Method and system for constructing ultra-large anchorage digital twinborn body

Abstract

The invention discloses an ultra-large anchorage digital twinborn body construction method and system, and the method comprises the following steps: arranging 16-hole inclinometer tubes to form a three-dimensional monitoring array, collecting data in real time through the inclinometer tubes, and generating accurate grid point data through an IDW interpolation algorithm; carrying out optimization processing on the digital model, wherein the optimization processing comprises structure optimization, efficient compression algorithm application and instantiation rendering; three-dimensional visualization is achieved based on the WebGL technology, and the three-dimensional visualization comprises model translucency, grid data accurate alignment, thermodynamic diagram rendering achieved through a custom shader and the like; a comprehensive digital twin system platform is constructed, and full-life-cycle management and real-time monitoring of the super-large anchorage structure are achieved. According to the invention, by arranging a sensor acquisition and real-time monitoring mode, the precision and comprehensiveness of data acquisition are remarkably improved, real-time monitoring of overall deformation of the anchorage is realized, high-precision, real-time and interactive digital twinborn visualization of the super-large anchorage structure is realized, and powerful support is provided for engineering monitoring and decision making.

Description

Technical Field

[0001] The present invention relates to the technical field of civil engineering monitoring and bridge structure digital twins, and in particular to a method and system for constructing a super-large anchorage digital twin. Background Art

[0002] With the rapid development of modern transportation infrastructure, large-scale bridge projects are becoming increasingly complex and intelligent. As the key load-bearing structure of the bridge, the safety and stability of the anchor are directly related to the service life and operational safety of the entire bridge. However, traditional monitoring methods have obvious limitations: high-precision monitoring methods are often inefficient and costly, while high-efficiency methods may lack accuracy and have large deviations. More importantly, these methods mostly obtain data from discrete points, which makes it difficult to fully reflect the overall properties of the anchor. In addition, many traditional methods also face problems such as lack of real-time performance, great influence of environmental factors, and difficulty in achieving long-term continuous monitoring, which cannot meet the needs of modern bridge projects for accurate, real-time, and comprehensive monitoring.

[0003] Therefore, it is urgent to develop an innovative monitoring system to meet the needs of efficient management of modern bridge projects. This system based on digital twin technology should have the characteristics of real-time monitoring of global properties, high-precision visualization, efficient data processing, intelligent early warning mechanism, remote monitoring capability and system integration. It can capture the overall deformation and stress distribution state of the anchor structure in real time, provide an intuitive and accurate three-dimensional visualization interface, quickly process and analyze a large amount of monitoring data, and realize intelligent early warning of potential risks. At the same time, it supports remote access and control, which is convenient for experts to conduct remote diagnosis and decision-making. This system is not only an inevitable trend of technological development, but will also provide strong technical support for the management of the entire life cycle of bridges, and promote the development of bridge projects in a smarter and safer direction. Summary of the invention

[0004] In view of the above defects or improvement needs of the prior art, the present invention aims to solve the limitations of traditional monitoring methods: high-precision monitoring methods are often inefficient and costly, while high-efficiency methods may lack accuracy and have large deviations. More importantly, these methods mostly obtain data from discrete points, which is difficult to fully reflect the overall state of the anchor. In addition, many traditional methods also face problems such as lack of real-time performance, great influence of environmental factors, and difficulty in achieving long-term continuous monitoring, and cannot meet the needs of modern bridge engineering for accurate, real-time, and comprehensive monitoring.

[0005] The present invention provides a method and system for constructing a super-large anchorage digital twin, comprising the following steps:

[0006] Step S1. A three-dimensional monitoring array of inclinometer tubes evenly distributed in the horizontal and vertical directions is set on the anchored ground-connected wall structure, and the collected displacement data is transmitted to the on-site data processing center in real time through an information transmission device;

[0007] Step S2. The three-dimensional structural features of the anchored ground-connected wall are converted into two-dimensional coordinates by using a 2D surface unfolding interpolation method. The two-dimensional plane inverse distance interpolation method is used in the unfolded two-dimensional coordinate system to complete the interpolation and expansion of the grid data, and the grid point displacement values ​​and their corresponding coordinate data are stored in a database in real time in a matrix form;

[0008] Step S3. Create a BIM model based on the structural data of the anchored ground-connected wall recorded during the design and construction process, optimize the model structure and compress the model file size, and then reduce the computer memory usage and rendering call times in model loading and rendering by repeating component instantiation to obtain a lightweight model;

[0009] Step S4. Initialize the 3D scene, configure the orthogonal camera to adapt to the anchoring of large structures; set the ambient light and parallel light source to enhance the three-dimensional effect; import the lightweight model in step S3 and configure the WEB 3D scene, align the grid data stored in the database according to step S2 with the BIM model coordinate system, and then process the grid data through the shader to complete the heat map rendering and visualization;

[0010] Step S5. Integrate multi-source heterogeneous data including real-time monitoring data from sensor networks, BIM model data, and historical construction data, perform integrated display of monitoring data, and provide a multi-dimensional visualization interface.

[0011] Furthermore, the three-dimensional monitoring array in step S1 is: a three-dimensional monitoring array of inclinometer tubes with no less than 16 holes evenly distributed in the horizontal and vertical directions, the depth of the inclinometer tube is consistent with the ground-connected wall structure, the guide groove of the inclinometer tube is perpendicular to the edge of the foundation pit and points to the center of the foundation pit, and the sensor collection frequency in the inclinometer tube is 1 time / hour.

[0012] Furthermore, the information transmission device in step S1 includes a distributed optical fiber sensor demodulator, an industrial-grade switch and an optical fiber network.

[0013] Furthermore, the 2D surface expansion interpolation method in step S2 includes:

[0014] (1) Establishment of coordinate system: The central axis of the ground-connected wall cylinder is set as the z-axis of the Cartesian coordinate system, and the center of the cylinder on the ground is the origin;

[0015] (2) Coordinate transformation: replace the Cartesian sensor coordinates and values ​​(x, y, z) with the cylindrical coordinate system (r, θ, z); where r is the distance from the sensor to the central axis, and θ is the angle of the sensor on the horizontal plane, θ∈[0,2π);

[0016] (3) Boundary condition continuity processing: To ensure the continuity at θ = 0 and θ = 2π, that is, the continuity of the left and right boundaries after the cylinder is unfolded, the data at θ = 0 is copied at θ = 2π;

[0017] (4) Coordinate normalization: normalize the angle θ according to the actual arc length of the cylindrical surface, that is, x_θ = θ × r, where the x-axis represents the distance along the circumference and the unit is the same as the z-axis (height);

[0018] (5) Using the two-dimensional plane inverse distance interpolation method, the calculation formula is:

[0019]

[0020]

[0021] In the formula is the estimated value of position x0, Z(x i ) is the known position x i The observed value of the unknown point (x0, y0) is the predicted value of the position are all known points (x i ,y i ) is the weighted sum of the values ​​Z of the unknown points. The weight of each known point is determined by the αth power of the inverse of its distance di from the unknown point; di represents the straight-line distance between the unknown point and the i-th known point.

[0022] Furthermore, the specific method for optimizing the model structure in step S3 includes: deleting unnecessary internal components, removing hidden surfaces and overlapping geometries, merging coplanar small patches, simplifying complex surfaces, reducing regional surfaces with small visual impact, organizing the model tree structure, and deleting redundant groups and subgroups to flatten the model structure.

[0023] Furthermore, the specific method for compressing the model file size in step S3 includes: using the Draco compression algorithm to quantize vertex coordinates, converting floating point numbers into integers, reducing data precision and maintaining visual effects; using predictive coding technology to compress vertex data, using the spatial relationship between adjacent vertices to reduce data redundancy; encoding the topological structure to reduce redundant information. Set the compression level to complete model size compression and maintain decoding speed.

[0024] Furthermore, the specific method of the shader processing data in step S4 includes:

[0025] (1) Data normalization processing, calculate the spatial coordinates and the given displacement normalization value v, assuming the minimum displacement value v_min and the maximum displacement value v_max, the normalization formula is:

[0026] t=(v-v_min) / (v_max-v_min)

[0027] (2) Adopting the solution of additive mixing, for each pixel in the overlapping area, a new color value C_result is assigned as follows:

[0028] C_result=min(C1+C2+...+Cn,1.0)

[0029] Among them, C1, C2, ..., Cn are the color values ​​of the n overlapping points at the pixel position;

[0030] Furthermore, the step S5 of providing a multi-dimensional visualization interface function includes: providing a multi-dimensional visualization interface, including a 3D model view, a 2D cross-section view, and a time series data chart.

[0031] As another aspect of the present invention, it also relates to a super-large anchorage digital twin construction system, including a large anchorage structure real-time data acquisition unit, a grid data generation unit, a BIM lightweight processing unit, a three-dimensional point cloud real-time rendering unit, and a digital twin platform unit;

[0032] The large anchor structure real-time data acquisition unit is used to collect original sensor displacement data in real time;

[0033] The grid data generation unit is used to expand the collected original data to enrich the grid point data set;

[0034] The BIM lightweight processing unit is used to construct a digital model of a large anchor structure and optimize the model to reduce model information redundancy, making the model light and efficient in actual operation;

[0035] The three-dimensional point cloud real-time rendering unit is used to color the digital information model of the constructed anchor structure;

[0036] The digital twin platform unit is used to integrate all multi-source heterogeneous data and generate visual graphics to achieve full life cycle management and real-time monitoring of super-large anchor structures;

[0037] As another aspect of the present invention, it also relates to a computer-readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to construct a method for constructing a digital twin of an ultra-large anchor for monitoring a large bridge project.

[0038] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0039] 1. The method for constructing a digital twin of a super-large anchor of the present invention significantly improves the accuracy and comprehensiveness of data acquisition and realizes real-time monitoring of the global deformation of the anchor by arranging sensors in a real-time monitoring manner and arranging a high-speed data acquisition and transmission network.

[0040] 2. The method for constructing a digital twin of a super-large anchorage of the present invention greatly reduces the amount of model data and improves rendering efficiency through BIM lightweight processing, so that complex models can run smoothly on ordinary equipment.

[0041] 3. The method for constructing a digital twin of a super-large anchorage of the present invention uses WebGL technology to achieve real-time rendering of large-scale three-dimensional point clouds, greatly improving the visualization effect and interactive experience; the Web-based visualization system facilitates multi-terminal access and remote monitoring, improving monitoring efficiency and decision-making speed; the constructed digital twin platform provides a strong guarantee for the safe operation of the bridge, and helps to timely discover and deal with potential risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the overall process of a preferred embodiment of the present invention;

[0043] Figure 2 A schematic diagram of a two-dimensional expansion method of a three-dimensional cylinder according to a preferred embodiment of the present invention;

[0044] Figure 3 A schematic diagram of a BIM model of a preferred embodiment of the present invention;

[0045] Figure 4 A schematic diagram showing the completion of three-dimensional rendering of a preferred embodiment of the present invention;

[0046] Figure 5 Schematic diagram of a digital twin platform system according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0048] Please refer to Figure 1 , an embodiment of the present invention provides a method for constructing a super-large anchorage digital twin, comprising:

[0049] 1) Data collection and grid data generation

[0050] According to the geometric characteristics and stress characteristics of the anchored ground-connected wall structure, a 16-hole inclinometer three-dimensional monitoring array is formed in the horizontal and vertical directions. The depth of the inclinometer tube is consistent with the ground-connected wall structure, and the guide groove is perpendicular to the edge of the foundation pit and points to the center of the foundation pit. The sensor acquisition frequency is 1 time / hour. A distributed optical fiber sensor demodulator is used to achieve synchronous high-speed acquisition of all sensors. The collected data is transmitted to the on-site data processing center in real time through industrial-grade switches and optical fiber networks.

[0051] In addition, in some other preferred embodiments, the types and functions of sensors are further expanded. In addition to the inclinometer tube sensor, more types of sensors are introduced, such as stress sensors, strain sensors, temperature sensors, etc., to obtain more comprehensive information on the state of the anchor structure. At the same time, the sensor layout strategy is optimized, and the optimal sensor distribution scheme is determined through numerical simulation or physical experiments to improve the accuracy and reliability of data collection. In addition, more advanced data transmission technologies are explored, such as 5G communication technology or low-power wide area network technology, to further improve the speed and stability of data transmission and ensure the delay-free transmission of real-time monitoring data.

[0052] Please refer to Figure 2 In the numerical processing method, 3D spatial interpolation is performed directly in the three-dimensional Cartesian coordinate system (x, y, z), which retains the original spatial information of the ground-connected wall. 2D surface unfolding interpolation is to "unfold" the cylindrical ground-connected wall into a two-dimensional plane. The circumferential coordinate is mapped to the x-axis, representing the position along the circumference, and the height direction is mapped to the y-axis, retaining the original height information. The surface interpolation is performed on this (x, y) plane. The consideration for using the 2D unfolding method is that the sensors are deployed on the continuous surface of the ground-connected wall. 2D unfolding can better capture the spatial relationship and constraints between sensors on the surface, which may be ignored in pure 3D interpolation. The 2D plane unfolding method is as follows:

[0053] (1) Establishment of coordinate system: The central axis of the ground-connected wall cylinder is set as the z-axis of the Cartesian coordinate system, and the center of the cylinder on the ground is the origin;

[0054] (2) Coordinate transformation: replace the Cartesian sensor coordinates and values ​​(x, y, z) with the cylindrical coordinate system (r, θ, z). Where r is the distance from the sensor to the central axis, which should be a constant in this case and can be omitted, and θ is the angle of the sensor on the horizontal plane, θ∈[0,2π);

[0055] (3) Boundary condition continuity processing: To ensure the continuity at θ = 0 and θ = 2π (i.e., the continuity of the left and right boundaries after the cylinder is unfolded), the data at θ = 0 is copied at θ = 2π;

[0056] (4) Coordinate normalization: Normalize the angle θ according to the actual arc length of the cylindrical surface, that is, x_θ = θ × r, where the x-axis represents the distance along the circumference and the unit is the same as the z-axis (height).

[0057] The two-dimensional plane inverse distance interpolation method is used, and the calculation formula is:

[0058]

[0059]

[0060] In the formula is the estimated value of position x0, Z(x i ) is the known position x i The observed value and the predicted value of the unknown point (x., y.) are all known points (x i ,y i ) is the weighted sum of the values ​​Z of the unknown points. The weight of each known point is determined by the αth power of the inverse of its distance di from the unknown point; di represents the straight-line distance between the unknown point and the i-th known point.

[0061] In this case, the circular curved ground-connected wall is divided into 25 parts vertically and 180 parts circumferentially. The grid size is approximately 2.5m×2.5m, with a total of 24*179=4296 grid points.

[0062] The grid point displacement values ​​and their corresponding coordinate data are stored in real time in the MongoDB database in matrix form.

[0063] (2) BIM lightweight processing

[0064] Please refer to Figure 3 , using Autodesk Revit software to create a high-precision BIM model of the anchorage. Optimize the model structure and delete unnecessary internal components, such as temporary supports and auxiliary structures for construction. Remove hidden faces and overlapping geometries to reduce redundant patches. Merge coplanar small patches to reduce the total number of patches. Simplify complex surfaces and use fewer polygons to approximate the surfaces while keeping the geometric accuracy within an acceptable range. For areas with little visual impact, the surface accuracy can be appropriately reduced. Optimize the model hierarchy and organize the model tree structure reasonably to facilitate subsequent selection and operation. Delete redundant groups and subgroups to flatten the model structure.

[0065] The open source Draco compression algorithm is used to quantize vertex coordinates and convert floating point numbers into integers, reducing data precision but maintaining visual effects. Predictive coding technology is used to compress vertex data, and the spatial relationship between adjacent vertices is used to reduce data redundancy. The topological structure is encoded to reduce redundant information. The appropriate compression level is set to strike a balance between file size and decoding speed. The compressed model file size can be reduced by about 70%-80% while maintaining acceptable visual quality.

[0066] Identify repeated components in the model, such as prestressed anchors, rebars and other standard components. Convert repeated components to instances, retaining a complete copy of the geometry data as a prototype. For other identical components, only the position, rotation and scale information is stored. Implement instance rendering technology, which uses a single draw call to render multiple instances during rendering, significantly reducing memory usage and the number of rendering calls.

[0067] The optimized BIM model is exported to the glTF (GL Transmission Format) format suitable for Web3D rendering. It supports compressed geometry data, can be efficiently streamed, and is widely supported by Web3D libraries such as Three.js.

[0068] In the lightweight processing of BIM models, more efficient model optimization algorithms and compression methods can be studied in depth. For example, combined with artificial intelligence technology, such as the model compression algorithm in deep learning, the BIM model can be automatically optimized and compressed while maintaining the accuracy and integrity of the model. For mesh data generation and storage, more accurate interpolation methods or numerical simulation methods can be considered to improve the accuracy and resolution of mesh data. In addition, it is also possible to study how to better integrate model data from different sources and different precisions to build a more accurate digital twin.

[0069] (3) Real-time rendering of 3D point cloud

[0070] 1) Scene initialization and BIM model import

[0071] Initialize the 3D scene based on the Three.js library of WebGL 2.0, and configure the orthographic camera to adapt to the anchoring of large structures. Set the ambient light and parallel light source to enhance the three-dimensional effect. Import the optimized BIM model and apply the MeshPhongMaterial material with a transparency of 0.5 to achieve the semi-transparent effect. Use THREE.GLTFLoader to load the model to ensure efficient rendering.

[0072] 2) Grid data processing and coordinate alignment

[0073] Convert the mesh data generated by monitoring into the BufferGeometry format of Three.js, and use Float32Array to store the position and displacement information. Implement an accurate coordinate transformation algorithm to align the mesh data with the BIM model coordinate system, with the error controlled within 1cm. Use Three.js's Matrix4 for coordinate transformation to ensure high-precision alignment.

[0074] 3) Heat map rendering and visualization

[0075] Please refer to Figure 4 , create ShaderMaterial to implement custom shading. The vertex shader processes the position and displacement data, calculates the spatial coordinates and the given displacement normalization value v, assuming the minimum value v_min and the maximum value v_max, the normalization formula is:

[0076] t=(v-v_min) / (v_max-v_min)

[0077] In point cloud heat map rendering, when multiple points with blurred edges overlap each other, simple transparency overlay may lead to color saturation or distortion, blurred boundaries of overlapping areas, and difficulty in accurately expressing the density of points.

[0078] To solve this problem, an additive blending solution is applied. For each pixel in the overlapping area, the new color value is:

[0079] C_result=min(C1+C2+...+Cn,1.0)

[0080] Among them, C1, C2, ..., Cn are the color values ​​of the n overlapping points at the pixel position.

[0081] The specific implementation method is: map the displacement value to the color space and set the appropriate transparency of 0.3. The rendering process: a. Enable additive blending mode, b. Render each point in sequence according to the spatial position of the point, c. For each pixel, the GPU automatically performs additive blending, d. Apply tone mapping or gamma correction to enhance the visual effect.

[0082] (4) Digital Twin System Platform

[0083] Please refer to Figure 5 ,The embodiment of the present invention provides a digital twin platform that integrates multi-source heterogeneous data such as real-time monitoring data from sensor networks, BIM model data, historical construction data, etc. Distributed data storage technology is adopted and NoSQL databases such as MongoDB are used to achieve efficient large-scale data management.

[0084] Based on WebGL technology, the integrated display of BIM models and monitoring data is realized. Provide a multi-dimensional visualization interface, including 3D model views, 2D profiles, time series data charts, etc. Support user-defined views and interactive operations, such as measurement point query, historical data playback, warning information display, etc. The server processes sensor data regularly and generates updated grid data. The client requests updated data in JSON format through FileLoader and dynamically refreshes BufferGeometry attributes. Use InstancedBufferGeometry to implement instantiation rendering of points, and Web Worker handles intensive computing tasks. Optimize rendering resolution and refresh strategy to improve performance on different devices.

[0085] The digital twin system platform can further expand its functions, such as adding a structural health assessment module to conduct real-time assessment and prediction of the health status of the anchor structure based on the collected data and model analysis results. At the same time, an intelligent decision support system is developed to automatically generate maintenance recommendations and decision plans based on the results of the structural health assessment. In addition, the integration and interaction with other bridge management systems should be strengthened to achieve data sharing and collaborative management, and improve the efficiency and level of bridge management throughout the life cycle. Virtual reality (VR) and augmented reality (AR) technologies can also be used to provide users with a more immersive interactive experience, making it easier for experts to conduct remote diagnosis and decision-making more intuitively.

[0086] Example 2

[0087] The embodiment of the present invention further provides a system for constructing a super-large anchorage digital twin, comprising:

[0088] Real-time data acquisition unit for large anchoring structures: This unit is the source of data for the entire system. It realizes data acquisition through a three-dimensional monitoring array of inclinometer tubes carefully set up on the anchoring ground-connected wall structure. The array is evenly distributed in the horizontal and vertical directions, with no less than 16 holes of inclinometer tubes. The depth of the inclinometer tubes is consistent with the ground-connected wall structure. Its guide groove is perpendicular to the edge of the foundation pit and points to the center of the foundation pit. The sensor acquisition frequency is 1 time / hour to ensure the scientificity and effectiveness of the collected data. At the same time, it is equipped with information transmission devices including distributed optical fiber sensor demodulators, industrial-grade switches and optical fiber networks, which can accurately and real-time transmit the collected original sensor displacement data to the on-site data processing center, providing basic data guarantee for subsequent data processing and analysis.

[0089] Grid data generation unit: This unit is mainly responsible for processing and expanding the collected raw data. It uses the 2D surface expansion interpolation method to first convert the Cartesian sensor coordinates and values ​​(x, y, z) into a cylindrical coordinate system (r, θ, z), and after processing the continuity of boundary conditions (copying the data at θ = 0 at θ = 2π) and coordinate normalization, it uses the two-dimensional plane inverse distance interpolation method in the expanded two-dimensional coordinate system. In this way, the grid data interpolation expansion is completed, the grid point data set is enriched, and the grid point displacement value and its corresponding coordinate data are stored in the database in real time in matrix form, laying a data foundation for the subsequent combination with the BIM model and visual display.

[0090] BIM lightweight processing unit: This unit is responsible for building and optimizing the digital model of large anchored structures in the system. After creating the BIM model based on the structural data of the anchored ground-connected wall recorded during the design and construction process, a series of optimization measures are taken to reduce the redundancy of model information. Specifically, it includes deleting unnecessary internal components, removing hidden surfaces and overlapping geometries, merging coplanar small patches, simplifying complex surfaces, reducing regional surfaces with small visual impact, organizing the model tree structure and deleting redundant groups and subgroups to flatten the model structure. In addition, the Draco compression algorithm is used to compress the model file size, quantize the vertex coordinates, convert floating point numbers to integers, reduce data accuracy while maintaining visual effects; use predictive coding technology to compress vertex data, and use the spatial relationship between adjacent vertices to reduce data redundancy; encode the topological structure to reduce redundant information, and set the appropriate compression level to make the model lightweight and efficient in actual operation, which is convenient for subsequent smooth operation and processing on different devices.

[0091] 3D point cloud real-time rendering unit: This unit focuses on coloring the digital information model of the built anchor structure. After initializing the 3D scene, configure the orthogonal camera to adapt to the large anchor structure, and set the ambient light and parallel light source to enhance the three-dimensional effect. Import the lightweight BIM model and configure the WEB 3D scene, and align the grid data stored in the database according to the grid data generation unit with the BIM model coordinate system. Process the data through a specific shader, including data normalization (calculating the spatial coordinates and the given displacement normalization value, using the normalization formula: t = (v-v_min) / (v_max-v_min) and the solution of additive mixing (for each pixel in the overlapping area, assign a new color value C_result = min (C1 + C2 + ... + Cn, 1.0) to complete the rendering and visualization of the heat map, and improve the visualization effect and interactive experience of the model.

[0092] Digital twin platform unit: The digital twin platform unit is the core integration part of the entire system. It can integrate multi-source heterogeneous data such as real-time monitoring data from sensor networks, BIM model data, historical construction data, and perform comprehensive processing on these data. By integrating and displaying these data, a multi-dimensional visual interface including 3D model views, 2D profiles, and time series data charts is provided to achieve full life cycle management and real-time monitoring of super-large anchor structures. This allows relevant personnel to fully understand the status of anchors from multiple angles, facilitate timely discovery of potential risks and make decisions, and provide strong guarantees for the safe operation of large bridge projects.

[0093] In the field of bridge engineering, the ultra-large anchorage digital twin construction system of the present invention can assist in design optimization and construction guidance in new bridge projects, and can achieve health monitoring and performance evaluation in the maintenance of existing bridges; in the field of civil engineering, it can be used for large-scale building foundations and underground engineering monitoring; in the field of water conservancy engineering, it is suitable for monitoring water conservancy facilities such as dams, sluices and pumping stations; in the field of geological engineering, it can be used for geological disaster monitoring and early warning, as well as underground resource development monitoring; and in the field of scientific research and teaching, it can be used as a scientific research experimental means and teaching training tool, with broad application prospects, and of great significance for ensuring the safety of various engineering structures and promoting the development of related disciplines.

[0094] The present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of any of the above-mentioned methods for constructing a digital twin of an ultra-large anchorage can be implemented.

[0095] The computer-readable storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0096] For an introduction to the computer-readable storage medium provided in this application, please refer to the above method embodiment, and this application will not go into details here.

[0097] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing a digital twin of a super-large anchor, characterized in that: The steps include: Step S1. A three-dimensional monitoring array of inclinometer tubes evenly distributed in the horizontal and vertical directions is set on the anchored ground-connected wall structure, and the collected displacement data is transmitted to the on-site data processing center in real time through an information transmission device; Step S2. The three-dimensional structural features of the anchored ground-connected wall are converted into two-dimensional coordinates by using a 2D surface unfolding interpolation method. The two-dimensional plane inverse distance interpolation method is used in the unfolded two-dimensional coordinate system to complete the interpolation and expansion of the grid data, and the grid point displacement values ​​and their corresponding coordinate data are stored in a database in real time in a matrix form; Step S3. Create a BIM model based on the structural data of the anchored ground-connected wall recorded during the design and construction process, optimize the model structure and compress the model file size, and then reduce the computer memory usage and rendering call times in model loading and rendering by repeating component instantiation to obtain a lightweight model; Step S4. Initialize the 3D scene and configure the orthogonal camera to adapt to the anchoring of large structures; Set ambient light and parallel light source to enhance the three-dimensional effect; import the lightweight model in step S3 and configure the 3D scene on the WEB side, align the grid data stored in the database according to step S2 with the BIM model coordinate system, and then process the grid data through the shader to complete the heat map rendering and visualization; Step S5. Integrate multi-source heterogeneous data including real-time monitoring data from sensor networks, BIM model data, and historical construction data, perform integrated display of monitoring data, and provide a multi-dimensional visualization interface.

2. The method for constructing a super-large anchorage digital twin for monitoring large bridge engineering according to claim 1, characterized in that: The three-dimensional monitoring array in step S1 is: a three-dimensional monitoring array of inclinometer tubes with no less than 16 holes evenly distributed in the horizontal and vertical directions. The depth of the inclinometer tube is consistent with the ground-connected wall structure. The guide groove of the inclinometer tube is perpendicular to the edge of the foundation pit and points to the center of the foundation pit. The sensor collection frequency in the inclinometer tube is 1 time / hour.

3. The method for constructing a super-large anchorage digital twin for monitoring large bridge engineering according to claim 2, characterized in that: In step S1, the information transmission device includes a distributed optical fiber sensor demodulator, an industrial-grade switch and an optical fiber network.

4. A method for constructing a super-large anchorage digital twin for monitoring large bridge engineering according to any one of claims 1 to 3, characterized in that ,The 2D surface expansion interpolation method in step S2 includes: (1) Establishment of coordinate system: The central axis of the ground-connected wall cylinder is set as the z-axis of the Cartesian coordinate system, and the center of the cylinder on the ground is the origin; (2) Coordinate transformation: replace the Cartesian sensor coordinates and values ​​(x, y, z) with the cylindrical coordinate system (r, θ, z); where r is the distance from the sensor to the central axis, and θ is the angle of the sensor on the horizontal plane, θ∈[0,2π); (3) Boundary condition continuity processing: To ensure the continuity at θ = 0 and θ = 2π, that is, the continuity of the left and right boundaries after the cylinder is unfolded, the data at θ = 0 is copied at θ = 2π; (4) Coordinate normalization: normalize the angle θ according to the actual arc length of the cylindrical surface, that is, x_θ = θ × r, where the x-axis represents the distance along the circumference and the unit is the same as the z-axis (height); (5) Using the two-dimensional plane inverse distance interpolation method, the calculation formula is: In the formula is the estimated value of position x0, Z(x i ) is the known position x i The observed value of the unknown point (x0, y0) is the predicted value of the position are all known points (x i ,y i )’s weighted sum; the weight of each known point is determined by the αth power of the inverse of its distance di from the unknown point; di represents the straight-line distance between the unknown point and the i-th known point.

5. A method for constructing a digital twin of a super-large anchorage for monitoring a large bridge project according to any one of claims 1 to 3, characterized in that: The specific method for optimizing the model structure in step S3 includes: deleting unnecessary internal components, removing hidden faces and overlapping geometries, merging coplanar small patches, simplifying complex surfaces, reducing regional surfaces with small visual impact, organizing the model tree structure, and deleting redundant groups and subgroups to flatten the model structure.

6. The method for constructing a super-large anchorage digital twin for monitoring large bridge engineering according to claim 5, characterized in that: The specific method for compressing the model file size in step S3 includes: using the Draco compression algorithm to quantize the vertex coordinates, converting floating-point numbers into integers, reducing data precision and maintaining visual effects; using predictive coding technology to compress vertex data and using the spatial relationship between adjacent vertices to reduce data redundancy; encoding the topological structure to reduce redundant information; setting the compression level to complete model size compression and maintain decoding speed.

7. A method for constructing a digital twin of a super-large anchorage for monitoring a large bridge project according to any one of claims 1 to 3, characterized in that: The specific method of the shader processing data in step S4 includes: (1) Data normalization processing, calculate the spatial coordinates and the given normalized displacement value v, assuming the minimum displacement value v_min and the maximum displacement value v_max, the normalization formula is: t=(v-v_min) / (v_max-v_min) (2) Adopting the solution of additive mixing, for each pixel in the overlapping area, a new color value C_result is assigned as follows: C_result=min(C1+C2+...+Cn,1.0) Among them, C1, C2, ..., Cn are the color values ​​of the n overlapping points at the pixel position.

8. The system and method for monitoring deformation and three-dimensional visualization of ground-connected wall for large foundation pit engineering according to any one of claims 1 to 3, characterized in that: Step S5 provides a multi-dimensional visualization interface function including: providing a multi-dimensional visualization interface, including a 3D model view, a 2D cross-section view, and a time series data chart.

9. A system for constructing a digital twin of a super-large anchor, characterized in that: include: Large anchor structure real-time data acquisition unit, used to collect raw sensor displacement data in real time; A grid data generation unit is used to expand the collected original data and enrich the grid point data set; BIM lightweight processing unit, used to build digital models of large anchor structures and optimize the models to reduce model information redundancy, making the models light and efficient in actual operation; A 3D point cloud real-time rendering unit is used to colorize the digital information model of the built anchor structure; And the digital twin platform unit is used to integrate all multi-source heterogeneous data and produce visual graphics to achieve full life cycle management and real-time monitoring of ultra-large anchor structures.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program is executed by a processor to implement the method for constructing a digital twin of a super-large anchor as described in any one of claims 1 to 8.

Images