A method and system for immersed tunnel construction based on digital twinning

Abstract

The application discloses a kind of based on digital twinning immersed tunnel construction method and system, the system includes physical entity module, digital twinning body module, perception transmission control module, digital twinning data management module, service application module, with immersed tunnel engineering full life cycle synchronization construction, throughout the survey, design, construction and operation and maintenance each stage of tunnel engineering, by means of intelligent sensing and internet of things, unmanned aerial vehicle and edge equipment etc., realize the integration and efficient cooperation of intelligent survey, intelligent design, intelligent construction, intelligent detection, the system of the application provides guiding scheme for the digital delivery of immersed tunnel, pipe section manufacturing, float and sinking etc. Intelligent construction and operation and maintenance management, in operation and maintenance, the data of real-time monitoring is corrected to behavior prediction model, and security is provided for the operation and maintenance management of tunnel.

Description

Technical Field

[0001] This invention belongs to the field of tunnel traffic engineering, and in particular relates to a method and system for constructing immersed tunnels based on digital twins. Background Technology

[0002] In recent years, with economic development, the construction of transportation projects has gradually increased. Tunnel engineering projects, in particular, span wide areas, involve complex terrain, and present significant construction challenges and numerous procedures. To ensure high-quality construction, builders have been continuously exploring better construction technologies and efficient management methods.

[0003] The development of transportation infrastructure construction technologies has gone through four stages: the manual stage, the mechanization and computerization stage, the informatization and automation stage, and the intelligent construction stage. In the manual stage, work was mainly carried out using manual techniques such as manual calculation and drawing. With the development of mechanization and computerization, construction projects used large-capacity and high-efficiency construction machinery and equipment, and simple design and progress analysis were carried out with computer assistance. In the informatization and automation stage, digital and information management systems were established, realizing the refined management of the construction process by automated equipment. In the intelligent construction stage, technologies such as three-dimensional modeling and simulation analysis, factory prefabrication, mechanized installation, precision measurement and control, structural safety and hygiene monitoring, construction environment perception, and intelligent information management and control platforms are mainly used to achieve high-quality and efficient construction throughout the entire process. The development of these technologies has promoted the continuous progress and improvement of transportation infrastructure construction (Chen Weile, Song Shenyou, Jin Wenliang, Xia Fengyong. Planning and Practice of Intelligent Construction System for Steel Shell Concrete Immersed Tunnel of Shenzhen-Zhongshan Bridge [J]. Tunnel Construction (Chinese and English), 2020, 40(04): 465-474.). Summary of the Invention

[0004] Currently, immersed tunnel engineering lacks collaborative methods for full lifecycle management, including intelligent construction, digital delivery, and intelligent operation and maintenance. This invention primarily establishes a digital twin system for immersed tunnels, fully integrating digital twins (DT) with next-generation advanced information technologies such as Building Information Modeling (BIM), Geological Information Modeling (GIS), Hydrological Information Modeling (HIM), Airflow Information Modeling (AIM), the Internet of Things (IoT), virtual simulation, drones, and edge computing devices. This allows for understanding the past, perceiving the present, and predicting the future, ultimately achieving truly intelligent industrialized construction of immersed tunnels.

[0005] The present invention is achieved by at least one of the following technical solutions.

[0006] A method for constructing immersed tunnels based on digital twins includes the following steps:

[0007] S01. Establish a three-dimensional visualized geological information model, hydrological information model, and air flow field information model, and use oblique photography to perform three-dimensional real-scene modeling of surrounding site buildings;

[0008] S02. Establish a three-dimensional visualized immersed tunnel structural information model based on design drawings;

[0009] S03. Establish analytical calculation models during the prefabrication, floating, and sinking of the pipe sections to analyze, calculate, and optimize the stress on the structure.

[0010] S04. Based on intelligent construction equipment, carry out automated and intelligent construction, and simultaneously embed sensing, transmission and control equipment;

[0011] S05. Using drones and edge computing equipment, data are collected on the floating path and movement attitude of tunnel sections and potential risks are identified.

[0012] S06. Through the monitoring system, continuously collect monitoring data, construct a behavior prediction model for the tunnel, and continuously optimize the behavior prediction model based on the error between the measured values ​​and the calculated values ​​of the behavior prediction model.

[0013] S07. Link databases, IoT and tunnel structures to achieve digital collaboration and build a system platform for tunnel operation and maintenance management.

[0014] The system for implementing the digital twin-based immersed tunnel construction method includes: a physical entity module, a digital twin module, a sensing, transmission and control module, a digital twin data management module, and a service application module.

[0015] The physical entity module is used to reflect the actual situation of the immersed tunnel structure. Taking the entire life cycle of the immersed tunnel as the timeline, the physical entities of each link interact with the digital twin, services, and twin data in real time. The physical entity module includes the foundation trench, piles, immersed tunnel structure, and tunnel ramps.

[0016] The digital twin module includes a geological information model, a hydrological information model, a tunnel structure information model, an analysis and calculation model, and a behavior prediction model. It is highly mapped one-to-one with the physical entity, and the functions of the physical entity module are iteratively optimized and simulated in real time in the digital twin according to user needs.

[0017] The sensing and transmission control module is used to collect and transmit data on the hydration heat of the precast concrete in the dry dock, the safety risks of the precast submerged section, the quality of the precast concrete, and the deformation and stress of the sections during the floating, sinking, docking and operation and maintenance processes. It is also used to obtain the construction data and status of the service application module through the model components.

[0018] The digital twin data management module is used to store and manage the monitoring data and model data of the sensing transmission control module, as well as to provide the service application module with an interface to view model information, component attributes and real-time monitoring data, and to forward the instructions from the service application module.

[0019] The service application module utilizes geological information models, hydrological information models, airflow field information models, and tunnel structure information models to present the three-dimensional spatial relationship between the tunnel and the weak foundation and the distribution of hydrogeology in a three-dimensional visualization manner. By analyzing and calculating models and behavioral prediction models, and linking monitoring data, calculation data, and construction documents with information models, the efficiency of operation and maintenance management is improved.

[0020] Furthermore, the digital twin module includes a geological information model, a hydrological information model, an airflow field information model, a tunnel structure information model, an analysis and calculation model, and a behavior prediction model. The geological information model is used to present the topography and lithology of the tunnel project environment in three dimensions; the hydrological information model is used to present the river water level and tide level of the tunnel project environment in three dimensions; the tunnel structure information model is used to present the tunnel structure in three dimensions and provide visualized construction technology briefings; the analysis and calculation model is used to perform mechanical analysis on the immersed tunnel structure and evaluate the structural safety and reliability of the immersed tunnel; the behavior prediction model combines the monitoring data and calculation data of the immersed tunnel to predict the development trend of tunnel defects and deformation, and provides timely early warning information.

[0021] Furthermore, based on the method of separating data from models, the relationship between the attribute parameters of geological information models, hydrological information models, air flow field information models, and tunnel structure information models and the sensor component codes is used to establish an attribute table and a monitoring data table for monitoring points in a relational database, and associate them through component codes to facilitate users in finding data on tunnel components.

[0022] Furthermore, during the precast tunnel construction, the sensing and transmission control module uses the Internet of Things and temperature sensors to synchronously monitor the temperature field of the precast tunnel section concrete and transmits the data to a database as a data benchmark for construction, operation and maintenance, and defect deformation analysis. It also uses drone aerial photography to conduct regular inspections of the tunnel's controlled precast construction and employs computer video and deep learning algorithms for risk identification and early warning management of the precast tunnel sections. Additionally, it utilizes drones for real-time video monitoring of the immersed tunnel's floating and transport, and installs edge equipment on the immersed tunnel to monitor the floating path and movement attitude of the tunnel sections in real time and provide real-time feedback, identifying potential risks.

[0023] Furthermore, the service application module utilizes geological information models, hydrological information models, airflow field information models, and tunnel structure information models to generate analysis and calculation models for stress analysis of the tunnel structure. It integrates and displays the calculation models with the tunnel structure information models, making the calculation results more intuitive. Real-time monitoring data is displayed in the system through visual charts, and by comparing and analyzing with monitoring indicator thresholds, abnormal tunnel conditions are promptly reported. Furthermore, using the tunnel structure information model and oblique photography real-scene models, it intelligently generates drone inspection route planning during the prefabrication of tunnel sections. During the floating and transporting of tunnel sections, remote operation controls the edge equipment installed on the tunnel sections to collect, process, and transmit data.

[0024] Furthermore, the digital twin data management module includes data drawings, geological 3D models, hydrological information models, airflow field information models, tunnel structure information models, attribute information and analysis calculation models, calculation results data of behavior prediction models, sensor monitoring data, IoT monitoring data, UAV aerial image data, and edge equipment measurement results data. It also enables the association of collected and monitored measurement data with information models to improve information management efficiency and document retrieval efficiency.

[0025] Furthermore, the service application module, with the help of the large amount of structured data and monitoring data integrated by the digital twin data management module, uses the embedded tunnel safety assessment algorithm to generate a behavior prediction model, conduct a safety assessment of the tunnel's condition, and store the prediction assessment structure in the database as a reference for operation and maintenance management.

[0026] Furthermore, the service application module adopts a C / S architecture, combining tunnel structure information model, geological information model, hydrological information model, air flow field information model, Internet of Things, drones and edge equipment to establish an intelligent construction and digital twin system platform for immersed tunnels, thereby improving the overall control level and quality of immersed tunnels.

[0027] Furthermore, the behavior prediction model employs a GA-BP neural network algorithm model combined with deep learning for generation and continuous optimization.

[0028] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0029] This invention proposes a method for intelligent construction of immersed tunnels and establishes a digital twin system, providing strong theoretical support and practical guidance for the intelligent construction of immersed tunnels. During segment prefabrication, drone inspection routes can be intelligently generated to identify and manage risks during segment prefabrication. During segment floating, drones and edge computing devices are used to monitor the floating path, attitude, deformation, and stress during docking in real time, improving the level of intelligent construction of immersed tunnels. In operation and maintenance, real-time monitoring data is used to correct the behavior prediction model, ensuring the safety of tunnel operation and maintenance management. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the intelligent construction method of the immersed tunnel digital twin system in an embodiment.

[0031] Figure 2 This is a schematic diagram of the modular structure of the immersed tunnel digital twin system as an example.

[0032] Figure 3 This is a schematic diagram illustrating the workflow of edge equipment in an immersed tunnel digital twin system, as an example.

[0033] Figure 4 This is a schematic diagram illustrating the model construction process and relationships of the digital twin module in the immersed tunnel digital twin system, as shown in the example. Detailed Implementation

[0034] Exemplary embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. However, it should be understood that the invention can be implemented in various forms and should not be limited to the exemplary embodiments set forth herein; rather, these embodiments are provided to enable a more thorough understanding of the invention and to fully convey the scope of the invention to those skilled in the art. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0035] Example 1

[0036] like Figure 1 As shown in the figure, this embodiment presents a digital twin-based immersed tunnel construction method. This method is used for the entire lifecycle management of immersed tunnel projects, including intelligent construction, digital delivery, and operation and maintenance. The intelligent construction process (which includes design, prefabrication, and floating transportation) includes the following steps:

[0037] S01. Search and sort out various geological survey data (including the natural geographical conditions and hydrogeological information of the tunnel project location), establish a three-dimensional visualized geological information model, hydrological information model, and air flow field information model, and use oblique photography to create a three-dimensional real-scene model of the surrounding site buildings.

[0038] In practice, based on the preliminary borehole exploration data of the immersed tunnel, borehole numbers and coordinates, top and bottom elevations of the strata, and ground elevations are entered into the database. The stored geological exploration data is read from the EVS software and converted to PGF format. Data matching with EVS parameters is completed, and corresponding GEO files are output according to the different geological bodies. Finally, GIS modeling is performed using the Kriging method as a spatial interpolation method in EVS software to establish a geological information model. Based on the preliminary exploration results, a topographic surface point cloud map is generated and fitted into a curved surface. Then, using the outer edge line of the water body and the riverbed water level line, Revit and Dynamo are used to create the water body model and air flow model, establishing a hydrological information model and an air flow field information model.

[0039] S02. Based on the design and construction drawings, a three-dimensional visualized immersed tunnel structural information model is established using parametric methods.

[0040] In practice, the design data, including the tunnel station number, elevation of each tunnel segment, section ID of the immersed tunnel component, and material of each component, is entered into the database. The construction drawings (CAD electronic drawings) in .dwg format are imported into Revit, and Dynamo is used to read the geometric parameters such as the tunnel segment elevation and section ID data from the database. The cross-section of the tunnel segment is determined, and the three-dimensional layout line required for model layout is fitted. The cross-section template is then used to create a solid model of the tunnel along the three-dimensional layout line.

[0041] S03. Based on the hydrological information model, air flow field information model and tunnel structure information model, generate corresponding analysis and calculation models (simulation modeling) during the prefabrication, floating and sinking of the tunnel sections, respectively, perform mechanical analysis on the tunnel, and perform structural stress analysis, calculation and optimization of the design scheme.

[0042] In practice, based on the established hydrological information model, airflow field information model, and tunnel structure information model, the fluid domain and pipe section surface model (surfaces, meshes, and geometric entities, etc.) are exported using Dynamo in Revit software and saved as a SAT file (binary file format). The exported SAT file is then imported into ANSYS FLUENT software for geometric model setting, member creation, mesh generation, and setting of analysis conditions and working conditions. A fluid-solid coupling simulation model (analysis and calculation model) for the pipe section immersion process is established. Numerical simulation calculations are performed on the analysis and calculation model to ensure the stress and stability of the pipe section during the construction phase, and to understand the hydrodynamic characteristics and stress conditions of the pipe section throughout the entire construction process from dry dock floating and transport to the completion of the immersed tunnel installation.

[0043] S04. Based on intelligent construction equipment, intelligent monitoring and measurement are carried out on the construction process of immersed tunnel, and the data is transmitted to the database for storage in real time, and sensing and transmission control equipment is embedded synchronously.

[0044] In practice, the Internet of Things (IoT) and sensors are used to collect data on the deformation and stress of the precast tunnel sections during the dry-dock prefabrication process, as well as to monitor the temperature field of the concrete hydration heat during prefabrication. This data is transmitted in real-time to the digital twin data management center via a 5G network. Using a database and the tunnel's structural information model, the data is stored and associated with specific components. During the factory prefabrication of the immersed tunnel sections, drones are used for aerial photography to document the controlled prefabrication construction. Computer video and deep learning algorithms (convolutional neural network algorithms) are employed for risk identification and early warning management during section prefabrication. Furthermore, intelligent drone flight path planning and oblique photography technology are used to conduct regular inspections of the concrete quality on the section surface.

[0045] S05. Using drones and edge computing equipment, data are collected on the floating path and movement attitude of tunnel sections and potential risks are identified.

[0046] In practice, during the floating and transporting of the tunnel segments, edge devices (capable of real-time sensing, data acquisition, and high-speed wireless transmission to the monitoring center) are pre-installed at the four corners and the middle (freeboard surface) of the immersed tunnel segment above the water. These edge devices are remotely controlled to collect attitude data in real time during the floating, sinking, and docking processes, and the data is transmitted to the monitoring center via a router floating on the water surface. Furthermore, drones are used to provide real-time video monitoring of the entire floating and transporting process, offering guidance for the segment's movement. During the floating, sinking, and docking of the immersed tunnel segments, Internet of Things (IoT) technology is employed to monitor and measure their deformation and stress in real time.

[0047] S06. Through the monitoring system, continuously collect monitoring data, use the tunnel monitoring data to build a tunnel behavior prediction model, and continuously update the monitoring data, compare it with the calculated values ​​of the behavior prediction model, and continuously modify the behavior prediction model.

[0048] In practice, sensor data measured on-site is transmitted to the digital twin data management module, integrated with the analysis and calculation model, to construct a behavior prediction model and generate a digital twin. By sensing current data, it predicts potential future safety hazards and displays the results in 3D visualization. Furthermore, the behavior prediction model is continuously refined using real-time collected data and analysis results, ultimately controlling its error value within an acceptable range. Changes in the monitored data are displayed in visual charts, and the data is compared and judged based on pre-set monitoring indicator thresholds to promptly report any abnormal conditions in the tunnel.

[0049] S07. Link databases, the Internet of Things and tunnel structures to achieve digital collaboration, establish a digital twin system for immersed tunnels, shift from intelligent construction to operation and maintenance management, and realize digital delivery of projects.

[0050] In practical implementation, a client / server architecture is adopted to establish a client for the immersed tunnel digital twin system. The client sends access commands to the database to browse the 3D visualization model, display visualized charts of monitoring data, and view real-time tunnel footage via the Internet of Things (IoT). The established digital twin system can coordinate the work of the owner, supervisor, design, and construction departments, making collaboration among all parties more efficient and convenient, and promoting the digital delivery of the project. It also shifts the focus from intelligent construction to intelligent operation and maintenance, enabling comprehensive management of project information and improving operation and maintenance efficiency.

[0051] Example 2

[0052] like Figure 2 As shown, a digital twin-based immersed tunnel construction system is established based on the above method. The system mainly consists of 5 modules, which have functional collaboration and interaction among themselves, including a service application module, a digital twin data management module, a sensing, transmission and control module, a physical entity module, and a digital twin module.

[0053] The service application module adopts a C / S architecture and combines technologies such as tunnel structure information model (BIM), geological information model (GIS), hydrological information model (HIM), air flow field information model (AIM), Internet of Things, drones and edge equipment to establish an intelligent construction and digital twin system platform for immersed tunnels, thereby improving the level and quality of whole-process control of immersed tunnels.

[0054] The client accesses data stored in the digital twin data management module. In surveying and design, 3D visualization browsing of Geological Information Models (GIS), Hydrological Information Models (HIM), Airflow Information Models (AIM), and Tunnel Structure Information Models (BIM) is possible, and relevant attribute information of components can be queried based on their location. During tunnel construction, sensors and IoT devices associated with tunnel model components allow for rapid monitoring of on-site construction status and progress. Numerical simulation results from analytical calculation models are overlaid on the tunnel information model, making data more intuitive and providing relevant prompts to ensure project schedule implementation and smooth construction. In operation and maintenance management, real-time on-site monitoring data is presented through visual sand table charts, and early warning prompts are issued based on set monitoring indicator thresholds. Furthermore, behavioral prediction models accurately predict potential future safety hazards. Throughout the tunnel's entire lifecycle, project documents and materials can be centrally managed, achieving information-based, digital, and paperless management.

[0055] Specifically, the service application module, with the help of a large amount of structured data and monitoring data integrated by the digital twin data management module, uses the embedded tunnel safety assessment algorithm (GA-BP neural network algorithm) to generate a behavior prediction model, which can perform a safety assessment of the tunnel's condition and store the prediction assessment structure in the database as a reference for operation and maintenance management.

[0056] The digital twin data management module stores and manages monitoring and model data from the transmission control module. It also provides interfaces for the service application module to view model information, component attributes, and real-time monitoring data. Furthermore, it provides interfaces for the service application module to interface with the sensing transmission control module and the digital twin module, and forwards instructions from the service application module. The digital twin data management module contains data including attribute information from drawings, geological 3D models, hydrological information models, airflow field information models, and tunnel structure information models; calculation results from analysis and calculation models and behavior prediction models; sensor monitoring data; IoT monitoring data; UAV aerial imagery data; and edge equipment measurement results. By building a relational database, it integrates and manages the surveyed hydrogeological information, designed tunnel structure information, and construction and maintenance monitoring data. The database consists of four tables: a model attribute information table, a document information table, a monitoring data table, and a prediction data table. Each table stores corresponding data and is linked through foreign keys. This allows for the association of collected and monitored measurement data with the information model, greatly improving information management and document retrieval efficiency.

[0057] As one embodiment, based on the method of separating data from models, the relationship between the attribute parameters of geological information models, hydrological information models, air flow field information models, and tunnel structure information models and the sensor component codes is used to establish an attribute table and a monitoring data table for monitoring points in a relational database, and associate them through component codes to facilitate users in finding data of tunnel components.

[0058] The sensing and transmission control module is used to collect and transmit data on the hydration heat of the precast concrete in the dry dock, the safety risks of the precast submerged section, the quality of the precast concrete, and the deformation and stress during the floating, sinking, docking, and operation and maintenance of the sections. It can also be used in the service application module to obtain its construction data and status through model components.

[0059] The perception, transmission, and control module specifically collects and monitors on-site data through sensors, the Internet of Things (IoT), and drones. Survey and design data is automatically collected and transmitted to the digital twin data management module. During the prefabrication of immersed tunnel segments, IoT and temperature sensors are used to synchronously monitor the temperature field of the prefabricated concrete and monitor the heat of hydration. Furthermore, the platform generates drone inspection flight plans for the segments using a tunnel information model and an oblique photography model (the tunnel information model is based on CAD immersed tunnel design drawings, using Revit+Dynamo modeling methods for automated, refined, and efficient BIM modeling of the immersed tunnel structure; the oblique photography model is based on drone low-altitude remote sensing technology, using drones as a platform to simultaneously acquire images from the sky at vertical and oblique angles to obtain building height and side textures, and using positioning technology, 3D modeling, and other technologies to generate a realistic 3D model). Data collection of the immersed tunnel construction scene is performed, and computer video and deep learning algorithms (convolutional neural network algorithms) are used for risk identification and early warning management during segment prefabrication. During the immersed tunnel floating and transportation process, drones are used for real-time video monitoring of the floating and transportation of the tunnel sections. Digital twin edge computing devices, pre-installed on the sections to sense their attitude, are used to collect real-time attitude data throughout the entire process. Internet of Things (IoT) technology is also used to monitor the deformation and stress of the tunnel sections during floating, sinking, and docking. In tunnel operation and maintenance, real-time monitoring data is combined with behavioral prediction models to predict future potential hazards. Sensors and IoT devices installed in the tunnel are used for real-time data acquisition and processing (data acquisition devices store data in databases or cloud platforms, then statistical analysis, machine learning, and other algorithms are used to analyze the monitoring data, identify anomalies, make predictions and provide early warnings, and visualize the monitoring data for users to understand the data more intuitively). This enables rapid monitoring of fault information, accurate location of fault causes, and assessment of structural condition, thereby achieving targeted predictive maintenance and effectively improving the safe operation of the tunnel.

[0060] Specifically, various sensors are installed inside the tunnel to collect data from various locations in real time. Big data technology is used to process, analyze, and model the collected real-time data. Machine learning algorithms are employed to intelligently analyze the data, identify potential fault modes, and provide timely warnings. A decision support system is established to compare real-time monitoring data with historical data, comprehensively analyze the causes of faults, accurately grasp the fault situation, and determine whether the tunnel requires maintenance or safety protection measures.

[0061] Digital twin edge devices refer to advanced computing devices or processors capable of real-time sensing, data acquisition, and high-speed wireless transmission to a monitoring center. They are typically embedded in physical or virtual devices to perform computing, storage, and networking functions. Digital twin edge devices usually operate in physical or virtual environments, providing a range of specific functions; these devices can provide real-time, fast, and reliable computing and storage capabilities when performing specific tasks. Edge devices include: high-performance IoT development boards; sensors and controllers; single-board computers such as Raspberry Pi; servers and storage devices in cloud computing and data centers; network edge computing devices such as switches; wireless base stations; and small mobile devices.

[0062] The physical entity module represents the hydrogeological conditions and geospatial information near the tunnel project, as well as the tunnel structure (including foundation trenches, piles, immersed tube structures, and tunnel ramps, etc.). It is used for the full life cycle management of tunnel engineering survey, design, construction, and operation and maintenance, reflecting the real situation of the immersed tube tunnel structure. Taking the full life cycle of the immersed tube tunnel as the timeline, the physical entities at each stage interact with the digital twin, services, and twin data in real time.

[0063] The digital twin module includes a physical model, a geometric model, a behavioral model, and a rule model. The physical model primarily describes the physical characteristics and attributes of the physical scene, such as the material, shape, size, density, mass, and elasticity of objects. It transforms the physical attributes of objects into digital data for calculation and simulation, thereby enabling the prediction and analysis of the physical behavior of objects. The geometric model primarily describes the spatial relationships of the physical scene, such as the relative positions, relative volumes, and shapes between objects. It digitizes the spatial relationships of the physical scene through 3D modeling of objects, facilitating visualization and interactive operation within the digital twin. The behavioral model primarily describes the motion behavior and dynamic characteristics of objects in the physical scene. It transforms physical principles into mathematical and computational models, enabling the prediction and simulation of object motion and interaction. The rule model primarily describes the interaction relationships and behavioral rules between objects in the physical scene. It transforms the dynamic laws, control laws, and constraint laws in the physical scene into computer programs and algorithms, enabling the prediction and simulation of object motion and interaction.

[0064] The digital twin module includes a geological information model, a hydrological information model, an airflow field information model, a tunnel structure information model, an analysis and calculation model, and a behavior prediction model. The geological information model, hydrological information model, and airflow field information model provide digital information on the geological and hydrological conditions around the tunnel and the movement of air, providing data support for the physical model of the digital twin. The geological and hydrological data contained in the geological and hydrological information models can be used to build digital models of the geological and hydrological scenes around the tunnel, and the airflow field information model can be used to predict airflow in the immersed tunnel, thus providing data input for the physical model of the digital twin. The tunnel structure information model is part of the geometric model of the digital twin, providing three-dimensional geometric information of the tunnel structure. The tunnel structure information model is used to construct the three-dimensional model of the tunnel in the digital twin, providing support for the visualization and interaction of the digital twin. The analysis and calculation model and the behavior prediction model are part of the behavior and rule models in the digital twin, used to analyze the structural characteristics and behavioral patterns of the tunnel. The analysis and calculation model can use techniques such as finite element analysis and computational fluid dynamics to analyze parameters such as strain and displacement of the tunnel structure under different loads, providing input for the behavior model of the digital twin. Behavioral prediction models can use machine learning and other technologies to predict the future behavior of tunnel structures, providing support for the rule model of digital twins.

[0065] Geological information models and hydrological information models are used for 3D visualization of the hydrogeological information of tunnel projects, including the topography and lithology of the tunnel environment. The 3D visualization also presents river and tidal levels in the tunnel environment, providing intuitive data for the intelligent construction of immersed tunnels. Airflow field information models can be used to study airflow during the construction and operation of immersed tunnels, helping to better understand phenomena, optimize design schemes, and improve engineering. Tunnel structure information models are used for 3D visualization of tunnel structures and to provide visualized construction technology briefings, effectively optimizing construction schemes and providing strong support for construction and operation management. Analysis and calculation models are used to perform mechanical analysis on the immersed tunnel structure and evaluate its structural safety and reliability. Behavioral prediction models serve as a data benchmark for operation and maintenance, combining monitoring and calculation data to predict the development trends of tunnel defects and deformations, and providing timely early warning information. Geological information models, hydrological information models, airflow field information models, tunnel structure information models, analytical calculation models, and behavioral prediction models are closely related to the physical models, geometric models, behavioral models, and rule models in digital twins. Together, they constitute the foundation and core technologies of digital twins.

[0066] like Figure 3The diagram illustrates the workflow of edge equipment. Edge equipment is constructed by installing multiple edge devices and deploying digital twins at the edge. Edge equipment installation and deployment: Based on the project scenario, the type of monitored data, and factors such as the device's communication and computing capabilities, suitable edge devices are selected. Appropriate operating systems and applications are installed on the devices, which are then installed in predetermined locations and connected to the network to ensure effective communication. Edge device security: Because edge devices are typically used to connect to networks, process sensitive data, or directly control physical devices, their security is paramount. This generally involves strengthening device management, ensuring device hardening and implementing strict management measures, including access control, firewalls, network isolation, strong passwords, security updates, and using asymmetric encryption algorithms to ensure the confidentiality and integrity of data transmission.

[0067] Edge devices operate and interact dynamically, enabling real-time communication and interaction with other devices and systems to achieve various functions and applications. Edge devices can collect information from their surroundings, such as temperature, strain, pressure, and acceleration, through sensors and other devices; they can also acquire data from other devices, such as controllers, computers, and cloud servers. Edge devices convert the collected data into computer-readable formats and perform preliminary processing, such as data cleaning, noise reduction, standardization, and normalization, to improve data quality and validity. They then use their built-in computing power for local data analysis and processing, such as IoT data analysis, image recognition, and machine learning, improving data processing efficiency and real-time performance. Edge devices then send the processed data to other devices or cloud servers for further analysis and processing, utilizing cloud computing and storage resources to analyze and mine the data, thereby generating more accurate predictions and modeling results. Simultaneously, edge devices can also receive instructions from other devices or cloud servers to perform automated tasks or respond to specific events. Edge devices can also connect to the internet for remote control and monitoring. Through the internet, users can monitor and control the device's operating status in real time, performing remote operation and adjustments. The cloud can also send remote commands and control signals to edge devices to respond to specific events and optimize the efficiency of device task execution, thus achieving closed-loop control.

[0068] Closed-loop control typically consists of four basic components: sensors, controllers, actuators, and a feedback loop. The signal output by the device (output quantity) is collected by the sensors and compared with a target signal (reference quantity). When a difference exists, the controller adjusts the output signal to bring it closer to the target signal. This adjustment signal is then sent to the actuator, which controls the device to adjust itself according to the input signal, achieving the desired control objective. Subsequently, the result of the execution is fed back to the sensors to further check the controller's effectiveness and adjust its parameters to optimize the control effect.

[0069] Before, during, and after the construction of immersed tunnel projects, data is collected and analyzed using edge equipment. Simultaneously, edge equipment is used to monitor crucial data such as settlement and horizontal displacement of the immersed tunnel sections in real time during construction. This data is automatically stored and processed on a cloud platform, and used for important predictions and simulation analyses through a digital twin model. The digital twin model can be adjusted based on data from actual construction, enabling real-time tracking of the immersed tunnel's operational status and providing intelligent risk warnings based on data analysis results. In the event of emergencies, the digital twin model can react immediately to the tunnel structure to avoid serious consequences. By simulating and optimizing the entire tunnel design and construction process, construction plans can be optimized, construction risks avoided, cost input reduced, and construction efficiency improved. In short, real-time data collected by edge equipment can update the digital twin model in real time, and through simulation analysis, the tunnel's condition can be predicted and risks assessed, problems can be identified promptly, and preventative maintenance can be carried out, improving the reliability of the immersed tunnel and enhancing project quality.

[0070] like Figure 4 The diagram shows the model construction process and relationship of the digital twin module. The digital twin module is constructed synchronously with the intelligent construction of the immersed tunnel.

[0071] A comprehensive survey of the immersed tunnel was conducted using methods such as geological survey and mapping, integrated geophysical exploration, geological drilling, hydrological measurement, and UAV aerial photography. Based on the survey data, geological information models and hydrological information models were established in geological modeling software such as EVS.

[0072] Based on the tunnel's construction drawings and other design information, the control parameters of the cross-section (such as the thickness of the top and bottom slabs, the thickness of the sidewalls, the net width and height inside the tunnel, etc.) and the coordinates of the tunnel centerline are read, and the tunnel structural information model is generated parametrically using 3D modeling software such as REVIT.

[0073] The geological and water body information model is integrated with the tunnel structure information model into a whole. Corresponding geometric and physical properties and other parameters are exported. Then, analysis software such as ANSYS is used to read the exported binary file data to generate a computational model. The model is then divided into elements, boundary conditions and working cases are set, and loads are applied. Finite element analysis is then performed to obtain the analytical computational model, providing a strong basis for the design and verification of immersed tunnel projects.

[0074] During the construction and operation of the tunnel, real-time monitored stress and strain data are compared with the calculated values ​​of the analysis and calculation model. Combined with deep learning to generate and continuously optimize the behavior prediction model (GA-BP neural network algorithm), the predicted data is made as close as possible to the real value, so as to predict the possible future situations of the immersed tunnel.

[0075] The service application module of this invention is directly user-oriented. It utilizes geological information models, hydrological information models, and tunnel structure information models to present the three-dimensional spatial relationship between the tunnel and the soft foundation and the distribution of hydrogeology in a three-dimensional visualization manner. Through analysis and calculation models and behavior prediction models, it provides a scientific basis for the safety of construction and operation and maintenance of immersed tunnels. It also links monitoring data, calculation data, and construction documents with information models to improve operation and maintenance management efficiency.

[0076] The preferred embodiments of the invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for constructing immersed tunnels based on digital twins, characterized in that, Includes the following steps: S01. Establish geological information models and hydrological information models, and use oblique photography to create three-dimensional real-scene models of surrounding buildings. S02. Establish a three-dimensional visualized immersed tunnel structural information model based on design drawings; S03. Establish analytical calculation models during the prefabrication, floating, and sinking of the pipe sections to analyze, calculate, and optimize the stress on the structure. S04. Based on intelligent construction equipment, carry out automated and intelligent construction, and simultaneously embed sensing, transmission and control equipment; S05. Using drones and edge computing equipment, data are collected on the floating path and movement attitude of tunnel sections and potential risks are identified. S06. Through the monitoring system, continuously collect monitoring data, construct a behavior prediction model for the tunnel, and continuously optimize the behavior prediction model based on the error between the measured values ​​and the calculated values ​​of the behavior prediction model. S07. Link databases, IoT and tunnel structures to achieve digital collaboration and build a system platform for tunnel operation and maintenance management.

2. A system for implementing the immersed tunnel construction method based on digital twins as described in claim 1, characterized in that, include: Physical entity module, digital twin module, sensing, transmission and control module, digital twin data management module, and service application module; The physical entity module is used to reflect the actual situation of the immersed tunnel structure. Taking the entire life cycle of the immersed tunnel as the timeline, the physical entities of each link interact with the digital twin, services, and twin data in real time. The physical entity module includes the foundation trench, piles, immersed tunnel structure, and tunnel ramps. The digital twin module includes a geological information model, a hydrological information model, an airflow field information model, a tunnel structure information model, an analysis and calculation model, and a behavior prediction model. It is highly mapped one-to-one with the physical entity, and the functions of the physical entity module are iteratively optimized and simulated in real time in the digital twin according to user needs. The sensing and transmission control module is used to collect and transmit data on the hydration heat of the precast concrete in the dry dock, the safety risks of the precast submerged section, the quality of the precast concrete, and the deformation and stress of the sections during the floating, sinking, docking and operation and maintenance processes. It is also used to obtain the construction data and status of the service application module through the model components. The digital twin data management module is used to store and manage the monitoring data and model data of the sensing transmission control module, as well as to provide the service application module with an interface to view model information, component attributes and real-time monitoring data, and to forward the instructions from the service application module. The service application module utilizes geological information models, hydrological information models, airflow field information models, and tunnel structure information models to present the three-dimensional spatial relationship between the tunnel and the soft foundation, the distribution of hydrogeology, and the airflow during the construction and operation of the immersed tunnel in a three-dimensional visualization manner. It analyzes and calculates models and behavioral prediction models, and associates monitoring data, calculation data, and construction documents with the information models.

3. The immersed tunnel construction system based on digital twins according to claim 2, characterized in that, The digital twin module includes a geological information model, a hydrological information model, an airflow field information model, a tunnel structure information model, an analysis and calculation model, and a behavior prediction model. The geological information model is used to visualize the topography and lithology of the tunnel project environment in three dimensions; the hydrological information model is used to visualize the river water level and tide level of the tunnel project environment in three dimensions; the airflow field information model is used to study the airflow during the construction and operation of the immersed tunnel; the tunnel structure information model is used to visualize the tunnel structure in three dimensions and provide visualized construction technology briefings; the analysis and calculation model is used to perform mechanical analysis on the immersed tunnel structure and evaluate the structural safety and reliability of the immersed tunnel. The behavioral prediction model combines monitoring and computational data of immersed tunnels to predict the development trend of tunnel defects and deformations, and provides timely early warning information.

4. The immersed tunnel construction system based on digital twins according to claim 3, characterized in that, Based on the method of separating data from models, the relationship between the attribute parameters of geological information models, hydrological information models, air flow field information models, and tunnel structure information models and sensor component codes is used to establish attribute tables and monitoring data tables for monitoring points in a relational database, and to associate them through component codes, so as to facilitate users to find data of tunnel components.

5. A digital twin-based immersed tunnel construction system according to claim 2, characterized in that, During the precast tunnel construction, the sensing and transmission control module uses IoT and temperature sensors to synchronously monitor the temperature field of the precast tunnel section concrete and transmits the data to a database as a data benchmark for construction, operation and maintenance, and defect deformation analysis. It also uses drone aerial photography to conduct regular inspections of the tunnel's controlled precast construction and employs computer video and deep learning algorithms for risk identification and early warning management of the precast tunnel sections. Furthermore, it utilizes drones for real-time video monitoring of the immersed tunnel's floating and transport, and installs edge equipment on the immersed tunnel to monitor the floating path and movement attitude of the tunnel sections in real time, providing real-time feedback and identifying potential risks.

6. The immersed tunnel construction system based on digital twins according to claim 3, characterized in that, The service application module uses geological information models, hydrological information models, air flow field information models, and tunnel structure information models to generate analysis and calculation models to perform stress analysis and calculation on the tunnel structure. It also integrates and displays the calculation model with the tunnel structure information model to make the calculation results more intuitive. Furthermore, it displays real-time monitoring data in the system through visualization charts and compares and analyzes the data with monitoring indicator thresholds to provide timely feedback on abnormal tunnel conditions. In addition, it uses the tunnel structure information model and oblique photography real-scene model to intelligently generate drone inspection route planning during pipe section prefabrication. During the floating and transporting of the tunnel segments, data acquisition, processing and transmission are carried out by remotely controlling the edge equipment installed on the tunnel segments.

7. The immersed tunnel construction system based on digital twins according to claim 3, characterized in that, The digital twin data management module includes data maps, geological 3D models, hydrological information models, airflow field information models, tunnel structure information models, attribute information and analysis calculation models, calculation results data of behavior prediction models, sensor monitoring data, IoT monitoring data, UAV aerial image data, and edge equipment measurement results data. It also enables the association of collected and monitored measurement data with information models to improve information management efficiency and document retrieval efficiency.

8. A digital twin-based immersed tunnel construction system according to claim 2, characterized in that, The service application module, with the help of a large amount of structured data and monitoring data integrated by the digital twin data management module, uses the embedded tunnel safety assessment algorithm to generate a behavior prediction model, conduct a safety assessment of the tunnel's condition, and store the prediction assessment structure in the database as a reference for operation and maintenance management.

9. A digital twin-based immersed tunnel construction system according to claim 3, characterized in that, The service application module adopts a C / S architecture and combines tunnel structure information model, geological information model, hydrological information model, air flow field information model, Internet of Things, drones and edge equipment to establish an intelligent construction and digital twin system platform for immersed tunnels, thereby improving the level and quality of whole-process control of immersed tunnels.

10. A digital twin-based immersed tunnel construction system according to any one of claims 2 to 8, characterized in that, The behavior prediction model is generated and continuously optimized using a GA-BP neural network algorithm model combined with deep learning.

Images