Digital reconstruction method of tunnel face based on multi-source data fusion

Abstract

The present invention discloses a method for digital reconstruction of a tunnel face based on multi-source data fusion. The method collects multi-source data such as geological, topographic, geomorphological exploration data, real-time data while drilling, laser scanning point cloud data, and infrared data of images taken by unmanned aerial vehicles, pre-processes and fuses the multi-source data to achieve deep integration of multi-source data fusion and digital technology. Based on the real-time collection of multi-source data, a crack propagation model is constructed through the combination of geometric modeling and mechanical analysis, and a digital reconstruction technology for the tunnel face is constructed. The morphology and stress state of the tunnel face are dynamically displayed through visual visualization to achieve accurate reconstruction and dynamic tracking of the tunnel face.

Description

Technical Field

[0001] The invention belongs to the field of data processing, and in particular to a method for digitally reconstructing a tunnel face based on multi-source data fusion. Background Art

[0002] The face refers to the front working face that is constantly advancing in the excavation of tunnels (such as coal mining, mining or tunnel engineering). The digital reconstruction of the face refers to the use of multi-source data (such as geological exploration data, laser scanning point cloud data, real-time data while drilling, drone images, etc.), combined with advanced data analysis, machine learning or deep learning and other technical means, to digitally reproduce the position, shape, geological characteristics and stress state of the face in real time, thereby forming a reconstruction process of accurate modeling, real-time monitoring and dynamic updating of the face change process. Digital reconstruction of the face is the basis for blasting operations, support design, and risk assessment, and is also a prerequisite for improving the efficiency and safety of tunnel excavation.

[0003] At present, for the digital reconstruction of the tunnel face, as described in the paper "Stability Analysis of Tunnel Face at Any Angle Based on MC Criterion" (Shan Shengbiao, Tan Zian. Journal of Underground Space and Engineering, 2024, 20(z1):41-53,58. DOI:10.20174 / j.JUSE.2024.S1.06.), a tunnel face stability analysis method based on the three-dimensional active and passive destruction model of the longitudinal inclined tunnel face is proposed, including: based on the linear Mohr-Coulomb strength criterion, using spatial discretization technology and "point-to-point" technology, considering the influence of the inclination angle on the stability of the tunnel face, improving the existing three-dimensional active and passive destruction model of the tunnel face, and realizing the solution of the active and passive destruction limit support pressure of the inclined tunnel face; establishing a three-dimensional active and passive destruction model of the inclined tunnel face under the linear Mohr-Coulomb criterion, generating the energy consumption equation of the corresponding active and passive destruction model; solving the limit support pressure of the longitudinal inclined tunnel face through the energy consumption equation, and realizing the stability analysis of the tunnel face. However, the above method relies on the three-dimensional rotation active and passive destruction model of the tunnel face under the traditional shield angle for digital reconstruction of the tunnel face. Faced with the actual complex environment, it is impossible to accurately construct the tunnel face reconstruction model under complex scenarios, thus affecting the accuracy and universality of the method.

[0004] Another example is the digital reconstruction of tunnel faces, such as the paper "Digital Identification Method and Application of Rock Mass Structure at Tunnel Face Based on Machine Vision 3D Reconstruction Technology" (Xie Zhuang, Chen Baolin, Fu Jinyang, et al. Journal of Railway Science and Engineering, 2019, 16(4): 1001-1007. DOI:10.19713 / j.cnki.43-1423 / u.2019.04.022.) discloses a method for digitally identifying rock mass structures at tunnel faces based on machine vision 3D reconstruction technology, including: collecting sequential images of tunnel excavation faces to generate a 3D point cloud model of the rock mass surface of the excavation faces; using a projection algorithm to obtain a high-definition image of a plane model of the rock mass structure of the tunnel face and the wall; and quantitatively describing the characteristics of rock joints based on the sketch results of the image surface structure traces; however, the above method reconstructs the face digitally by constructing a 3D point cloud model based on machine vision 3D reconstruction technology. In the face of geological and construction data in complex environments, it is impossible to effectively and accurately reconstruct and dynamically track, thereby affecting the efficiency and universality of the method.

[0005] In summary, the problems existing in the prior art are: the accuracy of digital reconstruction of the tunnel face is not high enough, the versatility is poor, and the efficiency is relatively low. Summary of the invention

[0006] The purpose of the present invention is to provide a method for digital reconstruction of a tunnel face based on multi-source data fusion, which has high accuracy, strong generalization ability and high efficiency.

[0007] The technical solution for achieving the purpose of the present invention is:

[0008] In a first aspect, a method for digital reconstruction of a tunnel face based on multi-source data fusion is provided, characterized in that the method comprises the following steps:

[0009] Based on actual working conditions, obtain multi-source data of the tunnel face; the multi-source data includes geological, topographic, and geomorphological exploration data, real-time data while drilling, laser scanning point cloud data, and infrared data of drone images;

[0010] Preprocess, align and fuse multi-source data to achieve multi-source data fusion;

[0011] The tunnel face is digitally reconstructed by combining the fused multi-source data.

[0012] As a further optimization solution of the present invention, the multi-source data acquisition step of the tunnel face includes:

[0013] Through geological exploration, drilling and remote sensing technology, using geographic information system GIS to obtain geological, topographic and geomorphic exploration data;

[0014] Use the MWD / LWD system to acquire real-time data while drilling, including temperature, pressure, vibration, and drill bit position information.

[0015] Obtain laser scanning point cloud data through a laser scanner;

[0016] Use the infrared thermal imaging instrument carried by the drone to obtain the infrared data of the images taken by the drone.

[0017] As a further optimization scheme of the present invention, the pretreatment step comprises:

[0018] S211, missing value processing: For missing geological, topographic and geomorphological exploration data, linear interpolation algorithm is used to fill in the missing values ​​to achieve missing value processing;

[0019] S212, Noise Removal: For the real-time data while drilling and the infrared data of the images taken by the UAV, the statistical filtering algorithm is used to remove the error or noise, and the neighborhood-based interpolation method is used to fill the laser scanning point cloud data.

[0020] As a further optimization scheme of the present invention, the data registration step includes:

[0021] Extract features from the pre-processed geological, topographic, geomorphological exploration data, real-time while drilling data, and laser scanning point cloud data, and match the extracted features; align data based on the feature matching results to achieve data registration in the same coordinate system;

[0022] The pre-processed laser scanning point cloud data and the infrared data of the drone-photographed images are registered to achieve spatial alignment between the two.

[0023] As a further optimization scheme of the present invention, the data fusion step includes:

[0024] Extract features from multi-source data after data registration and perform standardization processing;

[0025] The corresponding standardized processing results except the laser scanning point cloud data are mapped to each point cloud of the laser scanning point cloud data as additional attributes, a fused feature point cloud is constructed, and point cloud enhancement is performed to achieve the fusion of multi-source data and obtain the three-dimensional point cloud data of the tunnel face.

[0026] As a further optimization solution of the present invention, the step of digitally reconstructing the tunnel face includes:

[0027] Using the 3D point cloud data of the tunnel face, combined with the Poisson surface reconstruction method and the adaptive mesh refinement technology, a local refined mesh is generated to achieve the reconstruction of the 3D geometric model of the tunnel face under high-precision mesh;

[0028] The crack propagation model is obtained by combining the 3D point cloud data of the tunnel face and the 3D geometric model of the tunnel face. By setting mechanical loads and boundary conditions and identifying high-risk areas, accurate reconstruction of cracks and damaged areas can be achieved.

[0029] As a further optimization scheme of the present invention, the steps of accurately reconstructing the crack and damaged areas include:

[0030] Extract geometric information features and mechanical information features from the three-dimensional point cloud data of the tunnel face to form geometric-mechanical joint information features;

[0031] Based on the 3D geometry model of the tunnel face, mechanical boundary conditions are set, finite element simulation is performed, and the geometric deformation displacement is reversely adjusted according to the simulation results, thereby constructing a bidirectional coupling iterative model of geometry and mechanics;

[0032] Based on the geometric and mechanical bidirectional coupling iterative model, combined with the thermodynamic characteristics of the three-dimensional point cloud data of the tunnel face, a multi-field coupling crack propagation model of thermal-mechanical-dynamic multi-field interaction is constructed;

[0033] Set static load and dynamic load to realize mechanical load setting, and set boundary conditions by setting fixed boundary and natural boundary;

[0034] Based on the finite element method, the stress analysis of cracks and damaged areas is carried out, the stress concentration areas are identified according to the results of the stress analysis, the high-risk areas are identified, and the cracks and damaged areas are accurately reconstructed.

[0035] As a further optimization scheme of the present invention, the geometric and mechanical bidirectional coupling iterative model is expressed as:

[0036] ,

[0037] ,

[0038] in, , Respectively represent the displacement field of the three-dimensional geometric model of the tunnel face at the n+1th and nth steps, is based on the stress at step n and strain The calculated geometric adjustment function, is the adjustment coefficient, Characterize the stiffness matrix, The external force applied.

[0039] As a further optimization scheme of the present invention, the multi-field coupled crack propagation model of thermal-mechanical-dynamic multi-field interaction is expressed as:

[0040] ,

[0041] ,

[0042] In the formula, Represents the component of the stress tensor, the subscript i corresponds to the direction of stress action, and j corresponds to the normal direction of the surface to which the stress distribution acts; represents the spatial coordinate corresponding to the direction of the i-th stress; Represents the external force acting on the unit volume of material corresponding to the direction of the i-th stress; momentum term , to characterize the transient dynamic effects caused by the blast; represents the displacement corresponding to the direction of the i-th stress; is the density of the material; is the elastic modulus tensor, subscripts i and j correspond to stress Component index of , k, l correspond to the component index of strain; represents the component of the strain tensor, the subscript k describes the direction of deformation, and l represents the normal direction of the surface where the deformation is located; represents the thermal stress caused by the temperature field, where is the coefficient of thermal expansion, is the temperature difference; represents the additional stress caused by the blast shock wave, is the material damping coefficient, is the shock wave pressure.

[0043] In a second aspect, a computer-readable storage medium storing one or more programs is also provided. The one or more programs include instructions, which, when executed by a computing device, enable the computing device to perform the reconstruction method as described above.

[0044] According to a third aspect, an electronic device is also provided, comprising one or more processors, one or more memories and one or more programs, wherein the one or more programs are stored in the one or more memories and are configured to be executed by the one or more processors, and the one or more programs include instructions for executing the reconstruction method as described above.

[0045] Compared with the prior art, the present invention has the following significant advantages:

[0046] 1. More accurate reconstruction: The digital reconstruction technology of the tunnel face based on multi-source data fusion in the present invention collects multi-source data, constructs a crack propagation model combining geometric information and mechanical information, and realizes the deep integration of multi-source data fusion and digital technology, thereby realizing accurate reconstruction of the tunnel face;

[0047] 2. Strong generalization ability and high efficiency: The present invention collects multi-source data and effectively integrates multi-source data such as geological, topographic, geomorphological exploration data, real-time data while drilling, laser scanning point cloud data and infrared data of drone images, so as to solve the problems of data accuracy and consistency, and construct a digital reconstruction technology for the face. It can effectively process large-scale and highly complex geological and construction data, with stronger generalization ability and higher reconstruction efficiency.

[0048] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is the main flow chart of the method for digital reconstruction of the tunnel face based on multi-source data fusion of the present invention.

[0050] Figure 2 It is a flow chart of the multi-source data collection steps.

[0051] Figure 3 It is a flow chart of the multi-source data fusion steps.

[0052] Figure 4 It is a flow chart of the steps for digital reconstruction of the tunnel face.

[0053] Figure 5 It is a flow chart of the tunnel face visualization steps. DETAILED DESCRIPTION

[0054] like Figure 1 As shown, the method for digital reconstruction of a tunnel face based on multi-source data fusion of the present invention comprises the following steps:

[0055] S10, Multi-source data acquisition: Based on actual working conditions and in combination with existing measurement techniques, obtain multi-source data of the tunnel face;

[0056] Among them, the multi-source data of the face include geological, topographic, geomorphological exploration data, real-time data while drilling, laser scanning point cloud data and infrared data of drone images.

[0057] like Figure 2 As shown, the multi-source data collection step includes:

[0058] S11. Acquisition of geological, topographic and geomorphological exploration data: Acquisition of geological, topographic and geomorphological exploration data using GIS (Geographic Information System) through geological exploration, drilling and remote sensing technology.

[0059] S12. Real-time data acquisition while drilling: Use the data acquisition system while drilling (MWD / LWD system) to record various physical parameters during the drilling process in real time and obtain real-time data while drilling (such as temperature, pressure, vibration, and drill bit position information).

[0060] S13. Acquisition of laser scanning point cloud data: Perform high-precision scanning on the ground using a laser scanner to acquire laser scanning point cloud data.

[0061] S14. Acquisition of infrared data of images taken by drones: Use drones equipped with infrared thermal imaging instruments to obtain thermal imaging images of the ground or target objects. Use image processing software to stitch, calibrate and three-dimensionally reconstruct drone images to obtain infrared data of images taken by drones.

[0062] S20. Multi-source data fusion: Multi-source data cleaning is achieved through pre-processing of multi-source data, and multi-source data fusion is achieved through data registration, thereby constructing the three-dimensional model point cloud data of the tunnel face.

[0063] like Figure 3 As shown, the multi-source data processing includes:

[0064] S21. Multi-source data preprocessing: Multi-source data preprocessing is achieved by performing missing value processing, noise removal, and outlier detection and processing on the actual collected data.

[0065] The multi-source data preprocessing includes:

[0066] S211, missing value processing: For missing geological, topographic and geomorphological exploration data, linear interpolation algorithm is used to fill in the missing values ​​to achieve missing value processing;

[0067] S212, Noise Removal: For the downhole data and infrared data of drone images, a statistical filtering algorithm (mean filtering) is used to remove sensor errors or environmental noise. At the same time, a neighborhood-based interpolation method (KNN interpolation) is used to fill in the laser scanning point cloud data to achieve noise removal from multi-source data.

[0068] S22. Data registration: Through feature-based registration and image and point cloud registration, the pre-processed multi-source data are spatially aligned to achieve registration of multi-source data in the same coordinate system.

[0069] The data registration step comprises:

[0070] S221. Feature-based registration: Through feature extraction and feature matching, geometric transformation of multi-source data is achieved, thereby realizing feature-based registration.

[0071] The feature-based registration step S221 includes:

[0072] S2211, Feature Extraction: Combine SIFT and SURF algorithms to extract stable and recognizable features of geological, topographic, geomorphological exploration data, real-time data while drilling, and laser scanning point cloud data to achieve feature extraction;

[0073] The stable and recognizable features are points, edges, corners, surfaces or regions with obvious physical or geometric characteristics in the data.

[0074] S2212, feature matching: combining a matching algorithm (SIFT), matching the feature points in the data source in S2211, and then determining the spatial transformation relationship between the various data sets to achieve feature matching;

[0075] S2213, geometric transformation: according to the feature matching result, the data is aligned in combination with geometric transformations such as rotation, translation, and scale transformation to achieve registration of the data sources in the same coordinate system.

[0076] S222. Image and point cloud registration: Image data taken by the drone are registered with point cloud data obtained by laser scanning to achieve spatial alignment of the image data and the point cloud data, thereby achieving image and point cloud data registration.

[0077] Through feature-based registration and image and point cloud registration, multi-source data can be registered in the same coordinate system;

[0078] S23. Multi-source data fusion: feature integration is performed on the registered multi-source data, and multi-source data fusion is achieved through point cloud enhancement to obtain the three-dimensional model point cloud data of the tunnel face.

[0079] The multi-source data fusion step includes:

[0080] S231, Data feature integration: Extract geometric and physical features that can represent the characteristics of various data sources from multi-source data, perform standardization processing, and thus achieve data feature integration;

[0081] The steps of extracting geometric and physical features that can represent the characteristics of various data sources are specifically as follows:

[0082] Extract rock and soil property characteristics from geological exploration data, including rock and soil type, rock and soil density, porosity, permeability, mechanical properties, thermal properties and other physical characteristics; extract spatial characteristics such as morphology, structure, and change from terrain and geomorphological exploration data, including slope and aspect, terrain undulation, geomorphological units and other characteristics; extract engineering data characteristics such as drill bit stress and torque, displacement and deformation, and drilling temperature from real-time drilling data; extract point information characteristics reflecting the geometric structure shape from laser scanning point clouds, and extract surface texture, cracks, thermal distribution and other characteristics from images taken by drones.

[0083] The specific steps of the standardization process are:

[0084] Adjust the eigenvalues ​​to a standard normal distribution (mean 0, standard deviation 1) to eliminate the bias in the distribution:

[0085]

[0086] in, Characterize the original features, Characterize the feature mean, Characterize feature variance;

[0087] S232, point cloud enhancement: mapping feature information of other data sources in the standardized multi-source data as additional attributes of the point cloud to each point of the laser scanning point cloud, constructing a fused feature point cloud, performing weighted averaging on the attribute information of the point cloud, combining geological features, construction data and geometric data, to obtain more accurate three-dimensional model point cloud data, thereby achieving point cloud enhancement;

[0088] The fused feature point cloud includes the point information of the laser scanning point cloud and the geometric features and physical features such as mechanics and thermodynamics extracted from other different data sources in the multi-source data;

[0089] S233, multi-source data fusion: through the data feature integration and point cloud enhancement steps, multi-source data fusion is achieved to obtain the three-dimensional model point cloud data of the tunnel face.

[0090] S30, Digital reconstruction of tunnel face: Combined with the 3D model point cloud data of the tunnel face, the Poisson surface reconstruction method is used to achieve reconstruction of the 3D geometric model of the tunnel face under high-precision grid. By building a multi-field coupled crack propagation model, setting mechanical loads and boundary conditions, and identifying high-risk areas, the appearance and mechanical response of the tunnel face and its crack area can be digitally reconstructed;

[0091] like Figure 4 As shown, the digital reconstruction step of the tunnel face includes:

[0092] S31, reconstruction of the 3D geometric model of the tunnel face: using the 3D model point cloud data, combined with the Poisson surface reconstruction method, and using adaptive mesh refinement technology, local refined meshes are generated to achieve reconstruction of the 3D geometric model of the tunnel face under high-precision meshes;

[0093] Among them, the Poisson surface reconstruction method and the adaptive grid refinement technology are both existing technical means.

[0094] S32. Accurate reconstruction of cracks and damaged areas: Combine the 3D model point cloud data and the 3D geometric model of the tunnel face to obtain a multi-field coupled crack propagation model. By setting mechanical loads and boundary conditions and identifying high-risk areas, accurate reconstruction of cracks and damaged areas can be achieved.

[0095] The steps of accurately reconstructing the cracks and damaged areas include:

[0096] S321. Acquisition of multi-field coupled crack propagation model: Extract geometric information features and mechanical information features from the three-dimensional model point cloud data to form geometric-mechanical joint information features. Based on the three-dimensional geometric model of the face, a geometric and mechanical bidirectional coupled iterative model is constructed. Based on the geometric and mechanical bidirectional coupled iterative model, combined with the thermodynamic and other physical features in the point cloud data, a multi-field coupled crack propagation model of thermal-mechanical-dynamic multi-field interaction is constructed.

[0097] The multi-field coupled crack extension model acquisition step comprises:

[0098] S3211. Construction of geometric and mechanical bidirectional coupling iterative model: Combine the point cloud data of the 3D model, extract the geometric information features and mechanical information features in the point cloud data, form geometric-mechanical joint information features, set mechanical boundary conditions based on the 3D geometric model of the tunnel face, perform finite element simulation, and reversely adjust the geometric deformation displacement according to the mechanical simulation response results (stress, strain), so as to construct a geometric and mechanical bidirectional coupling iterative model;

[0099] The geometric information includes the geometric characteristics of the cracks and the overall geometric characteristics of the tunnel face, and the mechanical information includes the stress distribution characteristics and the local deformation and displacement field characteristics of the tunnel face.

[0100] The geometric and mechanical bidirectional coupling iterative model is:

[0101]

[0102]

[0103] in, Characterize the displacement field of the three-dimensional geometric model of the tunnel face at step n, Based on the current stress and strain The calculated geometric adjustment function, is the adjustment coefficient used to control the geometry update step size, The stiffness matrix that represents the system describes the response of the structure to external forces. is the externally applied force;

[0104] The initial 3D geometry model of the tunnel face is adjusted by introducing point cloud data in real time in each iteration step to make it closer to the actual project.

[0105] S3212. Construction of a multi-field coupled crack extension model: Based on the geometric and mechanical bidirectional coupled iterative model, a multi-field (stress field, temperature field, blasting shock wave field) coupled crack extension model is constructed by combining the thermal-mechanical-dynamic multi-field interaction:

[0106]

[0107]

[0108] In the formula, Represents the component of the stress tensor, the subscript i corresponds to the direction of stress action, and j corresponds to the normal direction of the surface to which the stress distribution acts; represents the spatial coordinate corresponding to the direction of the i-th stress; Represents the external force acting on the unit volume of material corresponding to the direction of the i-th stress; momentum term , to characterize the transient dynamic effects caused by the blast; represents the displacement corresponding to the direction of the i-th stress; is the density of the material; is the elastic modulus tensor, subscripts i and j correspond to stress Component index of , k, l correspond to the component index of strain; represents the component of the strain tensor, the subscript k describes the direction of deformation, and l represents the normal direction of the surface where the deformation is located; represents the thermal stress caused by the temperature field, where is the coefficient of thermal expansion, is the temperature difference; represents the additional stress caused by the blast shock wave, is the material damping coefficient, is the shock wave pressure.

[0109] S322. Mechanical load and boundary condition setting: Mechanical load setting is achieved by setting static load and dynamic load, and boundary condition setting is achieved by setting fixed boundary and natural boundary.

[0110] S323, high-risk area identification: Based on the finite element method (FEM), the stress analysis of cracks and damaged areas is carried out, and the stress concentration areas are identified according to the results. It should be noted that the identification of high-risk areas is an existing technology and will not be repeated here.

[0111] S324, accurate reconstruction of cracks and damaged areas: through the steps S321 to S323, accurate reconstruction of cracks and damaged areas is achieved;

[0112] S40, tunnel face visualization: obtain real-time data through data collection, and realize tunnel face visualization through digital reconstruction of the tunnel face;

[0113] like Figure 5 As shown, the tunnel face visualization step includes:

[0114] S41, real-time data collection: real-time collection of multi-source data, through the steps S10 to S20, to achieve the fusion of real-time collected data;

[0115] S42. Face visualization: Through the digital reconstruction of the face, real-time mechanical data feedback is achieved. Mechanical changes (such as displacement, stress, etc.) are monitored in real time through sensors, and these data are transmitted to the mechanical model to reflect the expansion of cracks or the formation of new damage. The dynamic display of crack expansion and visualization of the face are achieved through the Python software.

[0116] The present invention collects multi-source data such as geological, topographic, and geomorphological exploration data, real-time data while drilling, laser scanning point cloud data, and infrared data of images taken by unmanned aerial vehicles, and pre-processes and fuses the multi-source data to achieve deep integration of multi-source data fusion and digital technology. Based on the real-time collection of multi-source data, the present invention constructs a crack propagation model and a digital reconstruction technology of the tunnel face through the combination of geometric modeling and mechanical analysis. The morphology and stress state of the tunnel face are dynamically displayed through visualization to achieve accurate reconstruction and dynamic tracking of the tunnel face.

[0117] Based on the same technical solution, the present invention also discloses a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, and when the instructions are executed by a computing device, the computing device executes the above-mentioned reconstruction method.

[0118] Based on the same technical solution, the present invention also discloses a computing device, comprising one or more processors, one or more memories and one or more programs, wherein the one or more programs are stored in the one or more memories and are configured to be executed by the one or more processors, and the one or more programs include instructions for executing the above-mentioned reconstruction method.

[0119] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0120] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0121] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0122] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0123] The above embodiments are only for illustrating the technical idea of ​​the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for digital reconstruction of a tunnel face based on multi-source data fusion, characterized in that: The steps include: Based on actual working conditions, obtain multi-source data of the tunnel face; the multi-source data includes geological, topographic, and geomorphological exploration data, real-time data while drilling, laser scanning point cloud data, and infrared data of drone images; Preprocess, align and fuse the multi-source data in sequence to achieve fusion of multi-source data; Combining the fused multi-source data, the tunnel face is digitally reconstructed; The steps of digital reconstruction of the tunnel face include: The 3D point cloud data of the tunnel face is obtained by fusion of multi-source data, combined with the Poisson surface reconstruction method and the adaptive mesh refinement technology to generate local refined meshes, thus realizing the reconstruction of the 3D geometric model of the tunnel face under high-precision meshes. Combine the 3D point cloud data of the tunnel face with the 3D geometric model of the tunnel face to obtain the crack propagation model. By setting mechanical loads and boundary conditions and identifying high-risk areas, accurate reconstruction of cracks and damaged areas can be achieved. The steps for accurate reconstruction of cracks and damaged areas include: Extract geometric information features and mechanical information features from the three-dimensional point cloud data of the tunnel face to form geometric-mechanical joint information features; Based on the 3D geometry model of the tunnel face, mechanical boundary conditions are set, finite element simulation is performed, and the geometric deformation displacement is reversely adjusted according to the simulation results, thereby constructing a bidirectional coupling iterative model of geometry and mechanics; Based on the geometric and mechanical bidirectional coupling iterative model, combined with the thermodynamic characteristics of the three-dimensional point cloud data of the tunnel face, a multi-field coupling crack propagation model of thermal-mechanical-dynamic multi-field interaction is constructed; Set static load and dynamic load to realize mechanical load setting, and set boundary conditions by setting fixed boundary and natural boundary; Based on the finite element method, the stress analysis of cracks and damaged areas is carried out, and the stress concentration areas are identified according to the stress analysis results, and the high-risk areas are identified to achieve accurate reconstruction of cracks and damaged areas; The geometric and mechanical bidirectional coupling iterative model is expressed as: ， ， in, , Respectively represent the displacement field of the three-dimensional geometric model of the tunnel face at the n+1th and nth steps, is based on the stress at step n and geometric adjustment functions for strain calculation, is the adjustment coefficient, Characterize the stiffness matrix, The external force applied.

2. The method according to claim 1, characterized in that: The steps for acquiring multi-source data of the tunnel face include: Through geological exploration, drilling and remote sensing technology, using geographic information system GIS to obtain geological, topographic and geomorphic exploration data; Use the MWD / LWD system to acquire real-time data while drilling, including temperature, pressure, vibration, and drill bit position information. Obtain laser scanning point cloud data through a laser scanner; Use the infrared thermal imaging instrument carried by the drone to obtain the infrared data of the images taken by the drone.

3. The method according to claim 1, characterized in that The preprocessing steps include: For geological, topographic and geomorphic exploration data, a linear interpolation algorithm is used to fill missing values; Statistical filtering algorithms are used to remove errors or noise from real-time data while drilling and infrared data from drone images. For laser scanning point cloud data, a neighborhood-based interpolation method is used to fill missing values.

4. The method according to claim 1, characterized in that The data registration steps include: Extract features from the pre-processed geological, topographic, geomorphic exploration data, real-time data while drilling, and laser scanning point cloud data, and match the extracted features; align the pre-processed geological, topographic, geomorphic exploration data, real-time data while drilling, and laser scanning point cloud data based on the feature matching results to achieve data registration in the same coordinate system; The pre-processed laser scanning point cloud data and the infrared data of the drone-photographed images are registered to achieve spatial alignment between the two.

5. The method according to claim 1, characterized in that The data fusion steps include: Extract features from multi-source data after data registration and perform standardization processing; The corresponding standardized processing results except the laser scanning point cloud data are mapped to each point cloud of the laser scanning point cloud data as additional attributes, a fused feature point cloud is constructed, and point cloud enhancement is performed to achieve the fusion of multi-source data and obtain the three-dimensional point cloud data of the tunnel face.

6. The method according to claim 1, characterized in that The multi-field coupled crack propagation model of thermal-mechanical-dynamic multi-field interaction is expressed as: ， ， In the formula, Represents the component of the stress tensor, the subscript i corresponds to the direction of stress action, and j corresponds to the normal direction of the surface to which the stress distribution acts; represents the spatial coordinate corresponding to the direction of the i-th stress; It represents the external force acting on the unit volume of material corresponding to the direction of the i-th stress; represents the displacement corresponding to the direction of the i-th stress, and t represents the time; is the density of the material; is the tensor of elastic modulus; represents the component of the strain tensor, the subscript k describes the direction of deformation, and l represents the normal direction of the surface where the deformation is located; is the coefficient of thermal expansion, is the temperature difference; is the material damping coefficient, is the shock wave pressure.

7. A computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, characterized in that: When the instructions are executed by a computing device, the computing device is caused to perform the method according to any one of claims 1 to 6.

Images