Road service life comprehensive evaluation and analysis method and system

Abstract

The invention discloses a road service life comprehensive assessment analysis method and system, and relates to the technical field of road health assessment, and the method comprises the steps: collecting and preprocessing road multi-dimensional data, and extracting road features based on the preprocessed multi-dimensional road data; performing space-time correlation analysis on the road features, and constructing a road life evaluation data framework; preliminarily calculating a road health index by using the road health prediction model based on the road life evaluation data framework; dynamically adjusting the comprehensive weight of the road features by combining the preliminarily calculated road health index and historical data; and according to the adjusted comprehensive weight of the road feature, updating a road health index, and evaluating a road safety level. According to the method, the problems of insufficient multi-dimensional data integration and lack of dynamic analysis are solved, the scientificity and the practical application value of an evaluation result are remarkably improved, and a reliable basis is provided for road management and safety decision making.

Description

Technical Field

[0001] The present invention relates to the technical field of road health assessment, and in particular to a method and system for comprehensive assessment and analysis of road service life. Background Art

[0002] With the rapid development of the transportation industry, the design, construction and maintenance of roads, as important infrastructure, are becoming increasingly important. However, with the continuous growth of traffic volume and the long-term effects of the natural environment, the service life of roads is affected by a variety of complex factors, including pavement diseases, aging of internal road structures, repeated effects of traffic loads, and cumulative effects of environmental erosion. Traditional road maintenance and assessment methods mainly rely on periodic inspections and manual inspections, which are not only inefficient, but also easily affected by subjective judgments, making it difficult to quantitatively analyze and scientifically predict the entire life cycle of roads. At the same time, in recent years, methods based on big data analysis and artificial intelligence technology have gradually been applied to the field of road detection, such as using image processing technology to identify pavement cracks, rutting and other disease characteristics, or using sensors to monitor traffic loads and internal structure data. However, most of the existing technologies only analyze single-dimensional features and cannot comprehensively consider the correlation between multidimensional data, especially the mutual influence of spatial distribution and time evolution laws, which makes the existing methods still have great limitations in the accuracy and comprehensiveness of road life assessment.

[0003] Existing technologies have the following deficiencies in the comprehensive assessment of road service life: First, it is difficult for current assessment methods to effectively integrate multi-dimensional road data, such as pavement damage, internal structure, traffic load and environmental factors, which leads to a lack of comprehensiveness and scientificity in road life assessment results; Second, existing methods usually lack in-depth mining of spatiotemporal correlations and cannot accurately reveal the trend of road characteristics changing over time and the regularity of their spatial distribution, thus limiting the ability to predict the future health status of roads. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a comprehensive evaluation and analysis method for road service life to solve the problems of insufficient multi-dimensional data integration and lack of in-depth analysis of spatio-temporal correlation in the prior art.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for comprehensive evaluation and analysis of road service life, which includes collecting and preprocessing multidimensional road data, and extracting road features based on the preprocessed multidimensional road data; the multidimensional road data includes pavement disease data, road internal structure data, traffic load data and environmental data; performing spatiotemporal correlation analysis on road features to construct a road life evaluation data framework; based on the road life evaluation data framework, using a road health prediction model, preliminarily calculating a road health index; combining the preliminarily calculated road health index with historical data, dynamically adjusting the comprehensive weight of road features; updating the road health index based on the adjusted comprehensive weight of road features, and evaluating the road safety level.

[0007] As a preferred solution of the comprehensive evaluation and analysis method for road service life of the present invention, the multi-dimensional data of the road is preprocessed, and the specific steps are as follows: The disease data is cleaned through noise filtering algorithms to remove blurry images and invalid data; Spectral analysis is used to filter out clutter signals from the road internal structure data; Standardize and normalize traffic load data; Data interpolation methods are used to fill missing values ​​in environmental data and remove abnormal data.

[0008] As a preferred solution of the comprehensive evaluation and analysis method for road service life of the present invention, the road features are extracted based on the pre-processed multi-dimensional road data, and the specific steps are as follows: The pavement disease features are extracted from the cleaned pavement disease data through image segmentation algorithm; the pavement disease features include crack width, crack length, rutting depth and disease area; The road internal structure features are extracted from the road internal structure data after clutter is filtered out by signal analysis method; the road internal structure features include the void ratio, layer depth and void area range of the internal structure data; Extracting traffic load characteristics from the standard normalized traffic load data through statistical analysis; the traffic load characteristics include traffic flow peak, vehicle load distribution and traffic pressure distribution mean; Environmental erosion characteristics are extracted from pre-processed environmental data by data analysis; environmental erosion characteristics include rainfall frequency, temperature and humidity variation, and freeze-thaw cycle times; Road characteristics include: pavement disease characteristics, road internal structure characteristics, traffic load characteristics and environmental erosion characteristics.

[0009] As a preferred solution of the comprehensive evaluation and analysis method for road service life of the present invention, the specific steps of performing spatiotemporal correlation analysis on road characteristics and constructing a road service life evaluation data framework are as follows: Based on the road characteristics, the spatial units are divided according to fixed standards, and the road characteristics are mapped to the divided spatial units; Perform spatial analysis on the road features mapped into the spatial units to obtain the spatial distribution of the road features; Conduct time correlation analysis on road characteristics to obtain the time variation patterns and interdependencies of road characteristics; The spatial distribution of road characteristics and the temporal variation and interdependence of road characteristics are quantified, and combined according to spatial units and time nodes to generate a feature vector for each road section at each time point, and construct a multidimensional feature matrix; the multidimensional feature matrix includes pavement disease characteristics, temporal variation, spatial distribution and environmental erosion characteristics; Based on the multidimensional feature matrix and combined with historical data, the life status of each road section is marked, and the multidimensional feature matrix and the life status of each road section are integrated to form a road life assessment data framework.

[0010] As a preferred solution of the comprehensive evaluation and analysis method for road service life of the present invention, the road health index is preliminarily calculated based on the road life evaluation data framework and using the road health prediction model. The specific steps are as follows: Taking the multidimensional feature matrix in the road life assessment data framework as input, the road health prediction model is used to analyze the contribution of the multidimensional feature matrix to the health status, and the road health index is preliminarily calculated. The expression is: ; in, is the preliminary calculated road health index, is the comprehensive weight of the road health index, is the severity of the i-th type of pavement disease characteristics, is the influence coefficient of the i-th type of pavement disease characteristics, is the time sensitivity coefficient of the i-th type of pavement disease characteristics, is the cumulative time span of the i-th type of pavement disease characteristics, is the influence coefficient of the internal structural characteristics of the road, is the fluctuation variance of traffic load characteristics, is the sensitivity coefficient of environmental erosion characteristics, is the comprehensive environmental erosion index, The spatial scale factor of the i-th type of pavement damage feature, n is the number of pavement damage features, and i is the index variable of the pavement damage feature.

[0011] As a preferred solution of the comprehensive evaluation and analysis method for road service life of the present invention, the comprehensive weight of road characteristics is dynamically adjusted by combining the preliminary calculated road health index and historical data. The specific steps are as follows: The road health index was calculated by and historical data, using a weighted time decay function and combining the road health index and characteristic historical data to dynamically adjust the comprehensive weight, the expression is: ; in, is the adjusted comprehensive weight, is the historical time span, is the time decay function, is the historical value of the i-th type of pavement disease characteristics at time t, are the elements of the disease feature correlation matrix, is the historical value of the j-th disease characteristic at time t, is the regularization factor of the i-th disease feature, represents a historical time point, j represents the disease feature index variable that is correlated with the current pavement disease feature i, Is a measure of road health index The degree of impact on road characteristics.

[0012] As a preferred solution of the comprehensive evaluation and analysis method for road service life of the present invention, the road health index is updated and the road safety level is evaluated according to the adjusted comprehensive weight of the road characteristics. The specific steps are as follows: Comprehensive weight based on adjusted road characteristics , optimize the road health prediction model, and update the road health index based on the optimized road health prediction model , get the updated road health index ; Based on the long-term collection of road health index and the statistical results of the actual service life of the road, combined with the road design specifications and safety assessment standards, the safety threshold R1, risk threshold R2 and danger threshold R3 are defined; when >R1, the current road service life is considered to be in a safe state; When R2≤ When ≤R1, the current road service life is considered to be in good condition; When R3≤ When <R2, it is considered that the current road service life is at risk; when When <R3, it is considered that the current road service life is in a dangerous state.

[0013] In a second aspect, the present invention provides a road service life comprehensive evaluation and analysis system, including a feature extraction module, a road feature analysis module, a road health index calculation module, a comprehensive weight adjustment module and a road safety evaluation module; A feature extraction module is used to collect and preprocess multi-dimensional road data, and extract road features based on the pre-processed multi-dimensional road data; the multi-dimensional road data includes road disease data, road internal structure data, traffic load data and environmental data; Road feature analysis module, used to conduct spatiotemporal correlation analysis on road features and build a data framework for road life assessment; A road health index calculation module is used to preliminarily calculate the road health index based on the road life assessment data framework and using the road health prediction model; A comprehensive weight adjustment module is used to dynamically adjust the comprehensive weight of road characteristics by combining the initially calculated road health index and historical data; The road safety assessment module is used to update the road health index and evaluate the road safety level according to the adjusted comprehensive weights of road characteristics.

[0014] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the comprehensive evaluation and analysis method for road service life as described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the comprehensive evaluation and analysis method for road service life as described in the first aspect of the present invention.

[0016] The beneficial effects of the present invention are as follows: through the steps of multi-dimensional data collection and preprocessing, spatio-temporal correlation analysis, health prediction model construction and dynamic weight adjustment, a comprehensive evaluation of the service life of the road is achieved. By introducing spatio-temporal correlation analysis, the spatial distribution and temporal variation of road characteristics are quantified, and a complete evaluation data framework is constructed; by combining the health index and historical data to dynamically adjust the feature weights, the accuracy and flexibility of health status prediction are optimized. Compared with the prior art, the present invention solves the problems of insufficient multi-dimensional data integration and lack of dynamic analysis, significantly improves the scientific nature of the evaluation results and the practical application value, and provides a reliable basis for road management and safety decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1This is a flow chart of the comprehensive evaluation and analysis method for road service life in Example 1.

[0019] Figure 2 Schematic diagram of the comprehensive evaluation and analysis system for road service life in Example 1. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0023] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a method for comprehensive evaluation and analysis of road service life, comprising the following steps: S1: Collect and preprocess multi-dimensional road data, and extract road features based on the pre-processed multi-dimensional road data, including: S1.1: Collect multi-dimensional road data; the multi-dimensional road data includes road surface disease data, road internal structure data, traffic load data and environmental data.

[0024] It should be noted that the above data can comprehensively obtain the key factors affecting the service life of roads, and provide scientific and accurate basic data support for subsequent feature extraction, life assessment and health status prediction.

[0025] S1.2: Preprocessing the multi-dimensional road data includes the following steps: The pavement disease data is cleaned by a noise filtering algorithm to remove blurred images and invalid data; for example, the pavement disease data is cleaned by a noise filtering algorithm to remove blurred crack images and invalid data, thereby ensuring the accuracy of disease feature extraction.

[0026] Spectrum analysis is used to filter out clutter signals from the road internal structure data; for example, by processing the road internal structure data using spectrum analysis, clutter signals can be effectively filtered out, thereby accurately reflecting structural features such as void ratio and layer thickness.

[0027] Standardize and normalize traffic load data; for example, standardize and normalize traffic load data to unify flow data collected by different monitoring equipment to ensure data comparability and analysis accuracy.

[0028] Data interpolation methods are used to fill missing values ​​in environmental data and remove abnormal data; for example, data interpolation methods are used to fill missing values ​​in environmental data and remove abnormal data, which can fully reflect environmental characteristics such as rainfall frequency or temperature and humidity changes.

[0029] S1.3: Extract road features based on the preprocessed multi-dimensional road data. The specific steps are as follows: S1.31: Extract pavement damage features from the cleaned pavement damage data through image segmentation algorithm; the damage features include crack width, crack length, rutting depth and damage area.

[0030] Furthermore, specifically as follows, first, the cleaned pavement disease data is processed using an image segmentation algorithm to accurately segment the diseased area from the non-disease area in the pavement image; then, for the segmented diseased area, pavement disease characteristics are extracted, including crack width, crack length, rutting depth and diseased area area; wherein the crack width and crack length are calculated by analyzing the geometric boundaries of the diseased area, the rutting depth is quantified by quantifying the depth information of the pavement disease data, and the diseased area is determined by calculating the total number of pixels in the diseased area and converting it into an actual area value.

[0031] S1.32: The road internal structure characteristics are extracted from the road internal structure data after clutter is filtered out by signal analysis method; the road internal structure characteristics include the void ratio, stratification depth and void area range of the internal structure data.

[0032] Specifically, the signal analysis method is used to analyze the processed (i.e., after filtering out clutter) road internal structure data to extract the road internal structure characteristics, including the void ratio, stratification depth, and void area range of the internal structure data; wherein, the void ratio of the internal structure data is calculated by analyzing the frequency and amplitude ratio of abnormal fluctuations in the signal, the stratification depth is determined by inverting the delay time and waveform characteristics of the signal, and the void area range is quantified and extracted through the spatial distribution area of ​​the abnormal signal.

[0033] S1.33: Traffic load characteristics are extracted from the standard normalized traffic load data through statistical analysis; the traffic load characteristics include traffic flow peak, vehicle load distribution and traffic pressure distribution mean.

[0034] Specifically, first, the traffic load data after standard normalization is sorted and classified to ensure the integrity and accuracy of the data coverage time period; then, the traffic load data is processed by statistical analysis methods to extract traffic load characteristics, including traffic flow peak, vehicle load distribution and traffic pressure distribution mean; among them, the traffic flow peak is extracted by identifying the maximum traffic flow value per unit time in the traffic load data, the vehicle load distribution is obtained by frequency statistics and distribution analysis of the load conditions of different types of vehicles, and the traffic pressure distribution mean is calculated by averaging the values ​​of traffic pressure distribution in different time periods.

[0035] S1.34: Extract environmental erosion characteristics from preprocessed environmental data through data analysis methods; environmental erosion characteristics include rainfall frequency, temperature and humidity variation, and number of freeze-thaw cycles.

[0036] Specifically, the environmental data are processed and counted using data analysis methods to extract environmental erosion characteristics, including rainfall frequency, temperature and humidity variation, and freeze-thaw cycle number. The rainfall frequency is calculated by counting the number of rainfall events per unit time in the environmental data, the temperature and humidity variation is obtained by analyzing the difference between the maximum and minimum values ​​of temperature and humidity in the environmental data, and the freeze-thaw cycle number is identified and counted based on the number of times the temperature in the environmental data crosses zero degrees Celsius.

[0037] In the above embodiment, the road characteristics include: road surface disease characteristics, road internal structure characteristics, traffic load characteristics and environmental erosion characteristics.

[0038] S2: Conduct spatiotemporal correlation analysis on road characteristics and construct a road life assessment data framework, including: S2.1: Based on the road characteristics, the spatial units are divided according to fixed standards, and the road characteristics are mapped to the divided spatial units.

[0039] Furthermore, the disease characteristics: according to the spatial distribution of the disease (such as crack location, subsidence range), the characteristic values ​​(such as crack density, disease area ratio) are mapped to the corresponding spatial units. Point disease characteristics can be expanded to range characteristics (such as through the buffer zone method).

[0040] Internal structural features: The internal features of the road (such as void ratio and layer thickness) are assigned to the corresponding spatial units. If the feature distribution is discontinuous, a complete spatial feature map is generated through interpolation or distribution model.

[0041] Traffic load characteristics: According to the location and coverage of monitoring equipment, load characteristics (such as traffic flow, proportion of heavy-loaded vehicles) are attributed to the corresponding spatial unit. If the load characteristics involve multiple road sections, the characteristic contribution value is decomposed through the allocation rules.

[0042] Environmental characteristics: Regional environmental characteristics (such as rainfall, freeze-thaw cycles, and temperature changes) are superimposed on the corresponding spatial units. If they span multiple regions, the characteristic values ​​are allocated according to the weights of the areas covered by the unit.

[0043] S2.2: Perform spatial analysis on the road features mapped into the spatial units to obtain the spatial distribution of the road features.

[0044] S2.3: Perform time correlation analysis on road characteristics to obtain the time variation patterns and interdependencies of road characteristics.

[0045] Furthermore, firstly, based on the long-term collected road feature data, the road features are segmented and counted at fixed time intervals to ensure the consistency of the time dimension of the data.

[0046] Then, using the time series analysis method, for each type of road characteristics, including pavement disease characteristics, road internal structure characteristics, traffic load characteristics and environmental erosion characteristics, the changing trend at continuous time nodes is analyzed one by one to obtain the time change law of road characteristics.

[0047] Finally, by calculating the time correlation index and time dependence coefficient between road characteristics, the interdependence of pavement disease characteristics, road internal structure characteristics, traffic load characteristics and environmental erosion characteristics over time is revealed, providing dynamic time characteristic support for road life assessment.

[0048] S2.4: Quantify the spatial distribution of road features, the temporal variation patterns of road features and their interdependence, and combine them according to spatial units and time nodes to generate the feature vector of each road section at each time point and construct a multi-dimensional feature matrix.

[0049] Furthermore, the multidimensional feature matrix expression is: ; Among them, M is a multidimensional feature matrix, N represents the number of spatial units, and Z represents the number of time nodes. Represents the feature vector of the Nth spatial unit at the Tth time node.

[0050] The multidimensional feature matrix includes pavement disease characteristics, temporal changes, spatial distribution and environmental erosion characteristics.

[0051] It should be noted that the multidimensional feature matrix The expression of integrates the spatial distribution of roads (N represents the number of spatial units) and temporal changes (Z represents the number of time nodes) in matrix form. The specific eigenvector It reflects the characteristic state of each spatial unit at each time node. These characteristic vectors include pavement disease characteristics (such as crack width, rutting depth, etc.), time change characteristics (such as the deterioration trend of diseases over time), spatial distribution characteristics (such as the geographical distribution relationship of each road section) and environmental erosion characteristics (such as the influence of environmental factors such as rainfall, freeze-thaw, etc.). The dynamic changes of road health status in space and time can be fully expressed in the form of matrix m, providing complete data support for road life assessment.

[0052] S2.5: Based on the multidimensional feature matrix and combined with historical data, the life status of each road section is marked, and the multidimensional feature matrix and the life status of each road section are integrated to form a road life assessment data framework.

[0053] Furthermore, firstly, according to the characteristic vector of each spatial unit at each time node in the multidimensional feature matrix, key indicators including pavement disease characteristics, road internal structure characteristics, traffic load characteristics and environmental erosion characteristics are extracted; secondly, according to the definition standard of road life status, the comprehensive performance of pavement disease characteristics, road internal structure characteristics, traffic load characteristics and environmental erosion characteristics is compared with the life status data in the historical data, and the life status of each spatial unit at each time node is marked. The life status includes clear levels such as healthy state, sub-healthy state, risk state and dangerous state; then, the marked life status data is integrated with the pavement disease characteristics, road internal structure characteristics, traffic load characteristics and environmental erosion characteristics in the multidimensional feature matrix to form a complete road life assessment data framework, which provides accurate data support for the subsequent road life prediction model and road health prediction model.

[0054] It should be noted that building a road life assessment data framework can more comprehensively and scientifically reflect the dynamic changes in road health status and its service life. For example, the life status of different road sections can be used as implicit constraints or correction parameters to provide more accurate input for the road health prediction model. The introduction of life status not only enhances the robustness and adaptability of the model, but also provides higher reliability for subsequent optimization and decision-making. Therefore, the purpose of building a framework is to combine short-term health status with long-term life assessment to improve the scientificity and accuracy of the overall assessment.

[0055] S3: Based on the road life assessment data framework, the road health prediction model is used to preliminarily calculate the road health index, including: S3.1: Taking the multidimensional feature matrix in the road life assessment data framework as input, the road health prediction model is used to analyze the contribution of the multidimensional feature matrix to the health status, and the road health index is preliminarily calculated. The expression is: ; in, is the preliminary calculated road health index, is the comprehensive weight of the road health index, is the severity of the i-th type of pavement disease characteristics, is the influence coefficient of the i-th type of pavement disease characteristics, is the time sensitivity coefficient of the i-th type of pavement disease characteristics, is the cumulative time span of the i-th type of pavement disease characteristics, is the influence coefficient of the internal structural characteristics of the road, is the fluctuation variance of traffic load characteristics, is the sensitivity coefficient of environmental erosion characteristics, is the comprehensive environmental erosion index, is the spatial scale factor of the i-th type of pavement damage feature, n is the number of pavement damage features, and i is the index variable of the pavement damage feature.

[0056] It should be noted that the expression , , Together, E reflects the health status or life status of the road in multiple dimensions. As a combination of historical data and multidimensional features, the life status provides key contextual information to describe the health trend and degradation law of the road section, and is an important part of the data framework. At the same time, the life status is the formula Provide basic data support to enable comprehensive assessment of road health and lifespan through mathematical modeling.

[0057] In this embodiment, the road health prediction model is constructed based on MTL-LSTM. The main task in MTL-LSTM analyzes the contribution of the multidimensional feature matrix to the health status, and then obtains the road health index.

[0058] Furthermore, the expression of the comprehensive environmental erosion index E is:

[0059] in, is the cumulative frequency of the erosion characteristics of the i-th type of environment, is the severity of the erosion characteristics of the i-th type of environment.

[0060] It should be noted that the spatial scale factor of the i-th type of pavement disease characteristics is , specifically indicating the spatial distribution range of the disease characteristics, such as the length of cracks or the length of road sections affected by rutting.

[0061] The severity of the erosion characteristics of the i-th type of environment The acquisition process is as follows: First, extract the original data related to the i-th type of environmental erosion characteristics based on the environmental data source (such as meteorological monitoring system or long-term observation data), such as temperature change amplitude, rainfall intensity or freeze-thaw cycle amplitude, etc.; then, select appropriate quantitative indicators (such as the maximum rainfall intensity, the mean value of temperature and humidity changes, the temperature peak difference of freeze-thaw cycle, etc.) according to the actual impact of the characteristics; then, normalize or standardize these quantitative indicators to uniformly measure the impact of different environmental erosion characteristics; finally, determine the severity of the i-th type of environmental erosion characteristics through statistical analysis or expert experience assignment. , which is used to characterize the erosion impact of this feature on the road health status.

[0062] S4: Combine the preliminary calculated road health index and historical data to dynamically adjust the comprehensive weight of road characteristics, including: S4.1: Preliminary calculation of road health index and historical data, using a weighted time decay function and combining the road health index and characteristic historical data to dynamically adjust the comprehensive weight, the expression is: ; in, is the adjusted comprehensive weight, is the historical time span, is the time decay function, is the historical value of the i-th type of pavement disease characteristics at time t, are the elements of the disease feature correlation matrix, is the historical value of the j-th disease characteristic at time t, is the regularization factor of the i-th disease feature, represents a historical time point, j represents the disease feature index variable that is correlated with the current pavement disease feature i, Is a measure of road health index The degree of impact on road characteristics.

[0063] Furthermore, the time decay function The expression is: ; in, is the weight decay rate of historical data.

[0064] Regularization factor of the i-th disease characteristic The expression is: ; in, is the historical variance of the i-th feature, It represents the historical fluctuation degree of the pavement disease characteristics of the i-th type, and measures the variation range of the pavement disease characteristics in the time dimension. It represents the historical fluctuation degree (such as variance) of the j-th type of pavement disease characteristics, and is used to sum with other pavement disease characteristics to measure the overall fluctuation distribution.

[0065] It should be noted that i specifically represents the index of the target pavement defect feature currently being calculated (such as crack severity), and j represents the index of other pavement defect features that are correlated with the target pavement defect feature i (such as the influence of rutting depth on crack severity).

[0066] S5: Update the road health index and evaluate the road safety level based on the adjusted comprehensive weights of road characteristics, including: S5.1: Comprehensive weights based on adjusted road characteristics , optimize the road health prediction model, and update the road health index based on the optimized road health prediction model , get the updated road health index .

[0067] For example, in the health assessment of a highway section, the multidimensional feature matrix M contains the crack width , rutting depth , rainfall erosion E and traffic load intensity T. The weights in the initial model Assuming that cracks and rutting features account for the majority and rainfall and traffic loads have low impact, the calculated health index is =68.5. However, through historical data analysis, it was found that the road section is located in a hot and humid area with frequent rainfall, the impact of rainfall erosion on road health is underestimated, and the impact of crack characteristics is relatively small.

[0068] To this end, adjust the comprehensive weight , increasing the weight of rainfall erosion E and reducing the crack characteristics For example, the rainfall weight is increased from 0.2 to 0.4, and the crack weight is reduced from 0.5 to 0.3. =0.85 is substituted into the optimized health prediction model and recalculated After optimization, the health index increased to =72.3, which more accurately reflects the actual impact of rainfall on road degradation.

[0069] Finally, by comparing the optimized Actual test data (such as crack expansion rate and rutting deepening rate after rainfall) verify that the model prediction value is more consistent with the actual status. The optimized model is more scientific and reliable for road health assessment in hot and humid climates, providing a basis for road maintenance decisions.

[0070] It should be noted that this process dynamically reflects the actual contribution of different pavement disease characteristics, road internal structure characteristics, traffic load characteristics and environmental erosion characteristics to the road health status, enabling the prediction model to more accurately describe the changing laws of road health status, thereby improving the road health index. The calculation accuracy can provide more scientific data support for subsequent road safety level assessment, risk warning and maintenance decision-making.

[0071] S5.2: Based on the long-term collected statistical results of road health index and actual road service life, combined with road design specifications and safety assessment standards, define the safety threshold R1, risk threshold R2 and danger threshold R3, and the safety threshold R1>risk threshold R2>danger threshold R3.

[0072] when >R1, it is considered that the current road service life is in a safe state.

[0073] When R2≤ When ≤R1, the current road service life is considered to be in good condition.

[0074] When R3≤ When <R2, it is considered that the current road service life is at risk.

[0075] when When <R3, it is considered that the current road service life is in a dangerous state.

[0076] Furthermore, first, the health index of roads at different stages of use is collected. Data, combined with the statistical results of road design life and actual service life, analysis The distribution of health indexes is statistically analyzed, with special attention paid to roads that are undergoing major or medium repairs, or are about to fail or close. Secondly, combining the road design specifications and safety assessment standards, extract key performance indicators (such as crack rate, rutting depth, etc.) and their corresponding limits, and compare these technical indicators with the health index. For example, the design specification stipulates that a crack rate below 10% is a "good state" and a crack rate of 30% is a "dangerous state". It is divided into safety threshold R1, risk threshold R2 and danger threshold R3. Finally, through statistical analysis, the health index The historical data distribution is combined with the performance indicator limits to define the threshold range, such as R1=80 (safe), R2=60 (risk), and R3=40 (dangerous), providing a scientific basis for road health status assessment.

[0077] The present embodiment also provides a road service life comprehensive evaluation and analysis system, including: a feature extraction module, a road feature analysis module, a road health index calculation module, a comprehensive weight adjustment module and a road safety evaluation module; the feature extraction module is used to collect and preprocess multi-dimensional road data, and extract road features based on the preprocessed multi-dimensional road data; the road feature analysis module is used to perform spatiotemporal correlation analysis on road features and construct a road life evaluation data framework; the road health index calculation module is used to preliminarily calculate the road health index based on the road life evaluation data framework and using a road health prediction model; the comprehensive weight adjustment module is used to dynamically adjust the comprehensive weight of road features in combination with the preliminarily calculated road health index and historical data; the road safety evaluation module is used to update the road health index and evaluate the road safety level according to the adjusted comprehensive weight of road features.

[0078] This embodiment also provides a computer device, which is suitable for the comprehensive evaluation and analysis method of road service life, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the comprehensive evaluation and analysis method of road service life proposed in the above embodiment.

[0079] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0080] The present embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for comprehensive evaluation and analysis of road service life proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0081] In summary, the present invention achieves a comprehensive evaluation of the service life of roads through the following steps: multi-dimensional data collection and preprocessing, spatio-temporal correlation analysis, health prediction model construction, and dynamic weight adjustment. By introducing spatio-temporal correlation analysis, the spatial distribution and temporal variation of road characteristics are quantified, and a complete evaluation data framework is constructed; by combining health indexes and historical data to dynamically adjust feature weights, the accuracy and flexibility of health status predictions are optimized. Compared with the prior art, the present invention solves the problems of insufficient multi-dimensional data integration and lack of dynamic analysis, significantly improves the scientific nature and practical application value of the evaluation results, and provides a reliable basis for road management and safety decision-making.

[0082] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of the comprehensive evaluation and analysis method of road service life are provided.

[0083] In order to verify the effectiveness of the feature extraction and health index calculation method based on multi-dimensional road data, and to compare the existing technology with the technical solution of the present invention, this experiment selected a certain urban road as the research object, and collected multi-dimensional road data including pavement disease data, road internal structure data, traffic load data and environmental data. The specific implementation process is as follows: First, this experiment selected three roads in a city as the research objects and collected multi-dimensional road data including pavement disease data, road internal structure data, traffic load data and environmental data. The original data was preprocessed through noise filtering, spectrum analysis, standardization processing and data interpolation to ensure the integrity and consistency of the data.

[0084] Secondly, advanced algorithms are used to extract key features from the preprocessed data, including crack width, crack length, void ratio, traffic flow peak, etc. At the same time, environmental erosion characteristics such as rainfall frequency, temperature and humidity changes are extracted in combination with environmental data, laying the foundation for the calculation of the health index.

[0085] Next, based on the extracted multidimensional features, the road health index was calculated using the regularized weighting method of the prior art and the MTL-LSTM dynamic weight optimization model of the present invention. The health index results of different schemes were compared in the experiment, and the feature extraction values ​​and health index values ​​under each technical scheme were recorded.

[0086] Finally, through data analysis, it was found that the solution of the present invention is significantly superior to the existing technology in terms of feature extraction accuracy and health index calculation accuracy. The experimental results show that the solution of the present invention can not only more accurately evaluate the road health status, but also provide more reliable data support for road maintenance and life prediction.

[0087] In this experiment, the existing technology uses the traditional regularized weighted method to calculate the road health index. Specifically, the features such as crack width, crack length, void rate, and traffic flow peak are linearly weighted and summed by fixed weights, and the health index is directly superimposed. In terms of feature extraction, the crack features are extracted using traditional image processing algorithms, the void rate is estimated based on signal amplitude and frequency, and the traffic load data is processed using the statistical mean method. Due to the lack of dynamic weight adjustment, the health index calculation results of the existing technology are relatively rough and have limited accuracy.

[0088] The details are shown in Table 1 below: Table 1 Comparison of multi-dimensional road data characteristics and health index

[0089] Through the data analysis of the above table, it can be clearly seen that the present invention is significantly superior to the prior art in terms of feature extraction accuracy and health index calculation accuracy. The present invention effectively reduces the extraction errors of key indicators such as crack width, crack length and void ratio through improved image segmentation algorithms and signal analysis methods, and at the same time improves the scientific nature of the health index through a dynamic weight optimization model. For example, in Road 3 (the solution of the present invention), the crack width is only 2.6 mm, the void ratio is 1.7%, and the health index reaches 86.5, which is significantly better than the health index of 66.7 of Road 3 (the prior art). This shows that the present invention can more accurately reflect the actual health status and service life of the road.

[0090] The comprehensive evaluation and analysis method for road service life of the present invention can not only improve the accuracy of key feature extraction and the consistency of data processing, but also accurately calculate the road health index and scientifically evaluate the service life of the road. This reflects the innovation and practicality of the present invention in data processing, dynamic weight optimization and health status prediction, and provides reliable technical support for road maintenance and decision-making.

[0091] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A comprehensive evaluation and analysis method for road service life, characterized by: include, Collecting and preprocessing multi-dimensional road data, and extracting road features based on the preprocessed multi-dimensional road data; The multi-dimensional road data includes road surface disease data, road internal structure data, traffic load data and environmental data; Conduct spatiotemporal correlation analysis on road characteristics and construct a data framework for road life assessment; Based on the road life assessment data framework, the road health index is preliminarily calculated using the road health prediction model; Combine the preliminary calculated road health index with historical data to dynamically adjust the comprehensive weight of road characteristics; Based on the adjusted comprehensive weights of road characteristics, the road health index is updated and the road safety level is evaluated.

2. The road service life comprehensive evaluation and analysis method according to claim 1, characterized in that: Preprocess the multidimensional road data. The specific steps are as follows: The pavement disease data is cleaned by using a noise filtering algorithm to remove blurry images and invalid data; Spectral analysis is used to filter out clutter signals from the road internal structure data; Standardize and normalize traffic load data; Data interpolation methods are used to fill missing values ​​in environmental data and remove abnormal data.

3. The road service life comprehensive evaluation and analysis method as claimed in claim 2 is characterized in that: The specific steps of extracting road features based on pre-processed multi-dimensional road data are as follows: The pavement disease characteristics are extracted from the cleaned pavement disease data through image segmentation algorithm; the pavement disease characteristics include crack width, crack length, rutting depth and disease area; The internal structure features of the road are extracted from the internal structure data of the road after the clutter is filtered out by using the signal analysis method; The internal structural characteristics of the road include the void ratio, layer depth and void area range of the internal structural data; Extracting traffic load characteristics from the standard normalized traffic load data through statistical analysis; the traffic load characteristics include traffic flow peak, vehicle load distribution and traffic pressure distribution mean; Environmental erosion characteristics are extracted from pre-processed environmental data by data analysis; environmental erosion characteristics include rainfall frequency, temperature and humidity variation, and freeze-thaw cycle times; Road characteristics include: pavement disease characteristics, road internal structure characteristics, traffic load characteristics and environmental erosion characteristics.

4. The road service life comprehensive evaluation and analysis method as claimed in claim 3 is characterized in that: The specific steps of performing spatiotemporal correlation analysis on road characteristics and constructing a road life assessment data framework are as follows: Based on the road characteristics, the spatial units are divided according to fixed standards, and the road characteristics are mapped to the divided spatial units; Perform spatial analysis on the road features mapped into the spatial units to obtain the spatial distribution of the road features; Conduct time correlation analysis on road characteristics to obtain the time variation patterns and interdependencies of road characteristics; The spatial distribution of road features, the temporal variation patterns of road features and their interdependence are quantified, and combined according to spatial units and time nodes to generate the feature vector of each road section at each time point and construct a multi-dimensional feature matrix; The multidimensional feature matrix includes pavement disease characteristics, temporal changes, spatial distribution and environmental erosion characteristics; Based on the multi-dimensional feature matrix and combined with historical data, the life status of each road section is marked; The multi-dimensional feature matrix and the life status of each road section are integrated to form a road life assessment data framework.

5. The road service life comprehensive evaluation and analysis method as claimed in claim 4 is characterized in that: Based on the road life assessment data framework, the road health prediction model is used to preliminarily calculate the road health index. The specific steps are as follows: Taking the multidimensional feature matrix in the road life assessment data framework as input, the road health prediction model is used to analyze the contribution of the multidimensional feature matrix to the health status, and the road health index is preliminarily calculated. The expression is: ; in, is the preliminary calculated road health index, is the comprehensive weight of the road health index, is the severity of the i-th type of pavement disease characteristics, is the influence coefficient of the i-th type of pavement disease characteristics, is the time sensitivity coefficient of the i-th type of pavement disease characteristics, is the cumulative time span of the i-th type of pavement disease characteristics, is the influence coefficient of the internal structural characteristics of the road, is the fluctuation variance of traffic load characteristics, is the sensitivity coefficient of environmental erosion characteristics, is the comprehensive environmental erosion index, is the spatial scale factor of the i-th type of pavement damage feature, n is the number of pavement damage features, and i is the index variable of the pavement damage feature.

6. The road service life comprehensive evaluation and analysis method as claimed in claim 5, characterized in that: The specific steps of dynamically adjusting the comprehensive weight of road characteristics by combining the initially calculated road health index and historical data are as follows: The road health index was calculated by And historical data, the weighted time decay function is used to dynamically adjust the comprehensive weight, the expression is: ; in, is the adjusted comprehensive weight, is the historical time span, is the time decay function, is the historical value of the i-th type of pavement disease characteristics at time t, are the elements of the disease feature correlation matrix, is the historical value of the j-th disease characteristic at time t, is the regularization factor of the i-th disease feature, represents a historical time point, j represents the disease feature index variable that is correlated with the current pavement disease feature i, Is a measure of road health index The degree of impact on road characteristics.

7. The road service life comprehensive evaluation and analysis method according to claim 6 is characterized in that: The road health index is updated according to the adjusted comprehensive weight of the road characteristics, and the road safety level is evaluated. The specific steps are as follows: Comprehensive weight based on adjusted road characteristics , optimize the road health prediction model, and update the road health index based on the optimized road health prediction model , get the updated road health index ; Based on the long-term collection of road health index and the statistical results of the actual service life of the road, combined with the road design specifications and safety assessment standards, the safety threshold R1, risk threshold R2 and danger threshold R3 are defined, and the safety threshold R1>risk threshold R2>danger threshold R3; when >R1, the current road service life is considered to be in a safe state; When R2≤ When ≤R1, the current road service life is considered to be in good condition; When R3≤ When <R2, it is considered that the current road service life is at risk; when When <R3, it is considered that the current road service life is in a dangerous state.

8. A road service life comprehensive evaluation and analysis system, based on the road service life comprehensive evaluation and analysis method according to any one of claims 1 to 7, characterized in that: include: Feature extraction module, road feature analysis module, road health index calculation module, comprehensive weight adjustment module and road safety assessment module; A feature extraction module is used to collect and preprocess multi-dimensional road data, and extract road features based on the pre-processed multi-dimensional road data; the multi-dimensional road data includes road disease data, road internal structure data, traffic load data and environmental data; Road feature analysis module, used to conduct spatiotemporal correlation analysis on road features and build a data framework for road life assessment; A road health index calculation module is used to preliminarily calculate the road health index based on the road life assessment data framework and using the road health prediction model; A comprehensive weight adjustment module is used to dynamically adjust the comprehensive weight of road characteristics by combining the initially calculated road health index and historical data; The road safety assessment module is used to update the road health index and evaluate the road safety level according to the adjusted comprehensive weights of road characteristics.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the road service life comprehensive evaluation and analysis method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the road service life comprehensive evaluation and analysis method described in any one of claims 1 to 7 are implemented.

Images