Airfield pavement crack identification method and system, electronic equipment and medium

Abstract

The invention discloses an airport pavement crack identification method and system, electronic equipment and a medium, and relates to the technical field of data processing. The method comprises the following steps: acquiring multi-temporal image data and flight operation data of an airport pavement; extracting the multi-temporal image data to obtain crack evolution characteristics of a plurality of regions, and calculating load response characteristics of each region; taking the crack evolution characteristics as node attributes, determining load transmission paths and transmission intensities between adjacent areas as edge weights, and generating a damage network structure; determining crack evolution rates of a plurality of damage nodes and evolution coefficients of corresponding areas through the damage network structure, and constructing a load-evolution incidence matrix; screening each damage node, determining a target damage node, and determining a propagation path and an influence range of the target damage node; the crack expansion trend in the target period is obtained, and a corresponding maintenance report is generated. By implementing the technical scheme provided by the invention, the accuracy of airfield pavement crack identification can be improved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, specifically to a method, system, electronic device, and medium for identifying cracks in airport pavement. Background Technology

[0002] With the rapid development of the air transport industry, airport pavements are bearing an increasing load from aircraft takeoffs and landings. The structural integrity of airport pavements is directly related to the safety of aircraft takeoffs and landings; therefore, timely detection and treatment of pavement cracks are of great significance to ensuring air transport safety.

[0003] Currently, airport pavement crack detection mainly employs a combination of manual inspection and image recognition. High-resolution cameras acquire pavement images, which are then processed to identify cracks, thus enabling the detection of pavement cracks.

[0004] However, in practical applications, the formation and propagation of airport pavement cracks is a continuously changing and dynamic process, making it difficult to detect potential changes in cracks in a timely manner using existing static detection methods. This can easily lead to a lag in assessing crack development, reducing the accuracy of airport pavement crack identification. Summary of the Invention

[0005] This application provides a method, system, electronic device, and medium for identifying cracks in airport pavement, which can improve the accuracy of airport pavement crack identification.

[0006] In a first aspect, this application provides a method for identifying cracks in airport pavement, including: Multi-temporal image data of the airport pavement and corresponding flight operation data for the same period are acquired. The multi-temporal image data is extracted using the temporal difference method to obtain the crack evolution characteristics of multiple regions in the airport pavement. The load response characteristics of each region are calculated based on the flight operation data. The crack evolution characteristics are used as node attributes, and the load transmission path and transmission intensity between adjacent regions are determined as edge weights based on the load response characteristics to generate a damage network structure. The evolution coefficients of crack evolution rate of multiple damage nodes and flight take-off and landing times, aircraft load and taxiing frequency in the corresponding area are determined by the damage network structure, and a load-evolution correlation matrix is ​​constructed based on the evolution coefficients. Based on the load-evolution correlation matrix, each of the damage nodes is screened to determine the target damage node, and the propagation path and influence range of the target damage node are determined through network topology analysis. Based on the propagation path and the range of influence, predict the crack propagation trend of the airport pavement within the target period, and generate a maintenance report corresponding to the crack propagation trend.

[0007] By employing the aforementioned technical solution, crack evolution characteristics are extracted from multi-temporal image data using the temporal difference method. Simultaneously, load response characteristics are calculated based on flight operation data, dynamically reflecting the development process of pavement cracks over time and under varying loads. Secondly, a damage network structure is constructed by using crack evolution characteristics as node attributes and load transfer characteristics as edge weights. Based on this network structure, the evolution coefficient between crack evolution rate and flight takeoff and landing parameters is determined, thereby establishing a load-evolution correlation matrix. This accurately quantifies the impact of load changes on crack propagation. Furthermore, by screening damage nodes and analyzing their propagation paths and influence ranges, combined with the load-evolution correlation matrix, crack propagation trends within a target period can be predicted, enabling timely detection of potential crack propagation risks. This achieves dynamic monitoring and early warning of airport pavement crack development, which, compared to existing static detection methods, more accurately identifies and predicts crack development trends, improving the accuracy of airport pavement crack identification.

[0008] Optionally, the multi-temporal image data is preprocessed, including image registration and grayscale correction; the preprocessed image data is arranged in a time series, and pixel difference values ​​are extracted using adjacent temporal image difference operations, and potential crack regions with pixel difference values ​​greater than a difference threshold are screened out; morphological processing and connected component analysis are performed on the potential crack regions to extract the length, width, direction, and distribution density of cracks as crack geometric features; based on the changes in the crack geometric features over time, the crack growth rate, propagation direction, and damage degree of the potential crack regions are calculated as crack evolution features.

[0009] Optionally, the wheel load distribution parameters and ground pressure coefficients of each aircraft type in the flight operation data are obtained, and the stress distribution and deformation response of different aircraft types in each area are calculated in combination with the material property parameters of the airport pavement; the cumulative load application times of each area are counted based on the takeoff and landing frequency and taxiing path in the flight operation data; and the stress distribution, deformation response and cumulative load application times are used as load response characteristics of each area.

[0010] Optionally, the crack evolution characteristics of each damaged node and the connection relationships of adjacent nodes are extracted from the damage network structure to establish a node-feature mapping table; the number of flight takeoffs and landings, aircraft load, and taxiing frequency in the region corresponding to each damaged node are used as input variables, and multiple linear regression analysis is used to calculate the correlation coefficient between each input variable and the crack evolution rate; based on the correlation coefficient and the network topology location of the damaged node, the evolution coefficient of each damaged node is determined by weighted calculation; using the damaged node as the row index and the number of flight takeoffs and landings, aircraft load, and taxiing frequency as the column index, the corresponding evolution coefficients are filled into the matrix elements to construct a load-evolution correlation matrix.

[0011] Optionally, based on the network topology location of each damaged node, the shortest path distance from each damaged node to the network center node is calculated, and the position weight coefficient corresponding to each shortest path distance is determined based on a preset distance decay function; the correlation coefficient and the position weight coefficient of the corresponding damaged node are weighted and averaged to obtain the evolution coefficient of each damaged node.

[0012] Optionally, a vector summation operation is performed on the load-evolution correlation matrix to calculate the total load influence of each damaged node, and damaged nodes whose total load influence exceeds a preset threshold are selected as target damaged nodes; the positions of the target damaged nodes in the damaged network structure are mapped to the corresponding rows and columns of the adjacency matrix; the reachable paths of each target damaged node along the decreasing load transmission intensity direction are traced through the connection relationships in the adjacency matrix, and the node connection sequence in the reachable path is used as the propagation path; the search is extended along the propagation path, and all damaged nodes on the propagation path and the set of damaged nodes directly connected to the damaged node in the adjacency matrix are used as the influence range of the corresponding target damaged node.

[0013] Optionally, the target damaged node is used as the starting point for expansion. Multiple prediction nodes are set at preset intervals along the propagation path within the influence range. The crack expansion probability of each prediction node at different time steps is calculated based on the evolution coefficient in the load-evolution correlation matrix. The crack expansion probability is randomly sampled and calculated to determine the cumulative damage probability distribution of each prediction node in the target period. Based on the cumulative damage probability distribution, the areas where the prediction nodes have a cumulative damage probability exceeding a probability threshold are selected as risk expansion areas. The risk expansion areas are sorted in descending order according to the cumulative damage probability to obtain the crack expansion trend of the airport pavement in the target period.

[0014] A second aspect of this application provides a crack identification system for airport pavement, the system comprising: The data acquisition module is used to acquire multi-temporal image data of the airport pavement and flight operation data for the corresponding time period. The multi-temporal image data is extracted using the temporal difference method to obtain the crack evolution characteristics of multiple areas in the airport pavement, and the load response characteristics of each area are calculated based on the flight operation data. The structure generation module is used to take the crack evolution characteristics as node attributes, determine the load transfer path and transfer intensity between adjacent regions as edge weights based on the load response characteristics, and generate a damage network structure. The matrix construction module is used to determine the evolution coefficients of the crack evolution rate of multiple damage nodes and the corresponding flight take-off and landing times, aircraft load and taxiing frequency in the corresponding area through the damage network structure, and to construct a load-evolution correlation matrix based on the evolution coefficients. The crack identification module is used to filter each of the damaged nodes based on the load-evolution correlation matrix, determine the target damaged node, and determine the propagation path and impact range of the target damaged node through network topology analysis; predict the crack expansion trend of the airport pavement within the target period based on the propagation path and the impact range, and generate a maintenance report corresponding to the crack expansion trend.

[0015] A third aspect of this application provides an electronic device including a memory, a processor, and a program stored in the memory and executable on the processor, the program being loaded and executed by the processor to implement a method for identifying cracks in airport pavement.

[0016] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement a method for identifying cracks in airport pavement.

[0017] In summary, one or more technical solutions provided in this application have at least the following technical effects or advantages: By employing the aforementioned technical solution, crack evolution characteristics are extracted from multi-temporal image data using the temporal difference method. Simultaneously, load response characteristics are calculated based on flight operation data, dynamically reflecting the development process of pavement cracks over time and under varying loads. Secondly, a damage network structure is constructed by using crack evolution characteristics as node attributes and load transfer characteristics as edge weights. Based on this network structure, the evolution coefficient between crack evolution rate and flight takeoff and landing parameters is determined, thereby establishing a load-evolution correlation matrix. This accurately quantifies the impact of load changes on crack propagation. Furthermore, by screening damage nodes and analyzing their propagation paths and influence ranges, combined with the load-evolution correlation matrix, crack propagation trends within a target period can be predicted, enabling timely detection of potential crack propagation risks. This achieves dynamic monitoring and early warning of airport pavement crack development, which, compared to existing static detection methods, more accurately identifies and predicts crack development trends, improving the accuracy of airport pavement crack identification. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a method for identifying cracks in airport pavement provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of an airport pavement crack identification system provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0019] Explanation of reference numerals in the attached drawings: 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0021] In the description of the embodiments of this application, the words "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design that is described as "for example" or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design options. Rather, the use of the words "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0022] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0023] This application provides a method for identifying cracks in airport pavement. In one embodiment, please refer to... Figure 1 , Figure 1 This is a flowchart illustrating the airport pavement crack identification method provided in this application embodiment. This method can be implemented using a computer program, which can be integrated into an application or run as a standalone utility application. The method can also be implemented using a microcontroller or run on an airport pavement crack identification system based on the von Neumann architecture. Specifically, the method may include the following steps: Step 101: Obtain multi-temporal image data of the airport pavement and corresponding flight operation data for the time period.

[0024] Multi-temporal image data refers to a sequence of airport pavement images acquired at different points in time. Specifically, a high-resolution camera array fixed on both sides of the airport runway continuously captures images of the same area of ​​the pavement at preset time intervals, thereby obtaining an image sequence reflecting the changes in the pavement surface condition over time. This image data contains damage features such as cracks, spalling, and wear on the pavement surface. By comparing and analyzing images from different time phases, the development and change process of cracks can be identified. The acquired image data is preprocessed and stored in a database for subsequent crack evolution feature extraction.

[0025] Flight operation data refers to flight-related information reflecting the actual operational status of an airport. Specifically, it includes information such as takeoff and landing times, aircraft load parameters, taxi routes, and parking positions for each flight. This data is directly extracted from the airport's flight management system. Among these, aircraft load parameters include the wheel load distribution characteristics and ground pressure coefficients of different aircraft types, and taxi route data records the specific flight trajectories on the pavement. This information can be used to calculate the actual load borne by different areas of the pavement. By analyzing flight operation data, the load characteristics of different areas of the pavement can be determined, providing a basis for assessing the correlation between crack development and load.

[0026] Specifically, to accurately identify the crack evolution characteristics of airport pavement, it is first necessary to acquire multi-temporal image data of the airport pavement and corresponding flight operation data for the corresponding time periods. A high-resolution camera array is used to continuously photograph the airport pavement at preset time intervals, acquiring image sequences containing pavement surface condition information as multi-temporal image data. The high-resolution camera array includes multiple image acquisition devices fixedly installed on both sides of the airport runway. These devices are connected to the central control system via a wireless network and can automatically complete image acquisition tasks according to a preset acquisition frequency (e.g., fixed time periods each day). Simultaneously, flight operation data for the corresponding time periods is extracted from the airport flight management system, including takeoff and landing times, aircraft type information, taxi routes, and parking positions for each flight. By correlating the multi-temporal image data with the flight operation data in the time dimension, a correspondence between pavement crack condition changes and flight load effects can be established. This data acquisition method ensures continuous monitoring of pavement condition changes and, by incorporating flight operation data, provides necessary load information support for subsequent analysis of crack evolution patterns. This approach not only overcomes the limitations of traditional manual inspection methods and enables automated monitoring of pavement condition, but also establishes a direct correlation between crack evolution and actual load conditions, laying a data foundation for accurately assessing pavement damage and predicting crack development trends.

[0027] Step 102: Extract the multi-temporal image data using the temporal difference method to obtain the crack evolution characteristics of multiple areas in the airport pavement, and calculate the load response characteristics of each area based on flight operation data.

[0028] In this application, crack evolution characteristics refer to quantitative indicators characterizing the development and changes of cracks on airport pavement. By performing temporal difference analysis on multi-temporal image data, characteristic parameters of crack changes over time are extracted, including crack growth rate, propagation direction, and damage degree. These characteristic parameters reflect the dynamic changes of pavement cracks over time and can be used to characterize crack development trends and damage states, providing important basis for evaluating pavement structural performance. In this application, load response characteristics refer to the characteristic parameters that characterize the loads exerted on different areas of the airport pavement by aircraft. By analyzing aircraft parameters and operational information from flight operation data, the stress distribution, deformation response, and cumulative load cycles of each area are calculated. These characteristic parameters reflect the stress state of different areas of the pavement under actual operating conditions and can be used to assess the impact of loads on the pavement structure, providing basic data for analyzing the correlation between crack development and loads.

[0029] Specifically, after acquiring multi-temporal image data and flight operation data, it is necessary to extract and analyze the evolution characteristics and load response characteristics of pavement cracks. First, the multi-temporal image data is processed using the temporal difference method. This method identifies changes in pavement condition by calculating the differences between adjacent images in a time series. Specifically, the acquired image data is arranged in chronological order, and the images at adjacent times are subjected to difference operations to extract dynamic changes in the pavement surface condition. In this way, the evolution process of cracks in different regions can be effectively identified, obtaining crack evolution characteristics that reflect the crack development state. Simultaneously, based on flight operation data, the load conditions on each region of the pavement are analyzed. By calculating the stress distribution and cumulative effect of different aircraft types on the pavement, load response characteristics characterizing the stress state of each region are obtained. This processing method establishes a correspondence between the evolution process of pavement damage and actual load conditions, not only accurately reflecting the development law of cracks but also revealing the correlation between load action and crack evolution. By simultaneously acquiring crack evolution characteristics and load response characteristics, necessary characteristic parameters are provided for subsequent construction of damage network structures and analysis of crack propagation trends, thereby enabling a comprehensive assessment of the pavement crack development status.

[0030] Based on the above embodiments, as an optional embodiment, in step 102: extracting multi-temporal image data using the temporal difference method to obtain the crack evolution characteristics of multiple regions in the airport pavement, this step may further include the following steps: Step 201: Preprocess the multi-temporal image data, including image registration and grayscale correction; arrange the preprocessed image data according to the time series, extract the pixel difference value by the difference operation of adjacent temporal images, and filter out the regions where the pixel difference value is greater than the difference threshold.

[0031] Specifically, to ensure the accuracy of multi-temporal image data analysis, the acquired images first need to be preprocessed. A feature point matching algorithm is used to register images from different time phases. By identifying fixed reference points in the images (such as pavement markings and curbs), spatial correspondences between images are established, eliminating image misalignment caused by camera position deviations. Simultaneously, grayscale histogram equalization is used to correct the grayscale levels of the images, compensating for brightness differences caused by variations in lighting conditions at different acquisition times. The preprocessed images are arranged in chronological order to form a time series. Pixel-level difference operations are performed on images from adjacent time points to calculate pixel difference values ​​reflecting changes in pavement surface condition. A difference threshold is set to filter the pixel difference values, marking areas with difference values ​​greater than the threshold as areas with potential crack risk; these areas represent locations where significant changes in pavement surface condition have occurred.

[0032] Step 202: Perform morphological processing and connected component analysis on the region to extract the length, width, direction, and distribution density of the cracks as the geometric features of the cracks.

[0033] Specifically, to accurately extract the geometric features of cracks, the screened potential crack regions undergo further processing. First, morphological operations (including dilation and erosion) are used to optimize the potential crack regions, removing noise and enhancing the continuity of crack edges. Then, connected component analysis algorithms are used to label and classify the processed regions, identifying complete crack outlines. Based on the crack outline information, geometric parameters such as crack length (obtained through skeleton extraction), width (determined through edge detection), direction (obtained through principal direction analysis), and distribution density (through statistical analysis of the number of cracks per unit area) are calculated. These parameters collectively constitute the crack geometric features characterizing the spatial distribution of cracks.

[0034] Step 203: Based on the changes in crack geometry over time, calculate the crack growth rate, propagation direction, and damage degree of the region as crack evolution characteristics.

[0035] Specifically, after acquiring the geometric features of the cracks, to quantitatively describe the development trend of pavement cracks, it is necessary to calculate the crack evolution characteristics based on the changes in the crack geometry over time. For each region, the crack growth rate, characterizing the speed of crack propagation, is obtained by calculating the difference in crack length between adjacent time points and dividing by the time interval. The propagation direction, reflecting the crack development trend, is determined by analyzing the displacement direction of the crack endpoint coordinates and the extension trend of the crack profile. Finally, the damage degree, characterizing the severity of the cracks, is calculated based on the increase in crack width and the degree of change in distribution density, combined with the area of ​​the region. The crack growth rate is obtained by calculating the increment of crack length per unit time, the propagation direction is determined by analyzing the relative positions of the crack endpoints at adjacent time points, and the damage degree is quantified by assessing the changes in crack width and density per unit area. This approach transforms the dynamic development process of cracks into quantifiable characteristic parameters, accurately reflecting the evolution of cracks over time and providing important evidence for evaluating pavement structural performance and predicting crack development trends. By acquiring these crack evolution characteristics, accurate assessment of pavement damage status can be achieved, providing data support for formulating pavement maintenance strategies.

[0036] Based on the above embodiments, as an optional embodiment, step 102, which calculates the load response characteristics of each region based on flight operation data, may further include the following steps: Step 204: Obtain the wheel load distribution parameters and ground pressure coefficient of each aircraft type from the flight operation data, and calculate the stress distribution and deformation response of different aircraft types in each area in combination with the material property parameters of the airport pavement.

[0037] Specifically, to accurately assess the effects of aircraft loads on the pavement, it is necessary to analyze the mechanical responses of different aircraft types. First, wheel load distribution parameters and ground pressure coefficients for each aircraft type are extracted from flight operation data. Wheel load distribution parameters include wheelbase, wheelbase, and wheel load size, reflecting the spatial distribution characteristics of aircraft loads. The ground pressure coefficient characterizes the pressure state when the aircraft tires contact the pavement. Simultaneously, material property parameters of the airport pavement are obtained, including elastic modulus and Poisson's ratio, characteristic values ​​reflecting the mechanical properties of the pavement material. After obtaining these parameters, contact stress is first calculated based on the tire contact area and wheel load size. Then, the stress distribution under multiple combined loads is calculated based on the stress superposition principle. Specifically, the stress influence range is determined by the load diffusion angle, and stress components at each depth, including vertical stress, shear stress, and bending tensile stress, are calculated using a stress attenuation function. Based on this, the deformation response of the pavement is calculated using the stress-strain relationship, combining the material's elastic modulus and Poisson's ratio, to obtain the vertical and horizontal displacements of each region. This method achieves a quantitative description of the effects of aircraft loads, providing fundamental data for assessing the stress state of different regions of the pavement.

[0038] Step 205: Calculate the cumulative load impact times for each area based on the takeoff and landing frequency and taxiing path in the flight operation data.

[0039] Specifically, to assess the cumulative load on different areas of the pavement, statistical analysis of flight operation data is required. This involves extracting takeoff and landing times and taxiing path information for each flight from the flight operation data, and then counting the number of load applications for each area. First, the pavement is divided into several grid cells. Based on the taxiing path of the flights, the grid cells through which the loads pass are determined, and the number of load applications within each grid cell is counted. Considering the differences in load amplitude among different aircraft types, the number of load applications for each aircraft type is equivalently converted using a standard axle load conversion factor, which is determined based on load magnitude, tire contact pressure, and application frequency. For each area, the equivalent number of load applications for each type of aircraft passing through that area is accumulated, resulting in the cumulative load application count, reflecting the actual stress level in that area. This statistical method achieves a quantitative description of the load intensity in different areas of the pavement, providing a basis for assessing the fatigue damage state of the pavement structure.

[0040] Step 206: Use stress distribution, deformation response, and cumulative load application number as load response characteristics for each region.

[0041] Specifically, to comprehensively characterize the stress state of each region of the pavement, it is necessary to consider three characteristic parameters: stress distribution, deformation response, and the number of cumulative load applications. Specifically, the calculated stress distribution parameters (including vertical stress, shear stress, and flexural tensile stress), deformation response parameters (including vertical displacement and horizontal displacement), and the number of cumulative load applications are combined to form load response characteristics reflecting the stress state of each region. The stress distribution parameters are characterized by the magnitude and distribution pattern of stress components, the deformation response parameters are described by the amplitude and direction of displacement values, and the number of cumulative load applications is quantified by the number of equivalent loads. These characteristic parameters describe the mechanical response of the pavement structure under actual operating conditions from three perspectives: static stress, deformation characteristics, and fatigue damage. By using them as load response characteristics, the stress state of each region of the pavement can be accurately reflected. This approach achieves a comprehensive quantitative description of the pavement's stress state, providing reliable characteristic data for analyzing the relationship between load application and crack development.

[0042] Step 103: Using crack evolution characteristics as node attributes, and determining the load transfer path and transfer intensity between adjacent regions as edge weights based on load response characteristics, a damage network structure is generated.

[0043] Among them, node attributes refer to the crack evolution characteristics of each region, namely the crack length growth rate, width growth rate, and depth growth rate. These growth rates are calculated from the data of two consecutive inspections to determine the change in crack size, and are obtained by fitting an exponential growth model to characterize the damage development status of the pavement area.

[0044] The load transfer path refers to the direction of load transfer between adjacent regions, which is determined by analyzing the stress continuity at the interface. Specifically, it is based on coordinate transformation of the stress tensor, calculating the normal and shear stress components at the interface, and determining the direction of load transfer between adjacent regions.

[0045] Transfer intensity refers to the degree to which a load is transferred from one region to an adjacent region. It is a comprehensive index obtained by weighting together the stress transfer coefficient, deformation transfer coefficient, and load correlation coefficient. The stress transfer coefficient is calculated based on the continuity of interface stress, the deformation transfer coefficient is calculated based on displacement compatibility, and the load correlation coefficient is calculated based on the spatial correlation of the cumulative number of load applications.

[0046] Edge weight, or transfer strength, is used to quantify the magnitude of load transfer between adjacent regions. When the transfer strength exceeds a preset threshold, an edge connection is established between adjacent regions, and the transfer strength is used as the weight value of that edge.

[0047] Damage network structure is a networked representation of the spatial correlation of pavement damage. Network nodes represent pavement regions, node attributes represent crack evolution characteristics within those regions, network edges represent significant load transfer between adjacent regions, and edge weights represent the intensity of load transfer. This network structure enables a quantitative description of pavement damage distribution and its developmental correlation.

[0048] Specifically, to characterize the spatial correlation of pavement damage states, a damage network structure reflecting crack development patterns and load transfer characteristics needs to be constructed. First, each region of the pavement is treated as a network node, with the crack evolution characteristics of each region serving as node attributes. These attributes include the crack length growth rate, width growth rate, and depth growth rate calculated from data from two consecutive inspections. These growth rate parameters are obtained by fitting an exponential growth model and are used to characterize the crack development state within the region.

[0049] Then, based on the load response characteristics, the load transfer path and intensity between adjacent regions are determined. Specifically, the stress distribution of adjacent regions is analyzed, and the normal stress and shear stress components at the interface are calculated through coordinate transformation of the stress tensor. The load transfer path is determined based on the stress continuity condition. Simultaneously, the deformation response of adjacent regions is analyzed, and the deformation transfer coefficient is determined by calculating the displacement gradient at the interface. Furthermore, considering the spatial distribution of the cumulative load application times, correlation analysis is used to calculate the degree of correlation of load effects. The stress transfer coefficient, deformation transfer coefficient, and load correlation coefficient are weighted and combined to obtain the transfer intensity characterizing the load transfer effect. When the transfer intensity exceeds a preset threshold, an edge connection is established between adjacent regions, and the transfer intensity is used as the edge weight. In the final generated damage network structure, nodes represent pavement areas, node attributes reflect the crack development state of the area, network edges indicate significant load transfer effects between adjacent regions, and edge weights quantify the load transfer intensity. This networked representation achieves a quantitative description of the spatial distribution of pavement damage, reflects the transfer law of damage development through edge connections, and can be used to analyze crack propagation trends and evaluate the overall performance of the pavement structure, providing a basis for pavement maintenance decisions.

[0050] Step 104: Determine the evolution coefficients of crack evolution rate of multiple damage nodes and the number of flight takeoffs and landings, aircraft load and taxiing frequency in the corresponding area through the damage network structure, and construct the load-evolution correlation matrix based on the evolution coefficients.

[0051] Among them, a damage node refers to a node in the damage network structure that represents a pavement area with cracks. Each damage node contains the crack evolution characteristics of the corresponding area as its attribute, that is, it refers to a network node in the damage network structure.

[0052] The crack evolution rate refers to the rate of change in crack development at the damaged node. It is represented by an eigenvector consisting of the crack length growth rate, width growth rate, and depth growth rate, which are calculated from continuous inspection data.

[0053] Flight takeoffs and landings, aircraft load, and taxiing frequency are parameters characterizing the load on the pavement area. Flight takeoffs and landings reflect the frequency of load application, aircraft load indicates the magnitude of the load acting on the pavement, and taxiing frequency indicates the temporal distribution characteristics of the load on the pavement.

[0054] The evolution coefficient refers to the mapping coefficient between the crack evolution rate and the load parameters obtained through multiple regression analysis. It is used to characterize the degree of influence of different load parameters on crack development.

[0055] The load-evolution correlation matrix is ​​a matrix with damage nodes as rows and load parameters as columns, and its matrix elements are the corresponding evolution coefficients. It is used to quantitatively characterize the correlation between crack development and load application.

[0056] Specifically, to reveal the correlation between pavement crack development and load application, it is necessary to analyze the correspondence between crack evolution patterns and load characteristics at each node in the damage network structure. First, crack evolution characteristics of each node are extracted from the damage network structure, including crack length growth rate, width growth rate, and depth growth rate. These growth rates are used to form a feature vector reflecting the crack evolution rate. Simultaneously, load parameters for the corresponding regions are obtained, including flight takeoffs and landings, aircraft load, and taxiing frequency. Using multivariate regression analysis, a mapping relationship between crack evolution rate and load parameters is established, yielding evolution coefficients that characterize the degree of load influence on crack development. Specifically, for each damaged node, the correlation coefficient between its crack evolution rate and each load parameter is calculated, and a weighted combination is used to determine the comprehensive evolution coefficient. The evolution coefficients of all damaged nodes are combined to construct a load-evolution correlation matrix reflecting the correlation between load application and crack development. In this matrix, rows represent damaged nodes, columns represent load parameters, and matrix elements are the corresponding evolution coefficients. This method achieves a quantitative characterization of the relationship between crack development patterns and load application, providing a data foundation for analyzing the causal mechanisms of pavement damage.

[0057] Based on the above embodiments, as an optional embodiment, in step 104: determining the evolution coefficients of the crack evolution rate of multiple damage nodes and the corresponding flight takeoffs and landings, aircraft load, and taxiing frequency in the corresponding area through the damage network structure, and constructing a load-evolution correlation matrix based on the evolution coefficients, this step may further include the following steps: Step 301: Extract the crack evolution features of each damaged node and the connection relationship between adjacent nodes from the damage network structure, and establish a node-feature mapping table.

[0058] Specifically, to systematically analyze the relationship between crack development patterns and load effects, it is necessary to first extract key information from the damage network structure. This involves traversing all damage nodes in the damage network structure, extracting the crack evolution characteristics (including crack length growth rate, width growth rate, and depth growth rate) for each node, and recording the connections between this node and its neighboring nodes, as well as the corresponding edge weights. The extracted information is then organized into a node-feature mapping table. Each row of this table corresponds to a damage node and includes the node number, crack evolution feature vector, a list of neighboring nodes, and the corresponding edge weights. This mapping table structure provides a unified expression of the node characteristics and topological relationships in the damage network, facilitating subsequent analysis of the spatial correlation of crack development.

[0059] Step 302: Using the number of flight takeoffs and landings, aircraft load, and taxiing frequency in the area corresponding to each damage node as input variables, multiple linear regression analysis is used to calculate the correlation coefficient between each input variable and the crack evolution rate.

[0060] Specifically, to quantitatively analyze the impact of load on crack development, a multiple linear regression method was used to establish a load-evolution mapping relationship. First, load data for the regions corresponding to each damaged node were acquired, including flight takeoffs and landings, aircraft load, and taxiing frequency. These parameters were used as input variables for the regression analysis. Then, crack evolution characteristics from the node-feature mapping table were used as the dependent variable to construct a multiple linear regression model. The coefficients of the regression equation were solved using the least squares method to obtain the correlation coefficients between each input variable and the crack evolution rate. These correlation coefficients reflect the degree of influence of different load parameters on crack development, providing basic data for the subsequent construction of the load-evolution correlation matrix.

[0061] Step 303: Based on the correlation coefficient and the network topology location of the damaged node, determine the evolution coefficient of each damaged node through weighted calculation.

[0062] Specifically, to characterize the combined impact of load and network structure on crack development, evolution coefficients need to be determined based on correlation coefficients and network topology characteristics. First, the correlation coefficients between each damaged node and load parameters (number of flight takeoffs and landings, aircraft load, and taxiing frequency) are obtained. These coefficients reflect the direct impact of load on crack development. Then, the topological location characteristics of damaged nodes in the network are analyzed, including calculating network characteristic parameters such as degree centrality, betweenness centrality, and eigenvector centrality. These parameters characterize the importance and influence of nodes in the network structure. A weighted combination method is used to calculate the correlation coefficients and network characteristic parameters. Specifically, a weighting function is constructed, and the weight coefficients of correlation and network characteristics, along with the normalized values ​​of each parameter, are substituted into the calculation to obtain evolution coefficients that comprehensively reflect the influence of load and network structure. This weighted calculation method based on multiple characteristic parameters considers both the direct impact of load on crack development and the indirect effects of node positions in the network structure. This allows the evolution coefficients to more comprehensively characterize the crack development pattern and provides a reliable data foundation for the subsequent construction of the load-evolution correlation matrix.

[0063] Based on the above embodiments, as an optional embodiment, step 303: determining the evolution coefficient of each damaged node through weighted calculation based on the correlation coefficient and the network topology location of the damaged node. This step may further include the following steps: Step 313: Based on the network topology location of each damaged node, calculate the shortest path distance from each damaged node to the network center node, and determine the location weight coefficient corresponding to each shortest path distance based on the preset distance decay function.

[0064] Specifically, to characterize the impact of the spatial location characteristics of damaged nodes in the network structure on crack development, it is necessary to calculate positional weight coefficients based on the network topology. First, the central node is identified in the damaged network by calculating the degree centrality, betweenness centrality, and proximity centrality of each node, selecting the node with the highest comprehensive score for these centrality indicators as the network center node. Then, Dijkstra's shortest path algorithm is used to calculate the shortest path distance from each damaged node to the center node, which is obtained by accumulating the weights of each edge on the path. Next, a distance decay function f(d) = exp(-λd) is introduced, where d is the shortest path distance and λ is the decay coefficient. This function reflects the degree to which the influence of the overall network structure on a node decreases as the distance to the center node increases. For each damaged node, its shortest path distance to the center node is substituted into the decay function to obtain the corresponding positional weight coefficient. This weight calculation method based on network distance can quantitatively characterize the degree of influence of the relative position of a damaged node in the network space on its development and evolution.

[0065] Step 323: Calculate the evolution coefficient of each damaged node by weighting the correlation coefficient and the position weight coefficient of the corresponding damaged node.

[0066] Specifically, to comprehensively consider the direct impact of load on crack development and the indirect effect of network location, a weighted average of the correlation coefficient and the location weight coefficient is calculated. For each damaged node i, the evolution coefficient Ei is calculated as follows: Ei = (w1 × ri1 + w2 × ri2 + w3 × ri3) × f(di), where ri1, ri2, and ri3 are the correlation coefficients between the node and the number of flight takeoffs and landings, aircraft type load, and taxiing frequency, respectively; w1, w2, and w3 are the weight parameters of each correlation coefficient, satisfying w1 + w2 + w3 = 1; and f(di) is the location weight coefficient of the node. Through this weighted calculation method, the correlation coefficient reflects the direct impact of load parameters on crack development, while the location weight coefficient characterizes the modulation effect of the node's position in the network structure on its evolution characteristics. The final evolution coefficient includes both the local characteristics of the load-response relationship and reflects the overall effect brought about by the network structure, providing a more comprehensive characterization method for accurately depicting the crack development law.

[0067] Step 304: Using the damaged node as the row index and the flight takeoff and landing times, aircraft type load, and taxiing frequency as the column index, fill the corresponding evolution coefficients into the matrix elements to construct the load-evolution correlation matrix.

[0068] Specifically, to characterize the relationship between load action and crack development, a load-evolution correlation matrix M is constructed. Specifically, N damage nodes are used as row indices in the matrix according to their network numbering order {1, 2, ..., N}, and flight takeoffs and landings, aircraft load, and taxiing frequency are used as column indices, forming an N×3 matrix framework. For each element M(i, j) in the matrix, i represents the i-th damage node, and j represents the j-th load parameter (j=1 represents takeoffs and landings, j=2 represents load, j=3 represents taxiing frequency). The evolution coefficients E corresponding to each damage node calculated in step 303 are filled into the corresponding positions in the matrix, i.e., M(i, j) = Ei, j. Simultaneously, the eigenvalues ​​and eigenvectors of the matrix are calculated to analyze the dominant factors of load action; the condition number of the matrix is ​​calculated to evaluate the stability of the load-evolution relationship; and key patterns are extracted based on singular value decomposition to identify typical load-evolution combination patterns. The resulting load-evolution correlation matrix not only intuitively reflects the influence of different load parameters on crack development at each damaged node through the numerical values ​​of the matrix elements, but also reveals the systematic correlation between load action and crack development through the overall characteristics of the matrix, providing reliable data support for subsequent damage prediction and maintenance decisions.

[0069] Step 105: Screen each damaged node based on the load-evolution correlation matrix to determine the target damaged node, and determine the propagation path and impact range of the target damaged node through network topology analysis.

[0070] In this application, the term "target damage node" specifically refers to a damage node with a large evolution coefficient in the load-evolution correlation matrix and an important position in the network structure. Specifically, it refers to damage locations on airport pavements that are significantly affected by load factors such as flight takeoffs and landings, aircraft load, and taxiing frequency, and where crack evolution rates are rapid. These nodes are often located in critical areas such as pavement intersections and curves, and the deterioration of their damage state can significantly affect the overall service performance of the pavement.

[0071] In this application, the propagation path specifically refers to the route by which a target damaged node, through network connections, may influence other damaged nodes. Due to the integrity of the pavement structure, damage at one location often affects the surrounding area through mechanisms such as stress transfer and deformation accumulation, forming a chain reaction of damage. By analyzing the edge connections in the network, possible paths for damage to propagate outward from the target node can be predicted.

[0072] In this application, the scope of impact specifically refers to the area that the target damaged node may affect through its propagation path. Specifically, it refers to the pavement area covered by damaged nodes that are directly or indirectly connected to the target node within a certain network distance. This scope reflects the spatial distribution characteristics of pavement performance degradation that a single damaged location may cause, and is of great significance for assessing the severity of the damage and determining maintenance priorities.

[0073] Specifically, to identify key nodes in a damaged network and assess their impact range, node screening and propagation analysis based on the load-evolution correlation matrix are required. First, the element values ​​in the load-evolution correlation matrix are sorted, and damaged nodes with larger element values ​​are selected as candidate target nodes. These nodes represent regions with high evolutionary activity under load. Then, network centrality indices for candidate nodes are calculated, including degree centrality, betweenness centrality, and eigenvector centrality, comprehensively evaluating the importance of nodes in the network. Candidate nodes with higher centrality indices are identified as target damaged nodes. For the identified target damaged nodes, a breadth-first search algorithm is used to analyze their possible propagation paths. By traversing the edges connected to the target node and its adjacent nodes, possible paths for damage propagation are obtained. Simultaneously, the impact range of the target node is analyzed based on propagation distance thresholds and network connectivity. Specifically, the spatial range of damage impact is determined by calculating the number and distribution characteristics of nodes reachable within a given propagation distance. This node screening method based on matrix analysis and network topology can effectively identify key damage locations and predict their development trends, providing a basis for decision-making in formulating targeted maintenance strategies.

[0074] Based on the above embodiments, as an optional embodiment, in step 105: screening each damaged node based on the load-evolution correlation matrix to determine the target damaged node, and determining the propagation path and influence range of the target damaged node through network topology analysis, this step may further include the following steps: Step 401: Perform vector summation on the load-evolution correlation matrix to calculate the total load influence of each damaged node, and select damaged nodes whose total load influence exceeds a preset threshold as target damaged nodes.

[0075] Specifically, to identify the most critical nodes significantly affected by loads from numerous damaged nodes, quantitative analysis and screening of the load-evolution correlation matrix are required. Specifically, for an N×3 load-evolution correlation matrix M, where N is the number of damaged nodes, 3 represents the three load parameters: flight takeoffs and landings, aircraft load, and taxiing frequency, and the element M(i,j) in the matrix represents the evolution coefficient between the i-th damaged node and the j-th load parameter, first, a vector summation operation is performed on each row of matrix M, i.e., for each damaged node i, its total load impact value Si = ∑(j=1 to 3)M(i,j). This total value reflects the comprehensive influence of the three load parameters on the damaged node. Then, based on the service performance requirements of the pavement structure, a threshold St for the total load impact value is set. This threshold can be determined through historical maintenance data statistics or expert experience, with a typical value being the 75th percentile of the total load impact value of all nodes. Damaged nodes with a total load impact value Si greater than the threshold St are screened out as target damaged nodes requiring focused attention. This matrix-operation-based screening method can quantitatively assess the impact of load on damage development, effectively identify the key locations in the pavement structure most prone to damage expansion, and provide a reliable basis for subsequent maintenance decisions. Furthermore, by adjusting the threshold value St, the number of target damaged nodes can be flexibly controlled, enabling the rational allocation of maintenance resources.

[0076] Step 402: Map the position of the target damage node in the damage network structure to the corresponding row and column of the adjacency matrix; trace the reachable path of each target damage node along the direction of decreasing load transmission intensity through the connection relationship in the adjacency matrix, and use the node connection sequence in the reachable path as the propagation path.

[0077] Specifically, to accurately analyze the propagation characteristics of the target damaged node, it is necessary to analyze its expansion path in the network based on the adjacency matrix. First, an N×N adjacency matrix A is constructed, where N is the total number of nodes in the damaged network, and the matrix element A(i, j) represents the connection strength between node i and node j, reflecting the load transfer capability between the two nodes. For each target damaged node k, its corresponding row and column are located in the adjacency matrix; these row and column elements describe the direct connection relationships between this node and other nodes. Then, a depth-first search algorithm is used, starting from the target node k, selecting the second-highest unvisited node in the current node's adjacency matrix row as the next-hop node, until no next-hop node satisfying the conditions can be found. During the search process, a load transfer strength attenuation threshold δ is set; extension stops when the connection strength between adjacent nodes is less than the threshold. The resulting node sequence {k, n1, n2, ..., nm} is recorded as the propagation path of the target node k, where n1 to nm are intermediate nodes on the path. This analysis method, based on adjacency matrix and path search, can effectively identify the dominant paths for damage propagation and reflect the spatial correlation characteristics of crack development under load. By analyzing the propagation paths of multiple target nodes, typical propagation patterns of pavement damage can be identified, providing a basis for developing preventative maintenance strategies. Furthermore, by adjusting the intensity attenuation threshold δ, the tracking depth of the propagation path can be controlled, enabling a reasonable assessment of the damage's impact range.

[0078] Step 403: Extend the search along the propagation path and take all damaged nodes on the propagation path and the set of damaged nodes directly connected to the damaged nodes in the adjacency matrix as the influence range of the corresponding target damaged node.

[0079] Specifically, to comprehensively assess the potential impact range of a target damaged node, an extended analysis is needed based on the established propagation path. For each target damaged node k's propagation path Pk={k, n1, n2, ..., nm}, all nodes along the path are first added to its impact range set Rk. Then, for each node i in path Pk, the non-zero elements in the i-th row of the adjacency matrix A are searched. The column indices of these non-zero elements represent the damaged nodes directly connected to node i. A connection strength threshold θ is set. When the adjacency matrix element A(i, j) is greater than the threshold θ, node j is added to the impact range set Rk. Thus, the impact range set Rk not only includes nodes along the propagation path but also surrounding nodes with strong connections to these nodes, forming an impact area extending outward from the propagation path as the main axis. This analysis method based on path extension and connection relationships can more accurately characterize the spatial distribution features of damage development and reflect the network effect of crack propagation under load. By calculating the spatial distribution characteristics of nodes in the impact range set Rk, such as node density and coverage area, the severity of the target damaged node can be quantitatively assessed. Meanwhile, by comparing the impact range of different target nodes, the key damage locations that have the greatest impact on the overall pavement performance can be identified, providing a scientific basis for determining the priority maintenance sequence.

[0080] Step 106: Predict the crack propagation trend of the airport pavement within the target period based on the propagation path and the scope of impact, and generate a maintenance report corresponding to the crack propagation trend.

[0081] In this application, the target period specifically refers to the predicted time frame determined by the airport pavement maintenance management department based on engineering experience and safety assessment requirements. This time frame is typically set based on the pavement's daily inspection cycle, maintenance plan cycle, and safety management regulations, and can be on different scales such as 3 months, 6 months, or 1 year, to determine the time range for damage prediction in order to promptly identify potential risks and conduct maintenance interventions.

[0082] In this application, the crack propagation trend specifically refers to the characteristics of crack size changes at various damaged nodes along the predicted propagation path and within the influence range, with the target damaged node as the source point, within the target period. This includes the development and changes in crack length, width, and depth, as well as the spatial distribution of crack propagation direction and rate. These characteristics reflect the temporal and spatial evolution of pavement damage and are used to assess the development trend of the damage.

[0083] In this application, the maintenance report specifically refers to a technical document based on the analysis results of crack propagation trends. It mainly includes the following: the specific location information and current damage status parameters of the target damaged nodes; crack size data of each key node along the propagation path at different time points; the area within the affected range that requires focused monitoring; the spatial distribution characteristics of the crack propagation rate; the maintenance priority level determined based on the severity of damage and the risk of propagation; and the recommended maintenance timeline. This report provides technical support for pavement maintenance decisions.

[0084] Specifically, to assess the damage development trend of airport pavements and provide a basis for maintenance decisions, crack propagation trend analysis is needed based on the propagation path and the scope of influence. First, starting from the target damaged node, a network of predicted nodes is deployed along the propagation path within the scope of influence. The crack propagation probability of each predicted node at different time steps is calculated based on the load-evolution correlation matrix. Then, a statistical analysis of the propagation probability is performed using a random sampling method to obtain the cumulative damage probability distribution of the predicted nodes. Next, high-risk propagation areas are selected based on a preset probability threshold and sorted in descending order of cumulative damage probability to form the pavement crack propagation trend.

[0085] After predicting the crack propagation trend of airport pavement, a standardized maintenance report needs to be generated to facilitate maintenance personnel's understanding of the damage status and the development of maintenance plans. First, the report records basic information about the target damaged nodes, including their spatial coordinates, current crack size parameters, and the type of pavement component they are located on. Then, it describes in detail the predicted risk propagation areas, recording the spatial boundary coordinates, average cumulative damage probability, and propagation path number for each area, along with its position in the risk level ranking. Based on the risk level, maintenance priorities are divided into high, medium, and low levels. High-risk areas are recommended to be repaired within one month, medium-risk areas within three months, and low-risk areas before the end of the target period. The report also includes a pavement plan view, using different colors to mark the location and extent of each risk propagation area, visually demonstrating the spatial distribution characteristics of crack propagation. This standardized maintenance report format clearly conveys the damage prediction results, enabling maintenance personnel to quickly grasp the pavement status and take corresponding maintenance measures, improving the targeting and efficiency of pavement maintenance.

[0086] Based on the above embodiments, as an optional embodiment, step 106, which predicts the crack propagation trend of the airport pavement within the target period according to the propagation path and the range of influence, may further include the following steps: Step 501: Using the target damage node as the starting point for expansion, set multiple prediction nodes along the propagation path within the influence range at preset intervals, and calculate the crack propagation probability of each prediction node at different time steps based on the evolution coefficients in the load-evolution correlation matrix.

[0087] Specifically, to quantitatively assess the crack propagation risk caused by the target damaged node, a detailed analysis is needed in both spatial and temporal dimensions. First, starting from the target damaged node, prediction nodes are uniformly set along the propagation path within its influence range at preset intervals of 1 meter. For each prediction node, the network distance to the target damaged node is determined based on its position on the propagation path, and the evolution coefficient at the corresponding position is extracted from the load-evolution correlation matrix. Then, the target period is divided into multiple time steps, such as 15 days per step. For each prediction node at each time step, the crack propagation probability is calculated based on its evolution coefficient, the network distance to the target node, and the current time step. Specifically, the crack propagation probability P is calculated by combining the evolution coefficient E with the network distance attenuation factor D and the time step coefficient T, where the distance attenuation factor decreases with increasing network distance, and the time step coefficient increases with time. This probability calculation method based on spatial grids and time steps can reflect the spatiotemporal evolution characteristics of crack propagation, providing a quantitative basis for subsequent propagation trend analysis. By analyzing the expansion probability distribution of different predicted nodes, the direction and area where cracks are most likely to expand can be identified, providing guidance for developing targeted monitoring and maintenance measures.

[0088] Step 502: Perform random sampling calculation on the crack propagation probability to determine the cumulative damage probability distribution of each predicted node within the target period.

[0089] Specifically, to accurately assess the cumulative effect of crack propagation, random sampling analysis of the propagation probability of each predicted node is required. First, the crack propagation probability sequence of each predicted node at different time steps is repeatedly sampled, with the sampling number set to 1000 times to ensure statistical reliability. In each sampling, a random number between 0 and 1 is generated based on the propagation probability of that node. If the random number is less than the propagation probability, crack propagation is considered to have occurred at that time step. For each predicted node, the cumulative propagation count across all time steps within the target period is calculated and divided by the total number of samplings to obtain the cumulative damage probability of that node. Then, the cumulative damage probabilities of all predicted nodes are arranged according to spatial location to form a probability distribution map, where a higher probability value indicates a higher likelihood of crack propagation at that location. This cumulative probability calculation method based on random sampling can simulate the randomness of the crack propagation process and avoid the bias that may arise from a single deterministic prediction. By analyzing the spatial distribution characteristics of the cumulative damage probability, high-risk areas prone to crack propagation in the pavement structure can be clearly identified, providing reliable data support for developing preventative maintenance strategies.

[0090] Step 503: Based on the cumulative damage probability distribution, select the areas where the predicted nodes have a cumulative damage probability exceeding the probability threshold as risk expansion areas, sort the risk expansion areas in descending order according to the cumulative damage probability, and obtain the crack expansion trend of the airport pavement within the target period.

[0091] Specifically, to identify high-risk areas in the pavement structure and determine maintenance priorities, it is necessary to divide and rank these areas based on the cumulative damage probability distribution. First, a cumulative damage probability threshold of 0.75 is set, determined based on pavement safety requirements. Areas containing predicted nodes with cumulative damage probabilities exceeding this threshold are marked as risk expansion areas. Adjacent risk prediction nodes are merged into the same risk expansion area based on their spatial continuity, and the average cumulative damage probability of that area is calculated. Then, all risk expansion areas are sorted in descending order of their average cumulative damage probability; the sorting result reflects the crack propagation risk level of different areas. For each risk expansion area, its spatial extent, cumulative damage probability, and propagation path are recorded. This information collectively constitutes the crack propagation trend of the airport pavement within the target period. This probability threshold-based area identification and ranking method can integrate discrete predicted nodes into risk areas with engineering significance, facilitating maintenance personnel's understanding and implementation of maintenance measures. By analyzing the spatial distribution and probability magnitude of risk expansion areas, maintenance priorities can be scientifically formulated, achieving optimal allocation of pavement maintenance resources.

[0092] Reference Figure 2 This application provides an embodiment of an airport pavement crack identification system, which includes: a data acquisition module, a structure generation module, a matrix construction module, and a crack identification module, wherein: The data acquisition module is used to acquire multi-temporal image data of the airport pavement and flight operation data for the corresponding time period; the multi-temporal image data is extracted using the temporal difference method to obtain the crack evolution characteristics of multiple areas in the airport pavement, and the load response characteristics of each area are calculated based on the flight operation data; The structure generation module is used to take crack evolution characteristics as node attributes, determine the load transfer path and transfer intensity between adjacent regions as edge weights based on load response characteristics, and generate a damage network structure. The matrix construction module is used to determine the evolution coefficients of the crack evolution rate of multiple damage nodes and the corresponding flight take-off and landing times, aircraft load and taxiing frequency in the corresponding area through the damage network structure, and to construct the load-evolution correlation matrix based on the evolution coefficients. The crack identification module is used to screen each damaged node based on the load-evolution correlation matrix, identify the target damaged node, and determine the propagation path and impact range of the target damaged node through network topology analysis; based on the propagation path and impact range, it predicts the crack propagation trend of the airport pavement within the target period and generates a maintenance report corresponding to the crack propagation trend.

[0093] Based on the above embodiments, the data acquisition module is also used to preprocess multi-temporal image data, including image registration and grayscale correction; arrange the preprocessed image data according to the time series, extract pixel difference values ​​by using adjacent temporal image difference operation, and filter out regions with pixel difference values ​​greater than the difference threshold; perform morphological processing and connected component analysis on the regions, and extract the length, width, direction and distribution density of cracks as crack geometric features; based on the changes in crack geometric features in the time dimension, calculate the crack growth rate, expansion direction and damage degree of the region as crack evolution features.

[0094] Based on the above embodiments, the data acquisition module is also used to acquire the wheel load distribution parameters and ground pressure coefficient of each aircraft type in the flight operation data, and calculate the stress distribution and deformation response of different aircraft types in each area in combination with the material property parameters of the airport pavement; count the number of cumulative loads applied to each area according to the take-off and landing frequency and taxiing path in the flight operation data; and use the stress distribution, deformation response and the number of cumulative loads applied as the load response characteristics of each area.

[0095] Based on the above embodiments, the matrix construction module is also used to extract the crack evolution characteristics of each damaged node and the connection relationship of adjacent nodes from the damage network structure, and establish a node-feature mapping table; taking the number of flight take-offs and landings, aircraft load, and taxiing frequency in the region corresponding to each damaged node as input variables, and using multiple linear regression analysis to calculate the correlation coefficient between each input variable and the crack evolution rate; based on the correlation coefficient and the network topology location of the damaged node, the evolution coefficient of each damaged node is determined by weighted calculation; using the damaged node as the row index and the number of flight take-offs and landings, aircraft load, and taxiing frequency as the column index, the corresponding evolution coefficients are filled into the matrix elements to construct the load-evolution correlation matrix.

[0096] Based on the above embodiments, the matrix construction module is also used to calculate the shortest path distance from each damaged node to the network center node based on the network topology location of each damaged node, and determine the position weight coefficient corresponding to each shortest path distance based on a preset distance decay function; and calculate the evolution coefficient of each damaged node by weighted averaging the correlation coefficient and the position weight coefficient of the corresponding damaged node.

[0097] Based on the above embodiments, the crack identification module is also used to perform vector summation on the load-evolution correlation matrix, calculate the total load influence of each damaged node, and screen out damaged nodes whose total load influence exceeds a preset threshold as target damaged nodes; map the position of the target damaged node in the damage network structure to the corresponding row and column of the adjacency matrix; track the reachable path of each target damaged node along the direction of decreasing load transmission intensity through the connection relationship in the adjacency matrix, and take the node connection sequence in the reachable path as the propagation path; extend the search along the propagation path, and take all damaged nodes on the propagation path and the set of damaged nodes directly connected to the damaged node in the adjacency matrix as the influence range of the corresponding target damaged node.

[0098] Based on the above embodiments, the crack identification module is also used to take the target damage node as the starting point of expansion, set multiple prediction nodes along the propagation path within the influence range at preset intervals, calculate the crack expansion probability of each prediction node at different time steps according to the evolution coefficient in the load-evolution correlation matrix, perform random sampling calculation on the crack expansion probability to determine the cumulative damage probability distribution of each prediction node in the target period, and select the area where the prediction node with the cumulative damage probability exceeds the probability threshold as the risk expansion area based on the cumulative damage probability distribution, sort the risk expansion areas in descending order according to the cumulative damage probability to obtain the crack expansion trend of the airport pavement in the target period.

[0099] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0100] This application also discloses an electronic device. (See reference...) Figure 3 , Figure 3 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application. The electronic device 300 may include: at least one processor 301, at least one network interface 304, a user interface 303, a memory 305, and at least one communication bus 302.

[0101] The communication bus 302 is used to enable communication between these components.

[0102] The user interface 303 may include a display interface and a camera interface. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0103] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0104] The processor 301 may include one or more processing cores. The processor 301 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 305, and by calling data stored in the memory 305. Optionally, the processor 301 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 301 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface graphics, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.

[0105] The memory 305 may include random access memory (RAM) or read-only memory. Optionally, the memory 305 may include a non-transitory computer-readable storage medium. The memory 305 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. (Refer to...) Figure 3 The memory 305, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for identifying cracks in airport pavement.

[0106] exist Figure 3 In the illustrated electronic device 300, the user interface 303 is mainly used to provide an input interface for the user and acquire user input data; while the processor 301 can be used to call an application program stored in the memory 305 for identifying cracks in airport pavement. When executed by one or more processors 301, the electronic device 300 performs one or more methods as described in the above embodiments. It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0107] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0108] In the various embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between apparatuses or units may be electrical or other forms.

[0109] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0110] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0111] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0112] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will be readily apparent to those skilled in the art upon consideration of the specification and practical disclosure.

[0113] This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are to be considered exemplary only.

Claims

1. A method for identifying cracks in airport pavement, characterized in that, include: Acquire multi-temporal image data of the airport pavement and corresponding flight operation data for the corresponding time periods; The multi-temporal image data is extracted using the temporal difference method to obtain the crack evolution characteristics of multiple regions in the airport pavement, and the load response characteristics of each region are calculated based on the flight operation data. The crack evolution characteristics are used as node attributes, and the load transmission path and transmission intensity between adjacent regions are determined as edge weights based on the load response characteristics to generate a damage network structure. The evolution coefficients of crack evolution rate of multiple damage nodes and flight take-off and landing times, aircraft load and taxiing frequency in the corresponding area are determined by the damage network structure, and a load-evolution correlation matrix is ​​constructed based on the evolution coefficients. Based on the load-evolution correlation matrix, each of the damage nodes is screened to determine the target damage node, and the propagation path and influence range of the target damage node are determined through network topology analysis. Based on the propagation path and the range of influence, predict the crack propagation trend of the airport pavement within the target period, and generate a maintenance report corresponding to the crack propagation trend.

2. The method for identifying cracks in airport pavement according to claim 1, characterized in that, The step of extracting crack evolution characteristics in multiple regions of the airport pavement by using the temporal difference method from the multi-temporal image data includes: The multi-temporal image data is preprocessed, including image registration and grayscale correction. The preprocessed image data is arranged in a time series, and pixel difference values ​​are extracted using adjacent temporal image difference operation. Regions with pixel difference values ​​greater than the difference threshold are then selected. Morphological processing and connected component analysis are performed on the region to extract the length, width, direction, and distribution density of the cracks as the geometric features of the cracks. Based on the changes in the geometric features of the crack over time, the crack growth rate, propagation direction, and damage degree of the region are calculated as crack evolution characteristics.

3. The method for identifying cracks in airport pavement according to claim 1, characterized in that, The calculation of load response characteristics for each region based on the flight operation data includes: The wheel load distribution parameters and ground pressure coefficient of each aircraft type in the flight operation data are obtained, and the stress distribution and deformation response of different aircraft types in each area are calculated in combination with the material property parameters of the airport pavement. The cumulative load impact count for each region is calculated based on the takeoff and landing frequency and taxiing path data in the flight operation data. The stress distribution, the deformation response, and the number of cumulative load applications are used as the load response characteristics of each region.

4. The method for identifying cracks in airport pavement according to claim 1, characterized in that, The process involves determining the evolution coefficients of crack evolution rates at multiple damage nodes and their corresponding regions' flight takeoffs and landings, aircraft load, and taxiing frequency using the damage network structure, and constructing a load-evolution correlation matrix based on these coefficients, including: Extract the crack evolution features of each damaged node and the connection relationship between adjacent nodes from the damage network structure, and establish a node-feature mapping table; The number of flight takeoffs and landings, aircraft load, and taxiing frequency in the regions corresponding to each damage node were used as input variables, and the correlation coefficient between each input variable and the crack evolution rate was calculated by multiple linear regression analysis. Based on the correlation coefficient and the network topology location of the damaged node, the evolution coefficient of each damaged node is determined by weighted calculation. Using the damaged nodes as row indices and the number of flight takeoffs and landings, aircraft type load, and taxiing frequency as column indices, the corresponding evolution coefficients are filled into the matrix elements to construct a load-evolution correlation matrix.

5. The method for identifying cracks in airport pavement according to claim 4, characterized in that, The evolution coefficient of each damaged node is determined by weighted calculation based on the correlation coefficient and the network topology location of the damaged node, including: Based on the network topology location of each damaged node, the shortest path distance from each damaged node to the network center node is calculated, and the position weight coefficient corresponding to each shortest path distance is determined based on a preset distance decay function. The evolution coefficient of each damaged node is obtained by weighting the correlation coefficient with the position weight coefficient of the corresponding damaged node.

6. The method for identifying cracks in airport pavement according to claim 4, characterized in that, The process of filtering each damaged node based on the load-evolution correlation matrix to determine the target damaged node, and determining the propagation path and impact range of the target damaged node through network topology analysis, includes: The load-evolution correlation matrix is ​​subjected to vector summation to calculate the total load influence of each damaged node, and damaged nodes whose total load influence exceeds a preset threshold are selected as target damaged nodes. Map the position of the target damaged node in the damaged network structure to the corresponding row and column of the adjacency matrix; The reachable path of each target damage node along the decreasing load transmission intensity direction is traced through the connection relationship in the adjacency matrix, and the node connection sequence in the reachable path is used as the propagation path; The search is extended along the propagation path, and all damaged nodes on the propagation path and the set of damaged nodes directly connected to the damaged nodes in the adjacency matrix are taken as the influence range of the corresponding target damaged node.

7. The method for identifying cracks in airport pavement according to claim 1, characterized in that, The step of predicting the expansion trend of the airport pavement within a target period based on the propagation path and the area of ​​influence includes: Using the target damage node as the starting point for expansion, multiple prediction nodes are set at preset intervals along the propagation path within the influence range. The crack expansion probability of each prediction node at different time steps is calculated based on the evolution coefficients in the load-evolution correlation matrix. The crack propagation probability is randomly sampled and calculated to determine the cumulative damage probability distribution of each predicted node within the target period; Based on the cumulative damage probability distribution, the regions where the predicted nodes have a cumulative damage probability exceeding the probability threshold are selected as risk expansion regions. The risk expansion regions are then sorted in descending order according to the cumulative damage probability to obtain the crack expansion trend of the airport pavement within the target period.

8. A crack detection system for airport pavement, characterized in that, The system includes: The data acquisition module is used to acquire multi-temporal image data of the airport pavement and flight operation data for the corresponding time period; the multi-temporal image data is extracted using the temporal difference method to obtain the crack evolution characteristics of multiple areas in the airport pavement, and the load response characteristics of each area are calculated based on the flight operation data; The structure generation module is used to take the crack evolution characteristics as node attributes, determine the load transfer path and transfer intensity between adjacent regions as edge weights based on the load response characteristics, and generate a damage network structure. The matrix construction module is used to determine the evolution coefficients of the crack evolution rate of multiple damage nodes and the corresponding flight take-off and landing times, aircraft load and taxiing frequency in the corresponding area through the damage network structure, and to construct a load-evolution correlation matrix based on the evolution coefficients. The crack identification module is used to filter each of the damaged nodes based on the load-evolution correlation matrix, determine the target damaged node, and determine the propagation path and impact range of the target damaged node through network topology analysis; predict the crack expansion trend of the airport pavement within the target period based on the propagation path and the impact range, and generate a maintenance report corresponding to the crack expansion trend.

9. An electronic device, characterized in that, The device includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to enable the electronic device to perform a method for identifying cracks in airport pavement as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform a method for identifying cracks in airport pavement as described in any one of claims 1-7.

Images