A smart construction and maintenance device for asphalt pavement in hot and humid areas, ensuring its long service life throughout its entire lifecycle.

By using multi-source data acquisition, risk identification, zoned prevention and control, and digital twin technology, the problem of difficult perception of coupled asphalt pavement diseases in hot and humid areas has been solved, enabling full-cycle collaborative maintenance decision-making and improving pavement durability and service life.

CN123118068APending Publication Date: 2026-11-17POLY CHANGDA ENGINEERING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611309622.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-27
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In hot and humid regions, asphalt pavements are subject to various environmental factors and traffic loads during their service life. As a result, pavement and subgrade defects are coupled together, making it difficult to detect the internal deterioration state in a timely manner. The lack of full-cycle coordination between construction monitoring and maintenance decisions leads to inaccurate judgment of the development trend of defects, untimely maintenance, or repeated repairs.

Method used

The system employs a multi-source data acquisition module to obtain environmental, traffic load, and pavement condition data. It identifies the types and levels of defects through a road segment risk identification module, implements targeted prevention and control measures using a zoned defect prevention and control module, constructs and updates pavement condition data using a digital twin module, determines maintenance timing and measures using a maintenance decision module, and processes and feeds back data using a maintenance execution module, thus forming a full-cycle intelligent construction and maintenance and long-life protection device.

Benefits of technology

It enables proactive perception of the internal deterioration state of asphalt pavement and accurate maintenance decisions, reducing repeated repairs and excessive repair scope, and improving pavement durability and long-term service capability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123118068A_ABST
    Figure CN123118068A_ABST
Patent Text Reader

Abstract

This application discloses a full-cycle intelligent construction and maintenance and long-life guarantee device for asphalt pavement in hot and humid areas, including a multi-source data acquisition module, a road section risk identification module, a zoned disease prevention and control module, a digital twin module, a maintenance decision-making module, and a maintenance execution and feedback module. By collecting environmental, traffic load, construction, and pavement and subgrade condition data, the device determines the type and degree of disease risk for each road section and implements corresponding disease prevention and control measures. It integrates multi-source data to construct and update a digital twin, predicts disease development trends, determines maintenance timing, areas, and measures, controls maintenance execution equipment to carry out maintenance, and feeds back processed condition monitoring data. This forms a full-cycle closed loop of condition perception, risk zoning, disease prevention and control, maintenance decision-making, and feedback updates, improving the accuracy of pavement internal deterioration perception and maintenance decision-making, achieving coordinated prevention and control of pavement and subgrade diseases, and enhancing the durability and long-term service guarantee capability of asphalt pavement.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of road engineering technology, and in particular to an intelligent construction and maintenance and long-life protection device for asphalt pavement in hot and humid areas. Background Technology

[0002] Asphalt pavement is widely used in highways, urban roads, and bridge deck paving projects due to its advantages such as good driving comfort, high construction efficiency, and relatively convenient maintenance. However, in hot and humid regions, asphalt pavement is subjected to the combined effects of high temperature, high humidity, continuous rainfall, and traffic loads during its service life. This is especially true in sections such as river-crossing bridges, transportation hubs, and high embankments, where the working environment of the pavement and subgrade structures is complex, making them prone to water damage, rutting, cracking, interlayer slippage, and uneven settlement. With the increasing traffic volume and the proportion of heavy vehicles, improving the structural durability, operational safety, and maintenance management of asphalt pavement has become an important research topic in the field of road construction and maintenance.

[0003] Current asphalt pavement construction and maintenance technologies typically employ various measures depending on the specific defects, such as waterproofing layers, asphalt material modification, fiber or mesh reinforcement, subgrade drainage, geosynthetic water-blocking, grouting reinforcement, and pavement milling repair. While these measures can mitigate the development of corresponding defects to some extent, they are generally relatively independent, lacking synergistic analysis among pavement materials, pavement structure, subgrade condition, and the external environment. Under continuous rainfall, groundwater level fluctuations, and repeated vehicle loads, moisture may penetrate the structure through pavement pores, cracks, or interlayer interfaces, making it difficult for existing detection methods to promptly determine the internal seepage status and abnormal locations. Under the combined effects of high temperatures and heavy loads, relying solely on a single modified material is insufficient to continuously suppress rutting and interlayer shear deformation. For high-fill sections, the subgrade moisture content and settlement deformation are mutually influential, making it difficult to reflect changes in the internal subgrade condition in a timely manner using single drainage, water-blocking, or reinforcement measures. In addition, existing road monitoring data, construction data and maintenance data are often stored and used separately, lacking a continuous connection across the design, construction, operation and maintenance stages. This leads to a heavy reliance on manual experience in judging the development trend of road defects and determining the timing of maintenance, which can easily result in untimely maintenance, repeated repairs or excessive repair scope.

[0004] Therefore, in the construction and maintenance of asphalt pavements in hot and humid regions, the combined effects of various environmental factors and traffic loads, the coupling of pavement and subgrade defects, the difficulty in timely detection of internal deterioration, and the lack of full-cycle coordination between construction monitoring and maintenance decisions have become urgent problems that need to be solved. Summary of the Invention

[0005] This application provides an intelligent construction and maintenance device for asphalt pavement in hot and humid areas, aiming to solve the problems of multiple environmental factors and traffic loads acting together, pavement and subgrade diseases being coupled, internal deterioration being difficult to detect in a timely manner, and lack of full-cycle coordination between construction monitoring and maintenance decisions in the construction and maintenance of asphalt pavement in hot and humid areas.

[0006] This application provides a device for the intelligent construction and maintenance of asphalt pavements throughout their entire lifecycle and for ensuring their long service life in hot and humid regions. The device includes: The multi-source data acquisition module is used to collect environmental data, traffic load data, construction data, and pavement and subgrade status data of the target asphalt pavement during the construction and operation phases. The road segment risk identification module is used to determine the type and degree of distress risk of each road segment in the target asphalt pavement based on the environmental data, traffic load data, construction data, and pavement and subgrade condition data, and to obtain the road segment risk zoning results; the distress risk types include hydrological erosion risk, mechanical damage risk, and geological environment coupled deterioration risk. The zoned disease prevention and control module includes a hydrological erosion prevention and control unit, a mechanical damage suppression unit, and a geological environment coupled deterioration prevention and control unit. It is used to implement disease prevention and control corresponding to the disease risk type for the corresponding road section based on the road section risk zoning results, and to collect pavement and subgrade condition monitoring data during the disease prevention and control process. The digital twin module is used to integrate the building information model, geographic information data, construction data, and pavement and subgrade condition monitoring data of the target asphalt pavement to construct and update the digital twin corresponding to the target asphalt pavement. The maintenance decision module is used to determine the current service status of each road segment based on the digital twin, predict the development trend of defects in each road segment, and determine the corresponding maintenance timing, maintenance area and maintenance measures based on the development trend of defects and the preset maintenance evaluation target, so as to obtain the maintenance decision result. The maintenance execution and feedback module includes maintenance execution equipment, which is used to control the maintenance execution equipment to carry out maintenance treatment on the corresponding road section according to the maintenance decision results, and to feed back the pavement and subgrade condition monitoring data after maintenance treatment to the digital twin module to update the digital twin and subsequent maintenance decision results.

[0007] Optionally, in the above scheme, the multi-source data acquisition module is used for: Rainfall, temperature and humidity, and water environment data of the area where the target asphalt pavement is located are collected to obtain environmental data; Traffic load data is obtained by collecting vehicle flow data, vehicle axle load data, and vehicle operation data acting on the target asphalt pavement. The construction data is obtained by collecting material preparation data, paving data, and compaction data during the construction process of the target asphalt pavement. By using sensing devices installed in the target asphalt pavement and subgrade, seepage pressure data, pavement temperature data, subgrade moisture content data and displacement data are collected to obtain pavement and subgrade state data; The environmental data, traffic load data, construction data, and road surface and subgrade status data are synchronized in time and correlated in space to obtain a multi-source road segment dataset corresponding to each road segment.

[0008] Optionally, in the above scheme, the road segment risk identification module is used for: Extract road segment state features that characterize rainfall erosion, heavy load, pavement temperature, subgrade humidity, and structural deformation from the multi-source road segment dataset; Based on the road segment condition characteristics, the environmental severity level corresponding to each road segment is determined, and the hydrological erosion risk characterization value, mechanical damage risk characterization value, and geological environment coupled deterioration risk characterization value corresponding to each road segment are determined respectively. Based on the relationship between each risk characterization value and the corresponding risk judgment conditions, the type and degree of road damage risk for each road section are determined. The target asphalt pavement is divided according to the type and degree of the damage risk to obtain the risk zoning results of the road section.

[0009] Optionally, in the above scheme, the hydrological erosion control unit includes a first modified asphalt preparation device, a permeable mixture preparation device, a seepage pressure sensor array, and a seepage state interpretation device; The first modified asphalt preparation equipment is used to prepare modified asphalt containing silane nanocapsules, and the modified asphalt containing silane nanocapsules is used to construct the anti-seepage and anti-corrosion layer of the corresponding road section. The permeable mixture preparation equipment is used to prepare a permeable mixture containing basalt fiber, and the permeable mixture containing basalt fiber is used to construct a permeable drainage layer on one side of the anti-seepage and anti-corrosion layer. The pressure sensor array is used to collect pressure data of the corresponding areas of the anti-seepage and anti-corrosion layer and the permeable drainage layer. The seepage state interpretation device is used to determine the seepage state of the corresponding road section based on the seepage pressure data, and to generate hydrological erosion early warning information when the seepage state meets the preset seepage risk conditions.

[0010] Optionally, in the above scheme, the seepage pressure sensor array includes multiple fiber optic seepage pressure sensors arranged longitudinally and laterally along the target asphalt pavement; The seepage state interpretation device is used to determine the spatial and temporal variation relationships of seepage pressure inside the target asphalt pavement based on the seepage pressure values, acquisition time, and spatial location collected by each fiber optic grating seepage pressure sensor. Based on the spatial and temporal variation relationships, the seepage pressure gradient inside the target asphalt pavement and the direction of change of the seepage pressure gradient are determined. Based on the seepage pressure gradient and the direction of its change, the seepage direction and seepage influence range of the corresponding road section are determined. Based on the seepage direction and the seepage influence range, the abnormal seepage path and the corresponding road segment location are determined, and the seepage path location result is obtained.

[0011] Optionally, in the above scheme, the mechanical damage suppression unit includes a composite aggregate preparation device, a second modified asphalt preparation device, a grid reinforcement layer laying device, an infrared thermal imaging device, and a rut prediction device; The composite aggregate preparation equipment is used to composite steel slag aggregate and graphene material to obtain steel slag-graphene composite aggregate. The second modified asphalt preparation equipment is used to prepare high-entropy alloy modified asphalt, the steel slag-graphene composite aggregate and the high-entropy alloy modified asphalt are used to prepare rutting-resistant asphalt mixture, and the rutting-resistant asphalt mixture is used to construct a rutting-resistant asphalt layer. The mesh reinforcement layer laying equipment is used to lay a three-dimensional fiberglass mesh reinforced bonding layer between adjacent pavement structural layers; The infrared thermal imaging device is used to collect road surface temperature field data for the corresponding road section; The rutting prediction device is used to predict the rutting development trend of a corresponding road segment based on the road surface temperature field data, the traffic load data, and historical rutting data, and to determine the timing of rutting maintenance for the corresponding road segment based on the rutting development trend.

[0012] Optionally, in the above scheme, the geological environment coupled deterioration prevention and control unit includes roadbed construction equipment, displacement sensor array, soil moisture sensor and grouting equipment; The roadbed construction equipment is used to form a composite roadbed structure including a permeable drainage layer, a capillary water blocking layer and a lightweight load-bearing layer, wherein the capillary water blocking layer is disposed between the permeable drainage layer and the lightweight load-bearing layer; The displacement sensor array is used to collect displacement data of the composite roadbed structure, and the soil moisture sensor is used to collect moisture data of the corresponding area of ​​the composite roadbed structure. Based on the displacement data and the moisture content data, determine the coupling state between the changes in roadbed moisture and settlement. When the settlement amount within the preset statistical period is greater than the preset settlement threshold, and / or the change in the moisture content data relative to the historical moisture content data is greater than the preset moisture content change threshold, it is determined that the coupling state meets the preset settlement early warning condition. Based on the spatial positions of the displacement sensor and soil moisture sensor that meet the preset settlement early warning conditions, the area of ​​the roadbed to be reinforced is determined, and the grouting equipment is controlled to perform grouting reinforcement on the area of ​​the roadbed to be reinforced.

[0013] Optionally, in the above scheme, the digital twin module is used for: A building information model of the target asphalt pavement is established, and the building information model is spatially registered with the geographic information data to obtain a basic digital model including the road segment location, pavement structure layer and subgrade structure layer. The construction data and the pavement and subgrade condition monitoring data are mapped to the basic digital model according to the collection time and collection location to obtain the construction status parameters and service status parameters corresponding to each road section. The service status parameters are updated based on the newly collected pavement and subgrade condition monitoring data; Based on the updated service status parameters, the state changes of each road segment under different environmental conditions and traffic load conditions are deduced to obtain the digital twin corresponding to the target asphalt pavement.

[0014] Optionally, in the above scheme, the maintenance decision module includes a federated learning unit and a multi-objective optimization unit; The federated learning unit is used to store asphalt pavement distress data from different regions in their respective local nodes, control each local node to train the model based on the local asphalt pavement distress data, and aggregate the model update parameters uploaded by each local node to obtain a distress development prediction model. Input the service status parameters corresponding to the digital twin into the disease development prediction model to obtain the disease development prediction results for each road section within a preset time range; The multi-objective optimization unit is used to generate multiple candidate maintenance schemes based on the disease development prediction results, and evaluates the multiple candidate maintenance schemes with the optimization objectives of minimizing maintenance costs and carbon emissions and maximizing service life, to obtain the maintenance decision results.

[0015] Optionally, in the above scheme, the maintenance execution equipment includes a recycled binder preparation equipment and a microwave activation and regeneration equipment; The recycled cementitious material preparation equipment is used to prepare geopolymer-based recycled cementitious materials using manganese slag. The microwave activation and regeneration equipment is used to activate old asphalt pavement material generated during the maintenance process by microwave heating, so as to obtain activated old material. Based on the maintenance decision results, at least one of the geological polymer-based recycled binder and the activated old material is selected for maintenance treatment of the corresponding road section; Obtain pavement performance data and material utilization data after maintenance treatment, and determine the corresponding carbon emission data based on the material utilization data and equipment energy consumption data during the maintenance treatment process; The pavement performance data, material utilization data, and carbon emission data are fed back to the digital twin module and the maintenance decision module to update the digital twin and subsequent maintenance decision results.

[0016] Compared with the prior art, this application has at least the following beneficial effects: This application, based on further analysis and research of existing technical problems, recognizes that existing technologies in the construction and maintenance of asphalt pavements in hot and humid regions suffer from multiple environmental factors and traffic loads acting together, coupled pavement and subgrade diseases, difficulty in timely detection of internal deterioration, and a lack of full-cycle coordination between construction monitoring and maintenance decisions. By collecting environmental data, traffic load data, construction data, and pavement and subgrade condition data of the target asphalt pavement during the construction and operation phases, the external effects, construction quality, and internal service status of the pavement can be uniformly perceived. Based on this, the application determines the type and degree of disease risk for each road segment and forms a risk zoning result. This allows for the separate identification of hydrological erosion risks, mechanical damage risks, and geological environment coupled deterioration risks faced by different road segments. Subsequently, the zoned disease prevention and control module implements disease prevention and control measures tailored to the specific road segment's risk type, avoiding insufficient prevention and control caused by using independent or uniform treatment methods for different diseases. Simultaneously... By integrating Building Information Modeling (BIM), geographic information data, construction data, and pavement and subgrade condition monitoring data to construct and continuously update a digital twin, data scattered across the construction, operation, and maintenance phases can be linked to corresponding road sections and their service status. This allows for the determination of the current service status of each road section and the prediction of disease development trends. Furthermore, based on disease development trends and pre-set maintenance evaluation objectives, maintenance timing, areas, and measures are determined, enabling maintenance to shift from passive repair after disease occurrence to proactive decision-making oriented towards the disease development process. Finally, by controlling maintenance execution equipment to implement maintenance treatments and feeding back pavement and subgrade condition monitoring data after maintenance to the digital twin module, the digital twin and subsequent maintenance decision results are updated. This improves the perception of internal deterioration states of asphalt pavements in hot and humid regions and the accuracy of maintenance decisions, achieving coordinated prevention and control of pavement and subgrade diseases, reducing untimely maintenance, repeated repairs, and excessive repair scope, and improving the durability and long-term service guarantee capability of asphalt pavements. Attached Figure Description

[0017] Figure 1 This is a block diagram of the modular architecture of an intelligent construction and maintenance and long-life protection device for asphalt pavement in hot and humid areas, provided in one embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] In one embodiment, such as Figure 1 As shown, a smart construction and maintenance device for asphalt pavement in hot and humid areas with a long service life is provided. It includes a multi-source data acquisition module, a road section risk identification module, a zoned disease prevention and control module, a digital twin module, a maintenance decision-making module, and a maintenance execution and feedback module. These modules can interact via wired communication networks, wireless communication networks, or a construction site data bus. Each prevention and control unit in the zoned disease prevention and control module is connected to corresponding sensing devices, material preparation equipment, construction equipment, and maintenance execution equipment to form a closed-loop construction and maintenance system covering the construction, operation, and maintenance phases.

[0020] The target asphalt pavement can be used for highways, ordinary roads, urban roads, or bridge decks in hot and humid regions. The target asphalt pavement can be divided into multiple sections according to station number, administrative division, structural form, surrounding hydrological conditions, traffic load level, and subgrade type. Different sections may have different major risks of road damage. For example, bridge sections crossing rivers are susceptible to continuous rainfall, dynamic water pressure, and chloride erosion; traffic hub sections are susceptible to high temperatures, heavy loads, vehicle braking, and steering; and high embankment sections are susceptible to rainfall infiltration, capillary migration, and embankment deformation.

[0021] In the early stages of construction, a multi-source data acquisition module collects meteorological, hydrological, traffic, geological, and existing road distress data for the target asphalt pavement area. The road segment risk identification module analyzes the formation conditions of water damage, rutting, and settlement under the combined effects of moisture and heat based on the collected data, determining the main distress risk types, severity levels, and key control measures for each road segment. In one optional implementation, the distress risk level can be divided into four levels according to environmental severity; the higher the environmental severity, the more significant the combined impact of rainfall, temperature, salt corrosion, axle load, or subgrade moisture changes on the corresponding road segment.

[0022] The zoned disease prevention and control module includes a hydrological erosion control unit, a mechanical damage suppression unit, and a geological environment coupled degradation control unit. For road sections with a high risk of hydrological erosion, the hydrological erosion control unit implements synergistic control through impermeable and corrosion-resistant materials, permeable drainage materials, and seepage pressure monitoring equipment. For road sections with a high risk of mechanical damage, the mechanical damage suppression unit implements synergistic control through anti-rutting materials, interlayer reinforcement structures, and rutting prediction equipment. For road sections with a high risk of geological environment coupled degradation, the geological environment coupled degradation control unit implements synergistic control through drainage and water-blocking structures, lightweight load-bearing structures, displacement monitoring equipment, moisture content monitoring equipment, and grouting equipment. When a road section has two or more disease risks simultaneously, two or more control units can be configured simultaneously.

[0023] The digital twin module, based on the building information model and geographic information data of the target asphalt pavement, maps data generated during material preparation, paving, compaction, sensor monitoring, and maintenance to corresponding road sections and structural layers, forming a digital twin of the target asphalt pavement. The digital twin is continuously updated with newly acquired data, reflecting the current service status of each road section and predicting the development of defects under different rainfall, temperature, vehicle axle load, and subgrade moisture conditions.

[0024] The maintenance decision-making module obtains the seepage risk, rutting risk, settlement risk, material degradation status, and remaining service capacity of each road segment based on digital twins, and predicts the deterioration trend of each segment within a preset time range. Preset maintenance evaluation objectives may include reducing maintenance costs, extending service life, reducing carbon emissions, shortening traffic closure time, and reducing road operation risks. Based on the deterioration trend and preset maintenance evaluation objectives, the maintenance decision-making module determines the road segments requiring maintenance, the maintenance start time, the maintenance method, the maintenance materials, and the configuration of maintenance equipment.

[0025] The maintenance execution and feedback module controls the maintenance execution equipment to perform waterproofing repairs, partial milling, rut repairs, interlayer reinforcement, subgrade grouting, or old material recycling based on the maintenance decision results. After the maintenance is completed, the multi-source data acquisition module collects pavement and subgrade condition monitoring data again and compares the condition data before and after maintenance. The condition monitoring data after maintenance is fed back to the digital twin module and the maintenance decision module to update the digital twin, correct the defect prediction results, and adjust subsequent maintenance decisions.

[0026] During the engineering verification phase, cross-river bridge sections, transportation hub sections, and high embankment sections can be selected as test sections for continuous performance monitoring over three years. Monitoring content can include chloride salt resistance, seepage status, rutting development rate, interlayer shear status, subgrade deformation, subgrade moisture content, and maintenance costs. Based on the verification results, material formulations, construction temperatures, compaction parameters, early warning thresholds, and model parameters are adjusted to create a continuous optimization loop between each control unit and the intelligent construction and maintenance process.

[0027] Through the above settings, differentiated disease prevention and control can be implemented based on the environmental, load, and geological conditions of different road sections in hot and humid regions. By continuously linking data from the construction, operation, and maintenance stages through digital twins, hydrological erosion, mechanical damage, and coupled deterioration of the geological environment can no longer be addressed independently. At the same time, through closed-loop control of disease trend prediction, maintenance decision-making, maintenance execution, and monitoring feedback, the accuracy of identifying internal deterioration states and determining maintenance timing can be improved, reducing repeated repairs and large-scale road closures. This is conducive to extending the service life of the target asphalt pavement to more than 30 years and reducing the total life cycle cost and the risk of structural overhaul.

[0028] In one embodiment, the multi-source data acquisition module collects environmental data, traffic load data, construction data, and pavement and subgrade condition data, respectively. These different types of data can be obtained from sensors, construction equipment controllers, traffic detection equipment, meteorological monitoring equipment, and external data platforms located at the road site.

[0029] Environmental data includes rainfall, temperature and humidity, and water environment data for the area where the target asphalt pavement is located. Rainfall data may include average annual rainfall, single rainfall event amount, rainfall duration, rainfall intensity, and number of consecutive rainy days. Temperature and humidity data may include ambient temperature, pavement surface temperature, relative humidity, and internal temperature of the structural layers. Water environment data may include one or more of the following: surface water level, groundwater level, chloride concentration in water bodies, pavement surface water depth, and water film duration. For bridge sections crossing rivers, the focus can be on collecting data on river water level changes and chloride concentration to reflect dynamic water pressure and salt corrosion.

[0030] Traffic load data includes vehicle flow data, vehicle axle load data, and vehicle operation data. Vehicle flow data can be statistically analyzed by vehicle type, lane, and time period. Vehicle axle load data can be obtained from dynamic weighing equipment and used to form a heavy-load axle load spectrum. Vehicle operation data can include vehicle speed, braking frequency, acceleration, steering status, and vehicle dwell time on corresponding road segments. For traffic hubs, the proportion of heavy-load vehicles, low-speed driving duration, and number of braking and steering maneuvers can be collected to characterize repeated shearing effects.

[0031] Construction data includes material preparation data, paving data, and compaction data. Material preparation data may include raw material batches, raw material quantities, mixing temperature, mixing time, material admixtures, and discharge temperature. Paving data may include paving location, paving speed, paving thickness, and paving temperature. Compaction data may include compaction equipment location, number of compaction passes, compaction speed, compaction temperature, and compaction trajectory. Construction equipment can transmit the above data to a multi-source data acquisition module in real time via a construction IoT system.

[0032] The pavement and subgrade condition data include seepage pressure data, pavement temperature data, subgrade moisture content data, and displacement data. Seepage pressure data can be collected by fiber optic grating seepage pressure sensors installed inside the pavement structural layers; pavement temperature data can be collected by temperature sensors or infrared thermal imaging equipment; subgrade moisture content data can be collected by soil moisture sensors buried at different depths; and displacement data can be collected by BeiDou displacement sensors, settlement gauges, or other displacement monitoring equipment.

[0033] Data from different data sources can have different sampling frequencies and coordinate systems. The multi-source data acquisition module adds acquisition time, acquisition location, device number, and data type identifier to each data point, and unifies the time of different data to a preset time base. For continuously sampled data, resampling can be performed according to a preset time window; for non-continuous data, interpolation can be performed based on adjacent sampling times or the most recent valid value can be retained.

[0034] During spatial correlation, the station number and lane number of the target asphalt pavement can be used as a linear spatial reference, and the locations of sensor devices, construction equipment trajectories, and defect detection locations can be converted to unified road coordinates. For data with burial depth, it can be further correlated to specific pavement or subgrade structural layers in the building information model. After completing time synchronization and spatial location correlation, a multi-source road segment dataset corresponding to each road segment is formed.

[0035] The data acquisition process can also perform validity checks. Validity checks include detecting missing data, abnormal jumps, exceeding the device's measurement range, and inconsistencies between adjacent sensor data. When a single sensing device malfunctions, temporary compensation can be performed using data from adjacent devices, historical data, or digital twin prediction data, and equipment maintenance information can be generated to prevent single-point failures from affecting risk identification.

[0036] Through the above processing, multi-source heterogeneous data scattered across meteorology, hydrology, transportation, construction, pavement, and roadbed can be unified to specific road sections, specific times, and specific structural layers. This provides a continuous and traceable data foundation for disease risk identification and digital twin updates, reducing state judgment errors caused by inconsistent data time, mismatched spatial locations, or fragmented monitoring links.

[0037] In one embodiment, the road segment risk identification module extracts road segment condition features from a multi-source road segment dataset. These features include rainfall erosion characteristics, heavy load characteristics, pavement temperature characteristics, subgrade humidity characteristics, and structural deformation characteristics.

[0038] Rainfall erosion characteristics can be determined based on rainfall intensity, duration of continuous rainfall, water level changes, chloride concentration, osmotic pressure changes, and water film duration. Heavy load characteristics can be determined based on vehicle traffic volume, proportion of heavy-load vehicles, axle load size, number of repeated axle loads, vehicle braking frequency, and low-speed operation time. Pavement temperature characteristics can be determined based on the highest surface temperature, temperature duration, and degree of temperature field inhomogeneity. Subgrade moisture characteristics can be determined based on moisture content, rate of moisture content change, and differences in moisture content at different depths. Structural deformation characteristics can be determined based on settlement amount, settlement rate, lateral displacement, rut depth, and interlayer relative displacement.

[0039] The road segment risk identification module determines the environmental severity level based on the condition characteristics of each road segment. The environmental severity level can be divided into four levels: Level 1, Level 2, Level 3, and Level 4. Level 1 indicates a relatively weak environmental impact, while Level 4 indicates the combined effect of multiple adverse factors and a high probability of disease occurrence. The grading boundaries of the environmental severity level can be determined based on historical disease data for the region, results of accelerated indoor testing, and monitoring results from actual engineering projects.

[0040] For hydrological erosion risk, the characteristic value can be determined based on rainfall erosion characteristics, pavement seepage pressure, chloride ionization, and the integrity of the waterproof structure. For mechanical damage risk, the characteristic value can be determined based on heavy load characteristics, pavement temperature characteristics, historical rutting development, and interlayer shear strength. For coupled geological environment degradation risk, the characteristic value can be determined based on subgrade humidity characteristics, structural deformation characteristics, embankment height, and the degree of groundwater influence.

[0041] Each risk characterization value can be obtained through weighted calculation, rule-based models, machine learning models, or a combination thereof. When using weighted calculation, weights can be set according to the degree of influence of each state feature on the corresponding disease. When using machine learning models, historical disease samples can be used to train the risk classification model, and the current road segment state features can be input into the risk classification model to output the disease risk type and disease risk level.

[0042] The road segment risk identification module compares each risk characterization value with its corresponding risk assessment criteria. When the hydrological erosion risk characterization value meets the hydrological erosion risk assessment criteria, the corresponding road segment is identified as a hydrological erosion risk segment; when the mechanical damage risk characterization value meets the mechanical damage risk assessment criteria, the corresponding road segment is identified as a mechanical damage risk segment; when the geological environment coupled deterioration risk characterization value meets the geological environment coupled deterioration risk assessment criteria, the corresponding road segment is identified as a geological environment coupled deterioration risk segment. The same road segment can be identified as one or more types of disease risk.

[0043] The target asphalt pavement is divided into risk zones based on the type and degree of damage risk, resulting in road segment risk zoning. For example, river-crossing bridge sections can be classified as high hydrological erosion risk zones; transportation hub sections can be classified as high mechanical damage risk zones; and high embankment sections can be classified as high geological environment coupled deterioration risk zones. In the application example of Guilin Outer Ring Road B section, the Lijiang Grand Bridge section K103+600 can be designated as a key area for hydrological erosion prevention and control, the Diecai East Hub section K110+200 as a key area for mechanical damage prevention and control, and the Lingchuan high embankment section K95+800 as a key area for geological environment coupled deterioration prevention and control.

[0044] The results of road segment risk zoning are used not only to determine control units, but also to determine the density of sensor deployment, monitoring frequency, material grade, construction parameters, and maintenance priorities. The higher the degree of road damage risk, the higher the density of sensor deployment and the data sampling frequency can be, and higher-performance control materials can be used.

[0045] Through the above treatment, the main defects and their severity in different road sections can be identified based on factors such as rainfall, salt erosion, temperature, heavy load, roadbed humidity and structural deformation. This ensures that the defects control targets, monitoring resources and construction resources are matched with the actual risks, avoiding insufficient control or waste of resources caused by using the same structure or the same maintenance standards for all road sections.

[0046] In one embodiment, the hydrological erosion control unit includes a first modified asphalt preparation device, a permeable mixture preparation device, a seepage pressure sensor array, and a seepage state interpretation device, which is suitable for cross-river bridges, waterfront roads, road sections with high groundwater levels, and road sections that are greatly affected by chloride erosion.

[0047] The first modified asphalt preparation equipment is used to prepare modified asphalt containing silane nanocapsules. During preparation, the silane active component is encapsulated to form silane nanocapsules. The base asphalt is heated and stirred, and the silane nanocapsules are added according to a preset mass ratio, ensuring uniform dispersion of the silane nanocapsules in the base asphalt. The preset mass ratio, stirring temperature, and stirring time can be determined based on tests for impermeability, chloride salt resistance, adhesion, and workability.

[0048] Modified asphalt containing silane nanocapsules is used to construct impermeable and corrosion-resistant layers. When micro-cracks appear in the impermeable and corrosion-resistant layer or when exposed to moisture and chloride salts, the silane active component can act as a hydrophobic agent and provide interfacial protection in the cracked areas, reducing the continued migration of moisture and chloride salts into the pavement structure. The impermeable and corrosion-resistant layer can be set between the bridge deck and the asphalt pavement layer or at the interface of structural layers requiring impermeability and corrosion protection.

[0049] The permeable mixture preparation equipment is used to prepare permeable mixtures containing basalt fibers. The basalt fibers can form a dispersed reinforcing structure within the mixture, improving its overall stability under repeated water action and vehicle loads. The gradation, porosity, basalt fiber length, and basalt fiber content of the permeable mixture can be determined based on the target permeability and mechanical properties.

[0050] Permeable mixtures containing basalt fibers are used to construct a permeable drainage layer on one side of the impermeable and corrosion-resistant layer. The permeable drainage layer guides water entering the pavement structure outwards in a predetermined direction, reducing water accumulation at interlayer interfaces or in localized voids. The impermeable and corrosion-resistant layer and the permeable drainage layer work together to achieve water blocking, corrosion resistance, and drainage.

[0051] During material performance verification, the material formulation can be determined through immersion tests, chloride erosion tests, permeability tests, and freeze-thaw splitting tests. In one application example, by adjusting the dosage of silane nanocapsules and the gradation of the permeable mixture, the permeability of the impermeable structure is made not greater than 0.5%, the permeability coefficient of the permeable drainage layer is not less than 1.2 mm / s, and the mass loss rate under five-year equivalent chloride erosion conditions is less than 1%.

[0052] The pressure sensor array is installed in the corresponding areas of the anti-seepage and anti-corrosion layer and the permeable drainage layer, which can monitor the pressure changes at different structural locations. The seepage status interpretation equipment receives the pressure data and, in combination with rainfall data, water level data and the spatial location of each pressure sensor, determines whether there is a continuous increase in pressure, an abnormal pressure gradient, or localized seepage concentration in the corresponding road section.

[0053] When the seepage condition meets the preset seepage risk conditions, the seepage condition interpretation equipment generates a hydrological erosion early warning information. The preset seepage risk conditions may include a seepage pressure value greater than a preset seepage pressure threshold, a seepage pressure increase rate greater than a preset increase threshold, multiple adjacent sensors showing consecutive anomalies, or an abnormal seepage path pointing to a weak area of ​​the waterproof structure. The hydrological erosion early warning information may include the risk section, risk level, abnormal location, and recommended treatment measures.

[0054] Through the above-mentioned setup, the anti-seepage and anti-corrosion layer can reduce the probability of chloride salts and moisture migrating into the structure, the permeable drainage layer can promptly drain water that has entered the structure, and the seepage pressure sensor array and seepage state interpretation equipment can continuously monitor the working status of the waterproof drainage structure, thereby achieving a combination of material protection and dynamic water pressure monitoring, reducing the risk of water damage to the cross-river bridge section under continuous rainfall and dynamic water pressure.

[0055] In one embodiment, the permeability sensor array includes multiple fiber Bragg grating permeability sensors. These multiple fiber Bragg grating permeability sensors are spaced longitudinally and laterally along the target asphalt pavement, and can be configured as a single or multiple sensor arrays according to the depth of the pavement structural layers.

[0056] The spacing of fiber Bragg grating (FBG) permeability sensors is determined based on the risk level of the road section, pavement width, structural layer thickness, and target positioning accuracy. Sensor spacing can be reduced at joints, drainage outlets, bridge deck connection areas, waterproofing layer overlap areas, or locations with historical water damage. Each FBG permeability sensor has a corresponding road station number, lateral position, and burial depth.

[0057] Fiber Bragg grating pressure sensors convert pressure changes into fiber Bragg grating wavelength changes, and demodulation equipment obtains the pressure value based on these wavelength changes. To reduce the impact of temperature changes on pressure measurements, temperature-compensated sensors can be incorporated into the sensor array, or temperature compensation can be achieved using adjacent uncompressed fiber Bragg gratings.

[0058] The seepage state interpretation equipment constructs the spatiotemporal distribution of seepage pressure inside the pavement based on the seepage pressure values, acquisition time, and spatial location of each fiber Bragg grating seepage pressure sensor. For the same acquisition time, the seepage pressure gradient can be determined based on the seepage pressure difference and spatial distance between adjacent sensors; for continuous acquisition times, the seepage pressure propagation time relationship can be determined based on the order in which the seepage pressure of each sensor reaches the abnormal threshold.

[0059] Based on the osmotic pressure gradient and its changing direction, the main direction of water migration from high-osmotic-pressure areas to low-osmotic-pressure areas can be determined. By combining the expansion process of anomaly areas at multiple sampling times, the extent of seepage influence can be determined. The seepage state interpretation equipment can also input the spatiotemporal distribution of osmotic pressure into a fluid dynamics calculation model. By adjusting candidate seepage inlets, structural layer permeability parameters, and seepage channel parameters, the difference between the model calculation results and sensor measurements can be reduced, thereby retrieving the abnormal seepage path.

[0060] The seepage path location results include the starting location, extension direction, affected length, affected width, and corresponding road segment location of the abnormal seepage path. The location results can be mapped to a digital twin, and the spatial relationship between the abnormal seepage path and the anti-corrosion layer, permeable drainage layer, and structural joints can be marked in the building information model.

[0061] During the calibration process in actual engineering projects, a known seepage path can be established through artificial water injection tests or controlled seepage tests. The position output by the seepage state interpretation equipment is then compared with the known position to correct the sensor layout parameters and the seepage location model. When the sensor density and signal quality meet the requirements, the location accuracy of abnormal seepage paths can reach 0.3m.

[0062] Through the above processing, the spatial distribution and propagation process of internal seepage pressure of the target asphalt pavement can be recovered from multiple discrete seepage pressure measuring points. Abnormal seepage paths can be located based on seepage pressure gradient, propagation time sequence and fluid dynamics inversion results, enabling maintenance personnel to carry out local repairs on the actual seepage location and avoid large-scale excavation based solely on surface water accumulation or appearance defects.

[0063] In one embodiment, the mechanical damage suppression unit includes a composite aggregate preparation device, a second modified asphalt preparation device, a grid reinforcement layer laying device, an infrared thermal imaging device, and a rut prediction device, and is suitable for traffic hub sections with a high proportion of heavy-duty vehicles, frequent vehicle braking and steering, or long periods of high temperature.

[0064] The composite aggregate preparation equipment is used to screen, clean, and stabilize steel slag aggregate, and then composite the treated steel slag aggregate with graphene materials. The graphene materials can be distributed on the surface and pore areas of the steel slag aggregate through dry dispersion, wet dispersion, or surface loading. The proportion of steel slag in the aggregate, the amount of graphene added, and the composite treatment time can be determined based on the aggregate crushing value, surface adhesion, thermal conductivity, and high-temperature stability tests of the mixture.

[0065] The second modified asphalt preparation equipment is used to prepare high-entropy alloy modified asphalt. During the preparation process, the base asphalt is heated, and the high-entropy alloy modifier is added to the base asphalt according to a preset dosage. The high-entropy alloy modifier is then uniformly dispersed through high-speed shearing and continuous stirring. The composition, particle size, and dosage of the high-entropy alloy modifier can be determined based on its softening point, ductility, viscosity, high-temperature rheological properties, and fatigue properties.

[0066] Steel slag-graphene composite aggregate, high-entropy alloy modified asphalt, mineral powder, and other graded aggregates are mixed according to the designed mix proportions to obtain rutting-resistant asphalt mixture. This rutting-resistant asphalt mixture is used to construct an rutting-resistant asphalt layer, synergistically combining the high strength of the steel slag aggregate, the reinforcing effect of the graphene material, and the high-temperature stability of the high-entropy alloy modified asphalt.

[0067] The mesh reinforcement layer laying equipment lays a three-dimensional fiberglass mesh reinforced bonding layer between adjacent pavement structural layers. During laying, the underlying layer is first cleaned and bonded with adhesive material, and then the three-dimensional fiberglass mesh is laid under a preset tension to prevent wrinkles and localized lifting of the fiberglass mesh. The three-dimensional fiberglass mesh is used to disperse the horizontal shear stress generated by vehicle braking, acceleration, and steering, thereby improving the interlayer shear strength between adjacent pavement structural layers.

[0068] After optimization of material and structural parameters, the dynamic stability of the anti-rutting asphalt layer at 60℃ can reach more than 12,000 cycles / mm, and the interlaminar shear strength of the three-dimensional fiberglass mesh reinforced bonding layer can reach 2.8MPa, thus inhibiting rutting development from both material deformation and interlaminar slip.

[0069] Infrared thermal imaging equipment is installed on the roadside, gantry, or mobile inspection vehicle to collect road surface temperature field data for the corresponding road segment. The road surface temperature field data includes the road surface temperature at different locations, the highest temperature area, the temperature gradient, and the duration of the high-temperature area. The rut prediction equipment performs radiometric correction, spatial registration, and outlier removal on the infrared thermal imaging data, correlating the temperature field data to the corresponding road segment.

[0070] The rutting prediction equipment uses road surface temperature field data, traffic load data, and historical rutting data as input. Historical rutting data can be obtained by laser rutting detection equipment, 3D pavement detection equipment, or manual inspection. The rutting prediction equipment extracts features such as the duration of high temperatures, the number of heavily loaded axles, vehicle operating status, historical rutting depth, and rutting development speed, and uses regression models, time-series prediction models, or machine learning models to predict the rutting depth and rutting development speed within a preset future time range.

[0071] When the predicted rut depth or rut development rate reaches the preset maintenance conditions, the rut prediction equipment determines the appropriate rut maintenance time for the corresponding road section and sends the prediction results to the maintenance decision module. The maintenance decision module can arrange local repairs before the ruts affect road safety. Through continuous monitoring of the test section and model correction, the annual rut development rate can be controlled below 2mm.

[0072] Through the above setup, steel slag-graphene composite aggregate, high-entropy alloy modified asphalt, and three-dimensional glass fiber mesh reinforced bonding layer can respectively improve the high-temperature stability and interlayer shear strength of the materials. Infrared thermal imaging equipment and rutting prediction equipment can predict the rutting development trend in advance based on temperature and heavy load, thus forming a mechanical damage suppression system that combines material reinforcement, structural reinforcement, and predictive maintenance.

[0073] In one embodiment, the geological environment coupled deterioration prevention and control unit includes roadbed construction equipment, displacement sensor array, soil moisture sensor and grouting equipment, and is suitable for road sections with high fill, weak foundation, large changes in groundwater level or significant capillary migration.

[0074] The roadbed construction equipment is used to form a composite roadbed structure comprising a permeable drainage layer, a capillary water blocking layer, and a lightweight load-bearing layer. The capillary water blocking layer is positioned between the permeable drainage layer and the lightweight load-bearing layer. The permeable drainage layer collects and drains water entering the roadbed, the capillary water blocking layer prevents water from migrating upwards from the lower layers, and the lightweight load-bearing layer bears the load of the upper pavement and reduces additional pressure on the underlying fill.

[0075] In a specific application example, the permeable drainage layer consists of a 30cm thick permeable gravel layer, the capillary water blocking layer uses an HDPE geomembrane, and the lightweight load-bearing layer uses a 40cm thick foamed concrete base layer. A permeable gravel laying machine lays and levels the permeable gravel according to the designed thickness. An HDPE geomembrane laying machine continuously lays the HDPE geomembrane between the permeable gravel layer and the foamed concrete base layer. Foamed concrete pouring equipment forms the foamed concrete base layer on one side of the HDPE geomembrane.

[0076] Adjacent HDPE geomembranes can be connected by hot-melt welding, and the joints should be tested for sealing to prevent capillary water from seeping through. Sealing connectors can be installed at the connection points between the HDPE geomembrane and edge drainage structures, roadbed slope protection structures, or other structures. The density, strength, and pouring thickness of the foamed concrete can be determined based on the pavement load and roadbed settlement control requirements.

[0077] The displacement sensor array can include multiple BeiDou displacement sensors, which are respectively installed at the top of the roadbed, the slope, and internal monitoring points with settlement risk. Soil moisture sensors are installed near the permeable drainage layer, below the capillary water blocking layer, and inside the roadbed soil to obtain moisture content data at different depths.

[0078] The geological environment coupled degradation prevention and control unit determines the roadbed settlement, settlement velocity, and lateral displacement based on displacement data, and determines the roadbed moisture content level, moisture content variation range, and moisture migration trend based on moisture content data. It also correlates moisture content changes and settlement changes within the same spatial area over time to determine the coupling state between roadbed moisture changes and settlement changes.

[0079] Preset settlement warning conditions can be determined based on historical monitoring data from the construction and operation periods. In one application example, when the settlement within a month exceeds 5 mm, or the change in moisture content relative to the historical stable moisture content exceeds 15%, the coupling state is determined to meet the preset settlement warning conditions. Different warning levels can also be set according to the settlement and moisture content change ranges, so that minor anomalies trigger intensified monitoring first, and continuous or severe anomalies trigger grouting reinforcement.

[0080] Based on the spatial locations of displacement sensors and soil moisture sensors that meet the preset settlement early warning conditions, the area of ​​the roadbed to be reinforced is determined. The area of ​​the roadbed to be reinforced may include its planar extent, reinforcement depth, and target grouting layer. The grouting equipment performs drilling and grouting according to the area of ​​the roadbed to be reinforced, and the grouting stop conditions are determined based on the grouting pressure, grouting volume, grout backflow status, and displacement changes after reinforcement.

[0081] In one application example, this composite subgrade structure and humidity-settlement coupled early warning system can reduce saturated soil deformation by approximately 65% ​​and control subgrade deformation to within 3 mm. After curing, displacement and moisture content data are continuously collected to assess the reinforcement effect and update the early warning model.

[0082] Through the above setup, the permeable drainage layer, capillary water blocking layer, and lightweight load-bearing layer can reduce the probability of deformation caused by moisture in high embankment subgrades from three aspects: drainage, water blocking, and load reduction. Displacement monitoring and moisture content monitoring can identify the correlation between moisture changes and settlement, and locate the areas to be reinforced and trigger grouting under abnormal conditions, so that subgrade defects are transformed from post-reinforcement to a control process that can be monitored, warned, and actively intervened in.

[0083] In one embodiment, the digital twin module first establishes a building information model (BIM) of the target asphalt pavement. The BIM includes objects such as the road centerline, road segment station numbers, lanes, pavement structure layers, subgrade structure layers, bridge structures, drainage structures, waterproofing structures, sensing devices, and maintenance equipment.

[0084] The digital twin module spatially registers the Building Information Model (BIM) with geographic information data. Geographic information data includes road spatial locations, topographic elevations, water system locations, land use information, geological zoning, and environmental zoning. Spatial registration can use unified coordinate transformation parameters to convert local engineering coordinates in the BIM to geographic coordinates, and establish the correspondence between station numbers, lanes, and structural layers, resulting in a basic digital model.

[0085] The digital twin module receives construction data generated during material preparation, paving, and compaction, and maps the construction data to the corresponding road sections and structural layers in the basic digital model based on the construction equipment trajectory, data acquisition time, and road station number. For example, the mixing temperature is mapped to the corresponding material batch, the paving temperature and paving thickness are mapped to the corresponding pavement structural layer, and the compaction trajectory and number of compaction passes are mapped to the corresponding construction area, forming construction status parameters.

[0086] The digital twin module also receives seepage pressure data, pavement temperature data, subgrade moisture content data, and displacement data. Based on the sensor device number, acquisition location, and burial depth, it maps the condition monitoring data to the basic digital model, forming service status parameters for each road section. These service status parameters can include seepage status, temperature status, rutting status, subgrade moisture status, settlement status, and material deterioration status.

[0087] When new pavement and subgrade condition monitoring data are obtained, the digital twin module updates the service status parameters of the corresponding road segment and structural layer. During the update process, weights can be assigned to different data based on the reliability of the sensing equipment and the data quality, and the measured data can be fused with the model prediction data to reduce the impact of single measurement anomalies on the digital twin.

[0088] The digital twin module, based on updated service status parameters, simulates changes in condition under different environmental and traffic load conditions. For example, it can simulate scenarios such as continuous heavy rainfall, rising river levels, sustained high temperatures, increased heavy vehicle traffic, or rising roadbed moisture content to simulate changes in seepage pressure, rutting development, and settlement. The simulation results include the location of damage, the rate of damage development, the estimated time to reach the maintenance threshold, and the extent of affected structures.

[0089] During the construction phase, the digital twin is used to record the quality status of materials and construction; during the operation phase, it receives real-time monitoring data and predicts the development of defects; during the maintenance phase, it simulates the expected effects of different maintenance measures. After maintenance is completed, the actual maintenance results are compared with the predicted results, and the material parameters, structural parameters, and defect development parameters are corrected.

[0090] The digital twin module can integrate Building Information Modeling (BIM), Geographic Information Systems (GIS), and the Internet of Things (IoT) for construction, forming a data foundation covering the construction, operation, and maintenance phases. Digital twins of different road sections can be stored in a unified data format, providing data for cross-section defect model training and maintenance decision-making.

[0091] Through the above processing, the road geometry, geographical environment, construction quality, and operational status can be mapped to a unified digital space and continuously updated with real-time monitoring data, enabling the location of defects to specific road sections and structural layers. At the same time, through dynamic simulations under different environmental and load scenarios, the development trend of defects can be obtained in advance, providing verifiable data for determining maintenance timing and measures.

[0092] In one embodiment, the maintenance decision-making module includes a federated learning unit and a multi-objective optimization unit. The federated learning unit is used to train a disease development prediction model when asphalt pavement distress data from different regions are not directly and centrally transmitted.

[0093] Local nodes are set up in different regions. For example, local nodes can be set up in Guangxi Zhuang Autonomous Region, Guangdong, Hainan, or other hot and humid regions. Each local node stores local environmental data, traffic load data, construction data, pavement and subgrade condition data, defect detection data, and maintenance records, and constructs local training samples according to unified data fields and labeling rules.

[0094] The central coordinating node sends initial model parameters to each local node. Each local node trains the initial model using local asphalt pavement distress data to obtain updated local model parameters. Local nodes do not upload the original road data, only the updated model parameters or privacy-preserving gradient information.

[0095] The federated learning unit performs weighted aggregation of the model update parameters uploaded by each local node based on the amount of data, model training quality, and data reliability, to obtain the updated global model parameters. The updated global model parameters are then sent back to each local node for the next round of training until the model performance meets preset conditions, resulting in a disease development prediction model.

[0096] Given the differences in climate conditions, material systems, and traffic load distribution in different regions, the local data of the target asphalt pavement can be used to make adaptive adjustments based on the global model. This allows the disease development prediction model to not only utilize the disease patterns of multiple regions but also adapt to the specific environmental conditions of the target road section.

[0097] By inputting the service status parameters corresponding to the digital twin into the disease development prediction model, the disease development prediction results for each road section within a preset time range are obtained. The disease development prediction results can include the probability of seepage risk, the expected rut depth, the expected settlement, the expected time for the disease to reach the maintenance threshold, and the remaining service life.

[0098] The multi-objective optimization unit generates multiple candidate maintenance schemes based on the predicted road damage development results. Candidate maintenance schemes include the road section to be maintained, maintenance start time, maintenance duration, maintenance measures, material type, equipment type, and traffic organization method. Maintenance measures may include waterproofing layer repair, drainage structure treatment, rut milling repair, interlayer reinforcement, subgrade grouting, and recycling of old materials.

[0099] The multi-objective optimization unit calculates the maintenance cost, carbon emissions, service life extension, road closure time, and degree of risk reduction for each candidate maintenance scheme. Maintenance costs can be determined based on material, equipment, labor, and traffic organization costs; carbon emissions can be determined based on material production, transportation, equipment energy consumption, and the recycling process of old materials; and service life can be determined based on digital twin simulation and disease development prediction models.

[0100] With the optimization objectives of minimizing maintenance costs and carbon emissions while maximizing service life, and with constraints set as budget ceilings, permissible road closure times, material supply capacity, equipment availability, and road safety requirements, a multi-objective evaluation is conducted on multiple candidate maintenance schemes. The multi-objective evaluation can employ Pareto optimization, weighted scoring, or hierarchical screening methods to obtain maintenance decisions that satisfy the constraints.

[0101] The maintenance decision results include recommended maintenance plans and alternative maintenance plans, and can output the corresponding costs, lifespans, carbon emissions, and road closure times for each plan. In one application example, by optimizing the timing of maintenance and the allocation of maintenance resources, the accuracy of defect identification can reach over 95%, road closure time can be reduced by about 50%, and maintenance costs can be reduced by about 35%.

[0102] Through the above processing, disease data from different regions can participate in model training without directly collecting the original data, thereby improving the adaptability of the disease development prediction model to the complex environment of hot and humid regions. At the same time, by comprehensively evaluating maintenance costs, carbon emissions, service life, and road closure time, it is possible to avoid determining maintenance plans based solely on a single disease indicator, and to make maintenance decisions take into account road safety, economy, durability, and low-carbon requirements.

[0103] In one embodiment, the maintenance execution equipment includes a recycled binder preparation device and a microwave activation and regeneration device. The recycled binder preparation device is used to prepare geopolymer-based recycled binder using manganese slag, and the microwave activation and regeneration device is used to activate the old asphalt pavement material generated during the maintenance process using microwave heating.

[0104] Manganese slag can be derived from local manganese industry solid waste. Before preparing geopolymer-based recycled binder, the manganese slag is dried, crushed, screened, and ground to ensure that the particle size meets the requirements for binder preparation. Alkali-activating materials and auxiliary cementitious materials are selected based on the chemical composition of the manganese slag and mixed and activated according to a predetermined ratio.

[0105] In one application example, the manganese slag content in the geopolymer-based recycled binder is not less than 50% of the total mass of the binder. The manganese slag content, activator dosage, water-cement ratio, and curing conditions can be determined based on tests for compressive strength, bonding performance, water resistance, durability, and workability.

[0106] Microwave-activated recycling equipment screens and pre-treats milled asphalt pavement material, removing impurities that do not meet recycling requirements. The microwave power, processing temperature, and processing time are determined based on the moisture content, old asphalt content, and degree of aging of the material. Microwave energy selectively heats the interior of the material, softening the aged asphalt and improving the bonding ability between the old material and recycled binders or new asphalt materials.

[0107] Microwave activation and regeneration equipment can be equipped with a temperature detection unit and a conveyor speed control unit. When the temperature of the old material reaches the preset activation temperature, the microwave power or conveyor speed is adjusted to avoid localized overheating and further aging of the asphalt. After reaching the preset activation state, activated old material is obtained.

[0108] When maintenance decisions indicate the adoption of low-carbon recycled maintenance, at least one of geopolymer-based recycled binder and activated recycled material can be selected for the maintenance treatment of the corresponding road section. For example, geopolymer-based recycled binder can be used in subgrade reinforcement or base course repair; activated recycled material can be used in pavement rutting repair or partial milling and repaving. Depending on the maintenance location and performance requirements, the two can also be used in combination.

[0109] Under the condition that the material properties meet the requirements, the utilization rate of old asphalt pavement materials can reach about 45%, and the fatigue life of recycled materials can be increased by about 30% compared with conventional new materials. Replacing some new materials with manganese slag and old asphalt pavement materials can reduce the consumption of virgin aggregates and traditional cementitious materials.

[0110] The maintenance execution and feedback module acquires pavement performance data and material utilization data after maintenance treatment. Pavement performance data may include smoothness, rut depth, permeability, interlayer shear properties, load-bearing capacity, and fatigue performance; material utilization data may include manganese slag usage, recycled material usage, new material usage, and the proportion of recycled materials used.

[0111] The maintenance execution and feedback module also acquires equipment energy consumption data during material preparation, transportation, microwave activation, and on-site construction. Based on material utilization data, equipment energy consumption data, and corresponding carbon emission factors, it determines the carbon emission data corresponding to the maintenance treatment. By comparing with the baseline maintenance method using the new materials, the carbon emission reduction can be determined. In one application example, the carbon emission reduction rate can reach 40%, and the annual reduction in carbon dioxide emissions per kilometer can reach 120 to 150 tons.

[0112] Road surface performance data, material utilization data, and carbon emission data are fed back to the digital twin module and the maintenance decision module. The digital twin module updates the service status of the corresponding road segment based on the performance data after maintenance, while the maintenance decision module corrects the multi-objective evaluation parameters based on the actual material utilization rate, actual carbon emissions, and actual lifespan improvement effect, thereby optimizing the maintenance plan for subsequent road segments.

[0113] Through the above setup, local manganese slag and old asphalt pavement materials generated during maintenance can be transformed into recycled materials that can be used for road maintenance, realizing the resource utilization of solid waste and the recycling of old materials. At the same time, by providing feedback on material utilization, equipment energy consumption, carbon emissions, and post-maintenance performance, the low-carbon recycling effect can be incorporated into the digital twin and maintenance decision-making process, reducing the consumption of new materials and carbon emissions throughout the entire life cycle while ensuring road performance and fatigue life.

[0114] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A device for the intelligent construction, maintenance, and long-life guarantee of asphalt pavement in hot and humid areas, characterized in that, include: The multi-source data acquisition module is used to collect environmental data, traffic load data, construction data, and pavement and subgrade status data of the target asphalt pavement during the construction and operation phases. The road segment risk identification module is used to determine the type and degree of disease risk of each road segment in the target asphalt pavement based on the environmental data, traffic load data, construction data and pavement and subgrade status data, and obtain the road segment risk zoning results. The types of disease risks include hydrological erosion risk, mechanical damage risk, and coupled deterioration risk of geological environment. The zoned disease prevention and control module includes a hydrological erosion prevention and control unit, a mechanical damage suppression unit, and a geological environment coupled deterioration prevention and control unit. It is used to implement disease prevention and control corresponding to the disease risk type for the corresponding road section based on the road section risk zoning results, and to collect pavement and subgrade condition monitoring data during the disease prevention and control process. The digital twin module is used to integrate the building information model, geographic information data, construction data, and pavement and subgrade condition monitoring data of the target asphalt pavement to construct and update the digital twin corresponding to the target asphalt pavement. The maintenance decision module is used to determine the current service status of each road segment based on the digital twin, predict the development trend of defects in each road segment, and determine the corresponding maintenance timing, maintenance area and maintenance measures based on the development trend of defects and the preset maintenance evaluation target, so as to obtain the maintenance decision result. The maintenance execution and feedback module includes maintenance execution equipment, which is used to control the maintenance execution equipment to carry out maintenance treatment on the corresponding road section according to the maintenance decision results, and to feed back the pavement and subgrade condition monitoring data after maintenance treatment to the digital twin module to update the digital twin and subsequent maintenance decision results.

2. The intelligent construction and maintenance and long-life guarantee device for asphalt pavement in hot and humid areas according to claim 1, characterized in that, The multi-source data acquisition module is used for: Rainfall, temperature and humidity, and water environment data of the area where the target asphalt pavement is located are collected to obtain environmental data; Traffic load data is obtained by collecting vehicle flow data, vehicle axle load data, and vehicle operation data acting on the target asphalt pavement. The construction data is obtained by collecting material preparation data, paving data, and compaction data during the construction process of the target asphalt pavement. By using sensing devices installed in the target asphalt pavement and subgrade, seepage pressure data, pavement temperature data, subgrade moisture content data and displacement data are collected to obtain pavement and subgrade state data; The environmental data, traffic load data, construction data, and road surface and subgrade status data are synchronized in time and correlated in space to obtain a multi-source road segment dataset corresponding to each road segment.

3. The intelligent construction and maintenance and long-life guarantee device for asphalt pavement in hot and humid areas according to claim 2, characterized in that, The road section risk identification module is used for: Extract road segment state features that characterize rainfall erosion, heavy load, pavement temperature, subgrade humidity, and structural deformation from the multi-source road segment dataset; Based on the road segment condition characteristics, the environmental severity level corresponding to each road segment is determined, and the hydrological erosion risk characterization value, mechanical damage risk characterization value, and geological environment coupled deterioration risk characterization value corresponding to each road segment are determined respectively. Based on the relationship between each risk characterization value and the corresponding risk judgment conditions, the type and degree of road damage risk for each road section are determined. The target asphalt pavement is divided according to the type and degree of the damage risk to obtain the risk zoning results of the road section.

4. The intelligent construction and maintenance and long-life guarantee device for asphalt pavement in hot and humid areas according to claim 1, characterized in that, The hydrological erosion control unit includes a first modified asphalt preparation device, a permeable mixture preparation device, a seepage pressure sensor array, and a seepage state interpretation device; The first modified asphalt preparation equipment is used to prepare modified asphalt containing silane nanocapsules, and the modified asphalt containing silane nanocapsules is used to construct the anti-seepage and anti-corrosion layer of the corresponding road section. The permeable mixture preparation equipment is used to prepare a permeable mixture containing basalt fiber, and the permeable mixture containing basalt fiber is used to construct a permeable drainage layer on one side of the anti-seepage and anti-corrosion layer. The pressure sensor array is used to collect pressure data of the corresponding areas of the anti-seepage and anti-corrosion layer and the permeable drainage layer. The seepage state interpretation device is used to determine the seepage state of the corresponding road section based on the seepage pressure data, and to generate hydrological erosion early warning information when the seepage state meets the preset seepage risk conditions.

5. The intelligent construction and maintenance and long-life guarantee device for asphalt pavement in hot and humid areas according to claim 4, characterized in that, The permeability sensor array includes multiple fiber optic permeability sensors spaced longitudinally and laterally along the target asphalt pavement; The seepage state interpretation device is used to determine the spatial and temporal variation relationships of seepage pressure inside the target asphalt pavement based on the seepage pressure values, acquisition time, and spatial location collected by each fiber optic grating seepage pressure sensor. Based on the spatial and temporal variation relationships, the seepage pressure gradient inside the target asphalt pavement and the direction of change of the seepage pressure gradient are determined. Based on the seepage pressure gradient and the direction of its change, the seepage direction and seepage influence range of the corresponding road section are determined. Based on the seepage direction and the seepage influence range, the abnormal seepage path and the corresponding road segment location are determined, and the seepage path location result is obtained.

6. The intelligent construction and maintenance and long-life guarantee device for asphalt pavement in hot and humid areas according to claim 1, characterized in that, The mechanical damage suppression unit includes a composite aggregate preparation device, a second modified asphalt preparation device, a grid reinforcement layer laying device, an infrared thermal imaging device, and a rut prediction device. The composite aggregate preparation equipment is used to composite steel slag aggregate and graphene material to obtain steel slag-graphene composite aggregate. The second modified asphalt preparation equipment is used to prepare high-entropy alloy modified asphalt, the steel slag-graphene composite aggregate and the high-entropy alloy modified asphalt are used to prepare rutting-resistant asphalt mixture, and the rutting-resistant asphalt mixture is used to construct a rutting-resistant asphalt layer. The mesh reinforcement layer laying equipment is used to lay a three-dimensional fiberglass mesh reinforced bonding layer between adjacent pavement structural layers; The infrared thermal imaging device is used to collect road surface temperature field data for the corresponding road section; The rutting prediction device is used to predict the rutting development trend of a corresponding road segment based on the road surface temperature field data, the traffic load data, and historical rutting data, and to determine the timing of rutting maintenance for the corresponding road segment based on the rutting development trend.

7. The intelligent construction and maintenance and long-life guarantee device for asphalt pavement in hot and humid areas according to claim 1, characterized in that, The geological environment coupled deterioration prevention and control unit includes roadbed construction equipment, displacement sensor array, soil moisture sensor and grouting equipment; The roadbed construction equipment is used to form a composite roadbed structure including a permeable drainage layer, a capillary water blocking layer and a lightweight load-bearing layer, wherein the capillary water blocking layer is disposed between the permeable drainage layer and the lightweight load-bearing layer; The displacement sensor array is used to collect displacement data of the composite roadbed structure, and the soil moisture sensor is used to collect moisture data of the corresponding area of ​​the composite roadbed structure. Based on the displacement data and the moisture content data, determine the coupling state between the changes in roadbed moisture and settlement. When the settlement amount within the preset statistical period is greater than the preset settlement threshold, and / or the change in the moisture content data relative to the historical moisture content data is greater than the preset moisture content change threshold, it is determined that the coupling state meets the preset settlement early warning condition. Based on the spatial positions of the displacement sensor and soil moisture sensor that meet the preset settlement early warning conditions, the area of ​​the roadbed to be reinforced is determined, and the grouting equipment is controlled to perform grouting reinforcement on the area of ​​the roadbed to be reinforced.

8. The intelligent construction and maintenance and long-life guarantee device for asphalt pavement in hot and humid areas according to claim 1, characterized in that, The digital twin module is used for: A building information model of the target asphalt pavement is established, and the building information model is spatially registered with the geographic information data to obtain a basic digital model including the road segment location, pavement structure layer and subgrade structure layer. The construction data and the pavement and subgrade condition monitoring data are mapped to the basic digital model according to the collection time and collection location to obtain the construction status parameters and service status parameters corresponding to each road section. The service status parameters are updated based on the newly collected pavement and subgrade condition monitoring data; Based on the updated service status parameters, the state changes of each road segment under different environmental conditions and traffic load conditions are extrapolated to obtain the digital twin corresponding to the target asphalt pavement.

9. The intelligent construction and maintenance and long-life guarantee device for asphalt pavement in hot and humid areas according to claim 8, characterized in that, The maintenance decision-making module includes a federated learning unit and a multi-objective optimization unit; The federated learning unit is used to store asphalt pavement distress data from different regions in their respective local nodes, control each local node to train the model based on the local asphalt pavement distress data, and aggregate the model update parameters uploaded by each local node to obtain a distress development prediction model. Input the service status parameters corresponding to the digital twin into the disease development prediction model to obtain the disease development prediction results for each road section within a preset time range; The multi-objective optimization unit is used to generate multiple candidate maintenance schemes based on the disease development prediction results, and evaluates the multiple candidate maintenance schemes with the optimization objectives of minimizing maintenance costs and carbon emissions and maximizing service life, to obtain the maintenance decision results.

10. The intelligent construction and maintenance and long-life guarantee device for asphalt pavement in hot and humid areas according to claim 1, characterized in that, The maintenance equipment includes a recycled binder preparation device and a microwave activation and regeneration device; The recycled cementitious material preparation equipment is used to prepare geopolymer-based recycled cementitious materials using manganese slag. The microwave activation and regeneration equipment is used to activate old asphalt pavement material generated during the maintenance process by microwave heating, so as to obtain activated old material. Based on the maintenance decision results, at least one of the geological polymer-based recycled binder and the activated old material is selected for maintenance treatment of the corresponding road section; Obtain pavement performance data and material utilization data after maintenance treatment, and determine the corresponding carbon emission data based on the material utilization data and equipment energy consumption data during the maintenance treatment process; The pavement performance data, material utilization data, and carbon emission data are fed back to the digital twin module and the maintenance decision module to update the digital twin and subsequent maintenance decision results.