparison curve of the reference point displacement in one embodiment; Figure 13 This is a time history comparison graph of acceleration at a reference point shown in one embodiment; Figure 14 One embodiment shows a graph illustrating the variation of peak displacement and acceleration with elevation; Figure 15 One embodiment shows a site modeling area diagram; Figure 16 This is a schematic diagram showing the octree mesh decomposition structure of a site finite element model in one embodiment; Figure 17 This is a structural block diagram showing a complex site finite element network generation system in one embodiment; Figure 18 This is a schematic structural diagram of a computer device in one embodiment. Detailed Implementation

[0019] The present application 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 merely illustrative of the present application and are not intended to limit the scope of the present application.

[0020] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when used in this specification, the words “comprising” and / or “including” indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the various embodiments of this application will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details have been provided in the embodiments of this application to facilitate a better understanding of the application. However, the technical solutions claimed in this application can be implemented even without these technical details and various variations and modifications based on the following embodiments. The reference numerals in the following embodiments are for descriptive convenience and should not constitute any limitation on the specific implementation of this application. The various embodiments can be combined with and referenced by each other without contradiction.

[0022] For ease of understanding, the system to which this application applies will first be described. The finite element network generation method for complex sites provided in this application can be applied to, for example... Figure 1 The system architecture shown includes a user-space file server 103 and a terminal device 101. The terminal device 101 communicates with the user-space file server 103 via a network. The user-space file server 103 can be a file server based on the NFSv3 / v4 protocol, running in a Linux environment. NFS (Network File System) is a network abstraction on top of a file system, allowing remote clients running on the terminal device 101 to access the file system over the network in a manner similar to a local file system. The terminal device 101 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, etc. The user-space file server 103 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0023] Figure 2 A flowchart illustrating a method for generating finite element networks for complex sites, provided in this application embodiment, includes the following steps: S201. Obtain borehole data for complex site areas; the borehole data includes the coordinates, elevation, and material information of each soil layer. S202. Convert borehole data in complex site areas into a set of geotechnical feature points; S203. Perform maximum probability scatter interpolation on the set of soil and rock feature points to form a site material distribution image; S204. Using the site material distribution image as input data, the site is automatically meshed based on the octree algorithm to obtain a finite element model for seismic response analysis.

[0024] In this embodiment, the acquired borehole data is analyzed and processed to generate a set of soil and rock feature points. A high-precision electronic site material distribution image is generated based on the soil and rock feature point set using scatter interpolation, providing a geometric model for mesh generation. The site material distribution image is then automatically meshed using an octree algorithm to construct a high-precision site finite element model. This achieves high-precision site finite element model generation without manual intervention, significantly improving the modeling efficiency and automation level of complex sites while ensuring the accuracy of numerical analysis.

[0025] The following is a detailed explanation of each step: Please refer to Figure 2 As shown, step S201: Obtain borehole data for complex site areas.

[0026] In this embodiment, it should be noted that the borehole data includes the borehole's coordinate data, elevation information, and material information for each soil layer interval within the borehole. The soil layer material information refers to the fact that the soil material at different depths within each borehole varies with the borehole depth.

[0027] Please continue to refer to Figure 2 As shown, step S202 involves converting borehole data from complex site areas into a set of soil and rock feature points.

[0028] In this embodiment, it should be noted that the borehole data is processed and converted into a set of soil and rock feature points. This mainly involves representing the borehole data in the form of a computer-readable sequence or matrix. Then, in each soil layer interval, an appropriate number and location of soil and rock feature points are selected according to the sampling method of center point plus boundary point to obtain the set of soil and rock feature points.

[0029] Specifically, a thickness threshold δ is set. When the soil layer thickness is ≤ δ, a soil-rock feature point is extracted at the center of the soil layer. When the soil layer thickness is > δ, multiple soil-rock feature points are extracted uniformly along the depth direction at intervals of δ / 2. Double-marked soil-rock feature points are extracted at each soil layer interface, and the material properties of the adjacent soil layers are recorded. The spatial coordinates, elevation, and material property information of each soil-rock feature point are stored as a structured set of soil-rock feature points. By fusing borehole and surface elevation data, the soil structure and surface morphology of the site can be described more accurately.

[0030] In one example, such as Figure 3As shown, based on the thickness threshold δ, the borehole contains four layered intervals: H1, H2, H3, and H4. Within each interval, the center point and two boundary points of adjacent soil layers are selected as the extracted soil and rock feature points. After automatic batch selection, and based on the elevation information of each soil and rock feature point, combined with the coordinate data of the corresponding borehole, the spatial coordinates of each soil and rock feature point are obtained, ultimately yielding the set of soil and rock feature points.

[0031] Please refer to Figure 2 , Figure 4 As shown, in step S203, the set of soil and rock feature points is processed by maximum probability scatter interpolation to form a site material distribution image; Step S2031: Number the different soil materials according to the soil layer information; In this embodiment, it should be noted that, in one example, such as Figure 5 As shown, the borehole is divided into four soil layers, which are corresponding to the material numbers 1, 2, 3, and 4.

[0032] Step S2032: Create an independent (0, 1) binary indicator variable field for each soil material, where 1 represents that a certain soil feature point belongs to the material and 0 represents that a certain soil feature point does not belong to the material.

[0033] In this embodiment, it should be noted that, according to the preset binary indicator variable field, when a soil feature point is located inside the corresponding soil layer, its probability of belonging to that soil material is 1; when a soil feature point is located outside the corresponding soil layer, its probability of belonging to that soil material is 0. In particular, when a soil feature point is located at the interface of soil layers, the probability of it belonging to the upper and lower layers is 0.5.

[0034] Step S2033: Perform three-dimensional spatial interpolation calculations for each indicator variable field; In this embodiment, it should be noted that interpolation is performed on each soil and rock feature point, and the entire interpolation process is carried out separately according to the corresponding soil material number to achieve multi-stage interpolation. During the multi-stage interpolation process, the probability of the soil material corresponding to each soil and rock feature point is determined according to the indicator variable field to obtain the probability distribution value of each soil material in the entire set of soil and rock feature points.

[0035] like Figure 5 As shown, in one example, the borehole is divided into four soil layers corresponding to material numbers 1, 2, 3, and 4. The overall interpolation process of the borehole can be performed in four steps. According to the preset binary indicator variable field, in the first interpolation, if the soil feature point is located at soil material number 1, the indicator variable field is marked as 1; otherwise, it is marked as 0. Based on the four interpolations, the probability values ​​of each soil feature point corresponding to different interpolation batches are obtained, thus obtaining the result of the interpolation.

[0036] Step S2034: By comparing the probability values ​​of each soil layer material in spatial location, the material type corresponding to the highest probability is taken as the material at that location.

[0037] In this embodiment, it should be noted that the probability distribution of all pixels in the overall space is obtained by selecting the soil material with the highest probability as the actual soil material at each soil feature point.

[0038] In one example, such as Figure 6 (a) Figure 6 (b) Figure 6 As shown in (c), the range of 0~50m along the x and y axes is selected as the interpolation region. Pixels with a size of 0.1953m are used to fill the entire region, meaning the image pixel size is set to 256×256. The material probability of each pixel is obtained using probability-based scatter interpolation. Specifically, in one example, the site contains 6 types of soil materials, so the overall interpolation process is performed 6 times, and the corresponding material probability distribution is as follows. Figure 6 As shown in (d). Finally, the results of the interpolation are compared, and the material corresponding to the maximum probability is selected as the actual material of each pixel.

[0039] Step S2035: Store the three-dimensional soil material distribution results as a three-dimensional array of site material distribution images.

[0040] In this embodiment, it should be noted that after comparing the results of the multiple interpolations and selecting the material corresponding to the maximum probability as the actual material of each pixel, all pixels are then mapped to material numbers, thus storing the soil layer image information as a two-dimensional matrix. Finally, each soil layer is assigned a corresponding color value based on its image information, realizing the display from array to image, and obtaining the site material distribution image. The soil layer imaging result of the two-dimensional case is as follows: Figure 6 As shown in (e), the process of three-dimensional site imaging is similar to that of two-dimensional imaging, but with an added dimension in space, resulting in a three-dimensional voxel image. A voxel image can be decomposed along a coordinate axis and represented as a series of two-dimensional images.

[0041] Reference Figure 2 , Figure 4 As shown, the finite element mesh generation method for complex sites further includes: when a high-precision surface model exists, spatially registering and cropping the high-precision surface model with the site material distribution image; when a high-precision surface model is lacking, if contour line data exists, extracting soil and rock feature points from the contour lines to construct a constrained Delaunay triangulation and reconstructing the surface model; if there is no contour line data, using borehole opening elevation points to construct a constrained Delaunay triangulation and reconstructing the surface model; and using the generated surface model to crop the site material distribution image and remove voxels from the area above the surface.

[0042] In this embodiment, it should be noted that, as Figure 7 As shown, for sites with uneven surfaces, the surface elevation can be integrated to adjust the site material distribution image generated based on borehole interpolation, thereby more realistically restoring the surface morphology and providing a more accurate data foundation for subsequent modeling and seismic analysis. Specifically, when a high-precision surface model is lacking, if contour data exists, an appropriate number of surface points are extracted from the contour data to reconstruct the surface surface; the vertical projection of the extracted appropriate number of surface points is triangulated, and the surface points are connected according to the triangulation relationship to reconstruct the surface surface; the reconstructed surface surface is used to segment and adjust the imaging result, setting different colors for pixels above the surface; the portion above the surface surface is removed during the subsequent discretization process, while the portion below the surface surface is retained and associated with the corresponding soil material to obtain the final site image.

[0043] In one example, such as Figure 8 As shown in (a), before adjusting the surface, it is necessary to first set the upper limit of the initial imaging area of ​​the site. A horizontal reference plane located above the surface is selected as the upper limit of the initial imaging area to ensure that the initial imaging result covers the entire modeling range. The material between the upper limit of the initial imaging area and the surface is temporarily preset to be consistent with the material of the first layer of soil on the site. This preset can be understood as using the first layer of soil to initially fill and level the surface to obtain the initial image. Figure 8 As shown in (b), based on the initial imaging of the site, the imaging result is segmented and adjusted by fitting the surface surface. Specifically, the elevation of each pixel or voxel unit is compared with the surface elevation at the corresponding vertical position. Through elevation comparison, pixels above the surface surface are updated to a new color, which can be removed during model discretization. Pixels below the surface surface retain their original color values ​​and correspond to the corresponding soil material, resulting in the final site material distribution image.

[0044] Reference Figure 2 , Figure 9 As shown, step S204 involves using the site material distribution image as input data and automatically meshing the data using an octree algorithm to obtain a site finite element model for seismic response analysis. Specifically, this includes: presetting the maximum size Lmax and minimum size Lmin of the mesh element; calculating the local subdivision index of the mesh element based on the differences in soil material properties; performing octree subdivision when the differences in soil material are significant and the mesh element size > the minimum size Lmin; generating pyramid-tetrahedral transition elements in the soil material interface region; and outputting the mesh element when Lmin ≤ mesh element size ≤ Lmax to obtain a non-uniform finite element mesh.

[0045] In this embodiment, it should be noted that the concept of finite element analysis is to replace a complex problem with a simpler one before solving it. Here, the site material distribution image is used as input data, and the model is discretized by automatic mesh generation using the octree algorithm, thereby constructing a finite element model of the complex site. Subsequently, seismic response analysis of the site is performed based on the finite element method to obtain the site's displacement and acceleration response.

[0046] Specifically, the maximum and minimum sizes of the decomposed mesh cells are preset; local subdivision indices of the mesh cells are calculated based on the differences in soil material properties, and the cells are recursively decomposed step by step according to preset data until all cells meet the standard of color intensity uniformity or reach the minimum size requirement. When the differences in soil material are significant and the mesh cell size is greater than the minimum size Lmin, octree subdivision is performed.

[0047] The quadtree (2D) / octree (3D) algorithm is a recursive decomposition technique based on hierarchical trees. The general idea is to first determine the square or cube bounding box containing the image, using this as the root unit for initial decomposition, forming the background mesh of the image. Starting from the background mesh, the mesh is recursively decomposed level by level according to certain subdivision criteria until the representation of the problem reaches the specified precision. During the subdivision process, the maximum difference between the subdivision levels of adjacent units is usually limited to 1, which is called the 2:1 rule, and the resulting mesh is called a balanced quadtree / octree mesh. Here, the maximum and minimum sizes of the decomposed mesh units are preset first, and the boundary range of the site material distribution image is determined to generate the background mesh. Then, the background mesh is recursively decomposed level by level according to the mesh unit size until the minimum unit of the background mesh reaches the minimum size requirement of the preset mesh unit. Finally, the mesh units with dangling nodes are subdivided again. When Lmin ≤ mesh unit size ≤ Lmax, the mesh unit is output, resulting in a non-uniform finite element mesh, forming a site numerical model that can be used for finite element analysis.

[0048] In one example, such as Figure 9 Before decomposing the two-dimensional site material distribution image shown, it is necessary to determine the maximum and minimum cell sizes. According to the quadtree partitioning principle, both the maximum and minimum sizes should be integer powers of 2 (in pixels). The determination of the maximum size should also meet the requirements of engineering fluctuation theory. A color threshold is used to control the uniformity of color intensity within a cell. If the color difference within a cell exceeds the threshold, the cell will be divided into four sub-cells until all cells meet the uniformity standard or reach the minimum size requirement. Specifically, the maximum cell size is set to 4, the minimum size to 1, and the color threshold to 0, resulting in the quadtree decomposition of the two-dimensional site material distribution image. Figure 10As shown, based on the geometric center of the mesh element, the mesh element with suspended nodes is subdivided twice to obtain the finite element model of the site; where suspended nodes refer to points located on the edge of the mesh element.

[0049] Specifically, in one example, such as Figure 10 , 11 As shown, the first step is to discretize the surfaces with suspended nodes. This step can be referenced for the finite element mesh generation of quadtree elements. Then, the octree elements that have been discretized on the surface are further subdivided by connecting the geometric center of the element to each node. The original octree elements are further subdivided into several tetrahedral or pentahedral elements, thereby realizing the generation of the finite element mesh and obtaining the finite element model of the site.

[0050] In addition, based on the obtained site finite element model, the vertical displacement around the site finite element model is constrained, and a horizontal ground motion is applied at the bottom to calculate the seismic response of the site.

[0051] In this embodiment, it should be noted that the vertical displacement around the constrained model is controlled, and a horizontal ground motion is applied at the bottom to calculate the seismic response of the site and analyze the results. Here, borehole data SH01, located at the relative center of the complex site, is used as reference data to establish a horizontally layered site model (simplified model), which is then compared with the constructed finite element model (refined model). Using the surface of borehole SH01 as the reference point, the displacement and acceleration time history curves of the simplified and refined models at the reference point are output, as shown below. Figure 12 , Figure 13 As shown, it can be seen that the refined model's displacement response at the reference point lags slightly behind that of the simplified model overall; reference Figure 13 The refined model exhibits a smaller acceleration response amplitude at the reference point than the simplified model; for example... Figure 14 As shown, the peak displacement response of the refined model is smaller near the bottom of the model than that of the simplified model, but larger near the surface. The simplified model exhibits a more pronounced amplification effect on acceleration, especially at the site surface. Based on the above comparative analysis, it is evident that site modeling provides higher precision and more accurate seismic response analysis.

[0052] Based on the above methods, a finite element model of a complex site is constructed for seismic analysis. The entire construction process is illustrated here through a specific example. Specifically, for a certain site area... Figure 15The area, indicated by the blue dashed box, measures 160m × 160m. The study area contains 38 borehole data points, each with a different drilling depth. Partial geological profiles provided in the exploration report show that the soil layers in the study area are undulating and generally exhibit an oblique distribution. Furthermore, the distribution of different soil materials is complex and involves alternating layers. A total of six soil materials are included in the study area, and their corresponding material parameters are shown in Table 1.

[0053] Table 1 Soil material parameters Based on the adjusted modeling coordinate system, borehole data is organized, and interpolated soil and rock feature points are obtained. Then, interpolation intervals are set according to the specified modeling area. The upper limit of the interpolation interval should be above the surface, with the material between it and the surface initially set as plain fill. The lower limit of the interpolation interval is below the bottom of all boreholes, and the material between it and the bottom can be defined as either moderately weathered sandstone or moderately weathered silty mudstone, with the corresponding material parameters taken as the average of these two materials. The coordinates and material probabilities of the soil and rock feature points are input, and a probability-based stepwise interpolation method is used to perform scattered interpolation on the site modeling area, forming a site material distribution image. Based on this, the surface is adjusted using the fitted surface curve to obtain the final site image. Subsequently, the final site image is used as input data, with the maximum and minimum cell sizes set to 8 and 1 respectively, and an octree mesh decomposition is performed according to a 2:1 rule, as shown below. Figure 16 As shown. Next, the mesh elements with suspended nodes are further subdivided to obtain the final finite element model of the site. The vertical displacement around the finite element model of the site is constrained, and a horizontal seismic motion is applied at the bottom to calculate the seismic response of the site and analyze the results.

[0054] The implementation principle of this embodiment is as follows: The above steps mainly involve obtaining borehole data and surface elevation data of complex sites and performing maximum probability scatter interpolation to obtain a site material distribution image. This accurately describes the soil structure and surface morphology of the site, providing a data foundation for the subsequent construction of a finite element model of the complex site. This significantly improves the modeling accuracy and automation of complex sites. The site material distribution image is then decomposed into a mesh using an octree algorithm. The mesh generation process is based on the site material distribution image and requires no manual intervention, improving work efficiency. The quadtree / octree element form ensures sufficiently high mesh quality. Combined with the finite element method, this ensures the accuracy and efficiency of numerical analysis, thereby guaranteeing more accurate seismic analysis in the future.

[0055] Reference Figure 17As shown, this application also provides a finite element network generation system for complex sites. This system may include: an acquisition module 301, a processing module 302, a mapping module 303, and a network partitioning module 304. The main functions of each component module are as follows: The acquisition module 301 is used to acquire borehole data in complex site areas; the borehole data includes borehole coordinate data, elevation information, and material information of each soil layer interval; Processing module 302 is used to convert borehole data in complex site areas into a set of geotechnical feature points; The mapping module 303 is used to perform maximum probability scatter interpolation on the set of soil and rock feature points to form a site material distribution image; The mesh generation module 304 is used to take the site material distribution image as input data, perform automated mesh generation based on the octree algorithm, and obtain the site finite element model for seismic response analysis.

[0056] like Figure 18 The diagram shown is a block diagram of a computer device according to an embodiment of this application. The term "computer device" is intended to represent various forms of digital computers or mobile devices. The digital computer may include a desktop computer, a portable computer, a workbench, a personal digital assistant, a server, a mainframe computer, and other suitable computers. The mobile device may include a tablet computer, a smartphone, a wearable device, etc.

[0057] like Figure 18 As shown, device 600 includes a computing unit 601, a ROM 602, a RAM 603, a bus 604, and an input / output (I / O) interface 605. The computing unit 601, ROM 602, and RAM 603 are interconnected via the bus 604. The input / output (I / O) interface 605 is also connected to the bus 604.

[0058] The computing unit 601 can execute various processes in the method embodiments of this application according to computer instructions stored in read-only memory (ROM) 602 or computer instructions loaded from storage unit 608 into random access memory (RAM) 603. The computing unit 601 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. The computing unit 601 can include, but is not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. In some embodiments, the methods provided in the embodiments of this application can be implemented as computer software programs, which are tangibly contained in a computer-readable storage medium, such as storage unit 608.

[0059] RAM 603 may also store various programs and data required for the operation of device 600. Part or all of the computer program may be loaded and / or installed on device 600 via ROM 602 and / or communication unit 609.

[0060] The input unit 606, output unit 607, storage unit 608, and communication unit 609 in device 600 can be connected to I / O interface 605. The input unit 606 can be, for example, a keyboard, mouse, touchscreen, or microphone; the output unit 607 can be, for example, a display, speaker, or indicator light. Device 600 can exchange information and data with other devices through the communication unit 609. It should be noted that the device may also include other components necessary for normal operation. It may also include only the components necessary for implementing the solution of this application, without necessarily including all the components shown in the figures.

[0061] Various implementations of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SOCs), payload programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof.

[0062] The computer instructions used to implement the methods of this application may be written in any combination of one or more programming languages. These computer instructions may be provided to the computing unit 601 such that when executed by the computing unit 601, such as a processor, the computer instructions cause the execution of the steps involved in the embodiments of the methods of this application.

[0063] The computer-readable storage medium provided in this application can be a tangible medium that can contain or store computer instructions for performing the steps involved in the method embodiments of this application. The computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, and other forms of storage media.

[0064] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for generating finite element networks for complex sites, characterized in that, Includes the following steps: Obtain borehole data in complex site areas; among which, The borehole data includes the borehole's coordinate data, elevation information, and material information for each soil layer within the borehole. The borehole data in complex site areas are converted into a set of soil and rock feature points. The set of soil and rock feature points is subjected to maximum probability scatter interpolation to form a site material distribution image; Using the site material distribution image as input data, automated mesh generation is performed based on the octree algorithm to obtain a site finite element model for seismic response analysis.

2. The method for generating finite element networks for complex sites according to claim 1, characterized in that, The process of converting the borehole data in complex site areas into a set of geotechnical feature points includes: Set a thickness threshold δ. When the soil layer thickness is ≤ δ, extract a soil feature point at the center of the soil layer. When the soil layer thickness is > δ, extract multiple soil feature points uniformly along the depth direction at intervals of δ / 2. Extract double-marked soil feature points at each soil layer interface and record the material properties of the adjacent soil layers above and below. Store the spatial coordinates, elevation, and material property information of each soil feature point as a structured set of soil feature points.

3. The method for generating finite element networks for complex sites according to claim 2, characterized in that, The step of performing maximum probability scatter interpolation on the set of soil and rock feature points to form a site material distribution image specifically includes: Different soil materials are assigned material numbers based on soil material information; Create an independent (0, 1) binary indicator variable field for each soil layer material, where 1 represents that a certain soil feature point belongs to the material and 0 represents that a certain soil feature point does not belong to the material; perform three-dimensional spatial interpolation calculation for each indicator variable field; by comparing the probability values ​​of each soil layer material at its spatial location, take the material type corresponding to the highest probability as the material at that location; store the three-dimensional soil layer material distribution results as a site material distribution image in the form of a three-dimensional array.

4. The method for generating finite element networks for complex sites according to claim 3, characterized in that, The step of performing three-dimensional spatial interpolation calculations for each of the indicator variable fields includes: Interpolation is performed on each of the soil and rock feature points, and the entire interpolation process is carried out separately according to the corresponding soil layer material number to achieve interpolation in stages. During the interpolation process, the probability of the soil material corresponding to each soil feature point is determined according to the indicator variable field, so as to obtain the probability distribution value of each soil material in the entire set of soil feature points.

5. The method for generating finite element meshes for complex sites according to any one of claims 1-4, characterized in that, Also includes: When a high-precision surface model exists, it is spatially registered and cropped with the site material distribution image; when a high-precision surface model is lacking... If contour data exists, extract soil and rock feature points from the contours to construct a constrained Delaunay triangulation and reconstruct the surface model; if contour data is unavailable, use borehole elevation points to construct a constrained Delaunay triangulation and reconstruct the surface model; use the generated surface model to crop the site material distribution image and remove voxels from the area above the surface.

6. The method for generating finite element networks for complex sites according to claim 1, characterized in that, The process involves using the site material distribution image as input data, performing automated mesh generation based on the octree algorithm, and obtaining a site finite element model for seismic response analysis, including: The maximum size Lmax and minimum size Lmin of the preset mesh cells are used; the local subdivision index of the mesh cells is calculated based on the differences in the properties of the soil material; when the differences in the soil material are significant and the mesh cell size is greater than the minimum size Lmin, octree subdivision is performed; pyramid-tetrahedral transition elements are generated in the interface region of the soil material; when Lmin ≤ mesh cell size ≤ Lmax, the mesh cells are output to obtain a non-uniform finite element mesh.

7. A finite element network generation system for complex sites, characterized in that, include: The acquisition module is used to acquire borehole data in complex site areas; wherein, the borehole data includes borehole coordinate data, elevation information, and material information of each soil layer interval; The processing module is used to convert the borehole data of complex site areas into a set of soil and rock feature points; The mapping module is used to perform maximum probability scatter interpolation on the set of soil and rock feature points to form a site material distribution image; The network meshing module is used to take the site material distribution image as input data, perform automated meshing based on the octree algorithm, and obtain a site finite element model for seismic response analysis.

Images