Asphalt pavement maintenance strategy intelligent generation method and system

Abstract

The invention discloses an asphalt pavement maintenance strategy intelligent generation method and system, and relates to the technical field of road engineering.The maintenance strategy generation method comprises the steps that traffic analysis data information and traffic load data information are extracted from a multi-source traffic monitoring system, and the traffic analysis data information and the traffic load data information are obtained through a laser scanning and image intelligent recognition algorithm; analyzing the pavement structural disease level in each grid unit; according to the road surface structural disease level, the road condition risk level of each grid unit under the traffic load and damage superimposed effect is analyzed, a risk response area is divided, and a service life loss analysis instruction is generated; analyzing an accelerated decline trend of pavement structure deterioration of each grid unit in the risk response area to determine a life reduction value of each grid unit in the risk response area; after comparison, generating a pavement maintenance strategy of a corresponding grade; the maintenance strategy generation method has the advantages of being high in real-time performance, fine in area recognition and dynamically adjustable in strategy response.

Description

Technical Field

[0001] The present invention relates to the technical field of road engineering, and in particular to an intelligent generation method and system for asphalt pavement maintenance strategies. Background Art

[0002] Against the backdrop of accelerating transportation and urbanization, asphalt roads, as the primary structure of modern urban road networks, face core challenges in transportation engineering management, including operational safety, structural durability, and maintenance efficiency. This is especially true in high-traffic areas such as high-grade highways, urban arterial roads, and industrial heavy-load logistics corridors. Asphalt pavements are susceptible to premature disease and reduced lifespans due to the long-term impact of rutting and crack accumulation. Therefore, the key challenges in the development of intelligent road operations and maintenance are identifying micro-damage in real time, accurately quantifying structural risks, and generating responsive maintenance decisions based on a lifespan prediction mechanism, all supported by multi-source sensing data.

[0003] However, existing asphalt pavement maintenance strategy generation methods typically rely on periodic road inspection results, manual identification of defects, or static road grade classification to make decisions. They lack the ability to integrate traffic data with structural degradation mechanisms for analysis, making it difficult to timely identify local structural degradation trends, especially in high-frequency and heavy-load areas. At the same time, maintenance strategies often adopt a unified solution, making it difficult to accurately locate grid-level risk areas. This results in overload-sensitive areas being ignored and normal areas being over-intervened, resulting in increased maintenance costs and uncertain repair results.

[0004] The above-mentioned deficiencies are mainly due to the weak perception of road operating conditions by traditional asphalt pavement maintenance strategy generation methods, the lack of modeling capabilities for the evolution mechanism of overload-induced damage, and the lack of a closed-loop design for data-driven, zoning decision-making, and feedback optimization at the strategy response level. In particular, in logistics channels and urban high-traffic arterial roads where heavy industrial vehicles pass through in large numbers, frequent and high-intensity axle load traffic will cause rapid expansion of microcracks in the asphalt structural layer, continuous deepening of rutting, and early fatigue damage. If not identified and intervened in time, it is very easy to evolve into permanent structural damage. In addition, the lack of a life decay prediction mechanism and scheduling classification standards makes it difficult to efficiently allocate construction resources. Construction periods often overlap with overload peaks, missing the optimal repair window, further exacerbating asphalt pavement failure and maintenance delays. Summary of the Invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method and system for intelligently generating asphalt pavement maintenance strategies, which solve the problems in the above-mentioned background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for intelligently generating an asphalt pavement maintenance strategy, comprising the following steps: S1. Extract traffic analysis data and traffic load data from the multi-source traffic monitoring system, and analyze the level of pavement structural damage within each grid cell through laser scanning and image intelligent recognition algorithms; S2. Analyze the road condition risk level of each grid unit under the combined effects of traffic load and damage based on the level of pavement structural damage, divide the risk response area, and generate life loss analysis instructions; S3. After receiving the life loss analysis instruction, analyzing the accelerated decline trend of the pavement structure deterioration of each grid unit in the risk response area to determine the life loss value of each grid unit in the risk response area; S4. Generate a pavement maintenance strategy of corresponding level according to the life reduction value of each grid unit in the risk response area.

[0007] Preferably, the specific steps of S1 include: S11. Perform feature recognition on relevant data recorded in the multi-source traffic monitoring system to obtain traffic analysis data information and traffic load data information, specifically: Taking urban asphalt roads as target roads, combined with the road GIS mapping model, the target road area is projected with geometric coordinates. Regular grid segmentation is performed within the target road range according to the set fixed grid scale. The continuous target road area is divided into multiple rectangular grid areas, each of which is defined as a grid unit, to obtain a number of grid units. The ETC weight-based charging system deployed on the road surface of the toll point will automatically weigh vehicles passing through the toll point according to the set statistical cycle, and screen out overloaded vehicles; The filtered heavy-loaded vehicles are located according to the passing timestamps and associated with the rectangular grid areas divided in the road GIS mapping model to obtain traffic analysis data information, which includes the number of heavy-loaded vehicles passing through each grid cell; A deep learning convolutional neural network model is used to perform real-time local feature recognition on vehicle images captured by the roadside video acquisition system, obtaining multi-dimensional appearance features such as vehicle outline, length, height, and axle spacing. The obtained results are then matched with known vehicle models, corresponding axle numbers, and load types in a historical calibration sample library. A regression inference algorithm is then used to obtain the single axle load value. The passage position of each vehicle is mapped to the corresponding grid cell in the road network GIS model. According to the set statistical period, the axle load values ​​of all vehicles traveling in each grid cell are summarized and averaged to obtain traffic axle load data information. The traffic axle load data information includes the vehicle axle load value of each grid cell.

[0008] Preferably, S12, based on laser scanning and image intelligent recognition algorithm, analyze the level of pavement structural damage in each grid unit, specifically: Based on the acquired grid cells, a mobile road damage scanning system is used to collect road surface spatial structure information and generate raw point cloud data. In the preprocessing stage, noise filtering, coordinate fitting, and contour projection are performed on the raw point cloud data to obtain two-dimensional projection data for each grid cell. Texture directionality analysis and local binarization processing are performed on the image information in the grid area of ​​the two-dimensional projection data of each grid cell to identify the linear crack structure, and the total length, maximum width and number of cracks in each grid cell are extracted using a gradient filter; According to the vertical distance difference between the road surface reference plane and the lowest point in the original point cloud data, the maximum rutting depth value of each grid unit is obtained; The total crack length, maximum crack width, number of cracks, and maximum rutting depth of each grid unit are correlated. After dimensionless processing, the pavement structural damage level within each grid unit is analyzed to obtain the structural damage coefficient of each grid unit, which is specifically: Where, represents the structural damage coefficient of the i-th grid cell, represents the total length of the crack in the i-th grid cell, represents the maximum width of the crack in the i-th grid cell, represents the number of cracks in the i-th grid cell, Indicates the maximum rutting depth of the i-th grid cell.

[0009] Preferably, the specific steps of S2 include: S21. Determine the overload impact level of each grid cell based on the traffic analysis data and traffic load data, specifically: Perform feature recognition on traffic analysis data and traffic load data to screen out the maximum number of heavy-loaded vehicles passing through and the maximum axle load of vehicles; According to the maximum number of heavy-loaded vehicles and the maximum axle load of the vehicles screened out, combined with the traffic analysis data and traffic load data, the heavy-load impact level of each grid unit is determined. Specifically, Where, represents the overload influence coefficient of the i-th grid cell, represents the number of heavy-loaded vehicles passing through the i-th grid cell, represents the vehicle axle load value of the i-th grid cell, Indicates the maximum number of heavy-load vehicles passing through. Indicates the maximum axle load of the vehicle.

[0010] Preferably, S22, analyzing the road condition risk level of each grid unit under the superposition effect of traffic load and damage, specifically: According to the heavy load influence of each grid unit and the level of pavement structural disease in the corresponding grid unit, the road condition risk level of each grid unit under the superposition effect of traffic load and damage is analyzed, and the road condition risk integer value of each grid unit is obtained. Specifically, Where, represents the integer value of the road risk of the i-th grid cell, represents the structural damage coefficient of the i-th grid cell, represents the overload influence coefficient of the i-th grid cell, Indicates the floor symbol, and takes the largest integer smaller than itself as the calculation result.

[0011] Preferably, S23, based on the acquired road condition risk integer value of each grid cell, a corresponding strategy guidance identification code is marked on the grid cell corresponding to the current road condition risk integer value, and a corresponding life loss analysis instruction is generated, specifically: When the road condition risk integer value of the corresponding grid cell is greater than the risk integer threshold, it indicates that the road condition of the grid cell corresponding to the current road condition risk integer value is abnormal. At this time, the grid cell corresponding to the current road condition risk integer value is divided into a risk response area, marked with a strategy guidance identification code, and a life loss analysis instruction is generated; When the road condition risk integer value of the corresponding grid cell is less than or equal to the risk integer threshold, it indicates that the road condition of the grid cell corresponding to the current road condition risk integer value is normal. At this time, the grid cell corresponding to the current road condition risk integer value is divided into a normal observation area, and a secondary pavement maintenance strategy is generated.

[0012] Preferably, the specific steps of S3 include: S31. After receiving the life loss analysis instruction, the correlation between the heavy load impact of each grid unit in the risk response area and the pavement structural disease level in the corresponding grid unit is analyzed, and the coupling effect coefficient of each grid unit in the risk response area is obtained. Specifically, Where, represents the coupling effect coefficient of the jth grid cell in the risk response area, represents the structural damage coefficient of the jth grid cell in the risk response area, represents the overload influence coefficient of the jth grid cell in the risk response area, where A sensitivity factor that represents the overload influence coefficient.

[0013] Preferably, S32, the coupling effect coefficient of each grid unit in the risk response area obtained in S31 is associated with the road condition risk integer value of the corresponding grid unit, the accelerated decline trend of the pavement structure deterioration of each grid unit in the risk response area is analyzed, and the life loss value of each grid unit in the risk response area is determined. Specifically, Where, represents the life loss value of the jth grid cell in the risk response area, represents the coupling effect coefficient of the jth grid cell in the risk response area, represents the integer value of the road condition risk of the jth grid cell in the risk response area, represents the life decay factor, An exponential function representing the growth of reload frequency.

[0014] Preferably, the specific steps of S4 include: S41. Based on the obtained life reduction values ​​of each grid cell in the risk response area, determine whether the corresponding grid cell in the risk response area has a rapid life reduction trend due to deterioration of the asphalt pavement structure under the influence of heavy loads, and classify the corresponding grid cells into regions to generate a pavement maintenance strategy of the corresponding level, specifically: When the life reduction value of the corresponding grid cell is greater than or equal to the life reduction threshold, it indicates that the corresponding grid cell in the risk response area has a rapid reduction trend in the life of the asphalt pavement structure due to heavy load, and the corresponding grid cell is divided into a maintenance priority area to generate a first-level pavement maintenance strategy. When the service life reduction value of the corresponding grid cell is less than the service life reduction threshold, it means that the corresponding grid cell in the risk response area does not have a rapid service life reduction trend due to the deterioration of the asphalt pavement structure under the influence of heavy load, and the corresponding grid cell is divided into a general observation area to generate a secondary pavement maintenance strategy; S42. The specific contents of the generated primary and secondary pavement maintenance strategies are as follows: The primary pavement maintenance strategy includes: When maintaining priority areas, avoid periods of high-frequency, heavy-load traffic; employ structural reinforcement measures, including overlaying medium-thick layers, localized excavation and patching with simultaneous sealing, and deep crack grouting repair; and generate a detailed list of construction parameters, including recommended material mixes, equipment configuration recommendations, construction windows, and energy consumption scheduling plans. The secondary pavement maintenance strategy includes: when maintaining the pavement in the general observation area, adopt surface functional maintenance and structural light intervention measures, including micro-surfacing, fog sealing, shallow crack sealing treatment, and layout of road condition monitoring points; at the same time, continue to carry out periodic monitoring.

[0015] An intelligent generation system for asphalt pavement maintenance strategies, including a multi-source acquisition module, a damage quantification module, a lifespan reduction analysis module, and a strategy generation module; The multi-source acquisition module is used to extract traffic analysis data and traffic load data from the multi-source traffic monitoring system, and analyze the level of pavement structural damage within each grid unit through laser scanning and image intelligent recognition algorithms; The damage quantification module is used to analyze the road condition risk level of each grid unit under the superposition effect of traffic load and damage based on the level of pavement structural disease, divide the risk response area, and generate life loss analysis instructions; The life loss analysis module is used to analyze the accelerated decline trend of the pavement structure deterioration of each grid unit in the risk response area after receiving the life loss analysis instruction, so as to determine the life loss value of each grid unit in the risk response area; The strategy generation module is used to generate a pavement maintenance strategy of the corresponding level according to the life loss value of each grid unit in the risk response area.

[0016] The present invention provides a method and system for intelligently generating asphalt pavement maintenance strategies, which have the following beneficial effects: (1) Through the fusion of multi-source traffic monitoring data, quantification of structural disease identification, modeling of heavy load-induced impacts and life assessment technical paths, a three-dimensional coupling mechanism of traffic load, structural status and life attenuation is constructed, realizing a closed loop of the entire process from raw data collection to intelligent strategy generation. Compared with traditional maintenance schemes based on regular inspections and manual experience, this method has the advantages of strong real-time performance, precise regional identification and dynamically adjustable strategy response. It can realize risk identification and classification control at the grid level, improve the targeting of maintenance strategies and the accuracy of resource allocation, effectively reduce maintenance costs, delay structural failures and improve the reliability of road network operation.

[0017] (2) By establishing a traffic load intensity identification mechanism based on the integration of traffic analysis data and traffic load data, a joint calculation model of heavy load influence coefficient and structural damage coefficient is constructed, and then a road condition risk integer value with grade division characteristics is generated, realizing the spatial and hierarchical expression of the road structure status; this mechanism breaks through the traditional coarse-grained logic of setting maintenance response according to road grade or fixed section, and enables targeted analysis of high-frequency heavy load areas, accurately identifying heavy load sensitive grid units as risk response areas, and issuing life loss analysis instructions in a timely manner, thereby avoiding maintenance delays, misjudgments or waste of resources, and significantly improving the response identification capability of risk areas and the scientific decision-making level of maintenance strategies.

[0018] (3) Based on the joint judgment model of coupling effect coefficient and risk integer value, the structural disease index and overload induced influence are comprehensively considered, and the life attenuation factor and overload frequency growth function are combined to dynamically calculate the life reduction value of each grid unit in the risk response area, thereby realizing the quantitative prediction and early warning identification of the structural degradation trend; this mechanism effectively makes up for the problem of lack of modeling of the structural life change process and lack of dynamic evaluation capability in the existing maintenance scheme, and can accurately identify high-risk areas with rapidly reduced life, and divide them into maintenance priority areas and general observation areas according to the degree of rapid reduction in life, providing decision support for the subsequent matching of maintenance intensity and construction technology, and improving the efficiency of maintenance resource allocation and the timeliness of intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a flow chart of an intelligent generation method for asphalt pavement maintenance strategy according to the present invention; Figure 2 This is a block diagram of an intelligent generation system for asphalt pavement maintenance strategies according to the present invention; Figure 3 This is a logical thinking diagram of an intelligent generation method for asphalt pavement maintenance strategy according to the present invention; Figure 4 Schematic diagram of the multi-source traffic monitoring system in S1 of the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] Example 1 See also Figure 1 and Figure 3 The present invention provides an intelligent generation method for asphalt pavement maintenance strategy, comprising the following steps: S1. Extract traffic analysis data and traffic load data from the multi-source traffic monitoring system, and analyze the level of pavement structural damage within each grid cell through laser scanning and image intelligent recognition algorithms; S2. Analyze the road condition risk level of each grid unit under the combined effects of traffic load and damage based on the level of pavement structural damage, divide the risk response area, and generate life loss analysis instructions; S3. After receiving the life loss analysis instruction, analyzing the accelerated decline trend of the pavement structure deterioration of each grid unit in the risk response area to determine the life loss value of each grid unit in the risk response area; S4. Generate a pavement maintenance strategy of corresponding level according to the life reduction value of each grid unit in the risk response area.

[0022] In this embodiment, by constructing a complete chain of traffic load perception, structural disease identification, life decline analysis and strategy output feedback, intelligent and data-driven management of asphalt pavement status is achieved; compared with the traditional strategy formulation method that relies on manual inspection and static grade division, this method can not only realize the raster analysis of roads in space and improve the recognition accuracy, but also logically realize the superposition evaluation of traffic pressure and structural diseases, significantly enhancing the dynamic perception ability of road operation status; in specific implementation, the traffic load characteristics of each grid unit are accurately extracted through the multi-source traffic data fusion algorithm, and combined with the laser point cloud and image Identification methods obtain structural information of defects, enabling real-time identification and quantification of early defects such as microcracks and rutting. Furthermore, through the life decay trend analysis model, the evolution rate of structural degradation is predicted, and a life loss value is scientifically generated for each grid, thereby promoting the output of maintenance strategies with precise response levels and matching intervention intensity. Its core advantage lies in the ability to perform graded responses based on the operating load and degradation characteristics of local grid units, realize differentiated, prioritized, and quantitative maintenance decisions, improve resource scheduling efficiency, avoid the problems of over-maintenance and missed repairs in traditional solutions, and have significant real-time, precise, and strategic guidance value.

[0023] Example 2 Please refer to Figure 1 and Figure 4 , specifically: S1 specific steps include: S11. Perform feature recognition on relevant data recorded in the multi-source traffic monitoring system to obtain traffic analysis data information and traffic load data information, specifically: Taking urban asphalt roads as target roads, combined with the road GIS mapping model, the target road area is projected with geometric coordinates. Regular grid segmentation is performed within the target road range according to the set fixed grid scale. The continuous target road area is divided into multiple rectangular grid areas, each of which is defined as a grid unit, to obtain a number of grid units. The road GIS mapping model refers to a spatial projection model that digitally expresses the actual spatial layout, coordinate information and attribute data of urban roads and constructs it in a geographic information system. It is usually generated through vectorization and coordinate alignment of high-precision remote sensing images, road CAD drawings or field measurement data.

[0024] The ETC weight-based charging system deployed on the road surface of the toll point will automatically weigh vehicles passing through the toll point according to the set statistical cycle, and screen out overloaded vehicles; The filtered heavy-loaded vehicles are located according to the passing timestamps and associated with the rectangular grid areas divided in the road GIS mapping model to obtain traffic analysis data information, which includes the number of heavy-loaded vehicles passing through each grid cell; The number of heavy-loaded vehicles passing through each grid cell refers to the total number of vehicles identified as heavy-loaded vehicles passing through the grid cell within a set statistical period. It is a basic indicator for measuring the extent of the impact of heavy-load traffic in an area. Heavy-loaded vehicles are usually identified and screened through the ETC weighing system deployed at road toll points. Their weight data reaches or exceeds the set heavy-load threshold and is recorded as a valid sample. Combining the passage timestamp and vehicle travel trajectory, the road GIS mapping model accurately locates the position of each vehicle in the divided rectangular grid area. The number of heavy-loaded vehicles falling into each grid cell is counted to obtain the number of heavy-loaded vehicles passing through the corresponding grid cell. A deep learning convolutional neural network model is used to perform real-time local feature recognition on vehicle images captured by the roadside video acquisition system, obtaining multi-dimensional appearance features such as vehicle outline, length, height, and axle spacing. The obtained results are then matched with known vehicle models, corresponding axle numbers, and load types in a historical calibration sample library. A regression inference algorithm is then used to obtain the single axle load value. The passage position of each vehicle is mapped to the corresponding grid cell in the road network GIS model. According to the set statistical period, the axle load values ​​of all vehicles traveling in each grid cell are summarized and averaged to obtain traffic axle load data information. The traffic axle load data information includes the vehicle axle load value of each grid cell.

[0025] The vehicle axle load value for each grid cell refers to the average of the estimated axle loads of all vehicles passing through the grid cell within a set statistical period, reflecting the average axial load level borne by the roads in the area over a period of time. Vehicle images are acquired through a roadside video acquisition system, and multi-dimensional vehicle appearance features are extracted using a deep learning convolutional neural network model. These features are then matched with the vehicle model and load parameters in a historical calibration sample library, and the axle load value of each vehicle is calculated using a regression inference algorithm. The vehicle's passage timestamp and spatial location data are used to locate it to the corresponding grid cell in the GIS road mapping model, and the axle load values ​​of all vehicles within the cell are aggregated and averaged to form the vehicle axle load value for each grid cell. This indicator is used to quantify the long-term cumulative impact of traffic load on local road structures. It is one of the core input parameters for subsequent heavy load impact assessment, road condition risk analysis, and life loss modeling, and plays an important role in analytical and decision-making support. Specifically, S12 analyzes the pavement structural damage level within each grid unit based on laser scanning and image intelligent recognition algorithms, specifically: Based on the acquired grid cells, a mobile road damage scanning system is used to collect road surface spatial structure information and generate raw point cloud data. In the preprocessing stage, noise filtering, coordinate fitting, and contour projection are performed on the raw point cloud data to obtain two-dimensional projection data for each grid cell. Texture directionality analysis and local binarization processing are performed on the image information in the grid area of ​​the two-dimensional projection data of each grid cell to identify the linear crack structure, and the total length, maximum width and number of cracks in each grid cell are extracted using a gradient filter; The total crack length, maximum crack width, and number of cracks in each grid cell are important geometric characteristic parameters reflecting structural defects on the road surface. The total crack length represents the cumulative length of all identifiable cracks within the grid cell, the maximum crack width represents the maximum width of a single crack, and the number of cracks represents the number of cracks detected in that grid cell. These three parameters are extracted from the raw point cloud data and image information acquired by the mobile road defect scanning system. First, the point cloud data is subjected to noise filtering and geometric fitting to generate a two-dimensional projection image. Texture directionality analysis and local binarization are then performed on the image to identify linear crack structures with continuity and geometric characteristics. Subsequently, an image gradient filter is used to accurately extract the key indicators of crack length, width, and number. These parameters are used to quantify the degree of structural damage to the road and serve as important inputs for the subsequent calculation of the structural damage coefficient. They help accurately identify the severity and spatial distribution characteristics of the defect and provide an objective basis for the development of intelligent maintenance strategies. According to the vertical distance difference between the road surface reference plane and the lowest point in the original point cloud data, the maximum rutting depth value of each grid unit is obtained; The maximum rutting depth value for each grid cell refers to the maximum value obtained within each grid area by analyzing the vertical distance difference between the road surface reference plane and the lowest point in the original point cloud data for that area. It is used to characterize the degree of local indentation caused by repeated vehicle loads and is a key indicator for evaluating asphalt pavement structural defects, especially the cumulative effects of plastic deformation and fatigue. This value is obtained by relying on high-precision three-dimensional point cloud data collected by a mobile road defect scanning system. First, a flat reference plane of an ideal road surface is established as a reference. Then, the lowest elevation point is identified in the point cloud data of the grid cell. The vertical distance between this point and the reference plane is calculated and the maximum value is taken to obtain the maximum rutting depth. This parameter is used in the calculation of the structural damage coefficient in this method to help determine the long-term structural load-bearing capacity degradation trend of the road surface under heavy vehicle traffic, thereby providing important support for subsequent risk classification and maintenance strategy formulation. The total crack length, maximum crack width, number of cracks, and maximum rutting depth of each grid unit are correlated. After dimensionless processing, the pavement structural damage level within each grid unit is analyzed to obtain the structural damage coefficient of each grid unit, which is specifically: Where, represents the structural damage coefficient of the i-th grid cell, represents the total length of the crack in the i-th grid cell, represents the maximum width of the crack in the i-th grid cell, represents the number of cracks in the i-th grid cell, Indicates the maximum rutting depth of the i-th grid cell.

[0026] The formula for calculating the structural damage coefficient integrates four key disease parameters of each grid cell: total crack length, maximum crack width, number of cracks, and maximum rutting depth. After dimensionless normalization and logarithmic transformation, it constructs a comprehensive quantitative expression of the intensity of pavement structural disease. This formula is logically designed to address the weak structural state perception ability, low disease identification accuracy, and lack of unified quantitative evaluation standards mentioned in the background technology. Through the multi-dimensional fusion of point cloud and image recognition results, it not only improves the comprehensiveness of structural disease identification, but also ensures the scale comparability and consistency of evaluation standards between different indicators. As one of the core basic variables of the system, the structural damage coefficient is directly involved in subsequent key processes such as heavy load coupling evaluation, road risk level classification, and life decay trend deduction. It plays a connecting role and is a key bridge to achieve a closed loop from perception to decision-making logic, significantly improving the accuracy and practicality of the system in local risk identification and strategy classification output. In this embodiment, an intelligent perception mechanism with multi-source traffic monitoring data fusion and high-precision identification of structural defects as its core is constructed in S1. Its main advantage is the high-resolution spatial mapping and dynamic modeling capabilities of traffic load and pavement structure status; by geometrically projecting the urban asphalt road area based on the GIS model and dividing it into multiple grid units at a fixed scale, standardized management of road space and data attribution of traffic load are achieved, so that the traffic frequency and axle load data can be accurately collected on each grid unit; in addition, the integration of the ETC weighing system and the deep learning video recognition algorithm can not only identify the weight grade and number of axles of passing vehicles, but also infer the axle load level of a single vehicle by comparing the appearance characteristics with the historical sample library, thereby avoiding the problem of misjudgment of traditional single sensors and improving the reliability of traffic load information. and stability; at the same time, combined with the point cloud and image information collected by mobile scanning equipment, high-precision quantitative identification of typical diseases such as grid-level cracks and rutting is achieved through contour projection, texture directionality analysis and crack width and length extraction; the structural damage coefficient constructed by the dimensionless normalization processing method can not only reflect the intensity of microstructural diseases, but also provide key parameter support for subsequent coupling analysis and strategic response; the value of this step lies in its integration of multi-source data drive, structural accuracy identification and spatial mapping binding composite perception capabilities, so that pavement maintenance has the basic guarantee of spatial refinement, load reality and damage quantification from the original data layer, laying a solid foundation for the subsequent construction of response mechanisms, generation of life trends and formulation of grade strategies, and significantly improving adaptability and effectiveness in complex road operation scenarios.

[0027] Example 3 Please refer to Figure 1 , specifically: S2 specific steps include: S21. Determine the overload impact level of each grid cell based on the traffic analysis data and traffic load data, specifically: Perform feature recognition on traffic analysis data and traffic load data to screen out the maximum number of heavy-loaded vehicles passing through and the maximum axle load of vehicles; According to the maximum number of heavy-loaded vehicles and the maximum axle load of the vehicles screened out, combined with the traffic analysis data and traffic load data, the heavy-load impact level of each grid unit is determined. Specifically, Where, represents the overload influence coefficient of the i-th grid cell, represents the number of heavy-loaded vehicles passing through the i-th grid cell, represents the vehicle axle load value of the i-th grid cell, Indicates the maximum number of heavy-loaded vehicles passing through. Indicates the maximum axle load of the vehicle.

[0028] The formula for calculating the heavy load impact coefficient integrates the number of heavy-loaded vehicles passing through each grid cell and the vehicle axle load value, and normalizes it with the global maximum frequency and maximum axle load value to construct a standardized heavy load impact index that can be used for spatial comparison and grade determination. This formula provides a solution to the problems mentioned in the background technology, such as the difficulty in quantifying the intensity of local heavy-load traffic in real time and the lack of modeling capabilities for the impact of traffic loads on structural risks. By integrating the dual factors of frequency and load intensity and introducing normalization logic to avoid the impact of data fluctuations in different regions, the traffic load status of different grids is objectively, comparable, and dynamically expressed. As one of the key input variables used by the system to identify risk response areas, deduce life decay trends, and generate trigger strategies, the heavy load impact coefficient can effectively reflect the distribution characteristics of heavy-load traffic in the spatial dimension and the load pressure concentration effect, and is the core bridge for realizing the construction logic of load-driven maintenance strategies.

[0029] Specifically, S22 analyzes the road risk level of each grid cell under the superposition effect of traffic load and damage, specifically: According to the heavy load influence of each grid unit and the level of pavement structural disease in the corresponding grid unit, the road condition risk level of each grid unit under the superposition effect of traffic load and damage is analyzed, and the road condition risk integer value of each grid unit is obtained. Specifically, Where, represents the integer value of the road risk of the i-th grid cell, represents the structural damage coefficient of the i-th grid cell, represents the overload influence coefficient of the i-th grid cell, Indicates the floor symbol, and takes the largest integer smaller than itself as the calculation result.

[0030] The road condition risk integer value calculation formula constructs a unified standardized integer indicator for zoning judgment and strategy triggering by weighting and fusing the structural damage coefficient and overload impact coefficient of the grid unit, performing linear combination and flooring processing. This formula directly addresses the shortcomings of the background technology, which is difficult to characterize the comprehensive risk level after the superposition of traffic load and disease status, and lacks a feasible classification and response judgment basis. By setting a fixed proportional factor to strengthen the weight influence of overload and structural factors respectively, it not only retains the cumulative effect of the two key risk sources, but also has comparability and execution adaptability. It is a bridge variable that connects the assessment model and strategy division logic. The road condition risk integer value serves as the core driving parameter for risk response zone determination, strategy guidance code generation, and life analysis instruction triggering. It can express the current comprehensive risk level of the road in a standardized and hierarchical form, thereby realizing an integrated closed loop from disease identification, load assessment to risk level, significantly improving the automation level of road operation status analysis and the accuracy of maintenance strategy response. Specifically, S23, based on the acquired road condition risk integer value of each grid cell, a corresponding strategy guidance identification code is marked on the grid cell corresponding to the current road condition risk integer value, and a corresponding life loss analysis instruction is generated, specifically: When the road condition risk integer value of the corresponding grid cell is greater than the risk integer threshold, it indicates that the road condition of the grid cell corresponding to the current road condition risk integer value is abnormal. At this time, the grid cell corresponding to the current road condition risk integer value is divided into a risk response area, marked with a strategy guidance identification code, and a life loss analysis instruction is generated; When the road condition risk integer value of the corresponding grid cell is less than or equal to the risk integer threshold, it indicates that the road condition of the grid cell corresponding to the current road condition risk integer value is normal. At this time, the grid cell corresponding to the current road condition risk integer value is divided into a normal observation area, and a secondary pavement maintenance strategy is generated.

[0031] The risk integer threshold is based on the need to differentiate between risk levels formed by weighted synthesis of the structural damage coefficient and the heavy load impact coefficient. As the boundary between the risk response zone and the general observation zone, the risk integer threshold is derived from a comprehensive definition of historical data statistical analysis and empirical modeling results. The risk integer threshold is in the upper-middle critical section of the integer score range, which can effectively identify grid cells with high disease severity or frequent heavy load traffic, while avoiding false alarms in areas with minor damage or low loads, thereby ensuring the precise allocation of maintenance resources. In this embodiment, by constructing a joint assessment mechanism of traffic load and structural defects, dynamic identification of the road condition risk level of asphalt pavement grid units and accurate division of response areas are achieved, which has significant data fusion capabilities, risk quantification capabilities and strategy triggering capabilities; based on the traffic analysis data and traffic axle load data obtained from S1, a normalized heavy load influence coefficient model is established, which can standardize the number of heavy loads and axle load intensity of each grid unit, solving the problem of traditional difficulty in quantifying the spatial distribution of heavy load pressure; secondly, the heavy load influence level is functionally coupled with the structural damage coefficient to generate a road condition risk integer value with structural load superposition risk significance, and the data stability and comparability are ensured by rounding down. It effectively constructs a hierarchical response identification logic for the execution layer; on this basis, it further automatically identifies high-risk areas as risk response areas through preset risk integer thresholds, and assigns corresponding grid strategy guidance identification codes, while generating life analysis instructions as strategy calculation trigger signals, forming a regional identification and strategy activation linkage mechanism driven by dynamic parameters; the advantage of this step is that it no longer relies on manual segmentation or administrative divisions to set the repair range, but through the construction and dynamic calculation of precise indicators at the grid level, it realizes the accurate zoning identification of the evolution trend of high-frequency and heavy-load-induced diseases, improves the spatial accuracy and judgment timeliness of strategy response, and is especially suitable for refined maintenance deployment under complex traffic structures such as logistics trunk roads and urban main roads.

[0032] Example 4 Please refer to Figure 1 , specifically: S3 specific steps include: S31. After receiving the life loss analysis instruction, the correlation between the heavy load impact of each grid unit in the risk response area and the pavement structural disease level in the corresponding grid unit is analyzed, and the coupling effect coefficient of each grid unit in the risk response area is obtained. Specifically, Where, represents the coupling effect coefficient of the jth grid cell in the risk response area, represents the structural damage coefficient of the jth grid cell in the risk response area, represents the overload influence coefficient of the jth grid cell in the risk response area, where A sensitivity factor that represents the overload influence coefficient.

[0033] The calculation formula for the coupling effect coefficient multiplies the structural damage coefficient of each grid cell in the risk response area with the heavy load influence coefficient, and introduces the heavy load influence sensitivity factor to form a nonlinear enhancement term, thereby constructing an indicator reflecting the coordinated degradation trend of load and disease. This formula provides a solution to the core problems raised in the background technology, such as the failure to effectively characterize the coupling relationship between heavy load traffic and structural diseases and the difficulty in identifying the accelerated area of ​​structural deterioration. It integrates the degree of structural disease with the heavy load induced effect into the model, avoiding the deviation caused by deterioration judgment from only a single perspective. At the same time, through the adjustable sensitivity factor of the heavy load influence coefficient, it introduces a differentiated expression of sensitivity to different road types or loads, and has cross-scenario adaptability. The coupling effect coefficient plays a core role in bridging the interaction mechanism between structural state and load pressure. It is the basic variable for life decay trend modeling and life loss value calculation. It can accurately identify the spatial location of accelerated accumulation of structural risks and is a key supporting parameter for building a predictive and data-driven maintenance strategy system. Specifically, S32, correlating the coupling effect coefficient of each grid unit in the risk response area obtained in S31 with the road condition risk integer value of the corresponding grid unit, analyzing the accelerated decline trend of the pavement structure deterioration of each grid unit in the risk response area, and determining the life reduction value of each grid unit in the risk response area. Specifically, Where, represents the life loss value of the jth grid cell in the risk response area, represents the coupling effect coefficient of the jth grid cell in the risk response area, represents the integer value of the road condition risk of the jth grid cell in the risk response area, represents the life decay factor, An exponential function representing the growth of reload frequency.

[0034] The life reduction value calculation formula constructs an indicator that can quantify the accelerated degradation trend of the structure by functionally linking the coupling effect coefficient of the grid unit with the integer value of its road condition risk and introducing the logarithmic amplification mechanism of the overload frequency growth. This effectively responds to the shortcomings of the background technology in lacking the ability to model the evolution of structural life and the difficulty in predicting the rate of structural decline. The formula combines the life attenuation factor to regulate the coupling effect results and The nonlinear amplification effect of rising risk levels on life loss is particularly useful for capturing the potential for rapid disease expansion in heavy-load, high-frequency areas. Life loss is a key parameter used to determine maintenance response levels. It not only expresses the current degradation intensity of the structure but also predicts the reduction in lifespan over the next period of time, serving as a critical bridge in the transition from state identification to trend prediction. This indicator can accurately identify rapidly deteriorating grid cells, prioritize resource allocation, and develop higher-level intervention strategies, significantly improving the proactive and forward-looking nature of maintenance strategies and avoiding structural disasters caused by delayed intervention. In this embodiment, by constructing a coupling effect model of structural disease and heavy load influence, and further introducing a life decay deduction mechanism, the dynamic quantification of the structural deterioration trend of each grid unit in the risk response area and the accurate assessment of the life loss degree are achieved, which significantly improves the foresight and scientific nature of the maintenance strategy formulation; specifically, after receiving the life loss analysis instruction generated by S2, the structural damage coefficient and heavy load influence coefficient of each risk response grid unit are used as input to comprehensively calculate its coupling effect coefficient, reflecting the intrinsic strength and sensitivity of the grid's disease evolution under traffic stress. By introducing the heavy load influence sensitivity factor, the response elasticity of different types of roads to heavy load pressure can be adjusted, thereby enhancing the adaptability of the model; on this basis, The coupling effect coefficient is functionally associated with the road condition risk value, and the life decay factor and exponential growth function are introduced to deduce the life reduction value of each grid unit, that is, the life loss years, which is used to express the acceleration of its structural life decline. The core advantage of this step is that it can not only evaluate the current structural state, but also predict its future evolution trend. It is a key logical bridge from static identification to dynamic deduction. Especially in the context of nonlinear development of heavy load-induced damage, the life reduction value can be used as an important criterion for fine-tuning the matching of maintenance level, timing and reinforcement strength, which significantly improves the accuracy and initiative of strategy generation, avoids the traditional passive mode of repairing as soon as it is discovered or repairing on schedule, and constructs an intelligent predictive maintenance path driven by structural evolution trend.

[0035] Example 5 Please refer to Figure 1 , specifically: S4 specific steps include: S41. Based on the obtained life reduction values ​​of each grid cell in the risk response area, determine whether the corresponding grid cell in the risk response area has a rapid life reduction trend due to deterioration of the asphalt pavement structure under the influence of heavy loads, and classify the corresponding grid cells into regions to generate a pavement maintenance strategy of the corresponding level, specifically: When the life reduction value of the corresponding grid cell is greater than or equal to the life reduction threshold, it indicates that the corresponding grid cell in the risk response area has a rapid reduction trend in the life of the asphalt pavement structure due to heavy load, and the corresponding grid cell is divided into a maintenance priority area to generate a first-level pavement maintenance strategy. When the service life reduction value of the corresponding grid cell is less than the service life reduction threshold, it means that the corresponding grid cell in the risk response area does not have a rapid service life reduction trend due to the deterioration of the asphalt pavement structure under the influence of heavy load, and the corresponding grid cell is divided into a general observation area to generate a secondary pavement maintenance strategy; The life reduction threshold is a critical value set based on empirical modeling of structural degradation rates and life reduction trends and risk response level classification standards. The life reduction value reflects the life loss intensity of grid cells under the influence of heavy loads and superimposed diseases, and the life reduction threshold means identifying grid cells whose structural life loss reaches a high-risk critical level as areas with rapid degradation trends and classifying them as maintenance priority areas, thereby triggering a first-level maintenance strategy. The advantage of the life reduction threshold is that the turning point of the nonlinear growth section in the calculated life reduction value can accurately distinguish units with sudden structural degradation characteristics, effectively avoiding slightly deteriorated areas from being misjudged as high-level intervention targets. At the same time, it also ensures the sensitivity of the response mechanism and the precise deployment of strategic resources. While controlling operation and maintenance costs, it has the ability to intercept high-risk evolution in advance, thereby improving overall maintenance efficiency and road service safety. S42. The specific contents of the generated primary and secondary pavement maintenance strategies are as follows: The primary pavement maintenance strategy includes: When maintaining priority areas, avoid periods of high-frequency, heavy-load traffic; employ structural reinforcement measures, including overlaying medium-thick layers, localized excavation and patching with simultaneous sealing, and deep crack grouting repair; and generate a detailed list of construction parameters, including recommended material mixes, equipment configuration recommendations, construction windows, and energy consumption scheduling plans. The secondary pavement maintenance strategy includes: when maintaining the pavement in the general observation area, adopt surface functional maintenance and structural light intervention measures, including micro-surfacing, fog sealing, shallow crack sealing treatment, and layout of road condition monitoring points; at the same time, continue to carry out periodic monitoring.

[0036] In this embodiment, a hierarchical, responsive pavement maintenance strategy generation mechanism is constructed in S4 based on the quantitative assessment results of grid cell life loss values, achieving an efficient closed-loop conversion from the quantitative results of structural evolution to strategy implementation plans. This step distinguishes the urgency of structural decline in grid areas by setting life loss value thresholds, dividing road areas in different states into maintenance priority areas and general observation areas, matching them with primary and secondary maintenance strategies respectively. This regional division based on structural decline rate makes maintenance resource allocation more scientific and precise, breaking through the traditional coarse-grained operation method of dividing by administrative divisions or cycle time. The primary pavement maintenance strategy adopts structural repair and reinforcement measures for areas with rapid structural deterioration and high frequency of heavy loads, and combines construction window avoidance arrangements and energy consumption scheduling recommendations to avoid peak construction interference and energy efficiency waste. The secondary maintenance strategy implements functional coverage or mild intervention in areas where structural degradation is not yet serious, reducing investment intensity and delaying the spread of diseases. In particular, this step also incorporates a mechanism for automatically generating a construction parameter list, enabling a refined transition of maintenance strategies from macro-instructions to micro-construction support, significantly improving the plan's operability and implementation efficiency. In summary, this step not only achieves differentiated and data-driven maintenance responses, but also offers significant advantages in terms of resource conservation, construction efficiency improvement, and strategy accuracy. It is a key link in the intelligent, closed-loop road maintenance decision-making system.

[0037] Example 6 Please refer to Figure 1 and Figure 2 ,Specifically: An intelligent generation system for asphalt pavement maintenance strategy ,includes a multi-source acquisition module, a damage quantification module, a life loss analysis module and a ,strategy generation module; The multi-source acquisition module is used to extract traffic analysis data and traffic load data from the multi-source traffic monitoring system, and analyze the level of pavement structural damage within each grid unit through laser scanning and image intelligent recognition algorithms; The damage quantification module is used to analyze the road condition risk level of each grid unit under the superposition effect of traffic load and damage based on the level of pavement structural disease, divide the risk response area, and generate life loss analysis instructions; The life loss analysis module is used to analyze the accelerated decline trend of the pavement structure deterioration of each grid unit in the risk response area after receiving the life loss analysis instruction, so as to determine the life loss value of each grid unit in the risk response area; The strategy generation module is used to generate a pavement maintenance strategy of the corresponding level according to the life loss value of each grid unit in the risk response area.

[0038] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligently generating asphalt pavement maintenance strategies, characterized by: The following steps are involved: S1. Extract traffic analysis data and traffic load data from the multi-source traffic monitoring system, and analyze the level of pavement structural damage within each grid cell through laser scanning and image intelligent recognition algorithms; S2. Analyze the road condition risk level of each grid unit under the combined effects of traffic load and damage based on the level of pavement structural damage, divide the risk response area, and generate life loss analysis instructions; S3. After receiving the life loss analysis instruction, analyzing the accelerated decline trend of the pavement structure deterioration of each grid unit in the risk response area to determine the life loss value of each grid unit in the risk response area; S4. Generate a pavement maintenance strategy of corresponding level according to the life reduction value of each grid unit in the risk response area.

2. The method for intelligently generating an asphalt pavement maintenance strategy according to claim 1, characterized in that: The specific steps of S1 include: S11. Perform feature recognition on relevant data recorded in the multi-source traffic monitoring system to obtain traffic analysis data information and traffic load data information, specifically: Taking urban asphalt roads as target roads, combined with the road GIS mapping model, the target road area is projected with geometric coordinates. Regular grid segmentation is performed within the target road range according to the set fixed grid scale. The continuous target road area is divided into multiple rectangular grid areas, each of which is defined as a grid unit, to obtain a number of grid units. The ETC weight-based charging system deployed on the road surface of the toll point will automatically weigh vehicles passing through the toll point according to the set statistical cycle, and screen out overloaded vehicles; The filtered heavy-loaded vehicles are located according to the passing timestamps and associated with the rectangular grid areas divided in the road GIS mapping model to obtain traffic analysis data information, which includes the number of heavy-loaded vehicles passing through each grid cell; A deep learning convolutional neural network model is used to perform real-time local feature recognition on vehicle images captured by the roadside video acquisition system, obtaining multi-dimensional appearance features such as vehicle outline, length, height, and axle spacing. The obtained results are then matched with known vehicle models, corresponding axle numbers, and load types in a historical calibration sample library. A regression inference algorithm is then used to obtain the single axle load value. The passage position of each vehicle is mapped to the corresponding grid cell in the road network GIS model. According to the set statistical period, the axle load values ​​of all vehicles traveling in each grid cell are summarized and averaged to obtain traffic axle load data information. The traffic axle load data information includes the vehicle axle load value of each grid cell.

3. The method for intelligently generating an asphalt pavement maintenance strategy according to claim 2, characterized in that: S12. Analyze the pavement structural damage level within each grid cell based on laser scanning and image intelligent recognition algorithms. Specifically: Based on the acquired grid cells, a mobile road damage scanning system is used to collect road surface spatial structure information and generate raw point cloud data. In the preprocessing stage, noise filtering, coordinate fitting, and contour projection are performed on the raw point cloud data to obtain two-dimensional projection data for each grid cell. Texture directionality analysis and local binarization processing are performed on the image information in the grid area of ​​the two-dimensional projection data of each grid cell to identify the linear crack structure, and the total length, maximum width and number of cracks in each grid cell are extracted using a gradient filter; According to the vertical distance difference between the road surface reference plane and the lowest point in the original point cloud data, the maximum rutting depth value of each grid unit is obtained; The total crack length, maximum crack width, number of cracks, and maximum rutting depth of each grid unit are correlated. After dimensionless processing, the pavement structural damage level within each grid unit is analyzed to obtain the structural damage coefficient of each grid unit, which is specifically: Where, represents the structural damage coefficient of the i-th grid cell, represents the total length of the crack in the i-th grid cell, represents the maximum width of the crack in the i-th grid cell, represents the number of cracks in the i-th grid cell, Indicates the maximum rutting depth of the i-th grid cell.

4. The method for intelligently generating an asphalt pavement maintenance strategy according to claim 3, characterized in that: The specific steps of S2 include: S21. Determine the overload impact level of each grid cell based on the traffic analysis data and traffic load data, specifically: Perform feature recognition on traffic analysis data and traffic load data to screen out the maximum number of heavy-loaded vehicles passing through and the maximum axle load of vehicles; According to the maximum number of heavy-loaded vehicles and the maximum axle load of the vehicles screened out, combined with the traffic analysis data and traffic load data, the heavy-load impact level of each grid unit is determined. Specifically, Where, represents the overload influence coefficient of the i-th grid cell, represents the number of heavy-loaded vehicles passing through the i-th grid cell, represents the vehicle axle load value of the i-th grid cell, Indicates the maximum number of heavy-loaded vehicles passing through. Indicates the maximum axle load of the vehicle.

5. The method for intelligently generating an asphalt pavement maintenance strategy according to claim 4, characterized in that: S22. Analyze the road risk level of each grid cell under the combined effects of traffic load and damage, specifically: According to the heavy load influence of each grid unit and the level of pavement structural disease in the corresponding grid unit, the road condition risk level of each grid unit under the superposition effect of traffic load and damage is analyzed, and the road condition risk integer value of each grid unit is obtained. Specifically, Where, represents the integer value of the road risk of the i-th grid cell, represents the structural damage coefficient of the i-th grid cell, represents the overload influence coefficient of the i-th grid cell, Indicates the floor symbol, and takes the largest integer smaller than itself as the calculation result.

6. The method for intelligently generating an asphalt pavement maintenance strategy according to claim 5, characterized in that: S23. Based on the acquired road risk integer values ​​of each grid cell, the grid cell corresponding to the current road risk integer value is marked with a corresponding strategy guidance identification code, and a corresponding life loss analysis instruction is generated, specifically: When the road condition risk integer value of the corresponding grid cell is greater than the risk integer threshold, it indicates that the road condition of the grid cell corresponding to the current road condition risk integer value is abnormal. At this time, the grid cell corresponding to the current road condition risk integer value is divided into a risk response area, marked with a strategy guidance identification code, and a life loss analysis instruction is generated; When the road condition risk integer value of the corresponding grid cell is less than or equal to the risk integer threshold, it indicates that the road condition of the grid cell corresponding to the current road condition risk integer value is normal. At this time, the grid cell corresponding to the current road condition risk integer value is divided into a normal observation area, and a secondary pavement maintenance strategy is generated.

7. The method for intelligently generating an asphalt pavement maintenance strategy according to claim 6, characterized in that: The specific steps of S3 include: S31. After receiving the life loss analysis instruction, the correlation between the heavy load impact of each grid unit in the risk response area and the pavement structural disease level in the corresponding grid unit is analyzed, and the coupling effect coefficient of each grid unit in the risk response area is obtained. Specifically, Where, represents the coupling effect coefficient of the jth grid cell in the risk response area, represents the structural damage coefficient of the jth grid cell in the risk response area, represents the overload influence coefficient of the jth grid cell in the risk response area, where A sensitivity factor that represents the overload influence coefficient.

8. The method for intelligently generating an asphalt pavement maintenance strategy according to claim 7, characterized in that: S32, correlating the coupling effect coefficient of each grid unit in the risk response area obtained in S31 with the road condition risk integer value of the corresponding grid unit, analyzing the accelerated decline trend of the pavement structure deterioration of each grid unit in the risk response area, and determining the life reduction value of each grid unit in the risk response area, specifically, Where, represents the life loss value of the jth grid cell in the risk response area, represents the coupling effect coefficient of the jth grid cell in the risk response area, represents the integer value of the road condition risk of the jth grid cell in the risk response area, represents the life decay factor, An exponential function representing the growth of reload frequency.

9. The method for intelligently generating an asphalt pavement maintenance strategy according to claim 8, characterized in that: The specific steps of S4 include: S41. Based on the obtained life reduction values ​​of each grid cell in the risk response area, determine whether the corresponding grid cell in the risk response area has a rapid life reduction trend due to deterioration of the asphalt pavement structure under the influence of heavy loads, and classify the corresponding grid cells into regions to generate a pavement maintenance strategy of the corresponding level, specifically: When the life reduction value of the corresponding grid cell is greater than or equal to the life reduction threshold, it indicates that the corresponding grid cell in the risk response area has a rapid reduction trend in the life of the asphalt pavement structure due to heavy load, and the corresponding grid cell is divided into a maintenance priority area to generate a first-level pavement maintenance strategy. When the service life reduction value of the corresponding grid cell is less than the service life reduction threshold, it means that the corresponding grid cell in the risk response area does not have a rapid service life reduction trend due to the deterioration of the asphalt pavement structure under the influence of heavy load, and the corresponding grid cell is divided into a general observation area to generate a secondary pavement maintenance strategy; S42. The specific contents of the generated primary and secondary pavement maintenance strategies are as follows: The primary pavement maintenance strategy includes: When maintaining priority areas, avoid periods of high-frequency, heavy-load traffic; employ structural reinforcement measures, including overlaying medium-thick layers, localized excavation and patching with simultaneous sealing, and deep crack grouting repair; and generate a detailed list of construction parameters, including recommended material mixes, equipment configuration recommendations, construction windows, and energy consumption scheduling plans. The secondary pavement maintenance strategy includes: when maintaining the pavement in the general observation area, adopt surface functional maintenance and structural light intervention measures, including micro-surfacing, fog sealing, shallow crack sealing treatment, and layout of road condition monitoring points; at the same time, continue to carry out periodic monitoring.

10. An intelligent generation system for asphalt pavement maintenance strategies, for implementing the intelligent generation method for asphalt pavement maintenance strategies according to any one of claims 1 to 9, characterized in that: Including multi-source acquisition module, damage quantification module, life reduction analysis module and strategy generation module; The multi-source acquisition module is used to extract traffic analysis data and traffic load data from the multi-source traffic monitoring system, and analyze the level of pavement structural damage within each grid unit through laser scanning and image intelligent recognition algorithms; The damage quantification module is used to analyze the road condition risk level of each grid unit under the superposition effect of traffic load and damage based on the level of pavement structural disease, divide the risk response area, and generate life loss analysis instructions; The life loss analysis module is used to analyze the accelerated decline trend of the pavement structure deterioration of each grid unit in the risk response area after receiving the life loss analysis instruction, so as to determine the life loss value of each grid unit in the risk response area; The strategy generation module is used to generate a pavement maintenance strategy of the corresponding level according to the life loss value of each grid unit in the risk response area.

Images