Highway maintenance digital twin application system construction method and system

Abstract

This invention discloses a method and system for constructing a digital twin application system for highway maintenance, relating to the field of digital twin technology for highway maintenance. The method includes the following steps: acquiring a 3D model, perception data, and control data of at least one target highway maintenance object corresponding to a target maintenance unit; constructing a digital twin scenario construction model for highway maintenance containing a digital twin model library and a data analysis model library, establishing the association between maintenance units and digital twins, and between data analysis models and digital twins; determining the target twin and target data analysis model based on the association and the aforementioned data; and obtaining a digital twin application system for highway maintenance management based on the target twin and target data analysis model. This invention improves the scalability and reusability of the digital twin application system for highway maintenance; achieves precise maintenance management through data association and model analysis, providing reliable technical support for highway maintenance, and demonstrating significant application value.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology for highway maintenance, and in particular to a method and system for constructing a digital twin application system for highway maintenance. Background Technology

[0002] Highway maintenance is a core component of ensuring the safety of transportation infrastructure. With the continuous increase in highway mileage and the gradual increase in service life in my country, problems such as road surface damage, bridge aging, and tunnel defects are becoming increasingly prominent, placing higher demands on the precision and efficiency of maintenance. Digital twin technology, as a comprehensive means integrating advanced technologies such as Building Information Modeling (BIM), the Internet of Things, and big data analytics, can construct a digital mapping of physical entities, providing visualized management and intelligent decision support for the entire lifecycle of highway maintenance. It has become an important development direction in the field of highway maintenance.

[0003] However, existing digital twin application systems for highway maintenance still have many limitations. Most systems are custom-developed, requiring separate systems for different highway maintenance objects. Expanding application scenarios necessitates extensive secondary development, resulting in poor system scalability, low reusability, long development cycles, and high costs. The patent application document with publication number CN116227010B, entitled "Method and System for Constructing a Digital Twin Application System for Highway Maintenance," mentions the construction of a digital twin model library and a data analysis model library. By establishing the association between maintenance units and digital twins, and between data analysis models and digital twins, a digital twin scenario construction model for highway maintenance is built. This achieves adaptation to multiple maintenance objects, improves system scalability and reusability, shortens the development cycle, and reduces costs. However, this technology still has significant shortcomings: First, the compatibility between the digital twin and the physical maintenance unit lacks dynamic quantitative assessment, relying solely on fixed relationships for matching. This makes it difficult to address compatibility deviations caused by dynamic factors such as environmental changes and structural aging, affecting the accuracy of digital twin mapping. Second, data fusion remains at the level of format conversion and basic association, failing to achieve deep collaboration among BIM, GIS geospatial data, IoT sensing data, and maintenance business data. Data value mining is insufficient, making it difficult to support refined maintenance decisions. Third, maintenance strategy generation relies on the results of a single analytical model, lacking comprehensive quantitative consideration of multiple factors such as risk level, performance degradation rate, and maintenance cost, resulting in insufficient strategy targeting and economic efficiency. Fourth, the system's iterative optimization mechanism is imperfect. When structural changes such as renovation or repair occur in the maintenance unit, it cannot quickly trigger digital twin reconstruction, limiting long-term applicability. Furthermore, data security and multi-role access control are not adequately considered, and the anomaly warning and emergency response mechanisms are relatively simple, failing to meet the actual needs of complex maintenance scenarios. These problems make it difficult for digital twin application systems to fully realize their effectiveness in actual maintenance work, and there is an urgent need for a more accurate, efficient and adaptive solution for building a digital twin application system for highway maintenance. Summary of the Invention

[0004] The present invention proposes a method and system for constructing a digital twin application system for highway maintenance, in order to solve the problems mentioned in the prior art.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for constructing a digital twin application system for highway maintenance, comprising the following steps: The maintenance objects are divided into layers and multi-source basic data are collected. The target highway is divided into primary maintenance objects according to structural type, and then into secondary maintenance units according to professional attributes. BIM 3D model, GIS geospatial, IoT sensing and maintenance business multi-source data of each unit are collected. The construction of digital twin model library and intelligent analysis model library will establish a twin library that integrates geometric, physical, business and rule models to truly reflect the status of maintenance units and establish a model library for disease identification, performance degradation, risk assessment and maintenance optimization to support maintenance analysis and decision-making. Multi-dimensional data fusion and dynamic mapping of digital twins involve cleaning, format conversion, and spatiotemporal alignment of multi-source data, establishing a spatial mapping relationship between BIM models and GIS data, and updating the status information of digital twins based on real-time sensing data. Twin adaptation optimization and intelligent adaptation analysis: Based on the structural characteristics, environmental conditions and maintenance needs of the maintenance unit, the model parameters and data weights of the twin are dynamically adjusted to establish an adaptation evaluation mechanism between the twin and the actual maintenance unit. The mapping accuracy is verified through multi-dimensional indicators. Intelligent assessment of maintenance status and dynamic strategy generation: It calls on the corresponding model in the intelligent analysis model library to analyze the multi-source data carried by the twin, identify diseases, predict performance, determine risks, and generate maintenance plans with maintenance timing, methods and resource allocation. Application system integration and visualization release: integrate digital twin model library, intelligent analysis model library and data fusion results to build digital twin application system. The system supports two-dimensional and three-dimensional integrated visualization display, provides interactive functions, and is officially released after preview and verification. It supports multi-terminal access and data sharing.

[0006] Furthermore, it also includes a step for quantifying the adaptation of the digital twin, which comprehensively evaluates the degree of adaptation between the digital twin and the physical maintenance unit through multi-dimensional indicators. The specific formula is as follows: in, For the overall adaptation of twins, For geometrically adapted weights, For physical adaptation weights, To adapt weights for business needs, and , For geometric fit, For physical compatibility, To determine the business compatibility, this quantitative calculation is used to assess the twin's compatibility status.

[0007] Furthermore, it also includes a dynamic assessment step for maintenance priorities, which comprehensively determines maintenance priorities based on the risk level, performance degradation rate, and maintenance cost of the maintenance unit. The specific formula is as follows: in, This is the maintenance priority coefficient. The safety risk level of the maintenance unit, For performance degradation rate, Maintenance cost per unit.

[0008] Furthermore, in the multi-source basic data collection step, IoT sensing data is acquired through sensors deployed at key locations in the maintenance unit, while maintenance business data is acquired by integrating the highway maintenance management system, inspection APP, and maintenance record ledger, and format conversion is performed using a unified data standard.

[0009] Furthermore, in the construction steps of the digital twin model library, the geometric model is built based on BIM technology, supporting the import and export of mainstream formats. The physical model is established using the finite element analysis method, simulating the stress and deformation characteristics of the entity through material mechanics parameters and structural mechanics equations. The business model is associated with maintenance plan formulation, inspection task allocation, and disease treatment process. The rule model has built-in disease judgment thresholds, performance evaluation standards, and maintenance cycle specifications, which are dynamically adjusted according to actual application scenarios.

[0010] Furthermore, in the multi-dimensional data fusion step, data cleaning uses outlier detection algorithms to remove invalid data caused by sensor malfunctions and transmission interference. Spatiotemporal alignment achieves multi-source data matching through timestamp synchronization and spatial coordinate transformation. Semantic association algorithms establish the association relationship between BIM models, GIS data and sensing data based on the unique code of maintenance units.

[0011] Furthermore, the application system integration steps also include a system iteration and optimization mechanism, real-time collection of user operation feedback, maintenance effect data and environmental change information, regular updates to the digital twin model library and intelligent analysis model library, optimization of data fusion algorithms and visualization display effects, and triggering the twin reconstruction process to update model parameters and related data when structural changes occur in the maintenance unit.

[0012] Furthermore, it includes the following modules: The maintenance object segmentation and data acquisition module is used to segment the target highway into multi-level maintenance objects and maintenance units, and to collect BIM 3D models, GIS geospatial data, IoT sensing data and maintenance business data. It supports the access, format conversion and preliminary verification of multi-source data. The twin and analysis model library construction module is used to build a multi-dimensional digital twin model library containing geometric, physical, business, and rule models, as well as an intelligent analysis model library covering disease identification, performance degradation, risk assessment, and maintenance optimization, and to establish the correlation mapping relationship between models; The multi-source data fusion and twin mapping module uses data cleaning, spatiotemporal alignment, and semantic association technologies to process multi-source data and drive real-time synchronization between the digital twin and the physical maintenance unit. The adaptation optimization and intelligent analysis module is used to calculate the overall adaptation of the twin, dynamically adjust the model parameters and data weights, call the intelligent analysis model to evaluate the status of the maintenance unit, and generate performance trend prediction and safety risk assessment results. The maintenance strategy generation and decision support module generates targeted maintenance strategies based on intelligent analysis results, and provides functions for simulation and comparative analysis of maintenance plans to support maintenance decision-making. The visualization integration and publishing module integrates various models and data to build a two-dimensional and three-dimensional integrated visualization interactive interface, enabling intuitive display of maintenance unit status, disease information, and maintenance strategies, and supporting system preview, verification, and multi-terminal publishing; The system iteration and maintenance module collects user feedback and actual maintenance data, regularly updates the model library and algorithms, triggers the twin reconstruction process, and ensures that the system continuously adapts to the dynamic changes in highway maintenance needs.

[0013] Furthermore, it also includes a data security and access control module, which uses encryption algorithms to encrypt and store and transmit collected multi-source data, model files and maintenance business data, sets multi-level user permissions, distinguishes the operation permissions of different roles, the administrator is responsible for system configuration and model updates, maintenance personnel can view the status of maintenance units in their assigned areas and perform inspection tasks, and decision-makers can obtain analysis reports and maintenance strategy suggestions.

[0014] Furthermore, it also includes an anomaly warning and emergency response module, which monitors the status data and intelligent analysis results of the digital twin in real time. When the level of damage exceeds the threshold, the performance degradation rate is abnormal, or the safety risk level is too high, it automatically triggers a graded warning, notifies relevant personnel, and links to the emergency maintenance plan to provide guidance on emergency response procedures and resource scheduling suggestions.

[0015] Compared with existing technologies, the beneficial effects of this invention are: First, the construction of multi-dimensional digital twins improves the accuracy of physical entity representation. The digital twin of this invention integrates geometric models, physical models, business models, and rule models. It not only restores the physical shape and size parameters of the maintenance unit, but also simulates its mechanical properties and environmental response laws, and associates maintenance processes with state judgment criteria. Compared with existing single-dimensional digital twins, it can more comprehensively and realistically map the state and behavior of physical entities, providing a reliable data foundation for maintenance analysis and simulation, and effectively avoiding decision-making biases caused by incomplete representation.

[0016] Secondly, dynamic adaptation and deep data fusion enhance system adaptability and data value. By quantitatively calculating the geometric, physical, and business compatibility between the digital twin and physical units, and dynamically adjusting model parameters and data weights, the system can accurately respond to the impact of dynamic factors such as environmental changes and structural aging. This solves the adaptation deviation problem caused by fixed relationships in existing technologies, ensuring the long-term accuracy of digital twin mapping. Simultaneously, through data cleaning, spatiotemporal alignment, and semantic association technologies, deep collaboration between BIM, GIS, IoT sensing, and business data is achieved, breaking down data barriers and fully exploring the correlation value of multi-source data. This provides more comprehensive data support for disease identification, performance prediction, and risk assessment, facilitating refined maintenance decisions.

[0017] Furthermore, quantifying maintenance priorities and improving decision-making mechanisms enhance maintenance efficiency. This invention comprehensively considers the safety risk level, performance degradation rate, and maintenance cost of maintenance units, determining maintenance priorities through quantitative calculations to avoid blind maintenance, achieve optimal allocation of limited maintenance resources, and improve the pertinence and economy of maintenance work. During the maintenance strategy generation process, the results of multiple models, including defect identification, performance degradation, and risk assessment, are integrated. Combined with maintenance resource constraints and actual needs, customized maintenance plans are generated, supporting plan simulation and comparative analysis. This provides a scientific basis for maintenance decisions, significantly reducing maintenance costs and extending the service life of highways.

[0018] Furthermore, system iteration and safety emergency mechanisms ensure long-term applicability and operational safety. A robust system iteration and optimization mechanism can collect user feedback and actual maintenance data in real time, regularly update the model library and algorithms, and quickly trigger twin reconstruction when structural changes occur in maintenance units, ensuring the system continuously adapts to dynamically changing maintenance needs. Data security and access control modules ensure data transmission and storage security and operational compliance. Anomaly warning and emergency response mechanisms can quickly identify sudden defects, and through tiered warnings and emergency plan linkage, provide process guidance and resource scheduling suggestions for handling emergencies, effectively improving highway operational safety and emergency response efficiency. Multi-terminal visual interaction functions enhance system usability, support multi-role collaborative work, comprehensively improve the efficiency and level of maintenance management, and have broad application prospects. Attached Figure Description

[0019] Figure 1 This is a schematic block diagram of the method for constructing a digital twin application system for highway maintenance proposed in this invention; Figure 2 This is a schematic block diagram of the construction system for the digital twin application system for highway maintenance proposed in this invention; Figure 3 A schematic diagram illustrating the construction process of a digital twin application system for highway maintenance; Figure 4 A bar chart comparing the system construction cycles for different maintenance objects; Figure 5 A line graph showing the change in system reuse rate over the number of applications. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0022] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly; for example, they may refer to a fixed connection, a detachable connection, or an integral connection; they may refer to a mechanical connection or an electrical connection; they may refer to a direct connection or an indirect connection through an intermediate medium; and they may refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The invention will now be described in further detail with reference to the accompanying drawings.

[0023] Reference Figures 1 to 5 A method for constructing a digital twin application system for highway maintenance includes the following steps: The maintenance objects are divided into layers and multi-source basic data are collected. The target highway is divided into primary maintenance objects such as pavement, bridge, and tunnel according to structural type, and then into secondary maintenance units such as roadbed, abutment, and lining according to professional attributes. BIM 3D model, GIS geospatial data, IoT sensing data and maintenance business data are collected for each maintenance unit. The BIM 3D model includes geometric dimensions, material properties and structural relationships. The GIS data covers road network topology and topographic information. The IoT sensing data includes real-time monitoring data such as temperature, humidity, displacement and stress. The maintenance business data includes historical inspection records, maintenance files and disease treatment results. The digital twin model library and intelligent analysis model library are constructed. The digital twin model library contains multi-dimensional twins corresponding to each type of maintenance unit. Each twin integrates a geometric model, a physical model, a business model, and a rule model. The geometric model restores the physical form of the entity, the physical model simulates mechanical properties and environmental response, the business model is associated with the maintenance process, and the rule model defines the state judgment criteria. The intelligent analysis model library contains a disease identification model, a performance degradation model, a risk assessment model, and a maintenance optimization model, which are used for data feature extraction, performance trend prediction, safety risk assessment, and maintenance plan generation, respectively. Multi-dimensional data fusion and dynamic mapping of digital twins are achieved by using data cleaning, format conversion, and spatiotemporal alignment technologies to process multi-source data. A spatial mapping relationship between BIM models and GIS data is established through semantic association algorithms. Based on real-time perception data, the status update of digital twins is driven to achieve real-time synchronization between physical maintenance units and digital twins. The synchronization dimensions include geometric status, physical parameters, operating status, and business processes. Twin adaptation optimization and intelligent adaptation analysis dynamically adjust the model parameters and data weights of the twin based on the structural characteristics, environmental conditions and maintenance needs of the maintenance unit, establish an adaptation evaluation mechanism between the twin and the actual maintenance unit, and verify the mapping accuracy through multi-dimensional indicators to ensure that the digital twin can accurately represent the state and behavior of the physical entity. The system intelligently assesses the maintenance status and generates dynamic strategies. It calls on the corresponding models in the intelligent analysis model library to conduct in-depth analysis of the multi-source data carried by the twin. The disease identification model identifies the types and levels of diseases such as cracks and settlement based on image data and perception data. The performance degradation model predicts the service life and performance change trend of the maintenance unit. The risk assessment model determines the safety risk level by combining environmental factors and structural status. Based on the analysis results, it generates targeted maintenance strategies, including maintenance timing, maintenance methods and resource allocation schemes. The application system is integrated and visualized for release. It integrates the digital twin model library, intelligent analysis model library and data fusion results to build a digital twin application system with functions such as status monitoring, disease early warning, maintenance planning and simulation. The system supports two-dimensional and three-dimensional integrated visualization display and provides interactive functions such as maintenance unit status query, disease tracing and strategy simulation. After previewing and verification, it is officially released and supports multi-terminal access and data sharing.

[0024] This invention also includes a twin adaptation quantification calculation step, which comprehensively evaluates the adaptation degree between the digital twin and the physical maintenance unit through multi-dimensional indicators. The specific formula is as follows: The overall fitness of the twins ranges from 0 to 1. For geometrically adapted weights, For physical adaptation weights, To adapt weights for business needs, and , Geometric fit is calculated based on the dimensional deviations and morphological similarity between the twin and the real entity. For physical fit, the deviation between simulated and measured values ​​is determined based on mechanical properties and environmental response. To assess business adaptability, the system evaluates the compatibility between the maintenance process associated with the twin and the actual business. This quantitative calculation enables accurate determination of the twin's adaptability status, providing data support for model optimization.

[0025] This invention also includes a dynamic evaluation step for maintenance priorities, which comprehensively determines maintenance priorities based on the risk level, performance degradation rate, and maintenance cost of the maintenance unit. The specific formula is as follows: in This is a maintenance priority coefficient; the higher the value, the higher the priority. The safety risk level of the maintenance unit is rated from 1 to 5. The performance degradation rate, expressed as a percentage per year, is calculated using a performance degradation model. The unit maintenance cost is calculated in ten thousand yuan. This calculation comprehensively considers the urgency of risk, the rate of performance deterioration, and economic costs to optimize the allocation of maintenance resources and improve the pertinence and efficiency of maintenance work.

[0026] In this invention, during the multi-source basic data acquisition step, IoT sensing data is acquired through sensors deployed at key locations within the maintenance unit. The temperature sensor has a range of -40℃ to 85℃ and an accuracy of ±0.5℃; the displacement sensor has a measurement range of 0-50mm and an accuracy of ±0.01mm; and the stress sensor has a range of 0-200MPa. The data acquisition frequency is 1-10Hz. Maintenance business data is acquired by integrating the highway maintenance management system, inspection APP, and maintenance record ledger. A unified data standard is used for format conversion to ensure data consistency and integrity.

[0027] In this invention, the digital twin model library construction steps involve building a geometric model based on BIM technology, supporting import and export of mainstream formats such as IFC and RVT, establishing a physical model using finite element analysis, simulating the stress and deformation characteristics of the entity through material mechanics parameters and structural mechanics equations, and linking business processes such as maintenance plan formulation, inspection task allocation, and disease treatment procedures. The rule model has built-in disease judgment thresholds, performance evaluation standards, and maintenance cycle specifications, which can be dynamically adjusted according to actual application scenarios.

[0028] In this invention, during the multi-dimensional data fusion step, the data cleaning uses an outlier detection algorithm to remove invalid data caused by sensor malfunctions and transmission interference. Spatiotemporal alignment achieves accurate matching of multi-source data through timestamp synchronization and spatial coordinate transformation. The semantic association algorithm establishes the association relationship between BIM model, GIS data and sensing data based on the unique code of the maintenance unit, ensuring that the data is organized in an orderly manner and accessed efficiently in the digital twin.

[0029] In this invention, the application system integration step also includes a system iterative optimization mechanism, which collects user operation feedback, maintenance effect data and environmental change information in real time, updates the digital twin model library and intelligent analysis model library regularly, optimizes data fusion algorithms and visualization display effects, and triggers the twin reconstruction process when the maintenance unit undergoes structural changes such as modification or repair, updates model parameters and related data, and ensures that the application system continuously adapts to actual maintenance needs.

[0030] This invention includes the following modules: The maintenance object segmentation and data acquisition module is used to segment the target highway into multi-level maintenance objects and maintenance units, and to collect BIM 3D models, GIS geospatial data, IoT sensing data and maintenance business data. It supports the access, format conversion and preliminary verification of multi-source data. The twin and analysis model library construction module is used to build a multi-dimensional digital twin model library containing geometric, physical, business, and rule models, as well as an intelligent analysis model library covering disease identification, performance degradation, risk assessment, and maintenance optimization, and to establish the correlation mapping relationship between models; The multi-source data fusion and twin mapping module uses data cleaning, spatiotemporal alignment, and semantic association technologies to process multi-source data, driving real-time synchronization between the digital twin and the physical maintenance unit, and ensuring that the twin accurately represents the physical state. The adaptation optimization and intelligent analysis module is used to calculate the overall adaptation of the twin, dynamically adjust the model parameters and data weights, call the intelligent analysis model to evaluate the status of the maintenance unit, and generate performance trend prediction and safety risk assessment results. The maintenance strategy generation and decision support module generates targeted maintenance strategies based on intelligent analysis results and combined with constraints such as maintenance resources and cost budgets. It provides functions for simulation and comparative analysis of maintenance plans to support maintenance decision-making. The visualization integration and publishing module integrates various models and data to build a two-dimensional and three-dimensional integrated visualization interactive interface, enabling intuitive display of maintenance unit status, disease information, and maintenance strategies, and supporting system preview, verification, and multi-terminal publishing; The system iteration and maintenance module collects user feedback and actual maintenance data, regularly updates the model library and algorithms, triggers the twin reconstruction process, and ensures that the system continuously adapts to the dynamic changes in highway maintenance needs.

[0031] This invention also includes a data security and access control module, which uses encryption algorithms to encrypt and store the collected multi-source data, model files, and maintenance business data. It sets up multi-level user permissions to distinguish the operation permissions of roles such as administrators, maintenance personnel, and decision-makers. Administrators are responsible for system configuration and model updates, maintenance personnel can view the status of maintenance units in their assigned areas and perform inspection tasks, and decision-makers can obtain analysis reports and maintenance strategy suggestions to ensure data security and operational standards.

[0032] This invention also includes an anomaly warning and emergency response module, which monitors the status data and intelligent analysis results of the digital twin in real time. When the level of damage exceeds the threshold, the rate of performance degradation is abnormal, or the level of safety risk is too high, it automatically triggers a graded warning and notifies relevant personnel through SMS, system pop-ups, etc. At the same time, it links with the emergency maintenance plan, provides guidance on emergency response procedures and resource scheduling suggestions, and improves the emergency response efficiency and handling capabilities of highway maintenance.

[0033] The following two examples further illustrate specific embodiments of the present invention: Example 1: Implementation of Highway Bridge Maintenance Scenarios This embodiment is applied to the maintenance of highway bridges. Given the characteristics of bridges, such as complex structures, harsh service environments, frequent loads, and easily concealed defects, the system needs to achieve accurate digital mapping of each maintenance unit of the bridge, real-time status monitoring, defect early warning, and scientific maintenance decision-making. It is adaptable to different bridge types such as long-span bridges and continuous beam bridges to ensure the safety of bridge operation and service life.

[0034] I. Core Implementation Details Layered Segmentation of Maintenance Objects and Multi-Source Basic Data Acquisition: The target highway bridge is treated as a primary maintenance object, further subdivided into secondary maintenance units based on professional attributes, such as abutments, piers, bearings, bridge deck pavement, and expansion joints. BIM 3D models, GIS geospatial data, IoT sensing data, and maintenance operational data are collected for each maintenance unit. The BIM 3D model is constructed by importing mainstream file formats and includes detailed information such as the geometric dimensions, material properties, reinforcement layout, and connection relationships of each component, with millimeter-level accuracy. The GIS geospatial data covers the road network topology, topography, and surrounding hydrogeological environment information of the bridge, enabling spatial correlation between the bridge and its surrounding environment. IoT sensing data is acquired through sensors deployed at key locations. Temperature sensors are placed at piers and supports, with a data acquisition range of -40℃ to 85℃ and a data acquisition frequency of 5Hz. Displacement sensors are installed at the bridge deck and pier connections, with a measurement range of 0-50mm, monitoring settlement and deformation in real time. Stress sensors are deployed at key sections of the main beam, with a range of 0-200MPa, capturing stress changes under load. Vibration sensors are installed at the bottom of the piers to monitor structural vibration response. Maintenance data integrates historical inspection records, repair files, defect treatment reports, load test data, etc., and is converted and standardized according to a unified standard.

[0035] The construction of digital twin model libraries and intelligent analysis model libraries: The digital twin model library constructs a multi-dimensional twin for each maintenance unit. The abutment twin integrates geometric, physical, business, and rule models. The geometric model restores the abutment's dimensions and structure; the physical model simulates concrete shrinkage and creep, and temperature stress response; the business model links to the abutment inspection process and maintenance records; and the rule model defines the thresholds for judging defects such as settlement and cracks. Twins for other maintenance units such as piers and bearings are constructed using the same logic to ensure comprehensive coverage of structural characteristics and maintenance needs. The intelligent analysis model library includes defect identification models, performance degradation models, risk assessment models, and maintenance optimization models. The defect identification model, based on image and sensor data, identifies the types and levels of defects such as cracks, spalling, and corrosion through feature extraction and pattern matching. The performance degradation model, combining material properties, environmental factors, and load data, predicts the remaining service life of components. The risk assessment model comprehensively considers structural status, environmental risks, and traffic flow to determine the safety risk level. The maintenance optimization model, based on cost-benefit analysis, generates the optimal maintenance plan.

[0036] Multi-dimensional data fusion and dynamic mapping of the digital twin: Data cleaning algorithms are used to remove invalid data caused by sensor failures and transmission interference. Format conversion unifies multi-source data into a standard format. Spatiotemporal alignment technology is employed, using GPS timestamps to achieve time synchronization between the perceived data and the BIM model. Coordinate transformation establishes a spatial mapping relationship between the BIM model and GIS data, ensuring accurate matching of data in both time and space dimensions. Real-time perceived data drives the digital twin's status updates. Temperature data is fed back to the twin's physical model in real time, dynamically simulating the structural temperature field distribution. Displacement and stress data are synchronously updated to the twin's geometric and physical models, enabling visualization of structural deformation and stress states. Maintenance business data is linked to the twin's business model, forming a complete maintenance lifecycle data chain, ensuring real-time synchronization between physical maintenance units and the digital twin.

[0037] Twin Adaptation Optimization and Intelligent Adaptation Analysis: Based on the structural characteristics, environmental conditions, and maintenance requirements of each bridge maintenance unit, the model parameters and data weights of the twin are dynamically adjusted. To address the significant impact of environmental corrosion on bridge piers, the weight of corrosion sensor data is increased; to address the aging characteristics of bearings, the driving effect of temperature and displacement data on the twin's state is strengthened. An adaptation evaluation mechanism between the twin and actual maintenance units is established, verifying the mapping accuracy from three dimensions: geometric, physical, and operational. Geometric adaptation is assessed through dimensional deviations and morphological similarity between the twin and the physical entity; physical adaptation is determined based on the deviation between simulated and measured values ​​of mechanical properties and environmental responses; and operational adaptation is judged based on the fit between the maintenance processes associated with the twin and actual operations, ensuring that the digital twin can accurately represent the state and behavior of the physical entity.

[0038] Intelligent assessment of maintenance status and dynamic strategy generation: The system utilizes corresponding models from the intelligent analysis model library to conduct in-depth analysis of multi-source data on the twin structure. The defect identification model, based on image data and vibration sensor data of the bridge deck pavement, identifies the location, length, and width of cracks. The performance degradation model, combining pier concrete strength data and environmental humidity data, predicts the strength degradation trend over the next five years. The risk assessment model integrates bearing displacement data, traffic load data, and extreme weather warning information to determine the safety risk level of the bearings. Based on the analysis results, targeted maintenance strategies are generated: For bridge deck pavement with minor cracks, local repair plans are developed, specifying repair materials, construction techniques, and maintenance windows. For bearings with higher risk levels, replacement plans are generated, including construction processes, traffic diversion schemes, and resource allocation plans, while providing multiple alternative options for decision-making.

[0039] Application System Integration and Visualization Release: Integrating digital twin model libraries, intelligent analysis model libraries, and data fusion results, a digital twin application system is built, encompassing status monitoring, disease early warning, maintenance planning, and simulation simulation functions. The system supports integrated 2D and 3D visualization, allowing users to view the 3D models, real-time status data, disease information, and maintenance records of each bridge maintenance unit through drag-and-drop, zoom, and rotation operations. Interactive functions such as maintenance unit status query, disease tracing, and strategy simulation are provided to simulate the implementation effects and cost inputs of different maintenance schemes. After previewing and verification by technical personnel to confirm the accuracy of system data and normal functionality, the system is officially released, supporting access from multiple terminals including computers and mobile devices. Maintenance personnel can view bridge status and early warning information in real time via mobile devices, while decision-makers can perform maintenance scheme simulations and approvals via computers, enabling multi-role collaborative work.

[0040] Table 1: Performance Comparison Table for Highway Bridge Maintenance Scenarios Table 1 clearly demonstrates the significant advantages of this invention in highway bridge maintenance. Existing traditional maintenance systems are mostly custom-developed, with long construction cycles, limited mapping accuracy for various bridge maintenance units, reliance on manual inspections for defect identification leading to significant delays, and a focus on experience-based maintenance plan formulation lacking data support. Furthermore, the system has low reusability, requiring redevelopment for different bridge types. This invention, through layered decomposition of maintenance objects, construction of multi-dimensional twins, and an intelligent analysis model library, achieves a digital and intelligent upgrade of bridge maintenance. The system construction cycle is significantly shortened, mapping accuracy reaches the millimeter level, and it can monitor structural status in real time and provide early warnings of defects. Based on multi-source data, it generates scientifically sound maintenance plans, and the system is adaptable to different bridge types with high reusability, effectively improving the efficiency and scientific nature of bridge maintenance while reducing maintenance costs and safety risks.

[0041] Example 2: Implementation of Urban Main Road Surface Maintenance Scenarios This embodiment is applied to the maintenance scenario of urban main roads. In view of the characteristics of urban main roads such as large traffic flow, complex load, diverse types of defects and short maintenance window, the system needs to realize rapid digital modeling of each maintenance unit of the road surface, dynamic status monitoring, accurate defect identification and efficient maintenance decision-making, and adapt to different road surface types such as asphalt pavement and cement concrete pavement to ensure urban traffic efficiency and road service quality.

[0042] I. Core Implementation Details Layered Segmentation of Maintenance Objects and Multi-Source Basic Data Acquisition: The target urban arterial road is treated as the primary maintenance object, and further segmented into secondary maintenance units based on structural hierarchy, including roadbed, base course, surface course, drainage system, and sidewalks. BIM 3D models, GIS geospatial data, IoT sensing data, and maintenance operational data are collected for each maintenance unit. The BIM 3D model is constructed using on-site scanning and modeling software, containing details such as pavement structural layer thickness, material type, drainage pipe layout, and sidewalk paving style, with centimeter-level accuracy. GIS geospatial data covers the road network distribution, intersection layout, surrounding buildings, and underground pipeline distribution information of the arterial road, realizing spatial correlation between the road surface and urban infrastructure. IoT sensing data is acquired through a combination of vehicle-mounted mobile detection equipment and fixed sensors. The vehicle-mounted detection equipment collects information on road surface smoothness, rut depth, and cracks at a speed of 50 km / h, with a data acquisition frequency of 10 Hz. Fixed sensors are deployed at key road sections and drainage outlets to monitor road surface temperature, humidity, and water depth. Groundwater level sensors are deployed below the roadbed to monitor changes in groundwater level. Maintenance business data integrates historical maintenance records, disease statistics reports, traffic flow data, material testing reports, etc., and is incorporated into the system after being standardized.

[0043] The construction of digital twin model libraries and intelligent analysis model libraries: The digital twin model library constructs a multi-dimensional twin for each maintenance unit. The surface layer twin integrates geometric models, physical models, business models, and rule models. The geometric model restores morphological parameters such as pavement smoothness, cross slope, and longitudinal slope; the physical model simulates the fatigue loss and temperature shrinkage characteristics of pavement materials; the business model relates to maintenance processes such as pavement inspection, repair, and repaving; and the rule model defines the criteria for judging defects such as excessive smoothness and excessive crack width. Twins of other maintenance units such as the subgrade and drainage system are constructed according to the same logic, comprehensively covering the pavement structure and functional characteristics. The intelligent analysis model library includes defect identification models, performance degradation models, risk assessment models, and maintenance optimization models. The pavement defect identification model identifies the types, locations, and severity of defects such as cracks, ruts, and potholes based on images and laser data collected by vehicle-mounted detection equipment; the performance degradation model combines pavement material characteristics, traffic flow, and environmental data to predict the trend of pavement performance changes; the risk assessment model comprehensively considers the severity of defects, traffic flow, and weather warnings to determine the pavement safety risk level; and the maintenance optimization model combines maintenance window periods, construction costs, and traffic impacts to generate the optimal maintenance plan.

[0044] Multi-dimensional data fusion and dynamic mapping of the digital twin: Data cleaning techniques are used to remove noise data caused by vibration and lighting from vehicle-mounted inspection equipment. Format conversion unifies BIM model data, GIS data, sensor data, and business data into a standard format. A spatiotemporal alignment algorithm is employed to achieve precise matching between vehicle-mounted inspection data and road surface location based on mileage markers. Timestamp synchronization enables time coordination between fixed sensor data and mobile inspection data. Real-time sensor data drives the digital twin's state updates. Road surface smoothness data is fed back to the digital twin's geometric model in real time, dynamically presenting changes in road surface undulations. Temperature and humidity data drive the physical model to simulate changes in the mechanical properties of road surface materials. Defect data is associated with the business model, forming a complete closed loop of "defect discovery-analysis-processing-recording," achieving real-time synchronization between the physical road surface and the digital twin, and accurately mapping the road surface state.

[0045] Twin Adaptation Optimization and Intelligent Adaptation Analysis: Based on the traffic characteristics, environmental conditions, and maintenance needs of urban arterial roads, the model parameters and data weights of the digital twin are dynamically adjusted. To address the high-temperature softening characteristic of asphalt pavements, the weight of temperature sensor data is increased; to address the cracking tendency of cement concrete pavements, the impact of crack detection data on the twin's state is strengthened. An adaptation evaluation mechanism is established to verify mapping accuracy from three dimensions: geometric, physical, and operational. Geometric adaptation is assessed through the smoothness and cross slope deviation between the twin and the actual pavement; physical adaptation is determined based on the deviation between simulated and measured material performance values; and operational adaptation is judged based on the fit between the maintenance processes associated with the twin and actual operations, ensuring that the digital twin can accurately represent the actual condition and performance of the pavement.

[0046] Intelligent assessment of maintenance status and dynamic strategy generation: The system utilizes corresponding models from the intelligent analysis model library to perform in-depth analysis of multi-source data on the twin. The defect identification model, based on vehicle-mounted detection images, identifies transverse cracks on a section of the main road, classifying their severity as moderate. The performance degradation model, combining traffic flow data and pavement material characteristics, predicts the performance degradation of this section over the next three years. The risk assessment model, considering crack severity, peak-hour traffic flow, and rainfall warnings, determines the safety risk level of this section to be moderate. Based on the analysis results, a targeted maintenance strategy is generated: A maintenance window is selected during periods of lower nighttime traffic flow. A maintenance method combining crack sealing and thin-layer overlay is employed, specifying material selection, construction procedures, traffic diversion plans, and quality acceptance standards. Cost budgets and schedules are also provided to ensure efficient maintenance while minimizing traffic disruption.

[0047] Application System Integration and Visualization Release: Integrating digital twin model libraries, intelligent analysis model libraries, and data fusion results, a digital twin application system is built, encompassing status monitoring, disease early warning, maintenance planning, and simulation simulation functions. The system supports integrated 2D and 3D visualization, intuitively presenting 3D models, real-time status data, disease distribution, and maintenance plans for each maintenance unit along the main road. Interactive functions such as disease query, trajectory tracing, and scheme simulation are provided, simulating the construction process, traffic impact, and cost input of different maintenance schemes. After previewing and verification, the system is officially released, supporting access from multiple stakeholders including urban maintenance management departments, construction units, and regulatory agencies. Maintenance management departments can issue maintenance tasks through the system, construction units can upload construction progress in real time, and regulatory agencies can monitor construction quality online, achieving full-process digital management of maintenance work.

[0048] Table 2: Performance Comparison Table for Urban Main Road Surface Maintenance Scenarios Table 2 data highlights the core advantages of this invention in urban arterial road pavement maintenance scenarios. Existing traditional maintenance systems rely on manual inspections to identify defects, resulting in low accuracy and slow efficiency. Maintenance plan development requires multi-party coordination, is time-consuming, and struggles to accurately match maintenance windows, easily causing significant traffic disruptions. Furthermore, incomplete data traceability hinders subsequent maintenance optimization. This invention, through multi-source data fusion, intelligent analysis models, and digital twin technology, achieves accurate and rapid defect identification and scientifically efficient maintenance plan development, significantly improving the utilization rate of maintenance windows, minimizing the impact on urban traffic, and enabling complete traceability of the entire maintenance process, providing reliable support for subsequent pavement maintenance optimization. This effectively solves the pain points of low efficiency and significant traffic impact in urban arterial road pavement maintenance.

[0049] refer to Figure 4 This diagram visually demonstrates that the present invention boasts a shorter system construction cycle across various maintenance scenarios. Its core advantage stems from the pre-built digital twin model library and data analysis model library. The present invention pre-packages digital twins and data analysis models for various maintenance units into the library. By pre-setting the relationships between maintenance units and twins, and between models and twins, it eliminates the need for repeatedly developing basic modules for different maintenance objects. It only requires access to corresponding multi-source data for rapid adaptation. In contrast, existing customized systems require separate design of models and data association logic for each maintenance object, resulting in a cumbersome development process with much repetitive work, significantly extending the construction cycle. The short-cycle characteristic of the present invention can quickly respond to the needs of different highway maintenance scenarios, lower the system deployment threshold, and accelerate the digital transformation process of highway maintenance.

[0050] refer to Figure 5The figure clearly demonstrates that the system reuse rate of this invention is far higher than that of existing technologies, and it continues to improve and stabilize with the number of applications. The core reason is that the digital twin model library of this invention covers basic unit twins for all scenarios of highway maintenance, and the data analysis model library integrates various general analysis algorithms. Through flexible configuration of relationships, it can quickly adapt to different maintenance objects and application scenarios. Existing systems lack a unified model and data reuse framework, requiring extensive customization and modification for each application, making it difficult to improve the reuse rate. The high reusability of this invention significantly reduces redundant development work, reduces the time and cost of multiple deployments, and is especially suitable for the phased digital transformation of large-scale highway networks, highlighting its value for large-scale applications.

[0051] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for constructing a highway maintenance digital twin application system, characterized in that, Includes the following steps: The maintenance objects are divided into hierarchical segments and multi-source basic data are collected. The target highway is divided into primary maintenance objects according to structural type, and then into secondary maintenance units according to professional attributes. BIM 3D model, GIS geospatial data, IoT sensing and maintenance business multi-source data of each unit are collected. The construction of digital twin model library and intelligent analysis model library will establish a twin library that integrates geometric, physical, business and rule models to truly reflect the status of maintenance units and establish a model library for disease identification, performance degradation, risk assessment and maintenance optimization to support maintenance analysis and decision-making. Multi-dimensional data fusion and dynamic mapping of digital twins involve cleaning, format conversion, and spatiotemporal alignment of multi-source data, establishing a spatial mapping relationship between BIM models and GIS data, and updating the status information of digital twins based on real-time sensing data. Twin adaptation optimization and intelligent adaptation analysis: Based on the structural characteristics, environmental conditions and maintenance needs of the maintenance unit, the model parameters and data weights of the twin are dynamically adjusted to establish an adaptation evaluation mechanism between the twin and the actual maintenance unit. The mapping accuracy is verified through multi-dimensional indicators. Intelligent assessment of maintenance status and dynamic strategy generation: It calls on the corresponding model in the intelligent analysis model library to analyze the multi-source data carried by the twin, identify diseases, predict performance, determine risks, and generate maintenance plans with maintenance timing, methods and resource allocation. Application system integration and visualization release: integrate digital twin model library, intelligent analysis model library and data fusion results to build digital twin application system. The system supports two-dimensional and three-dimensional integrated visualization display, provides interactive functions, and is officially released after preview and verification. It supports multi-terminal access and data sharing.

2. The method of claim 1, wherein, It also includes a step for quantifying the adaptation of the digital twin, which comprehensively evaluates the degree of adaptation between the digital twin and the physical maintenance unit through multi-dimensional indicators. The specific formula is as follows: wherein is a twin synthesis fitness, is a geometric fitness weight, is a physical fitness weight, is a business fitness weight, and , is a geometric fitness, is a physical fitness, is a business fitness, by which the determination of the twin fitness state is achieved.

3. The method of claim 1, wherein, The method further comprises a maintenance priority dynamic evaluation step, which determines the maintenance priority based on the risk level, performance degradation rate and maintenance cost of the maintenance unit, and the specific formula is: wherein, is a maintenance priority coefficient, is a safety risk level of the maintenance unit, is a performance degradation rate, is a unit maintenance cost.

4. The method of claim 1, wherein, In the multi-source basic data collection process, IoT sensing data is acquired through sensors deployed at key locations in the maintenance unit, while maintenance business data is acquired by integrating the highway maintenance management system, inspection APP, and maintenance record ledger, and format conversion is performed using a unified data standard.

5. The method of claim 1, wherein, In the construction steps of the digital twin model library, the geometric model is built based on BIM technology, supporting the import and export of mainstream formats. The physical model is established using the finite element analysis method, simulating the stress and deformation characteristics of the entity through material mechanics parameters and structural mechanics equations. The business model is associated with maintenance plan formulation, inspection task allocation, and disease treatment process. The rule model has built-in disease judgment thresholds, performance evaluation standards, and maintenance cycle specifications, which are dynamically adjusted according to actual application scenarios.

6. The method of claim 1, wherein, In the multi-dimensional data fusion process, data cleaning uses outlier detection algorithms to remove invalid data caused by sensor malfunctions and transmission interference. Spatiotemporal alignment achieves multi-source data matching through timestamp synchronization and spatial coordinate transformation. Semantic association algorithms establish the association between BIM models, GIS data and sensing data based on the unique code of maintenance units.

7. The method of claim 1, wherein, The application system integration steps also include a system iteration and optimization mechanism, real-time collection of user operation feedback, maintenance effect data and environmental change information, regular updates to the digital twin model library and intelligent analysis model library, optimization of data fusion algorithms and visualization effects, and triggering the twin reconstruction process when structural changes occur in the maintenance unit to update model parameters and related data.

8. A system for constructing a digital twin application for highway maintenance according to the method of any one of claims 1-7, characterized in that, Includes the following modules: The maintenance object segmentation and data acquisition module is used to segment the target highway into multi-level maintenance objects and maintenance units, and to collect BIM 3D models, GIS geospatial data, IoT sensing data and maintenance business data. It supports the access, format conversion and preliminary verification of multi-source data. The twin and analysis model library construction module is used to build a multi-dimensional digital twin model library containing geometric, physical, business, and rule models, as well as an intelligent analysis model library covering disease identification, performance degradation, risk assessment, and maintenance optimization, and to establish the correlation mapping relationship between models; The multi-source data fusion and twin mapping module uses data cleaning, spatiotemporal alignment, and semantic association technologies to process multi-source data and drive real-time synchronization between the digital twin and the physical maintenance unit. The adaptation optimization and intelligent analysis module is used to calculate the overall adaptation of the twin, dynamically adjust the model parameters and data weights, call the intelligent analysis model to evaluate the status of the maintenance unit, and generate performance trend prediction and safety risk assessment results. The maintenance strategy generation and decision support module generates targeted maintenance strategies based on intelligent analysis results, and provides functions for simulation and comparative analysis of maintenance plans to support maintenance decision-making. The visualization integration and publishing module integrates various models and data to build a two-dimensional and three-dimensional integrated visualization interactive interface, enabling intuitive display of maintenance unit status, disease information, and maintenance strategies, and supporting system preview, verification, and multi-terminal publishing; The system iteration and maintenance module collects user feedback and actual maintenance data, regularly updates the model library and algorithms, triggers the twin reconstruction process, and ensures that the system continuously adapts to the dynamic changes in highway maintenance needs.

9. The system for building a digital twin application for highway maintenance according to claim 8, wherein, It also includes a data security and access control module, which uses encryption algorithms to encrypt and store the collected multi-source data, model files and maintenance business data, sets multi-level user permissions, distinguishes the operation permissions of different roles, the administrator is responsible for system configuration and model updates, the maintenance personnel can view the status of the maintenance units in their assigned area and perform inspection tasks, and the decision-makers can obtain analysis reports and maintenance strategy suggestions.

10. The system for building a digital twin application system for highway maintenance according to claim 8, wherein, It also includes an anomaly warning and emergency response module, which monitors the status data and intelligent analysis results of the digital twin in real time. When the level of the disease exceeds the threshold, the performance degradation rate is abnormal, or the safety risk level is too high, it automatically triggers a graded warning, notifies relevant personnel, and links to the emergency maintenance plan to provide guidance on emergency response procedures and resource scheduling suggestions.

Images