A digital twin and fine modeling system and method for a tunnel long pipe roof

Abstract

The present invention discloses a digital twin and fine modeling system and method for a tunnel long pipe roof. The system includes four modules: a long pipe roof physical entity data perception and transmission module, a digital twin model creation module, a digital twin module, and a digital twin data management module. Through seamless connection between the modules and the combination of the four functional modules, the digital twin of the tunnel long pipe roof and the physical entity are synchronized to run, interact with each other, simulate and iteratively optimize, so as to achieve fine modeling of the construction benchmark state of the tunnel long pipe roof, perceive the post-construction evolution throughout the process, and accurately predict the construction quality and reinforcement effect. The method is based on the full collection and research of engineering data, and constructs a digital twin synchronously with the entire life cycle of the tunnel long pipe roof project, realizing three-dimensional visualization of the actual grouting effect of the tunnel long pipe roof, perceiving the physical entity data of the long pipe roof throughout the process, and predicting the behavior of the long pipe roof construction process throughout the process, so as to efficiently guide the construction of the tunnel long pipe roof.

Description

Technical Field

[0001] The present invention belongs to the field of tunnel engineering, and in particular relates to a digital twin and fine modeling system and method for a long tunnel pipe roof. Background Art

[0002] In order to realize the visual engineering management of the tunnel long pipe roof construction process, the tunnel long pipe roof will be modeled during the tunnel BIM management process, and the construction process will be simulated and analyzed accordingly.

[0003] Traditional modeling methods often use manual processing to perform modeling. The data transmission between the collected model data and the modeled model relies on the modeler as a data interaction bridge, which makes the traditional modeling method inefficient, labor-intensive, and has low modeling accuracy. For complex and cumbersome models, traditional modeling methods have great limitations, and the created models often cannot fit the actual situation well.

[0004] In the traditional modeling process of creating the grouting model, the diffusion radius of the grouting is first calculated according to the following formula The grouting model is drawn using the most ideal grouting state, namely a cylinder plus hemisphere structure. This idealized grouting model is used for special rock formations with low porosity or when cavities appear within the rock formation. On the one hand, the grouting volume obtained through the idealized grouting model differs significantly from the actual grouting volume, which cannot effectively guide construction cost control and material management. On the other hand, the actual grouting effect does not match the idealized model, and the actual grouting effect cannot be evaluated in a three-dimensional intuitive manner. (Liu Xiaotong. Research on Calculation of Surface Grouting Volume in Ultra-Shallow Tunnel Sections Based on BIM Technology [J]. Construction Technology, 2020, 49(S1): 505-507.) Summary of the Invention

[0005] The present invention discloses a digital twin and fine modeling system and method for a long pipe roof in a tunnel. The present invention adopts a modular design and divides the modules according to their functionality, including four modules: a long pipe roof physical entity data perception and transmission module, a model creation module, a digital twin module, and a digital twin data management module. Through the seamless connection between the modules and the combination of the four functional modules, a fine modeling process from the design of the long pipe roof in the tunnel to the export of BIM models and engineering quantities is realized. The model information finally generated covers the outlines of each structural layer or functional layer of the tunnel, the materials of each structural layer or functional layer of the tunnel, the elevations of each section of the tunnel model, the starting and ending coordinates of the pipe roof grouting drilling, the relevant parameters of the steel flower pipe for grouting the pipe roof, the type of steel grille and its layout coordinates in the tunnel, the type of tunnel pipe roof anchor rods and the plum blossom drilling layout, etc. The engineering quantity information covers the amount of grouting slurry for the long pipe roof in the tunnel.

[0006] The present invention is achieved through at least one of the following technical solutions.

[0007] A digital twin and fine modeling method for a long tunnel pipe roof, characterized by comprising the following steps:

[0008] S01. Comprehensively collect tunnel exploration data and establish a three-dimensional visual geological information model;

[0009] S02. Create a 3D visual tunnel long pipe roof structure information model based on the design drawings;

[0010] S03. Combine the geological information model and the tunnel long pipe roof structure information model to establish a corresponding analysis and calculation model;

[0011] S04. Combine the long pipe shed construction method and process, and use finite element software to simulate the entire process flow of long pipe shed grouting;

[0012] S05. Continuously collect on-site detection data. By comparing it with real-time detection data, use the artificial intelligence deep learning algorithm to modify the analysis and calculation model of the tunnel long pipe shed to ensure that the grouting effect is consistent with the actual situation. Import the data from the long pipe shed grouting simulation process into the digital twin data management module for storage;

[0013] S06. Export the tunnel grouting model from the digital twin data management module to achieve digital delivery of the tunnel long pipe shed grouting model.

[0014] Furthermore, the geological information model includes surrounding rock information and geological, hydrological, topographic and geomorphological information of the tunnel long pipe roof construction area.

[0015] Furthermore, the tunnel long pipe roof structural information model includes the long pipe roof, initial support, secondary lining, and grouting body.

[0016] The tunnel long pipe shed structural information model also includes the long pipe shed geometric dimensions, steel arch frame geometric dimensions information, plum blossom-shaped drilling geometric information, mechanical properties of steel pipe component materials, slurry physical and chemical characteristics and constitutive relationships.

[0017] Furthermore, the analysis and calculation model includes the geometric information, physical information, mechanical properties and constitutive relationships of the long pipe roof components, the geometric information of the plum blossom-shaped drilling holes, the geometric properties and constitutive relationships of the surrounding rock and geological body, as well as construction factors such as the loading time, loading pressure, and physical and chemical properties of the slurry of the long pipe roof grouting.

[0018] Furthermore, the digital twin data management module is built using a relational database model, which is used to receive and store monitoring data during the excavation of the tunnel long pipe shed, initial long pipe shed design data, and calculation analysis and simulation data of the digital twin module, and provide a corresponding interface for model export during the grouting process of the tunnel long pipe shed.

[0019] Furthermore, the geological information model and the tunnel long pipe roof structure information model are converted into an analysis and calculation model through the software GBMDT.

[0020] A system for implementing the digital twin and fine modeling method of a tunnel long pipe roof comprises a long pipe roof physical entity data perception and transmission module, a digital twin model creation module, a digital twin module, and a digital twin data management module;

[0021] The long pipe shed physical entity data perception and transmission module is used to obtain surrounding rock information and geological, hydrological, topographic and geomorphological information of the tunnel long pipe shed construction area, and to monitor in real time the mechanical state and grouting state of the surrounding rock, the actual position of the tunnel long pipe shed, the slurry density and diffusion range, and other data during the construction of the tunnel long pipe shed, and transmit the monitoring and measurement data to the digital twin data management module;

[0022] The digital twin model creation module is used to create a geological information model and a tunnel long pipe roof structure information model, and transmit the geological information model and the tunnel long pipe roof structure information model to the digital twin module;

[0023] The digital twin module is used to simulate the long pipe shed grouting process in stages and to modify the analysis model based on the real-time data collected by the long pipe shed physical entity data perception and transmission module. It also uses an artificial intelligence-based deep learning algorithm to predict the behavior of the long pipe shed construction process, thereby obtaining a behavior prediction model. The behavior prediction model and related analysis data are then transmitted to the digital twin data management module.

[0024] The digital twin data management module is used to store and manage system data, as well as export system data and export tunnel long pipe roof grouting models.

[0025] Furthermore, the long pipe roof physical entity data perception and transmission module is also used to collect the monitoring data, construction quality data, grouting pressure and range parameters of the physical entity in real time, and upload the collected data to the digital twin data management module to realize real-time data collection, and use the collected real-time data to correct the analysis and calculation model and predict behavior, so as to obtain analysis results that are consistent with reality.

[0026] Furthermore, the tunnel long pipe roof structure information model is based on the tunnel long pipe roof structure information library of design drawings and is modeled using a visual programming modeling method.

[0027] Furthermore, the model correction of the digital twin module mainly compares the real-time construction big data collected by the long pipe roof physical entity data perception and transmission module with the staged long pipe roof finite element simulation data, thereby correcting the surrounding rock classification and surrounding rock mechanical properties of the original geological information model, and feeding it back to the analysis and calculation model for further analysis and prediction, and performing multiple cycles of identification and correction to obtain a behavior prediction model.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] From forward design of the tunnel pipe roof to BIM model and quantity export, the system sequentially establishes four modules: the pipe roof physical entity data perception and transmission module, the model creation module, the digital twin module, and the digital twin data management module. This improves modeling efficiency and accuracy while reducing modeling labor intensity. Furthermore, through the synchronous operation of the tunnel pipe roof's digital twin and the physical entity, virtual-reality interaction, simulation, and iterative optimization, it achieves detailed modeling of the tunnel pipe roof's construction baseline state, full perception of post-construction evolution, and accurate prediction of construction quality and reinforcement effects, facilitating BIM5D construction and engineering management of tunnel projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a flow chart of a digital twin and fine modeling method for a long tunnel pipe roof according to an embodiment of the present invention;

[0031] Figure 2 This is a schematic diagram of the module composition structure of a tunnel long pipe roof digital twin and fine modeling system in an embodiment;

[0032] Figure 3 This is a schematic diagram of the model construction process and relationship of a tunnel long pipe roof digital twin and a fine modeling system in an embodiment. DETAILED DESCRIPTION

[0033] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0034] Example 1

[0035] A digital twin and fine modeling method for a long tunnel pipe roof includes the following steps:

[0036] S1. Comprehensively collect tunnel exploration data and establish a three-dimensional visual geological information model; the geological information model includes surrounding rock information and geological, hydrological, topographic and geomorphological information of the tunnel long pipe shed construction area.

[0037] S2. Create a three-dimensional visual information model of the tunnel long pipe shed structure based on the design drawings; the tunnel long pipe shed structure information model includes the long pipe shed geometry, steel arch geometry, plum blossom borehole geometry, mechanical properties of steel pipe component materials, and slurry physical and chemical properties and constitutive relationships. The analytical calculation model includes the long pipe shed component geometry, physical information, mechanical properties and constitutive relationships of the material, plum blossom borehole geometry, surrounding rock and geological body geometry and constitutive relationships, and construction factors such as long pipe shed grouting loading time, loading pressure, and slurry physical and chemical properties.

[0038] S3. Use the GBMDT software to transform the geological information model and the tunnel long pipe shed structure information model, and establish a corresponding analysis and calculation model by combining the geological information model and the tunnel long pipe shed structure information model;

[0039] S4. Combine the long pipe shed construction method and process, and use finite element software to simulate the entire process of long pipe shed grouting;

[0040] S5. Continuously conduct on-site detection to collect data such as the actual position of the tunnel long pipe shed, slurry density, and diffusion range. By comparing the real-time detection data, the analysis and calculation model of the tunnel long pipe shed is corrected using an artificial intelligence deep learning algorithm to make the grouting effect consistent with the actual situation, and the data during the long pipe shed grouting simulation process is imported into the digital twin data management module for storage; the digital twin data management module is built using a relational database model to receive and store monitoring data during the construction process of the tunnel long pipe shed, initial long pipe shed design data, and calculation analysis and simulation data of the digital twin module, and provide a corresponding interface for model export during the tunnel long pipe shed grouting process.

[0041] S6. Export the tunnel grouting model from the digital twin data management module to achieve digital delivery of the tunnel long pipe shed grouting model.

[0042] Example 2

[0043] like Figure 1 The figure shows a flow chart of a digital twin and fine modeling system and method for a long tunnel tube roof. The method includes the following steps:

[0044] S01. Comprehensively collect tunnel exploration data, including necessary information such as the tunnel's physical geography, engineering geology, and hydrogeology, and establish a three-dimensional visual geological information model;

[0045] In specific implementation, based on the information obtained from comprehensive geological surveys, three-dimensional geological modeling technology can be used to combine spatial information management, geological interpretation, spatial analysis and prediction, geostatistics, entity content analysis, and graphic visualization tools in a three-dimensional environment to establish a three-dimensional visual geological information model.

[0046] In specific implementation, airborne, vehicle-mounted, or ground-based 3D laser scanning can be used to survey the terrain near the long pipe shed construction site. By rapidly acquiring a dense surface point cloud and employing appropriate data processing methods, a detailed 3D terrain surface model can be constructed. On-site investigation of the geological structure of the surrounding rock mass within the long pipe shed construction section can be conducted using borehole sonic logging, geological radar, gamma-ray detection, or core sampling. This provides geological information such as the composition and classification of the surrounding rock mass within the long pipe shed construction area. Combining the 3D terrain surface model with an AI-based deep learning algorithm can be used to delineate the surrounding rock mass within different sections, resulting in a 3D geological model that aligns with actual conditions.

[0047] S02. Create a 3D visual tunnel long pipe roof structure information model based on the design drawings;

[0048] During the specific implementation, based on the design drawings provided by the designer, relevant parameters such as the long tube shed arch section, the three-dimensional coordinates of the head and tail of the steel flower pipe, the three-dimensional coordinates of the steel arch frame positioning, the initial support, and the secondary lining are extracted into the tunnel long tube shed structure information database, and transmitted to the digital twin data management module, and the tunnel shed structure information model is modeled using visual programming modeling.

[0049] The visual programming modeling is to call the data in the tunnel long pipe roof structure information database to carry out the lofting modeling of the long pipe roof from the arch outline to the arch entity, taking into account the operations such as the drilling and placement of steel flower pipes, the positioning and placement of advance small guide pipes, anchor rods and steel arch frames, so as to improve the model accuracy and modeling efficiency; if the design drawings need to be changed due to safety factors such as changes in the surrounding rock, landslides, and water gushing during the construction process, modifications can be made based on the original tunnel long pipe roof structure information database, and the model parameters can be quickly and accurately changed through the original programming program.

[0050] As a preference, the visual programming modeling software may be Dynamo.

[0051] The information in the tunnel long pipe roof structure information database is divided into four categories according to type: geometric model information, spatial position information, material mechanical parameter information and slurry physical and chemical characteristic parameter information;

[0052] The geometric model information includes the long pipe shed arch geometric model, steel pipe geometric model, steel arch frame geometric model, and grouting small pipe geometric model;

[0053] Spatial position information includes the spatial coordinate values ​​of the first and last endpoints of the steel flower pipe, the placement position and angle of the grouting small pipe, and the placement coordinates of the steel arch frame;

[0054] Material mechanics information includes the compressive strength of the long tube shed arch, the compressive strength of the steel flower tube, and the compressive strength of the steel arch frame;

[0055] The physicochemical property parameter information of the slurry includes the anti-seepage and water-stopping properties of the slurry, the strength of the slurry after solidification, etc.

[0056] S03. Combine the geological information model and the tunnel long pipe roof structure information model to establish a corresponding analysis and calculation model;

[0057] During specific implementation, the geological information model and the tunnel long pipe shed structure information model created in the model creation module are imported into the finite element analysis software in the form of an analysis model through the software GBMDT (Digital Twin Software for Highway Standardization and Beautification Construction Project). After the import is completed, the physical information, geological information and mechanical properties of the corresponding model components need to be added. At the same time, according to the long pipe shed grouting method, the loading time, loading pressure, slurry physical and chemical properties and other construction factors are determined to determine the construction load during the long pipe shed construction process, the boundary conditions during the construction process, and the size of the entity unit division. Alternatively, the format conversion method is used to import the tunnel long pipe shed structure information model and the geological information model into the numerical analysis software, and divide the units to generate the corresponding analysis and calculation model. For example, the rvt file containing the entity model is converted into a sat file, and then imported into the numerical analysis software NERAP.

[0058] S04. Combine the long pipe shed construction method and process, and use finite element software to simulate the entire process flow of long pipe shed grouting;

[0059] Based on the long pipe roof grouting process, a staged simulation of the process is performed within the digital twin module. This staged simulation primarily simulates construction parameters such as the grouting method, grouting pressure, grouting concentration, and grouting time during the long pipe roof construction process. Using the construction process provided in the current tunnel construction manual or the extended finite element method, the diffusion range of the grouting slurry in the surrounding rock and the stress and strain changes in the surrounding rock after grouting are simulated to assess and verify the strength and stability of the surrounding rock after grouting.

[0060] The extended finite element method (FEM) was used to simulate the grouting process, with finite element analysis performed on the splitting grouting behavior of the grouting slurry within the rock mass. There are two primary approaches to simulating crack formation within the FEM: the discrete crack model and the diffuse crack model. The diffuse crack model simulates crack initiation and development, determines the initial surrounding rock strength parameters based on the geological information model, and describes the mechanical properties of the cracked surrounding rock units using an anisotropic elastic constitutive model.

[0061] During the staged grouting simulation, to accommodate long pipe shed construction conditions, the principles of "outside first, inside later," "skip-hole grouting," and "from thin to thick" were followed. The grouting process was divided based on grouting sequence, slurry concentration, and grouting pressure. Flow control was employed during the simulation, and the fluid volume method was used to determine the slurry-water boundary at any given moment, thereby generating grouting models for different time periods.

[0062] Based on the analysis and calculation results, the design factors affecting the reinforcement effect, such as the spatial position and diameter of the steel tubes in the tunnel long pipe roof structural information model, were modified and optimized. At the same time, the construction factors such as the grouting time and grouting pressure, grouting pore filling rate, and slurry type in the tunnel long pipe roof construction plan were reasonably adjusted. The two were simultaneously modified to the corresponding analysis and calculation model and re-analyzed and calculated to ensure a safe and economical grouting effect.

[0063] S05. Continuously conduct on-site detection to collect data such as the actual location of the tunnel long pipe shed, slurry density, and diffusion range. By comparing real-time detection data, use artificial intelligence deep learning algorithms to modify the analysis and calculation model of the tunnel long pipe shed to ensure that the grouting effect is consistent with the actual situation. The data from the long pipe shed grouting simulation process is imported into the digital twin data management module for storage;

[0064] In specific implementation, the top and side surfaces of the tunnel face behind the long pipe shed arch are selected as monitoring points. When grouting operations are carried out in the long pipe shed, the elastic wave three-dimensional imaging method is used to monitor the monitoring points regularly. Multiple seismic source points are buried in the tunnel face, and seismic waves are stimulated by artificial hammering or mechanical impact sources, thereby detecting the grouting situation in front of the tunnel face; or a position close to the steel flower pipe is selected as a monitoring point on the tunnel face, and stress and strain sensors or bedrock displacers or multi-point displacement meters are buried in the tunnel face to record the stress and strain conditions of the tunnel face at the prescribed monitoring frequency for comparison with the analysis model results; or high-density resistivity imaging is used for detection to directly generate a real-time three-dimensional model of the entire grouting process, thereby obtaining the penetration of the grouting slurry in the surrounding rock; or geological radar and a supporting antenna system are used to continuously detect the working face during construction, and the actual position of the tunnel long pipe shed, slurry density and diffusion range data are obtained based on the feedback radar images, and the data are transmitted to the digital twin data management module through wired or wireless network transmission equipment.

[0065] Based on the construction data collected in real time, the analysis and calculation model is adjusted and corrected using an artificial intelligence-based deep learning algorithm, thereby correcting the surrounding rock classification and surrounding rock mechanical properties of the original geological information model, and feeding it back into the analysis and calculation model for further analysis and prediction. Through multiple cycles of identification and correction, the error between the numerical value obtained through analysis and the measured value is controlled within a reasonable range. Based on the analysis and calculation model, the action of the next step of the tunnel long pipe roof construction is predicted, and the relevant surrounding rock stress and strain data and analysis and calculation model are stored and summarized in the digital twin data management module.

[0066] S06. Export the tunnel grouting model from the digital twin data management module to achieve digital delivery of the tunnel long pipe shed grouting process.

[0067] During implementation, the secondary development interface of the digital twin data management module aggregates the grouting models for each stage of the long pipe shed within the module and ultimately converts them into a file format accepted by common BIM modeling software, enabling 3D visualization of the grouting effects. This phased export of the models allows for the simulation of the entire construction process, ultimately completing the digital delivery of the tunnel long pipe shed grouting process.

[0068] Example 3

[0069] like Figure 2 As shown in the figure, a digital twin and fine modeling system for a tunnel long pipe shed is constructed based on the above-mentioned long pipe shed modeling method flow. The module composition structure diagram of this system is used for tunnel long pipe shed modeling, digital delivery and full life cycle management. It includes a long pipe shed physical entity data perception and transmission module, a digital twin model creation module, a digital twin module, and a digital twin data management module.

[0070] The long pipe shed physical entity data perception and transmission module is used to obtain surrounding rock information and geological, hydrological, topographic and geomorphological information of the tunnel long pipe shed construction area, and to monitor in real time the mechanical state and grouting state of the surrounding rock, the actual position of the tunnel long pipe shed, the slurry density and diffusion range, and other data during the construction of the tunnel long pipe shed, and transmit the monitoring and measurement data to the digital twin data management module;

[0071] Furthermore, the long pipe roof physical entity data perception and transmission module is also used to collect the monitoring data, construction quality data, grouting pressure and range parameters of the physical entity in real time, and upload the collected data to the digital twin data management module to realize real-time data collection, and use the collected real-time data to correct the analysis and calculation model and predict behavior, so as to obtain analysis results that are consistent with reality.

[0072] The data sensing and transmission equipment includes geological information collection equipment for obtaining geological information in preliminary surveys and sensing equipment during the construction process, including stress and strain sensors, piezometers, multi-point displacement meters, bedrock displacers, cross-section deformation monitoring and measurement devices, geological radars and supporting antenna systems, etc. embedded in the structure during construction, as well as wired or wireless network transmission equipment for transmitting sensing data.

[0073] The digital twin model creation module is used to create a geological information model and a tunnel long pipe roof structure information model, wherein the geological information model should include the surrounding rock information and geological, hydrological, topographic and geomorphological information of the tunnel long pipe roof construction section; the tunnel long pipe roof structure information model should include the long pipe roof, initial support, secondary lining and grouting body; after the completion of this module, the geological information model and the tunnel long pipe roof structure information model are transferred to the digital twin module.

[0074] The geological information model is used to express geological survey information in a three-dimensional visual manner, providing a basis for the surrounding rock geology for the phased analysis of the tunnel long pipe shed grouting construction;

[0075] The tunnel long pipe roof structure information model is used to evaluate and optimize the design scheme of the tunnel long pipe roof structure and to generate an analysis and calculation model;

[0076] The analytical calculation model is used to evaluate and verify the grouting effect of the tunnel long pipe shed and the surrounding rock structure after grouting in terms of mechanics and stability during the construction process of the tunnel long pipe shed.

[0077] The digital twin module is used to simulate the long pipe shed grouting process in stages and to perform model correction based on the real-time data collected by the long pipe shed physical entity data perception and transmission module. It also uses an artificial intelligence-based deep learning algorithm to predict the behavior of the long pipe shed construction process, thereby obtaining a behavior prediction model, and passing the behavior prediction model and related analysis data to the digital twin data management module.

[0078] The model correction of the digital twin module mainly compares the real-time construction data collected by the long pipe roof physical entity data perception and transmission module with the staged long pipe roof simulation data using an artificial intelligence deep learning algorithm. This data is then corrected to the surrounding rock classification and mechanical properties of the original geological information model. This data is then fed back into the analysis and calculation model for further analysis and prediction. Through multiple cycles of identification and correction, a behavior prediction model is obtained. Combined with the existing grouting volume, the presence of cavities in the surrounding rock and the degree of diffusion of the grouting slurry in the surrounding rock are predicted, thereby timely adjusting the grouting volume and grouting pressure during the long pipe roof construction process to ensure the grouting effect and maximize economic benefits.

[0079] The behavior prediction model is used to predict the stress, deformation, safety status, etc. of the surrounding rock and structure of the long pipe shed at a certain point in the future during the long pipe shed construction process, and to predict possible grouting defects or dangerous situations.

[0080] The digital twin data management module is built based on a relational database model and is used to receive and store monitoring data during the excavation process of the tunnel long pipe shed, initial long pipe shed design data, and calculation, analysis, and simulation data of the digital twin module, and provides a corresponding interface for model export during the grouting process of the tunnel long pipe shed.

[0081] A digital twin and detailed modeling system for tunnel pipe roofs creates a digital twin that is fully equivalent to the historical state of the pipe roof's physical entity during construction. Construction data of the pipe roof's physical entity is collected through the pipe roof's physical entity data perception and transmission module. The digital twin module uses this data to modify the analytical calculation model, assessing and predicting the deformation and stress characteristics of the pipe roof during construction, as well as the grouting effect. The digital twin data management module stores data related to pipe roof structural analysis, corrections, and predictions, enabling decision-makers to determine the appropriate grouting solution. It also provides an interface for exporting the pipe roof's detailed model.

[0082] like Figure 3 The figure shows the model construction process and relationship diagram of the digital twin module. The digital twin module is constructed synchronously with the survey, design and construction of the tunnel project.

[0083] Using a variety of exploration methods, including geological radar, laser scanning, magnetotellurics, and borehole sonic logging, a multi-technical exploration plan is developed to obtain relatively detailed and accurate geological exploration information. Utilizing 3D geological modeling technology, spatial information management, geological interpretation, spatial analysis and prediction, geostatistics, entity content analysis, and graphical visualization tools are combined in a 3D environment to create a 3D visualized geological information model. For example, laser scanning technology is used to obtain a 3D terrain point cloud model of the long pipe roof construction section of the tunnel, which is then converted into a 3D terrain surface model. Based on geological surrounding rock data collected by sensing equipment, combined with the 3D terrain surface model and artificial intelligence-based deep learning algorithms, the surrounding rock is divided into different sections, resulting in a 3D geological model that conforms to reality.

[0084] Based on the design drawings provided by the designer, relevant parameters such as the long pipe shed lining section, the three-dimensional coordinates of the beginning and end of the steel flower pipe, and the three-dimensional coordinates of the steel arch frame positioning are extracted into the tunnel long pipe shed structure information database and transmitted to the digital twin data management module. The tunnel shed structure information model is modeled using visual programming modeling. The visual programming modeling calls the data in the database to perform operations such as the lofting modeling of the long pipe shed from the arch outline to the arch entity, the placement of steel pipe drilling holes, the positioning and placement of advance small guide tubes, anchor rods and steel arch frames, which can improve the model accuracy and modeling efficiency. If the design drawings need to be changed during the construction process due to safety factors such as changes in the surrounding rock, landslides, and water gushing, modifications can be made based on the original tunnel long pipe shed structure information database, and the model parameters can be quickly and accurately changed through the original programming program.

[0085] The geological information model and the tunnel pipe shed structural information model are imported into finite element analysis software as analytical models using the GBMDT software, or converted into numerical analysis software after format conversion. Physical information, geological information, and material constitutive relationships of the corresponding model components are added. Simultaneously, construction factors such as loading time, loading pressure, and slurry properties are determined based on the pipe shed grouting method. Construction loads, boundary conditions, and the size of the entity elements during the pipe shed construction process are also determined, thereby forming an analytical calculation model. This analytical calculation model can be used to verify and analyze the structural design during the design and construction phases. If the design results do not meet the requirements, the tunnel pipe shed structural information model can be modified and optimized.

[0086] During the tunnel pipe shed grouting construction process, analytical and computational models were used to predict the grouting effect, surrounding rock stress, and deformation. Real-time construction big data collected by the pipe shed's physical entity data perception and transmission module was compared with staged pipe shed finite element simulation data using an artificial intelligence-based deep learning algorithm. This data was then used to correct the surrounding rock classification and mechanical properties of the original geological information model. This data was then fed back into the analytical and computational model for further analysis and prediction. Multiple cycles of identification and correction were performed to ensure that the error between the analyzed values ​​and the measured values ​​remained within a reasonable range. The analytical and computational model was then used to predict the next step in the tunnel pipe shed construction process, thereby generating a behavioral prediction model. This model enabled a visual representation of the tunnel pipe shed grouting process and the prediction and assessment of the surrounding rock mechanical conditions after grouting. Through the synchronous operation of the tunnel pipe shed's digital twin and the physical entity, virtual-real interaction, simulation, and iterative optimization, detailed modeling of the tunnel pipe shed's construction baseline state was achieved, full perception of post-construction evolution, and accurate prediction of construction quality and reinforcement effectiveness.

[0087] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A digital twin and fine modeling method for a long tunnel pipe roof, characterized by: The following steps are involved: S01. Comprehensively collect tunnel exploration data and establish a three-dimensional visualized geological information model; the geological information model includes surrounding rock information and geological, hydrological, topographic and geomorphological information of the tunnel long pipe shed construction area; S02. Creating a three-dimensional visualized tunnel long pipe shed structure information model based on the design drawings; the tunnel long pipe shed structure information model includes the long pipe shed, primary support, secondary lining, and grouting body; S03. Establish a corresponding analysis and calculation model by combining the geological information model and the tunnel long pipe roof structural information model; the analysis and calculation model includes geometric information and physical information of the long pipe roof components, mechanical properties and constitutive relationships of the materials, geometric information of the plum blossom-shaped boreholes, geometric properties and constitutive relationships of the surrounding rock and geological body, and construction factors including loading time, loading pressure, and physical and chemical properties of the grouting fluid for the long pipe roof; S04. Using finite element software to simulate the entire process of long pipe shed grouting, in combination with the long pipe shed construction method and process; simulating the long pipe shed grouting process in stages in the digital twin module according to the long pipe shed grouting process; the staged simulation of the digital twin module simulates the grouting method, grouting pressure, grouting concentration, and grouting time construction parameters during the long pipe shed construction process; During the staged grouting simulation, to meet the on-site construction conditions of the long pipe shed, the principles of outside-first-inside, skip-hole grouting, and grouting from dilute to concentrated were followed. The grouting process was divided based on the grouting sequence, grouting slurry concentration, and grouting pressure. Flow control was used during the simulation, and the fluid volume method was used to determine the slurry-water boundary at any time, thereby obtaining grouting models for different time periods. Based on the analysis and calculation results, the design factors affecting the reinforcement effect, such as the spatial position and diameter of the steel tubes in the tunnel long pipe shed structural information model, were modified and optimized. At the same time, the construction factors of the tunnel long pipe shed construction plan, such as the grouting time and grouting pressure, grouting pore filling rate, and slurry type, were rationally adjusted. These two factors were simultaneously modified and incorporated into the corresponding analysis and calculation model, and re-analyzed and calculated to ensure safe and economical grouting results. S05. Continuously collect on-site detection data. By comparing it with real-time detection data, use the artificial intelligence deep learning algorithm to modify the analysis and calculation model of the tunnel long pipe shed to ensure that the grouting effect is consistent with the actual situation. Import the data from the long pipe shed grouting simulation process into the digital twin data management module for storage; The top and side surfaces of the tunnel face behind the long pipe shed are selected as monitoring points. When grouting is carried out in the long pipe shed, the elastic wave three-dimensional imaging method is used to monitor the monitoring points regularly. Multiple seismic source points are buried in the tunnel face, and seismic waves are stimulated by artificial hammering or mechanical impact source methods, thereby detecting the grouting situation in front of the tunnel face. Alternatively, a position close to the steel flower pipe is selected as a monitoring point on the tunnel face, and stress and strain sensors or bedrock displacement meters or multi-point displacement meters are buried in the tunnel face to record the stress and strain conditions of the tunnel face at the specified monitoring frequency for comparison with the analysis model results. Alternatively, high-density resistivity imaging is used for detection to directly generate a real-time three-dimensional model of the entire grouting process, thereby obtaining the penetration of the grouting slurry in the surrounding rock. Alternatively, a geological radar and a supporting antenna system are used to continuously detect the working face during construction, and the actual position of the tunnel long pipe shed, slurry density and diffusion range data are obtained based on the feedback radar images, and the data are transmitted to the digital twin data management module through wired or wireless network transmission equipment. Based on the construction data collected in real time, the analysis and calculation model is adjusted and corrected using an artificial intelligence-based deep learning algorithm. This corrects the surrounding rock classification and mechanical properties of the original geological information model, and feeds it back into the analysis and calculation model for further analysis and prediction. Through multiple cycles of identification and correction, the error between the values ​​obtained through analysis and the measured values ​​is controlled within a reasonable range. The analysis and calculation model is used to predict the next step of the tunnel long pipe roof construction process, and the relevant surrounding rock stress and strain data and analysis and calculation model are stored and summarized in the digital twin data management module. S06. Export the tunnel grouting model from the digital twin data management module to achieve digital delivery of the tunnel long pipe shed grouting model.

2. The digital twin and fine modeling method for a long tunnel tube roof according to claim 1 is characterized by: The digital twin data management module is built using a relational database model to receive and store monitoring data from the tunnel long pipe shed excavation process, initial long pipe shed design data, and calculation, analysis, and simulation data from the digital twin module, and provides a corresponding interface for model export during the tunnel long pipe shed grouting process.

3. The digital twin and fine modeling method for a long tunnel pipe roof according to any one of claims 1 to 2, characterized in that: GBMDT is used to transform the geological information model and the tunnel long pipe roof structure information model into an analysis and calculation model.

4. A tunnel long pipe roof digital twin and fine modeling system, used to implement the tunnel long pipe roof digital twin and fine modeling method described in claim 3, characterized by: It includes the long pipe shed physical entity data perception and transmission module, digital twin model creation module, digital twin module, and digital twin data management module; The long pipe shed physical entity data perception and transmission module is used to obtain surrounding rock information and geological, hydrological, topographic and geomorphological information of the tunnel long pipe shed construction area, and monitor the mechanical state and grouting state of the surrounding rock, the actual position of the tunnel long pipe shed, the slurry density and diffusion range data in real time during the construction of the tunnel long pipe shed, and transmit the monitoring and measurement data to the digital twin data management module; The digital twin model creation module is used to create a geological information model and a tunnel long pipe roof structure information model, and transmit the geological information model and the tunnel long pipe roof structure information model to the digital twin module; The digital twin module is used to simulate the long pipe shed grouting process in stages and to modify the analysis model based on the real-time data collected by the long pipe shed physical entity data perception and transmission module. It also uses an artificial intelligence-based deep learning algorithm to predict the behavior of the long pipe shed construction process, thereby obtaining a behavior prediction model. The behavior prediction model and related analysis data are then transmitted to the digital twin data management module. The digital twin data management module is used to store and manage system data, as well as export system data and export tunnel long pipe roof grouting models.

5. The digital twin and fine modeling system for a long tunnel tube roof according to claim 4 is characterized by: The long pipe roof physical entity data perception and transmission module is also used to collect real-time monitoring data, construction quality data, grouting pressure and range parameters of the physical entity, and upload the collected data to the digital twin data management module to realize real-time data collection, and use the collected real-time data to correct the analysis and calculation model and predict behavior, so as to obtain analysis results that are consistent with reality.

6. The digital twin and fine modeling system for a long tunnel tube roof according to claim 4 is characterized by: The tunnel long pipe roof structure information model is based on the tunnel long pipe roof structure information library of design drawing data and is modeled using a visual programming modeling method.

7. The digital twin and fine modeling system for a long tunnel tube roof according to claim 4 is characterized by: The model correction of the digital twin module uses the real-time construction big data collected by the long pipe roof physical entity data perception and transmission module to compare the data with the staged long pipe roof finite element simulation data, thereby correcting the surrounding rock classification and surrounding rock mechanical properties of the original geological information model, and feeding it back to the analysis and calculation model for further analysis and prediction, and performing multiple cycles of identification and correction to obtain a behavior prediction model.

Images