Intelligent road maintenance method self-matching system

Abstract

The invention relates to the technical field of road maintenance method self-matching, in particular to an intelligent road maintenance method self-matching system. According to the technical scheme, the intelligent road maintenance method self-matching system comprises a multivariate heterogeneous data fusion module, a dynamic damage evolution simulation module, an intelligent maintenance decision center module, a self-adaptive maintenance execution module, a user cognition enhancement module and a self-evolution learning module. According to the scheme, climate, traffic and material characteristic data are fused through multiple devices such as satellite remote sensing, a temperature and humidity sensor and a radar spectrum analyzer, a high-precision digital twinborn model is constructed, real-time dynamic updating of the roadbed state is achieved, metric climate data are automatically converted into British units, the international model standard is compatible, and real-time dynamic updating of the roadbed state is achieved. And the data utilization rate and the model generalization ability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of self - matching of road maintenance methods, and particularly to an intelligent road maintenance method self - matching system. Background Art

[0002] In 2004, AASHTO and the National Cooperative Highway Research Program (NCHRP) in the United States jointly launched the Mechanistic - Empirical Pavement Design Guide (MEPDG). This is a design method that combines a large amount of actual road performance observation data with mechanical methods, which is a theoretical analysis and experience - combined design method to evaluate the influence of factors such as materials, structures, and environments on road performance and lifespan.

[0003] Existing road maintenance systems usually rely on a single data source, with limited data dimensions, making it difficult to construct a comprehensive and real - time roadbed model, resulting in lagging and inaccurate maintenance decisions. At the same time, static damage models are mostly used, without considering the coupling effect of multiple environmental factors, and there is a large difference between the prediction results and the actual damage evolution.

[0004] Aiming at the problems of existing road maintenance systems that usually rely on a single data source, have limited data dimensions, are difficult to construct a comprehensive and real - time roadbed model, resulting in lagging and inaccurate maintenance decisions, this solution integrates climate, traffic, and material property data through multiple devices such as satellite remote sensing, temperature - humidity sensors, and radar spectrometers to construct a high - precision digital twin model, realizing real - time dynamic update of the roadbed state. At the same time, it automatically converts metric climate data into imperial units, is compatible with international model standards, and improves data utilization and model generalization ability. Summary of the Invention

[0005] In order to overcome the problems of the traditional road maintenance method self - matching system, where a single data source leads to insufficient model accuracy, unable to comprehensively reflect the roadbed state, without considering the coupling effect of multiple environmental factors, resulting in damage prediction deviating from reality, and at the same time, the maintenance plan depends on experience, lacks personalization and multi - objective optimization, and cannot be dynamically adjusted according to the real - time environment.

[0006] The technical solution of the present invention is: an intelligent road maintenance method self - matching system, including the following modules: Multi - heterogeneous data fusion module: used to integrate multi - dimensional data and construct a high - precision digital twin roadbed model; Dynamic damage evolution simulation module: used to realize the full - life - cycle prediction of pavement performance based on the improved MEPDG model; Intelligent maintenance decision - making center module: used to generate a personalized maintenance plan based on the prediction results; Adaptive maintenance execution module: used for the intelligent conversion of the maintenance plan into construction parameters; User cognition enhancement module: used to construct an immersive decision - making support environment; Self-evolution learning module: used for continuously optimizing the system performance.

[0007] Preferably, the multi-source heterogeneous data fusion module includes: A101: Climate perception unit, including satellite remote sensing climate receiver, road structure temperature and humidity sensor array, and EICM climate file converter, for collecting and converting climate data and establishing a dynamic model of the road surface environment; A102: Traffic load analysis unit, including piezoelectric-millimeter wave radar axle recognition system, axle load spectrum conversion algorithm engine, and traffic flow spatio-temporal distribution modeler, for obtaining vehicle axle load spectrum and predicting traffic flow spatio-temporal distribution; A103: Material property detection unit, including mobile road surface radar spectrum analyzer, nanoindentation material tester, and infrared thermal imaging interlayer bonding state monitor, for establishing a material performance database by combining the collected data of multiple instruments.

[0008] Preferably, the dynamic damage evolution simulation module includes: A201: Environment-coupled damage unit, including freeze-thaw cycle-humidity field coupling simulator, ultraviolet aging-temperature gradient superposition analyzer, and dynamic load-water damage synergy algorithm, for analyzing the coupling damage effect of multiple environmental factors on the road surface according to multi-scenario simulations; A202: Multi-physical field coupling solution unit, including finite element-discrete element hybrid calculation framework, GPU-accelerated parallel calculation engine, and multi-scale progressive damage analysis module, for providing a calculation framework and engine for complex simulations; A203: Performance prediction visualization unit, including three-dimensional road surface damage evolution sand table, virtual reality road condition roaming interface, and maintenance benefit dynamic deduction module, for generating a three-dimensional road surface damage evolution sand table, supporting virtual reality road condition roaming, dynamically deducing maintenance benefits, and intuitively displaying prediction results.

[0009] Preferably, when the dynamic damage evolution simulation module is working, it includes the following steps: S101: The platform automatically receives real-time data from climate stations, traffic counting stations, and road surface intelligent inspection systems, including air temperature, rainfall, wind speed, sunshine rate, and relative humidity, as well as vehicle axle load and classified traffic data, and road surface material physical and mechanical properties, structural layer information; S102: Convert data from different sources and formats into a unified standard, and convert the metric units in climate data into the imperial units required by the EICM climate file; S103: Select a preset damage evolution model according to road type and material properties, or customize a model based on historical data; S104: Define damage criteria using the Hashin criterion, including fiber tensile damage, compressive damage, and matrix shear damage. Introduce continuum damage mechanics, establish a damage evolution equation, and set stiffness degradation parameters; S105: Define model parameters in the compiled file and generate an executable file for the material fatigue threshold and environmental sensitivity coefficient through the Fortran compilation language; S106: Use virtual pavement technology to generate a digital twin model consistent with the test section, simulate pavement responses under different climate and traffic load conditions, and calculate the damage accumulation process; S107: Generate a report containing crack propagation and material deterioration prediction results, indicating the location, time, and degree of damage occurrence; S108: Display the simulation results through 3D rendering or heat maps.

[0010] Preferably, the intelligent maintenance decision-making central module includes: A301: Method generation unit, including a material-process matching knowledge graph, a whole-life cost optimization algorithm, and a maintenance timing planner, used to construct a material-process matching knowledge graph, combine the whole-life cost optimization algorithm and the timing planner to generate a preliminary maintenance plan; A302: Multi-objective optimization unit, including an economy-durability balance model, a traffic impact minimization algorithm, and a carbon emission constraint optimizer, used to balance economy, durability, and traffic impact, and optimize the environmental friendliness of the system through the carbon emission constraint optimizer; A303: Method verification unit, including digital twin maintenance effect preview, multi-scheme comparison sand table deduction, and sensitivity analysis module, used to use digital twin to preview maintenance effects, conduct multi-scheme comparison deduction and sensitivity analysis, and verify the feasibility of the plan.

[0011] Preferably, when the intelligent maintenance decision-making central module is working, it includes the following steps: S201: Receive the prediction results of the dynamic damage evolution simulation platform in real time, and combine the maintenance resource database and historical maintenance records; S202: Analyze the correlation between damage and climate, traffic, and material properties through machine learning algorithms to identify key influencing factors; S203: Prioritize maintenance tasks according to the damage severity and traffic impact range, and automatically generate a preliminary plan including maintenance measures, material selection, and equipment allocation; S204: Calculate the total maintenance cost based on the material usage, equipment usage time, and labor cost in the plan; S205: Optimize the plan details according to resource availability and urgency, output a maintenance plan including specific tasks, schedules, and responsible persons, and dock with the adaptive maintenance execution system; S206: Automatically trigger a warning for potential major risks and initiate the emergency response process.

[0012] Preferably, the adaptive maintenance execution module includes: A401: Construction parameter generation unit, including a milling depth intelligent calculation module, a paving speed-temperature collaborative controller, and a compaction process parameter optimizer, used to intelligently calculate the milling depth, paving speed-temperature collaborative parameters, and compaction process parameters, and optimize the construction process; A402: Intelligent equipment collaboration unit, including an unmanned paving vehicle control system interface, an intelligent compaction robot path planner, and a 3D printing road repair equipment control protocol, used to control the unmanned paving vehicle, intelligent compaction robot, and 3D printing repair equipment, and plan the optimal path; A403: Quality monitoring unit, including a mobile road laser texture scanner, an intelligent compaction quality cloud monitoring system, and a maintenance effect AI inspection terminal, used to monitor the maintenance quality in real time through laser texture scanning, cloud monitoring system, and AI inspection terminal, and generate an evaluation report.

[0013] Preferably, when the adaptive maintenance execution module is working, it includes the following steps: S301: Obtain the maintenance plan from the intelligent maintenance decision-making center module, and analyze the task type, execution location, and time requirements; S302: Automatically check the status of on-vehicle equipment, and plan the optimal driving route according to the task location and real-time traffic information; S303: Real-time collect road surface images and material performance data through high-resolution videos and sensors, and compare them with the expected data in the plan; S304: If unforeseen damages are detected, automatically adjust the maintenance measures, and adjust the construction parameters according to the real-time environmental data; S305: Store the images, sensor data, and operation logs during the maintenance process; S306: Compare the road surface performance data before and after maintenance, generate an effect evaluation report, and feedback it to the intelligent maintenance decision-making center module.

[0014] Preferably, the user cognition enhancement module includes: A501: Three-dimensional visualization interaction unit, including a digital twin road holographic projection, a multi-dimensional data fusion dashboard, and a scheme comparison AR sand table, used to provide a digital twin road holographic projection, a multi-dimensional data dashboard, and an AR sand table to support interactive decision-making; A502: Intelligent decision support unit, including an expert system rule base, a real-time Q&A robot, and a scheme deduction digital sand table, used to integrate the expert system rule base, the real-time Q&A robot, and the scheme deduction sand table to assist users in making quick decisions; A503: Prediction and Forecasting Notification Unit, including a multi-channel early warning push system, an emergency response plan library, and a road health warning light system, which is used to push early warnings, emergency response plans, and road health warning lights through multiple channels to improve the user response efficiency.

[0015] Preferably, the self-evolution learning module includes: A601: Model Update Unit, including an online learning module, a transfer learning framework, and a model interpreter, which is used to dynamically update the prediction model through online learning, transfer learning, and the model interpreter; A602: Data Quality Enhancement Unit, including an abnormal data intelligent cleaning system, a missing data generative adversarial network, and a data distribution adaptive calibrator, which is used to clean abnormal data, generate missing data, and calibrate the data distribution to improve data reliability; A603: Performance Evaluation Unit, including a multi-model comparison test platform, a prediction accuracy tracking dashboard, and a user satisfaction feedback analyzer, which is used to compare the performance of multiple models, track the prediction accuracy, analyze user feedback, and form a closed-loop optimization.

[0016] Advantages of the present invention: 1. Compared with the existing road maintenance systems, which usually rely on a single data source with limited data dimensions and are difficult to build a comprehensive and real-time roadbed model, resulting in lagging and inaccurate maintenance decisions. This solution integrates climate, traffic, and material property data from multiple devices such as satellite remote sensing, temperature and humidity sensors, and radar spectrometers to build a high-precision digital twin model, realizing real-time dynamic updates of the roadbed state. At the same time, it automatically converts metric climate data into imperial units to be compatible with international model standards, improving data utilization and model generalization capabilities; 2. Compared with the existing road maintenance systems, which mostly adopt static damage models and do not consider the coupling effects of multiple environmental factors, resulting in significant differences between the prediction results and the actual damage evolution. This solution combines algorithms such as freeze-thaw cycle - humidity field coupling and ultraviolet aging - temperature gradient superposition to simulate the pavement damage process under complex environments, improving the prediction accuracy. At the same time, it generates a digital twin model consistent with the test section to support response simulations under different climate and load conditions, realizing full-life cycle dynamic prediction; 3. Compared with the existing road maintenance systems, the construction parameters are more dependent on preset rules and cannot be dynamically adjusted according to the real-time environment, affecting the maintenance effect. This solution optimizes the construction process through modules such as milling depth calculation and paving speed - temperature coordinated control. At the same time, it monitors the maintenance quality in real time, compares the data before and after maintenance, automatically adjusts measures, and generates an evaluation report to form a closed-loop optimization. Description of the Drawings

[0017] Figure 1The figure shows a schematic diagram of the module process of a self-matching system for an intelligent road maintenance method of the present invention; Figure 2 The figure shows a schematic diagram of the working process of the dynamic damage evolution simulation module of a self-matching system for an intelligent road maintenance method of the present invention; Figure 3 The figure shows a schematic diagram of the working process of the intelligent maintenance decision-making center module of a self-matching system for an intelligent road maintenance method of the present invention; Figure 4 The figure shows a schematic diagram of the working process of the adaptive maintenance execution module of a self-matching system for an intelligent road maintenance method of the present invention; Figure 5 The figure shows a schematic diagram of the process of a self-matching system for an intelligent road maintenance method of the present invention. Detailed implementation manners

[0018] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0019] Please refer to Figure 1 , the present invention provides an embodiment: a self-matching system for an intelligent road maintenance method, including the following modules: Multi-source heterogeneous data fusion module: used to integrate multi-dimensional data and construct a high-precision digital twin roadbed model; Dynamic damage evolution simulation module: used to realize the full life cycle prediction of pavement performance based on the improved MEPDG model; Intelligent maintenance decision-making center module: used to generate personalized maintenance plans based on the prediction results; Adaptive maintenance execution module: used for the intelligent conversion of maintenance plans into construction parameters; User cognition enhancement module: used to construct an immersive decision support environment; Self-evolving learning module: used to continuously optimize the system performance.

[0020] Please refer to Figures 2 - 4 , in this embodiment, the multi-source heterogeneous data fusion module includes: A101: Climate perception unit, including a satellite remote sensing climate receiver, a road structure temperature and humidity sensor array, and an EICM climate file converter, used to collect and convert climate data and establish a dynamic pavement environment model; A102: Traffic load analysis unit, including a piezoelectric-millimeter wave radar axle identification system, an axle load spectrum conversion algorithm engine, and a traffic flow spatio-temporal distribution modeler, used to obtain the vehicle axle load spectrum and predict the traffic flow spatio-temporal distribution; A103: Material property detection unit, including a mobile pavement radar spectrum analyzer, a nanoindentation material tester, and an infrared thermal imaging interlayer bonding state monitor, used to establish a material performance database by combining the acquisition data of multiple instruments.

[0021] As described above, the multi - heterogeneous data fusion module integrates multi - dimensional data such as climate, traffic, and material properties to construct a high - precision digital twin subgrade model, breaking the limitations of traditional single - data sources, making the model closer to the actual road conditions, and at the same time realizing the unified standardization processing of multi - source heterogeneous data, providing comprehensive and real - time data support for damage prediction and maintenance decision - making, and avoiding decision - making biases caused by information fragmentation.

[0022] Preferably, the dynamic damage evolution simulation module includes: A201: Environmental coupling damage unit, including a freeze - thaw cycle - humidity field coupling simulator, an ultraviolet aging - temperature gradient superposition analyzer, and a dynamic load - water damage synergistic action algorithm, used to analyze the coupling damage effect of multi - environmental factors on the road surface according to multi - scenario simulations; A202: Multi - physical - field coupling solution unit, including a finite element - discrete element hybrid calculation framework, a GPU - accelerated parallel calculation engine, and a multi - scale progressive damage analysis module, used to provide a calculation framework and engine for complex simulations; A203: Performance prediction visualization unit, including a three - dimensional road surface damage evolution sand table, a virtual reality road condition roaming interface, and a maintenance benefit dynamic deduction module, used to generate a three - dimensional road surface damage evolution sand table, support virtual reality road condition roaming, dynamically deduce maintenance benefits, and intuitively display prediction results.

[0023] Preferably, when the dynamic damage evolution simulation module is working, it includes the following steps: S101: The platform automatically receives real - time data from climate stations, traffic counting stations, and road surface intelligent inspection systems, including air temperature, rainfall, wind speed, sunshine rate, and relative humidity, as well as vehicle axle loads and classified traffic data, and road surface material physical and mechanical properties, structural layer information; S102: Convert data from different sources and formats into a unified standard, and convert the metric units in climate data into the imperial units required by the EICM climate file; S103: According to the road type and material properties, select a preset damage evolution model or customize a model based on historical data; S104: Apply the Hashin criterion to define damage criteria, including fiber tensile damage, compressive damage, and matrix shear damage, introduce continuum damage mechanics, establish a damage evolution equation, and set stiffness degradation parameters; S105: Define model parameters in the compiled file, and generate an executable file for the material fatigue threshold and environmental sensitivity coefficient through the Fortran compilation language; S106: Use virtual pavement technology to generate a digital twin model consistent with the test section, simulate pavement responses under different climate and traffic load conditions, and calculate the damage accumulation process; S107: Generate a report containing the prediction results of crack propagation and material deterioration, indicating the location, time, and degree of damage occurrence; S108: Display the simulation results through 3D rendering or heat maps.

[0024] As described above, the dynamic damage evolution simulation module is based on the improved MEPDG model, combines multiple environmental factors and real-time dynamic data, simulates the performance evolution of the pavement throughout its life cycle, improves prediction accuracy, and at the same time formulates a maintenance plan in advance by predicting the time, location, and degree of damage such as crack propagation and material deterioration, reduces sudden damage and maintenance costs, and extends the service life of the road.

[0025] Preferably, the intelligent maintenance decision-making central module includes: A301: Method generation unit, including a material-process matching knowledge graph, a whole-life cost optimization algorithm, and a maintenance timing planner, used to construct a material-process matching knowledge graph, combine the whole-life cost optimization algorithm and the timing planner to generate a preliminary maintenance plan; A302: Multi-objective optimization unit, including an economy-durability balance model, a traffic impact minimization algorithm, and a carbon emission constraint optimizer, used to balance economy, durability, and traffic impact, and optimize the environmental friendliness of the system through the carbon emission constraint optimizer; A303: Method verification unit, including digital twin maintenance effect preview, multi-scheme comparison sand table deduction, and sensitivity analysis module, used to use digital twin to preview maintenance effects, conduct multi-scheme comparison deduction and sensitivity analysis, and verify the feasibility of the plan.

[0026] Preferably, when the intelligent maintenance decision-making central module is working, it includes the following steps: S201: Receive the prediction results of the dynamic damage evolution simulation platform in real time, and combine the maintenance resource database and historical maintenance records; S202: Analyze the correlation between damage and climate, traffic, and material performance through machine learning algorithms to identify key influencing factors; S203: Prioritize maintenance tasks according to the severity of damage and the scope of traffic impact, and automatically generate a preliminary plan including maintenance measures, material selection, and equipment allocation; S204: Calculate the total maintenance cost based on the material usage, equipment usage time, and labor cost in the plan; S205: Optimize the details of the plan according to resource availability and urgency, output a maintenance plan including specific tasks, schedules, and responsible persons, and interface with the adaptive maintenance execution system; S206: Automatically trigger a warning for potential major risks and initiate an emergency response process.

[0027] As described above, the intelligent maintenance decision-making center module generates a personalized maintenance plan that takes into account economy, durability, and environmental protection based on the damage prediction results, in combination with the material-process matching knowledge graph and the whole-life cost optimization algorithm. At the same time, it analyzes the correlation between damage and multiple factors through machine learning, identifies key influencing factors, prioritizes maintenance tasks, optimizes the allocation of materials, equipment, and manpower, and reduces resource waste and traffic congestion.

[0028] Preferably, the adaptive maintenance execution module includes: A401: A construction parameter generation unit, including a milling depth intelligent calculation module, a paving speed-temperature collaborative controller, and a compaction process parameter optimizer, for intelligently calculating the milling depth, paving speed-temperature collaborative parameters, and compaction process parameters, and optimizing the construction process; A402: An intelligent equipment collaboration unit, including an interface for the control system of an unmanned paving vehicle, a path planner for an intelligent compaction robot, and a control protocol for 3D printing pavement repair equipment, for controlling an unmanned paving vehicle, an intelligent compaction robot, and 3D printing repair equipment, and planning the optimal path; A403: A quality monitoring unit, including a mobile pavement laser texture scanner, an intelligent compaction quality cloud monitoring system, and a maintenance effect AI inspection terminal, for real-time monitoring of maintenance quality and generating an evaluation report through laser texture scanning, a cloud monitoring system, and an AI inspection terminal.

[0029] Preferably, when the adaptive maintenance execution module is working, it includes the following steps: S301: Obtain the maintenance plan from the intelligent maintenance decision-making center module and parse the task type, execution location, and time requirements; S302: Automatically check the status of on-vehicle equipment and plan the optimal driving route according to the task location and real-time traffic information; S303: Real-time collect pavement images and material performance data through high-resolution videos and sensors, and compare them with the expected data in the plan; S304: If unforeseen damage is detected, automatically adjust the maintenance measures and adjust the construction parameters according to the real-time environmental data; S305: Store the images, sensor data, and operation logs during the maintenance process; S306: Compare the pavement performance data before and after maintenance, generate an effect evaluation report, and feedback it to the intelligent maintenance decision-making center module.

[0030] As described above, the adaptive maintenance execution module automatically converts the maintenance plan into specific construction parameters, supports the collaborative operation of unmanned equipment and intelligent robots, improves construction efficiency and quality, and at the same time automatically adjusts maintenance measures and construction parameters by real-time monitoring of the road surface conditions and environmental changes, adapts to complex and changeable maintenance scenarios, and reduces manual intervention and misoperations.

[0031] Preferably, the user cognition enhancement module includes: A501: Three-dimensional visualization interaction unit, including digital twin road holographic projection, multi-dimensional data fusion dashboard and scheme comparison AR sand table, for providing digital twin road holographic projection, multi-dimensional data dashboard and AR sand table to support interactive decision-making; A502: Intelligent decision support unit, including expert system rule base, real-time Q&A robot and scheme deduction digital sand table, for integrating expert system rule base, real-time Q&A robot and scheme deduction sand table to assist users in making quick decisions; A503: Prediction and notification unit, including multi-channel warning push system, emergency response plan library and road health warning light system, for pushing warnings, emergency response plans and road health warning lights through multiple channels to improve user response efficiency.

[0032] As described above, the user cognition enhancement module helps users intuitively understand complex data and schemes by providing interactive tools such as digital twin holographic projection, multi-dimensional data dashboard and AR sand table, improves decision-making efficiency and accuracy, and at the same time notifies potential major risks in a timely manner through the multi-channel warning push system and emergency response plan library, starts the emergency process, and reduces the impact of road failures on traffic.

[0033] Preferably, the self-evolution learning module includes: A601: Model update unit, including online learning module, transfer learning framework and model interpreter, for dynamically updating the prediction model through online learning, transfer learning and model interpreter; A602: Data quality enhancement unit, including abnormal data intelligent cleaning system, missing data generative adversarial network and data distribution adaptive calibrator, for cleaning abnormal data, generating missing data and calibrating data distribution to improve data reliability; A603: Performance evaluation unit, including multi-model comparison test platform, prediction accuracy tracking dashboard and user satisfaction feedback analyzer, for comparing the performance of multiple models, tracking prediction accuracy, analyzing user feedback, and forming a closed-loop optimization.

[0034] As described above, the self-evolving learning module dynamically updates the prediction model through online learning, transfer learning, and the model interpreter to adapt to new data patterns and new scenario requirements, improving system stability and prediction accuracy. Meanwhile, combined with data quality enhancement and performance evaluation, it forms a closed-loop optimization of "data - model - decision - execution - feedback" to promote the long-term evolution of the system. Embodiment

[0035] Step 1: Select a test section for data acquisition and preprocessing.

[0036] Climate data: The data file comes from the detected climate data of regional climate stations and data from websites such as the China Meteorological Network and the Geographical Remote Sensing Ecological Network Platform, and the data is converted into the imperial units required for the EICM climate file.

[0037] Field survey: Obtain soil samples at different depths of the road slope and conduct water content analysis to judge the soil moisture condition of the road. Through on-site measurement and analysis, generate data calibration parameters, and obtain more accurate data results through computer processing. Introduce a comprehensive climate model, including hourly data of five main climate factors in the location of the road. And through section measurement, provide the temperature and humidity conditions of each relevant layer of the road surface. Corresponding calculations will be carried out during each design analysis period, which will be used to estimate the material properties of the roadbed and pavement structure layers during the entire design life.

[0038] 3. For traffic data, set up traffic counting stations, use electronic devices with piezoelectric and scanning sensors to count and classify vehicles, and analyze relevant highway data stored by the local transportation department to obtain data such as the length of trucks, the number of axles, axle weight, the distance between axles, and the total weight of trucks.

[0039] 4. The system combines the road surface image data with the collected climate, traffic, and road surface material data, uses machine learning and deep learning techniques to identify road surface diseases and structural features through images, and conducts comprehensive analysis in combination with climate, traffic, and material data, etc., to achieve the prediction of road conditions and the formulation of maintenance plans.

[0040] The embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited to the above embodiments, and various changes can be made without departing from the spirit of the present invention within the scope of knowledge possessed by those skilled in the art.

Claims

1. An intelligent road maintenance method self-matching system; characterized in that: It includes the following modules: Multi-source heterogeneous data fusion module: used to integrate multi-dimensional data and build a high-precision digital twin roadbed model; Dynamic damage evolution simulation module: used to realize the full life cycle prediction of pavement performance based on the improved MEPDG model; Intelligent maintenance decision-making center module: used to generate personalized maintenance plans based on the prediction results; Adaptive maintenance execution module: used for the intelligent conversion of maintenance plans to construction parameters; User cognition enhancement module: used to build an immersive decision-making support environment; Self-evolving learning module: used to continuously optimize the system performance.

2. The self-matching system of an intelligent road maintenance method according to claim 1, characterized in that: The multi-source heterogeneous data fusion module includes: A101: Climate perception unit, including satellite remote sensing climate receiver, road structure temperature and humidity sensor array, and EICM climate file converter; A102: Traffic load analysis unit, including piezoelectric-millimeter wave radar axle identification system, axle load spectrum conversion algorithm engine, and traffic flow spatio-temporal distribution modeler; A103: Material property detection unit, including mobile pavement radar spectrometer, nanoindentation material tester, and infrared thermal imaging interlayer bond state monitor.

3. The self-matching system of an intelligent road maintenance method according to claim 1, characterized in that: The dynamic damage evolution simulation module includes: A201: Environment-coupled damage unit, including freeze-thaw cycle-humidity field coupling simulator, ultraviolet aging-temperature gradient superposition analyzer, and dynamic load-water damage synergy algorithm; A202: Multi-physical field coupling solution unit, including finite element-discrete element hybrid calculation framework, GPU-accelerated parallel calculation engine, and multi-scale progressive damage analysis module; A203: Performance prediction visualization unit, including three-dimensional pavement damage evolution sand table, virtual reality road condition roaming interface, and maintenance benefit dynamic deduction module.

4. An intelligent road maintenance method self-matching system according to claim 3, characterized in that: When the dynamic damage evolution simulation module is working, it includes the following steps: S101: The platform automatically receives real-time data from climate stations, traffic counting stations, and pavement intelligent inspection systems, including air temperature, rainfall, wind speed, sunshine rate, and relative humidity, as well as vehicle axle loads and classified traffic data, and pavement material physical and mechanical properties, structural layer information; S102: Convert data from different sources and formats into a unified standard, and convert the metric units in the climate data into the imperial units required by the EICM climate file; S103: According to the road type and material properties, select a preset damage evolution model or customize a model based on historical data; S104: Apply the Hashin criterion to define damage criteria, including fiber tensile damage, compression damage, and matrix shear damage, introduce continuum damage mechanics, establish a damage evolution equation, and set stiffness degradation parameters; S105: Define model parameters in the compiled file, and generate an executable file for the material fatigue threshold and environmental sensitivity coefficient through Fortran compilation language; S106: Use virtual pavement technology to generate a digital twin model consistent with the test section, simulate the pavement response under different climate and traffic load conditions, and calculate the damage accumulation process; S107: Generate a report containing the prediction results of crack propagation and material deterioration, indicating the location, time, and degree of damage occurrence; S108: Display the simulation results through three-dimensional rendering or heat maps.

5. The intelligent road maintenance method self-matching system according to claim 1, wherein: The intelligent maintenance decision-making center module includes: A301: Method generation unit, including a material-process matching knowledge graph, a whole-life cost optimization algorithm, and a maintenance time sequence planner; A302: Multi-objective optimization unit, including an economy-durability balance model, a traffic impact minimization algorithm, and a carbon emission constraint optimizer; A303: Method verification unit, including digital twin maintenance effect preview, multi-scheme comparison sand table deduction, and sensitivity analysis module.

6. An intelligent road maintenance method self-matching system according to claim 5, characterized in that: When the intelligent maintenance decision-making center module is working, it includes the following steps: S201: Receive the prediction results of the dynamic damage evolution simulation platform in real time, and combine with the maintenance resource database and historical maintenance records; S202: Analyze the correlation between damage and climate, traffic, and material properties through machine learning algorithms to identify key influencing factors; S203: According to the damage severity and traffic impact range, prioritize the maintenance tasks, and automatically generate a preliminary plan including maintenance measures, material selection, and equipment allocation; S204: Calculate the total maintenance cost based on the material usage, equipment usage time, and labor cost in the plan; S205: Optimize the plan details according to resource availability and urgency, output a maintenance plan including specific tasks, schedules, and responsible persons, and dock with the adaptive maintenance execution system; S206: Automatically trigger early warnings for potential major risks and initiate the emergency response process.

7. An intelligent road maintenance method self-matching system according to claim 1, characterized in that: The adaptive maintenance execution module includes: A401: Construction parameter generation unit, including a milling depth intelligent calculation module, a paving speed-temperature collaborative controller, and a compaction process parameter optimizer; A402: Intelligent equipment collaboration unit, including an interface for the control system of an unmanned paving vehicle, a path planner for an intelligent compaction robot, and a control protocol for 3D printing pavement repair equipment; A403: Quality monitoring unit, including a mobile pavement laser texture scanner, an intelligent compaction quality cloud monitoring system, and an AI inspection terminal for maintenance effects.

8. An intelligent road maintenance method self-matching system according to claim 7, characterized in that: When the adaptive maintenance execution module is working, it includes the following steps: S301: Obtain the maintenance plan from the intelligent maintenance decision-making center module, and parse the task type, execution location, and time requirements; S302: Automatically check the status of on-vehicle equipment, and plan the optimal driving route according to the task location and real-time traffic information; S303: Real-time collect pavement images and material property data through high-resolution videos and sensors, and compare them with the expected data in the plan; S304: If unforeseen damage is detected, automatically adjust the maintenance measures, and adjust the construction parameters according to real-time environmental data; S305: Store the images, sensor data, and operation logs during the maintenance process; S306: Compare the pavement performance data before and after maintenance, generate an effect evaluation report, and feedback it to the intelligent maintenance decision-making center module.

9. An intelligent road maintenance method self-matching system according to claim 1, characterized in that: The user cognition enhancement module includes: A501: 3D visualization interaction unit, including digital twin road holographic projection, multi-dimensional data fusion dashboard, and scheme comparison AR sand table; A502: Intelligent decision support unit, including an expert system rule base, a real-time Q&A robot, and a scheme deduction digital sand table; A503: Predicted Forecasting Notification Unit, including a multi-channel early warning push system, an emergency response plan library, and a road health warning light system.

10. An intelligent road maintenance method self-matching system according to claim 1, characterized in that: Self-evolution learning module, including: A601: Model update unit, including an online learning module, a transfer learning framework, and a model interpreter; A602: Data quality enhancement unit, including an intelligent abnormal data cleaning system, a missing data generative adversarial network, and a data distribution adaptive calibrator; A603: Performance evaluation unit, including a multi-model comparison test platform, a prediction accuracy tracking dashboard, and a user satisfaction feedback analyzer.

Images