Highway traffic volume axle load dynamic high-precision detection method

Abstract

The invention relates to the technical field of traffic detection, in particular to a highway traffic volume axle load dynamic high-precision detection method. The method comprises the following steps: acquiring highway traffic volume original axle load detection data; performing space-time synchronization and noise suppression processing on the highway traffic volume original axle load detection data, and performing load-response mapping estimation at the same time to obtain a highway traffic volume axle load preliminary estimation value; obtaining a vehicle type database, carrying out vehicle axle load verification and time-phased distribution statistics, and meanwhile, carrying out pavement axle load accumulated damage assessment to obtain an expressway pavement axle load accumulated damage value; and dynamic high-precision detection is carried out based on the highway axle load distribution data in different periods and the highway pavement axle load accumulated damage value, and a corresponding highway pavement axle load maintenance scheme is formulated. According to the invention, real-time acquisition and dynamic updating of highway axle load data can be realized, and high-precision data support is provided for road maintenance and traffic management.

Description

Technical Field

[0001] This invention relates to the field of traffic detection technology, and in particular to a method for dynamic high-precision detection of axle load in highway traffic volume. Background Technology

[0002] With the continuous development of the social economy and the acceleration of urbanization, the traffic volume on highways is constantly increasing, and traffic management and road maintenance are facing unprecedented challenges. Dynamic monitoring of traffic volume and vehicle axle load is an important means to ensure the safe operation of highways and optimize traffic management. In highway dynamic detection, vehicle axle load weight is one of the key data for assessing road load, predicting pavement damage, and formulating maintenance strategies. However, traditional detection methods based on static or simplified dynamic models cannot fully reflect the dynamic load characteristics of vehicles under high-speed driving conditions, resulting in low accuracy of detection results. At the same time, they cannot provide stable detection data under changing environmental conditions, especially under weather changes, heavy traffic, or high vehicle speeds, and are often subject to interference, resulting in unstable detection data and the inability to plan road maintenance in a timely manner. Summary of the Invention

[0003] Therefore, it is necessary for the present invention to provide a method for dynamic high-precision detection of axle load in highway traffic volume, so as to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a method for dynamic high-precision detection of axle load in highway traffic volume includes the following steps:

[0005] Step S1: Collect dynamic response signals and vehicle traffic characteristic data corresponding to the highway road surface through a multi-type sensor array to obtain the original axle load detection data of highway traffic volume; perform spatiotemporal synchronization and noise suppression processing on the original axle load detection data of highway traffic volume to generate a standardized axle load detection dataset.

[0006] Step S2: Based on the standardized axle load detection dataset, identify vehicle axle type and locate axle position to extract axle group distribution characteristics and axle spacing parameters, and generate highway vehicle axle system feature data; based on the highway vehicle axle system feature data and combined with the attenuation characteristics corresponding to the road surface response signal, perform load-response mapping estimation to obtain a preliminary estimate of highway traffic volume axle load.

[0007] Step S3: Obtain the vehicle type database, and perform vehicle axle load verification and time-segmented distribution statistics on the standardized axle load detection dataset based on the preliminary estimate of highway traffic volume axle load and the vehicle type database to obtain highway time-segmented axle load distribution data; obtain highway pavement structure parameters, and perform pavement axle load cumulative damage assessment based on highway pavement structure parameters combined with highway time-segmented axle load distribution data to obtain highway pavement axle load cumulative damage value.

[0008] Step S4: Perform dynamic high-precision detection based on the time-segmented axle load distribution data of the expressway and the cumulative axle load damage value of the expressway pavement to generate a dynamic detection report of expressway traffic volume axle load; formulate a corresponding expressway pavement axle load maintenance plan based on the dynamic detection report of expressway traffic volume axle load.

[0009] Furthermore, step S1 includes the following steps:

[0010] Step S11: By deploying strain sensors and acceleration sensors at different depths on the highway surface, and deploying a multi-type sensor array consisting of laser profile sensors and video recognition equipment on the side of the highway, the strain sensors collect the dynamic response signal corresponding to the road surface strain when a vehicle passes, the acceleration sensors collect the dynamic response signal corresponding to the road surface vibration, the laser profile sensors collect the vehicle length and wheelbase geometric parameters, and the video recognition equipment collects the vehicle type and travel trajectory, the original axle load detection data of highway traffic volume is obtained.

[0011] Step S12: Based on the time synchronization module corresponding to the multi-type sensor array, the acquisition timestamps of each sensor are unified to the same clock reference, and the error is controlled within the preset range to generate multi-source data time synchronization parameters.

[0012] Step S13: Establish the coordinate mapping relationship of each sensor detection point according to the spatial layout of each sensor in the multi-type sensor array, and associate the dynamic response signal in the original axle load detection data of highway traffic volume with the vehicle traffic characteristic data to the same coordinate mapping relationship to generate multi-source data spatial calibration parameters.

[0013] Step S14: Based on the time synchronization parameters and spatial calibration parameters of multi-source data, the original axle load detection data of highway traffic volume is spatiotemporally synchronized and integrated to generate a spatiotemporally aligned detection dataset of highway axle load containing time and space.

[0014] Step S15: Perform noise suppression and standardization on the highway axle load spatiotemporal alignment detection dataset. Wavelet threshold denoising is used to remove high-frequency interference noise and standardize the data to generate a standardized axle load detection dataset.

[0015] Furthermore, step S2 includes the following steps:

[0016] Step S21: Perform peak value statistics on the corresponding dynamic response signals in the standardized axle load detection dataset to statistically analyze the peak occurrence time, peak amplitude, and peak signal spectrum of each peak value of the highway axle load dynamic signal, and generate a highway axle load signal peak value sequence.

[0017] Step S22: Perform signal peak feature analysis based on the peak sequence of highway axle load signal to obtain the peak time interval, amplitude change rate and spectral energy distribution corresponding to the dynamic signal of highway axle load;

[0018] Step S23: Based on the peak time interval, amplitude change rate, and spectral energy distribution of the dynamic signal of highway axle load, and combined with the corresponding vehicle type and traffic trajectory in the standardized axle load detection dataset, perform vehicle axle type identification analysis to identify and distinguish vehicle axle type groups corresponding to single axle, parallel dual axle, and parallel triple axle, and obtain the highway traffic volume axle type group classification results.

[0019] Step S24: Based on the classification results of highway traffic volume axle type groups and combined with the peak time interval and amplitude change rate corresponding to the highway axle load dynamic signal, perform vehicle axle position positioning statistics on the standardized axle load detection dataset to generate highway vehicle axle system feature data.

[0020] Step S25: Based on the axle system characteristic data of highway vehicles and combined with the attenuation characteristics of the road surface response signal, perform load-response mapping estimation to obtain a preliminary estimate of the axle load of highway traffic volume.

[0021] Furthermore, step S24 includes the following steps:

[0022] Step S241: Obtain the corresponding highway vehicle speed by matching the vehicle type and travel trajectory within the standardized axle load detection dataset;

[0023] Step S242: Based on the peak time interval and amplitude change rate corresponding to the dynamic signal of highway axle load, and combined with the speed of highway vehicles, the vehicle axle spacing is converted to obtain the initial parameters of highway vehicle axle spacing;

[0024] Step S243: Based on the vehicle length and wheelbase geometric parameters corresponding to the standardized axle load detection dataset, perform axle spacing comparison and correction calculation on the initial parameters of highway vehicle axle spacing to obtain highway axle spacing comparison and correction parameters;

[0025] Step S244: Based on the classification results of axle type groups of highway traffic volume and combined with the highway vehicle speed, calculate the axle position of each axle group to locate the corresponding highway pavement coordinates and generate a dataset of axle position coordinates for each axle group; perform axle group distribution statistics based on the dataset of axle position coordinates for each axle group to analyze and extract the number and distribution of axles in each axle group, and obtain the distribution characteristics of each axle group of the highway.

[0026] Step S245: Based on the highway axle spacing comparison correction parameters and the distribution characteristics of each axle group on the highway, integrate the vehicle axle system features to generate highway vehicle axle system feature data containing the number of axles, axle position and axle spacing of each axle group.

[0027] Furthermore, step S25 includes the following steps:

[0028] Step S251: Perform signal attenuation characteristic analysis on the pavement response signals corresponding to pavement strain and pavement vibration in the standardized axle load test dataset to calculate the amplitude attenuation rate and phase shift of the pavement response signal at different propagation distances, and generate highway pavement response attenuation characteristic parameters.

[0029] Step S252: Based on the axle coordinate information in the highway vehicle axle system feature data, determine the road surface response signal acquisition points corresponding to each axle group, and perform axle position-attenuation correlation analysis based on the road surface response signal acquisition points corresponding to each axle group and the highway road surface response attenuation feature parameters to obtain highway vehicle axle position-response attenuation correlation data.

[0030] Step S253: Based on the number of axles and axle spacing in the axle system characteristic data of highway vehicles and combined with the response attenuation characteristic parameters corresponding to the axle position-response attenuation correlation data of highway vehicles, perform load-response nonlinear mapping analysis to obtain the nonlinear mapping relationship between highway vehicle axle load and road surface response signal attenuation.

[0031] Step S254: Based on the nonlinear mapping relationship between highway vehicle axle load and road surface response signal attenuation, and combined with highway vehicle axle system characteristic data and highway road surface response attenuation characteristic parameters, construct the axle load-response mapping function to generate the highway vehicle axle load-response mapping function;

[0032] Step S255: Perform load-response mapping estimation based on the highway vehicle axle load-response mapping function to obtain a preliminary estimate of highway traffic volume axle load.

[0033] Furthermore, step 255 includes the following steps:

[0034] By deploying environmental sensors on the corresponding highway surface to synchronously collect corresponding temperature, humidity, road surface dryness and wetness status and rainfall data, a highway surface environmental parameter dataset is generated.

[0035] Based on the environmental parameters in the highway pavement environmental parameter dataset, an environmental-axle load characteristic influence analysis is performed on the highway vehicle axle system characteristic data to generate the highway environmental-axle load characteristic influence distribution curve.

[0036] Based on the distribution curve of the influence of highway environment-axle load characteristics, the corresponding highway axle load influence correction factor under different environmental condition combinations is calculated. Based on the highway axle load influence correction factor and the highway vehicle axle load-response mapping function, the load-response mapping correction estimation is performed to obtain the preliminary estimate of highway traffic volume axle load.

[0037] Furthermore, step S3 includes the following steps:

[0038] Step S31: Obtain the vehicle type database, which includes the standard axle load range, typical axle group configuration and axle weight distribution for different vehicle types;

[0039] Step S32: Based on the preliminary estimate of highway traffic volume axle load and the vehicle type database, perform vehicle axle load verification on the standardized axle load detection dataset to generate verified highway traffic volume axle load data;

[0040] Step S33: Perform time-segmented distribution statistics on the verified highway traffic volume axle load data according to the preset time period, so as to calculate the number of vehicles, total axle load and proportion corresponding to different axle load intervals in each time period, so as to obtain the highway time-segmented axle load distribution data.

[0041] Step S34: Obtain the structural parameters of the highway pavement, including the thickness, material type, compressive strength, flexural tensile strength and fatigue characteristics of each pavement structural layer;

[0042] Step S35: Based on the highway pavement structure parameters and the highway time-segmented axle load distribution data, conduct a pavement axle load cumulative damage assessment to obtain the highway pavement axle load cumulative damage value.

[0043] Furthermore, step S32 includes the following steps:

[0044] The standard axle load parameters for matching the corresponding vehicle type axle group are obtained from the vehicle type database, and the deviation is calculated by comparing the preliminary estimated value of highway traffic volume axle load with the standard axle load parameters to obtain the highway traffic volume axle load deviation rate.

[0045] The standard axle load test dataset corresponding to the axle load deviation rate of highway traffic volume is selected by comparing and judging between the axle load deviation rate of highway traffic volume and the preset axle load deviation range, so as to obtain the highway axle load data with deviation rate that does not meet the requirements.

[0046] Obtain the wheelbase and tire quantity constraints corresponding to the vehicle type, and perform constraint correction and adjustment verification on the highway axle load data with non-compliant deviation rates based on the wheelbase and tire quantity constraints corresponding to the vehicle type. Then, integrate the verified highway axle load data with the highway axle load data with compliant deviation rates to generate verified highway traffic volume axle load data.

[0047] Furthermore, step S35 includes the following steps:

[0048] Step S351: Based on the material type and fatigue characteristic parameters in the highway pavement structure parameters, perform axle load conversion matching analysis to determine the equivalent axle number conversion coefficients corresponding to different axle loads and generate a highway fatigue axle load equivalent conversion table.

[0049] Step S352: Based on the highway fatigue axle load equivalent conversion table and the highway time-sharing axle load distribution data, calculate the equivalent axle number corresponding to different axle load intervals in each time period to generate highway time-sharing equivalent axle number data.

[0050] Step S353: Construct a pavement axle load damage assessment model based on the thickness, compressive strength and flexural tensile strength of each pavement structural layer in the pavement structural parameters of the highway, and input the equivalent axle data of the highway in different time periods into the pavement axle load damage assessment model to calculate the cumulative damage corresponding to each pavement structural layer according to the time series, and calculate the cumulative damage summation based on the cumulative damage corresponding to each pavement structural layer to obtain the cumulative damage value of the highway pavement axle load.

[0051] Furthermore, step S4 includes the following steps:

[0052] Step S41: Obtain the number and proportion of highway vehicle axle loads and the cumulative axle load damage value of the highway pavement in different axle load intervals within each time period by using the highway time-segmented axle load distribution data and the highway pavement axle load cumulative damage value.

[0053] Step S42: Based on the number and proportion of axle loads of highway vehicles in different axle load intervals within each time period and the cumulative axle load damage value borne by the road surface, perform axle load-damage correlation analysis to analyze and identify the correlation between vehicle axle load distribution and cumulative damage within each time period, and generate highway vehicle axle load-cumulative damage correlation features.

[0054] Step S43: Obtain real-time axle load detection data of the highway, and perform dynamic high-precision detection and correction of real-time traffic volume axle load based on the highway vehicle axle load-cumulative damage correlation characteristics to generate a dynamic detection report of highway traffic volume axle load.

[0055] Step S44: Based on the cumulative axle load damage value of the road surface corresponding to the dynamic detection report of axle load of highway traffic volume, compare and evaluate the road surface damage with the preset road surface design service life threshold to obtain the current damage level of the highway road surface.

[0056] Step S45: Based on the current damage level of the highway pavement and combined with the dynamic detection report of highway traffic volume and axle load, formulate corresponding vehicle axle load limits and maintenance timing plans, and generate a highway pavement axle load maintenance plan that includes maintenance process, maintenance scope and implementation priority.

[0057] The beneficial effects of this invention are:

[0058] The proposed method for dynamic high-precision detection of axle load in highway traffic volume, compared with existing technologies, offers several advantages. Firstly, by deploying multi-type sensor arrays to collect dynamic response signals of highway pavement and vehicle traffic characteristic data, it can acquire raw axle load detection data of highway traffic volume in real time and comprehensively. Different types of sensors can capture different traffic information, such as vehicle axle load, speed, and location, ensuring data diversity and integrity. This provides ample raw data support for subsequent data processing. Spatiotemporal synchronization processing effectively unifies and coordinates sensor data from different locations and times, eliminating spatiotemporal errors and ensuring data timeliness and accuracy. Secondly, noise suppression processing removes noise caused by environmental, equipment, or other interference factors, ensuring a high signal-to-noise ratio in the final dataset and improving the reliability of subsequent analysis. This, in turn, promotes accurate assessment of traffic flow and pavement damage. Secondly, based on the standardized axle load detection dataset, vehicle axle type identification and axle position localization can provide more detailed and specific information for traffic volume estimation. By accurately identifying the axle type and position of each vehicle, the axle group distribution characteristics and axle spacing parameters can be extracted, providing a deeper basis for understanding the loads exerted by vehicles on the road surface. This data not only reflects the characteristics of traffic flow but also reflects vehicle weight and distribution, thus more accurately assessing the road surface load-bearing capacity at different time periods. Simultaneously, by combining the attenuation characteristics of the road surface response signal for load-response mapping estimation, a preliminary estimate of the axle load for highway traffic volume can be calculated, providing a reliable load basis for further analysis. Then, by acquiring and utilizing a vehicle type database, and combining the preliminary estimate of highway traffic volume axle load with the vehicle type database for axle load verification, not only can the previous estimation results be verified, but the traffic flow distribution data for each time period can be further refined. This process effectively improves the accuracy of vehicle axle load data, and the statistical analysis of different time periods can reveal the fluctuation trend of traffic flow, helping decision-makers understand the changes in road usage intensity at different times, thus providing a data basis for the formulation of road maintenance plans. Furthermore, by combining the pavement structure parameters of highways to conduct pavement axle load cumulative damage assessment, the cumulative damage of the road under different traffic intensities can be quantified. Environmental impact factors are also introduced into the assessment, which can provide more accurate and stable detection data under variable environmental conditions. This assessment result can reflect the degree of pavement damage in long-term use, help predict the need for pavement maintenance in advance, and provide a reference for the formulation of scientific maintenance strategies.Finally, by conducting dynamic high-precision detection based on the time-segmented axle load distribution data and the cumulative axle load damage value of the highway pavement, a more accurate dynamic traffic volume and axle load detection report can be obtained. This report not only provides real-time dynamic traffic volume data but also reflects the damage changes of the pavement under the actual traffic flow, thus providing comprehensive data support for the formulation of maintenance plans. Dynamic detection can help relevant departments monitor traffic volume and pavement conditions in real time, promptly identify potential damage risks, and formulate targeted emergency response plans. This enables more scientific planning of road maintenance work, thereby avoiding the decline in road function caused by sudden damage or uneven traffic load, and also providing stronger protection for road safety. Attached Figure Description

[0059] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0060] Figure 1 This is a schematic diagram of the steps in the method for dynamic high-precision detection of axle load in highway traffic volume according to the present invention.

[0061] Figure 2 for Figure 1 A detailed flowchart of step S1;

[0062] Figure 3 for Figure 1 A detailed flowchart of step S2. Detailed Implementation

[0063] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0064] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0065] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0066] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides a method for dynamic high-precision detection of axle load in highway traffic volume, the method comprising the following steps:

[0067] Step S1: Collect dynamic response signals and vehicle traffic characteristic data corresponding to the highway road surface through a multi-type sensor array to obtain the original axle load detection data of highway traffic volume; perform spatiotemporal synchronization and noise suppression processing on the original axle load detection data of highway traffic volume to generate a standardized axle load detection dataset.

[0068] Step S2: Based on the standardized axle load detection dataset, identify vehicle axle type and locate axle position to extract axle group distribution characteristics and axle spacing parameters, and generate highway vehicle axle system feature data; based on the highway vehicle axle system feature data and combined with the attenuation characteristics corresponding to the road surface response signal, perform load-response mapping estimation to obtain a preliminary estimate of highway traffic volume axle load.

[0069] Step S3: Obtain the vehicle type database, and perform vehicle axle load verification and time-segmented distribution statistics on the standardized axle load detection dataset based on the preliminary estimate of highway traffic volume axle load and the vehicle type database to obtain highway time-segmented axle load distribution data; obtain highway pavement structure parameters, and perform pavement axle load cumulative damage assessment based on highway pavement structure parameters combined with highway time-segmented axle load distribution data to obtain highway pavement axle load cumulative damage value.

[0070] Step S4: Perform dynamic high-precision detection based on the time-segmented axle load distribution data of the expressway and the cumulative axle load damage value of the expressway pavement to generate a dynamic detection report of expressway traffic volume axle load; formulate a corresponding expressway pavement axle load maintenance plan based on the dynamic detection report of expressway traffic volume axle load.

[0071] In the embodiments of this invention, please refer to Figure 1 The diagram shown is a flowchart illustrating the steps of the dynamic high-precision detection method for axle load of highway traffic volume according to the present invention. In this example, the dynamic high-precision detection method for axle load of highway traffic volume includes the following steps:

[0072] Step S1: Collect dynamic response signals and vehicle traffic characteristic data corresponding to the highway road surface through a multi-type sensor array to obtain the original axle load detection data of highway traffic volume; perform spatiotemporal synchronization and noise suppression processing on the original axle load detection data of highway traffic volume to generate a standardized axle load detection dataset.

[0073] In this embodiment of the invention, a multi-type sensor array is deployed along the K10+000 to K10+500 section of a two-way four-lane highway, with a set of detection points set every 10m along the longitudinal direction for each lane. Each set includes a fiber optic strain gauge (range -2000 to +2000με, accuracy ±2με, sampling frequency 1kHz) buried 5cm below the road surface, a piezoelectric accelerometer (range ±50g, sensitivity 100mV / g, sampling frequency 2kHz) 10cm below the surface, a laser profile sensor (scanning frequency 100Hz, measurement range 0.5-10m, accuracy ±2mm) installed 1.5m from the roadside, and a high-definition camera (resolution 1920×1080, frame rate 25fps) fixed on a 6m high pole. When a six-axle truck passed at 60 km / h, strain gauges collected a strain signal of 1200 με, accelerometers recorded 8g of vibration data, laser profile sensors measured a vehicle length of 12m and a wheelbase of 1.8m, and cameras identified it as a heavy truck and recorded its trajectory. The timestamps of each sensor were unified to ±1ms using a GPS timing module (synchronization accuracy 10ns). Spatial mapping was established according to the lane coordinate system (X-axis along the driving direction, Y-axis laterally, origin at the center of lane K10+000) to ensure that the strain, vibration, and geometric parameters of the same vehicle were associated with the same coordinate points. Noise reduction was performed using a 5-level decomposition with a db4 wavelet basis, removing strain peaks >2000 με and vibration anomalies >50g. The data was standardized to the [-1,1] interval, generating a standardized axle load detection dataset containing timestamps, coordinates, strain, vibration, vehicle length, wheelbase, and vehicle type.

[0074] Step S2: Based on the standardized axle load detection dataset, identify vehicle axle type and locate axle position to extract axle group distribution characteristics and axle spacing parameters, and generate highway vehicle axle system feature data; based on the highway vehicle axle system feature data and combined with the attenuation characteristics corresponding to the road surface response signal, perform load-response mapping estimation to obtain a preliminary estimate of highway traffic volume axle load.

[0075] In this embodiment of the invention, axle type identification is performed on the standardized axle load detection dataset: the peak sequence of strain signals of a six-axle truck is extracted. The first peak appears at 1620000000.123s (1200με), the second at 1620000000.303s (1100με), with an interval of 0.18s and an amplitude decrease of 8.3%, indicating a parallel dual-axle configuration; the third and fourth peaks have an interval of 0.12s (910με, 868με), with a decrease of 4.6%, also indicating a parallel dual-axle configuration; the fifth and sixth peaks have an interval of 0.24s (812με, 742με), with a decrease of 8.6%, also indicating a parallel dual-axle configuration, thus determining the axle type group to be 6×4. Based on the vehicle speed of 16.67 m / s, the axle coordinates are calculated as follows: Axle 1 X = 10.00 m, Axle 2 X = 13.00 m, Axle 3 X = 17.00 m, Axle 4 X = 19.00 m, Axle 5 X = 22.00 m, Axle 6 X = 26.00 m, with a Y-axis of 2.8 m. Axle spacings are 3.00 m, 4.00 m, 2.00 m, 3.00 m, and 4.00 m, generating axle system characteristic data. The pavement response attenuation characteristics are analyzed: A 100-120 kN axle load results in a strain of 1200 με at X = 10 m, which attenuates to 720 με (attenuation rate 40%) at X = 15 m. A mapping function is established: Axle load (kN) = strain (με) × 0.05 ÷ (1 - attenuation rate / 100). Substituting the data for each axis: first axis 1200με×0.05÷1=60kN, second axis 1100με×0.05÷0.85≈64.71kN, third axis 910με×0.05÷0.7≈65kN, fourth axis 868με×0.05÷0.65≈66.77kN, fifth axis 812με×0.05÷0.6≈67.67kN, sixth axis 742με×0.05÷0.55≈67.45kN, we obtain the preliminary estimated values ​​for the axle loads.

[0076] Step S3: Obtain the vehicle type database, and perform vehicle axle load verification and time-segmented distribution statistics on the standardized axle load detection dataset based on the preliminary estimate of highway traffic volume axle load and the vehicle type database to obtain highway time-segmented axle load distribution data; obtain highway pavement structure parameters, and perform pavement axle load cumulative damage assessment based on highway pavement structure parameters combined with highway time-segmented axle load distribution data to obtain highway pavement axle load cumulative damage value.

[0077] In this embodiment of the invention, the standard axle load range for a 6×4 six-axle truck in the vehicle type database is: 65-75kN for the first and second axles, 95-105kN for the third and fourth axles, and 110-120kN for the fifth and sixth axles. Comparing the preliminary estimated values ​​with the standard, the deviation for the third to sixth axles exceeds 20%, so constraint corrections are made: the third and fourth axles are corrected to 98kN and 102kN (dual-axle balance, total 200kN), the fifth and sixth axles are corrected to 113kN and 117kN (total 230kN), and the first and second axles are retained at 60kN and 64.71kN (deviation < 5kN), generating the verified data. Statistics were compiled based on the following time periods: morning peak (7:00-9:00), off-peak (9:00-17:00), and evening peak (17:00-19:00): During the morning peak, 100 trucks carried 200 axles with a load of 50-70 kN (total 12471 kN, accounting for 22%), and 400 axles with a load of 100-120 kN (total 43000 kN, accounting for 78%). During the off-peak, 80 trucks carried 160 axles (9976.8 kN) and 320 axles (34400 kN). During the evening peak, 120 trucks carried 240 axles (14965.2 kN) and 480 axles (51600 kN). The axle load distribution data for each time period was obtained. Road surface structural parameters were obtained: 18cm asphalt surface layer (compressive strength 4.5 MPa, flexural strength 1.0 MPa, fatigue life 5.0 × 10⁻⁶). 6 (Time), cement-stabilized crushed stone base course 36cm (compressive strength 3.0MPa, flexural strength 0.8MPa, fatigue life 3.0×10⁻⁶). 6 Damage was calculated using Miner's theory: 354.65 equivalent axial cycles during the early peak, and 7.09 × 10⁻⁶ surface layer damage. -5 1.18×10 at the grassroots level -4 Peak hours: 210.32; Surface layer: 4.21 × 10⁻⁶ -5 7.01×10 at the grassroots level -5 Evening peak: Train 380.17, surface level: 7.60×10 -5 1.27×10 at the grassroots level -4 The total cumulative damage value is 5.04 × 10⁻⁶. -4 .

[0078] Step S4: Perform dynamic high-precision detection based on the time-segmented axle load distribution data of the expressway and the cumulative axle load damage value of the expressway pavement to generate a dynamic detection report of expressway traffic volume axle load; formulate a corresponding expressway pavement axle load maintenance plan based on the dynamic detection report of expressway traffic volume axle load.

[0079] In this embodiment of the invention, the following data was extracted from time-segmented data and cumulative damage values: 200 shafts with 50-70kN load during the morning peak (accounting for 22%, with a damage of 5.04 × 10⁻⁶ kN). -5), 400 shafts with a capacity of 100-120kN (accounting for 78%, with 2.565×10⁻⁶ damage). -4 ); 160 (3.85×10) corresponding to off-peak periods. -5 ), 320 (1.978×10) -4 ); 240 during the evening rush hour (5.78 × 10 -5 ), 480 (3.08×10) -4 Analysis revealed that shafts operating in the 100-120kN range contributed 83%-85% of the damage, with the damage during the evening peak being 1.2 times that of the morning peak, establishing a load-damage correlation characteristic. Real-time detection revealed 500 shafts operating in the 100-120kN range (a 25% increase) and 200 shafts operating in the 50-70kN range. The damage value, corrected for an evening peak factor of 1.2, was 4.351 × 10⁻⁶. -4 A dynamic report is generated, containing 700 axes, representing 71.4% / 28.6% of the total, with a deviation rate of +25%. The preset lifespan threshold is 1.0 (15 years), and the current damage level is 5.04 × 10⁻⁶. -4 ×365×3≈0.552 (base layer accounts for 62.5%), remaining lifespan≈2.4 years. Maintenance plan: During evening rush hour, restrict 30% of vehicles with axle loads of 100-120kN, and limit speed to 80km / h; the base layer should be infiltrated with emulsified asphalt within one month (dosage 3.5kg / m²). 2 (Spray at 160℃), range K10+000-K15+000; micro-surfacing of the surface layer within 3 months (3-5mm stone, oil-stone ratio 6.0%), full coverage, ensuring that the damage rate of the base layer is reduced by 40% and the service life of the surface layer is extended by 2 years.

[0080] Furthermore, as an embodiment of the present invention, reference is made to... Figure 2 As shown, Figure 1 A detailed flowchart of step S1 is shown below. In this embodiment, step S1 includes the following steps:

[0081] Step S11: By deploying strain sensors and acceleration sensors at different depths on the highway surface, and deploying a multi-type sensor array consisting of laser profile sensors and video recognition equipment on the side of the highway, the strain sensors collect the dynamic response signal corresponding to the road surface strain when a vehicle passes, the acceleration sensors collect the dynamic response signal corresponding to the road surface vibration, the laser profile sensors collect the vehicle length and wheelbase geometric parameters, and the video recognition equipment collects the vehicle type and travel trajectory, the original axle load detection data of highway traffic volume is obtained.

[0082] In this embodiment of the invention, multiple types of sensor arrays are deployed within the road surface structure layer of the highway. The strain sensors are embedded fiber optic strain gauges (range -2000 to +2000 με, accuracy ±2 με), buried at depths of 5 cm, 10 cm, and 15 cm below the road surface, arranged in groups of three every 10 m along the longitudinal direction of the lane. Each group consists of three sensors arranged in an equilateral triangle (50 cm apart) to collect the dynamic strain response signal generated when a vehicle passes (sampling frequency 1 kHz). The accelerometers are piezoelectric accelerometers (range ±50 g, sensitivity 100 mV / g), deployed at the same depth as the strain sensors to collect road vibration acceleration signals (sampling frequency 2 kHz). Laser profile sensors (scanning frequency 100 Hz, measurement range 0.5-10 m, accuracy ±2 mm) are deployed at a height of 1.5 m above the road surface along the roadside, installed at 3 m intervals along the transverse direction of the lane, to collect geometric parameters such as vehicle length (error ±5 cm) and wheelbase (error ±3 cm). The video recognition equipment consists of a high-definition camera (1920×1080 resolution, 25fps frame rate), mounted on a roadside pole (6m high), with the lens facing the direction of oncoming traffic. It collects vehicle type (distinguishing between cars, trucks, and buses) and travel trajectory (positioning accuracy ±10cm). When a six-axle truck passes through the detection area at 60km / h, the strain sensor outputs a 1200με strain signal, the accelerometer outputs an 8g vibration signal, and the laser profile sensor measures the vehicle length as 12m and the wheelbase as 1.8m. The video recognition equipment identifies it as a heavy truck with a trajectory deviation ≤5cm. All data is aggregated to form the original axle load detection data for highway traffic volume.

[0083] Step S12: Based on the time synchronization module corresponding to the multi-type sensor array, the acquisition timestamps of each sensor are unified to the same clock reference, and the error is controlled within the preset range to generate multi-source data time synchronization parameters.

[0084] In this embodiment of the invention, a time synchronization module is configured using a multi-type sensor array, employing GPS time synchronization (synchronization accuracy 10ns) combined with a local temperature-controlled crystal oscillator (frequency stability 1×10⁻⁶). -9A clock reference is established. The timestamps from the strain sensor and accelerometer are connected to the synchronization module via an RS485 bus, while the laser profile sensor and video recognition device achieve time synchronization via Ethernet (IEEE 1588PTP protocol). The synchronization module calibrates the clocks of each sensor every 100ms, controlling the timestamp error within ±1ms. For example, the timestamp of the strain signal acquired by the strain sensor for a truck is 1620000000.123s, the timestamp of the corresponding vibration signal from the accelerometer is 1620000000.124s, the timestamp of the truck detected by the laser profile sensor is 1620000000.123s, and the timestamp of the truck captured by the video recognition device is 1620000000.125s. After calibration by the synchronization module, all timestamps are unified to 1620000000.123s ± 0.5ms, generating multi-source data time synchronization parameters to ensure consistent timestamps for data from different sensors triggered by the same vehicle.

[0085] Step S13: Establish the coordinate mapping relationship of each sensor detection point according to the spatial layout of each sensor in the multi-type sensor array, and associate the dynamic response signal in the original axle load detection data of highway traffic volume with the vehicle traffic characteristic data to the same coordinate mapping relationship to generate multi-source data spatial calibration parameters.

[0086] In this embodiment of the invention, a coordinate mapping relationship is established based on the spatial deployment positions of the sensors. The origin (0,0,0) is the lane start point, the longitudinal direction of the lane is the X-axis (accuracy ±1cm), the transverse direction is the Y-axis (accuracy ±1cm), and the vertical direction is the Z-axis (depth, accuracy ±0.5cm). Strain sensors are located at (10m, 2.5m, 5cm), (10m, 3.0m, 10cm), and (10m, 3.5m, 15cm), accelerometers are located at (10m, 3.0m, 5cm), (10m, 3.0m, 10cm), and (10m, 3.0m, 15cm), laser contour sensors are located at (10m, 0m, 1.5m), and video recognition equipment is located at (10m, 5m, 6m). When the front wheel of a vehicle passes through the point X=10m, the strain sensor generates a strain signal at X=10m, the laser profile sensor detects the front wheel track at X=10m, and the video recognition device records the front wheel trajectory at X=10m. Through coordinate mapping, these data are associated with the same lateral section (Y-axis range 0-5m) at X=10m, generating multi-source data spatial calibration parameters. This clarifies the spatial correspondence between the dynamic response signal (at X=10m, Y=3.0m) and the vehicle's traffic characteristics (front wheel at X=10m, Y=2.8m), with the error controlled within ±5cm.

[0087] Step S14: Based on the time synchronization parameters and spatial calibration parameters of multi-source data, the original axle load detection data of highway traffic volume is spatiotemporally synchronized and integrated to generate a spatiotemporally aligned detection dataset of highway axle load containing time and space.

[0088] In this embodiment of the invention, the original axle load detection data is spatiotemporally integrated based on multi-source data time synchronization parameters (timestamps unified to ±0.5ms) and spatial calibration parameters (coordinate mapping error ±5cm). A truck passes through a location at X=10m at 1620000000.123s. At this time, strain sensors (10m, 3.0m, 10cm) collect a strain of 1200με, acceleration sensors (10m, 3.0m, 10cm) collect a vibration of 8g, a laser profile sensor measures the wheelbase at this location as 1.8m, and video recognition equipment records the passage as the front wheels of a six-axle truck. During integration, these data are marked as associated data from the same time (1620000000.123s) and the same spatial location (X=10m), and arranged sequentially according to the vehicle's direction of travel (incrementing along the X-axis). For the continuous positions of the vehicle (X=10m, X=10.5m, X=11m...), the sensor data at the corresponding time points are integrated to form a spatiotemporal aligned detection dataset of highway axle load that includes timestamps, X / Y / Z coordinates, strain values, acceleration values, vehicle length, wheelbase, and vehicle type, ensuring that the full-journey data of the same vehicle is continuously correlated in the spatiotemporal dimension.

[0089] Step S15: Perform noise suppression and standardization on the highway axle load spatiotemporal alignment detection dataset. Wavelet threshold denoising is used to remove high-frequency interference noise and standardize the data to generate a standardized axle load detection dataset.

[0090] In this embodiment of the invention, noise suppression and standardization are performed on the spatiotemporal alignment detection dataset of highway axle loads. Wavelet threshold denoising is used: a db4 wavelet basis is selected, and the data is decomposed into 5 layers. Hard thresholding is applied to the high-frequency coefficients of each layer (the threshold is 1.5 times the root mean square error) to remove high-frequency interference (such as instantaneous spikes caused by road particle vibration) >2000με in the strain signal and outliers (such as sensor noise) >50g in the acceleration signal. During standardization, the strain data is converted to the [-1,1] interval (mean 800με, standard deviation 300με) according to (measured value - mean) / standard deviation, and the acceleration data is converted to the [-1,1] interval (mean 5g, standard deviation 2g). Geometric parameters such as vehicle length and wheelbase are standardized according to the ratio of actual size to the maximum possible value (vehicle length 20m, wheelbase 2.5m). After processing, the strain data of a truck was converted from 1200με to 1.33, the acceleration data from 8g to 1.5, the vehicle length from 12m to 0.6, and the wheelbase from 1.8m to 0.72. All data were retained to three decimal places to generate a standardized axle load test dataset, eliminating the influence of dimensions to facilitate subsequent analysis.

[0091] Furthermore, as an embodiment of the present invention, reference is made to... Figure 3 As shown, Figure 1 A detailed flowchart of step S2 is shown below. In this embodiment, step S2 includes the following steps:

[0092] Step S21: Perform peak value statistics on the corresponding dynamic response signals in the standardized axle load detection dataset to statistically analyze the peak occurrence time, peak amplitude, and peak signal spectrum of each peak value of the highway axle load dynamic signal, and generate a highway axle load signal peak value sequence.

[0093] In this embodiment of the invention, the peak value of the axle load signal is statistically analyzed by analyzing the dynamic response signal (strain signal range -2000 to +2000 με, sampling frequency 1 kHz) within the standardized axle load detection dataset. The strain signal sequence of a six-axle truck passing by was selected, and six valid peaks were identified by a peak detection algorithm (threshold set to 500με). The first peak appeared at 1620000000.123s with an amplitude of 1200με, and the main peak frequency of the spectrum was 10Hz after Fourier transform. The second peak appeared at 1620000000.303s with an amplitude of 1100με and a main peak frequency of 12Hz. The third to sixth peaks appeared at 1620000000.543s, 1620000000.663s, 1620000000.843s, and 1620000001.083s, respectively, with amplitudes of 910με, 868με, 812με, and 742με, and main peak frequencies of 9Hz, 11Hz, 10Hz, and 12Hz. Arrange these data in chronological order to generate a peak sequence of highway axle load signals, including the precise timestamp (down to milliseconds), amplitude (down to 1 με), and spectral parameters of each peak, ensuring that the peak characteristics are completely recorded.

[0094] Step S22: Perform signal peak feature analysis based on the peak sequence of highway axle load signal to obtain the peak time interval, amplitude change rate and spectral energy distribution corresponding to the dynamic signal of highway axle load;

[0095] In this embodiment of the invention, signal peak characteristic analysis is performed based on the peak sequence of highway axle load signals. Peak time intervals are calculated as follows: the interval between the first and second peaks is 0.18s, between the second and third is 0.24s, between the third and fourth is 0.12s, between the fourth and fifth is 0.18s, and between the fifth and sixth is 0.24s. Amplitude change rate is calculated as follows: the decrease from the first to the second peak is 8.3% ((1200-1100) / 1200×100%), the decrease from the second to the third is 17.3%, the decrease from the third to the fourth is 4.6%, the decrease from the fourth to the fifth is 6.4%, and the decrease from the fifth to the sixth is 8.6%. Spectral energy distribution is analyzed as follows: the energy proportion of the 10Hz frequency component is 35%, 11Hz is 25%, 12Hz is 30%, and other frequencies account for 10%, with energy mainly concentrated in the 9-12Hz frequency band. By plotting time interval curves, amplitude change rate curves, and spectral energy histograms, the peak characteristic parameters corresponding to the highway axle load dynamic signal are obtained, clarifying the temporal and energy correlation between each peak.

[0096] Step S23: Based on the peak time interval, amplitude change rate, and spectral energy distribution of the dynamic signal of highway axle load, and combined with the corresponding vehicle type and traffic trajectory in the standardized axle load detection dataset, perform vehicle axle type identification analysis to identify and distinguish vehicle axle type groups corresponding to single axle, parallel dual axle, and parallel triple axle, and obtain the highway traffic volume axle type group classification results.

[0097] In this embodiment of the invention, vehicle axle type identification is performed based on peak time interval, amplitude change rate, and spectral energy distribution, combined with vehicle type (six-axle truck) and traffic trajectory (straight driving, offset < 5cm) within the standardized axle load detection dataset. The first and second peaks have an interval of 0.18s (corresponding to an axle spacing of 3.00m), similar amplitudes (decreasing by 8.3%), and consistent spectral characteristics (10Hz, 12Hz), thus identifying them as parallel dual-axle configurations. The third and fourth peaks have an interval of 0.12s (axle spacing of 2.00m), an amplitude decrease of 4.6%, and both contain an 11Hz component, thus identifying them as parallel dual-axle configurations. The fifth and sixth peaks have an interval of 0.24s (axle spacing of 4.00m), an amplitude decrease of 8.6%, and matching spectral characteristics, thus identifying them as parallel dual-axle configurations. Combining the vehicle type, the axle type group is confirmed as "3 groups of parallel dual-axle configurations," i.e., 6×4 axle type groups. Highway traffic volume axle type group classification results are generated, with each axle type group labeled with its corresponding peak group number and feature parameters to ensure consistency between axle type identification and actual structure.

[0098] Step S24: Based on the classification results of highway traffic volume axle type groups and combined with the peak time interval and amplitude change rate corresponding to the highway axle load dynamic signal, perform vehicle axle position positioning statistics on the standardized axle load detection dataset to generate highway vehicle axle system feature data.

[0099] In this embodiment of the invention, vehicle axle position positioning statistics are performed on the standardized axle load detection dataset based on the axle type group classification results (6×4 axle type groups), peak time intervals, and amplitude change rates. Given a vehicle speed of 60.0 km / h (16.67 m / s), when the first peak occurs, the front wheels of the vehicle are located at X = 10.00 m. The coordinates of each axle are calculated according to the time intervals: Second axle X = 10.00 + 16.67 × 0.18 = 13.00 m, Third axle X = 13.00 + 16.67 × 0.24 = 17.00 m, Fourth axle X = 17.00 + 16.67 × 0.12 = 19.00 m, Fifth axle X = 19.00 + 16.67 × 0.18 = 22.00 m, Sixth axle X = 22.00 + 16.67 × 0.24 = 26.00 m, with a Y-coordinate of 2.8 m (track center). The system counts 6 axles with axle spacings of 3.00m, 4.00m, 2.00m, 3.00m, and 4.00m, generating characteristic data of the highway vehicle axle system. This data includes axle position coordinates (accurate to 0.01m), axle spacing (accurate to 0.01m), and axle type groupings, fully reflecting the spatial distribution of the vehicle axle system.

[0100] Step S25: Based on the axle system characteristic data of highway vehicles and combined with the attenuation characteristics of the road surface response signal, perform load-response mapping estimation to obtain a preliminary estimate of the axle load of highway traffic volume.

[0101] In this embodiment of the invention, load-response mapping estimation is performed based on highway vehicle axle system characteristic data (axle position, axle spacing) and road surface response signal attenuation characteristics (15% attenuation every 5m). The first axle is located at X = 10.00m (attenuation rate 0%), with a strain of 1200με, calculated as 60kN according to the mapping relationship (axle load = strain × 0.05 ÷ (1 - attenuation rate / 100)); the second axle is located at X = 13.00m (attenuation rate 15%), with a strain of 1100με, calculated as 1100 × 0.05 ÷ 0.85 ≈ 64.71kN; the third axle is located at X = 17.00m (attenuation rate 30%), with a strain of 910με, calculated as 910 × 0.05 ÷ 0.7 ≈ 65.00kN; the fourth axle is located at X = 19.00m... For the fifth axle (X = 22.00m, attenuation rate 35%), the strain is 868με, calculated as 868×0.05÷0.65≈66.77kN; for the sixth axle (X = 26.00m, attenuation rate 40%), the strain is 812με, calculated as 812×0.05÷0.6≈67.67kN; for the sixth axle (X = 26.00m, attenuation rate 45%), the strain is 742με, calculated as 742×0.05÷0.55≈67.45kN. The preliminary estimates of axle loads for highway traffic volume are retained to two decimal places to provide basic data for subsequent corrections.

[0102] Furthermore, step S24 includes the following steps:

[0103] Step S241: Obtain the corresponding highway vehicle speed by matching the vehicle type and travel trajectory within the standardized axle load detection dataset;

[0104] In this embodiment of the invention, the vehicle speed on the highway is obtained by combining the vehicle type (six-axle truck) and traffic trajectory data within a standardized axle load detection dataset. The traffic trajectory data includes continuous position records of the vehicle in the X-axis direction, with a timestamp interval of 0.1 seconds. The time taken for a six-axle truck to pass through X = 10m is 1620000000.123s, and the time taken to pass through X = 20m is 1620000000.723s. ​​The distance between the two points is 10m, and the time difference is 0.6 seconds. The calculated speed = 10m ÷ 0.6s = 16.67m / s, which is equivalent to 60km / h (1m / s = 3.6km / h). The average of the speed calculation results from five consecutive locations (59.8 km / h, 60.2 km / h, 60.0 km / h, 59.9 km / h, and 60.1 km / h) was taken, and the final vehicle speed was determined to be 60.0 km / h, with the error controlled within ±0.5 km / h, to ensure that the speed data accurately reflects the actual driving status of the vehicle.

[0105] Step S242: Based on the peak time interval and amplitude change rate corresponding to the dynamic signal of highway axle load, and combined with the speed of highway vehicles, the vehicle axle spacing is converted to obtain the initial parameters of highway vehicle axle spacing;

[0106] In this embodiment of the invention, the axle spacing is calculated by combining the peak time interval and amplitude change rate of the dynamic signal of highway axle load with the vehicle speed of 60.0 km / h (16.67 m / s). In the dynamic signal of the six-axle truck collected by the strain sensor, the peak time interval between two adjacent axles is 0.3 seconds, and the amplitude change rate decreases from 1200 με to 1000 με (a decrease of 16.7%). The axle spacing is calculated as: speed × time interval = 16.67 m / s × 0.3 s = 5.00 m. The peak time intervals (0.3 s, 0.4 s, 0.5 s, 0.3 s, 0.4 s) of all adjacent axle pairs are calculated sequentially to obtain the initial axle spacing parameters of 5.00 m, 6.67 m, 8.33 m, 5.00 m, and 6.67 m. Each value is rounded to two decimal places to ensure that the axle spacing conversion result is consistent with the actual axle distribution trend of the vehicle.

[0107] Step S243: Based on the vehicle length and wheelbase geometric parameters corresponding to the standardized axle load detection dataset, perform axle spacing comparison and correction calculation on the initial parameters of highway vehicle axle spacing to obtain highway axle spacing comparison and correction parameters;

[0108] In this embodiment of the invention, the initial axle spacing parameters are compared and corrected based on the vehicle length of 12m (normalized value 0.6) and wheelbase of 1.8m (normalized value 0.72) in the standardized axle load detection dataset. The relationship between the theoretical vehicle length and the sum of the axle spacings of a six-axle truck is: total axle spacing = vehicle length - front overhang - rear overhang (front overhang 1.5m, rear overhang 1.5m, total axle spacing = 12 - 1.5 - 1.5 = 9m). The initial sum of axle spacings = 5.00 + 6.67 + 8.33 + 5.00 + 6.67 = 31.67m, which far exceeds the theoretical value, indicating a multi-axle recognition error. Combining the dual-axle parallel feature corresponding to a wheelbase of 1.8m, adjacent duplicate axle signals are merged during correction. The corrected axle spacings are 3.00m, 4.00m, and 2.00m (total 9.00m), consistent with the theoretical total axle spacing. This yields the axle spacing comparison and correction parameters, ensuring that each axle spacing conforms to the structural characteristics of a six-axle truck.

[0109] Step S244: Based on the classification results of axle type groups of highway traffic volume and combined with the highway vehicle speed, calculate the axle position of each axle group to locate the corresponding highway pavement coordinates and generate a dataset of axle position coordinates for each axle group; perform axle group distribution statistics based on the dataset of axle position coordinates for each axle group to analyze and extract the number and distribution of axles in each axle group, and obtain the distribution characteristics of each axle group of the highway.

[0110] In this embodiment of the invention, axle positioning is calculated based on the traffic volume axle group classification results (six-axle trucks belong to axle group 6×4) and the vehicle speed of 60.0 km / h. Starting from X = 10m, the passage time of the first axle is 1620000000.123s, and the passage time of the second axle is 1620000000.123s + 3.00m ÷ 16.67m / s = 1620000000.303s, corresponding to an X coordinate of 10m + 3.00m = 13.00m. Similarly, calculate the coordinates of the third axis (X=17.00m), the fourth axis (X=19.00m), the fifth axis (X=22.00m), and the sixth axis (X=26.00m), generating the following axis coordinate datasets: (10.00m, 2.8m), (13.00m, 2.8m), (17.00m, 2.8m), (19.00m, 2.8m), (22.00m, 2.8m), and (26.00m, 2.8m). The axis group distribution statistics show that this axis group contains 6 axes, arranged in a "2+2+2" pattern (first 2 axes, middle 2 axes, last 2 axes).

[0111] Step S245: Based on the highway axle spacing comparison correction parameters and the distribution characteristics of each axle group on the highway, integrate the vehicle axle system features to generate highway vehicle axle system feature data containing the number of axles, axle position and axle spacing of each axle group.

[0112] In this embodiment of the invention, the axis system features are integrated based on the axis spacing comparison correction parameters (3.00m, 4.00m, 2.00m) and the axis group distribution characteristics ("2+2+2" arrangement). The first axis group (2 axes) has an axis spacing of 3.00m and axis position coordinates of (10.00m, 2.8m) and (13.00m, 2.8m); the second axis group (2 axes) has an axis spacing of 4.00m and axis position coordinates of (13.00m, 2.8m) and (17.00m, 2.8m); the third axis group (2 axes) has an axis spacing of 2.00m and axis position coordinates of (19.00m, 2.8m), (22.00m, 2.8m), (22.00m, 2.8m), and (26.00m, 2.8m). After integration, the axle system feature data of highway vehicles is generated, including the number of axles in each axle group (all are 2 axles), axle position coordinates (accurate to 0.01m), and axle spacing (accurate to 0.01m), which fully reflects the axle system structure of a six-axle truck and provides accurate vehicle structural parameters for axle load dynamic detection.

[0113] Furthermore, step S25 includes the following steps:

[0114] Step S251: Perform signal attenuation characteristic analysis on the pavement response signals corresponding to pavement strain and pavement vibration in the standardized axle load test dataset to calculate the amplitude attenuation rate and phase shift of the pavement response signal at different propagation distances, and generate highway pavement response attenuation characteristic parameters.

[0115] In this embodiment of the invention, by performing attenuation characteristic analysis on the road surface response signal within the standardized axle load detection dataset, the road surface strain signal (1200 με) and vibration signal (8g) when a six-axle truck passes are selected as the original signals. Sensors are deployed at propagation distances of 0m (at the signal source), 5m, 10m, 15m, and 20m to collect signals, and the amplitude attenuation rate and phase shift are calculated: at a distance of 5m, the strain amplitude decreases to 1020 με (attenuation rate 15%), and the phase shift is 0.02π; at 10m, the strain amplitude is 840 με (attenuation rate 30%), and the phase shift is 0.05π; at 15m, the strain amplitude is 720 με (attenuation rate 40%), and the phase shift is 0.08π; at 20m, the strain amplitude is 600 με (attenuation rate 50%), and the phase shift is 0.1π. The vibration signal amplitude is 6.8g (attenuation rate 15%) at 5m, 5.6g (attenuation rate 30%) at 10m, 4.8g (attenuation rate 40%) at 15m, and 4g (attenuation rate 50%) at 20m, with phase shift consistent with the strain signal. The generated highway pavement response attenuation characteristic parameters include attenuation rates (increasing by 15% every 5m) and phase shifts (increasing by 0.03π every 5m) corresponding to different distances, clarifying the signal propagation law.

[0116] Step S252: Based on the axle coordinate information in the highway vehicle axle system feature data, determine the road surface response signal acquisition points corresponding to each axle group, and perform axle position-attenuation correlation analysis based on the road surface response signal acquisition points corresponding to each axle group and the highway road surface response attenuation feature parameters to obtain highway vehicle axle position-response attenuation correlation data.

[0117] In this embodiment of the invention, the road surface response signal acquisition points corresponding to each axle group are determined based on the axle coordinates (10.00m, 2.8m), (13.00m, 2.8m), etc., in the vehicle axle system characteristic data. The first axle corresponds to the sensor at X=10m, the second axle corresponds to the sensor at X=13m, and so on. Combining the road surface response attenuation characteristic parameters, the strain signal of the first axle at X=10m is calculated to be 1200με, which attenuates to 720με (attenuation rate 40%) when it propagates to X=15m; the strain signal of the second axle at X=13m is 1100με, which attenuates to 660με (attenuation rate 40%) when it propagates to X=18m. By associating the axis position with the attenuation rate through coordinate mapping, we obtain the axis position-response attenuation correlation data: the first axis (10.00m) corresponds to an attenuation rate of 0% (at the acquisition point), the second axis (13.00m) corresponds to an attenuation rate of 15% (propagation 5m), and the third axis (17.00m) corresponds to an attenuation rate of 30% (propagation 10m), ensuring that the response signal attenuation value of each axis position is available.

[0118] Step S253: Based on the number of axles and axle spacing in the axle system characteristic data of highway vehicles and combined with the response attenuation characteristic parameters corresponding to the axle position-response attenuation correlation data of highway vehicles, perform load-response nonlinear mapping analysis to obtain the nonlinear mapping relationship between highway vehicle axle load and road surface response signal attenuation.

[0119] In this embodiment of the invention, load-response nonlinear mapping analysis is performed based on shaft system characteristic data (6 axes, shaft spacing 3.00m, 4.00m, 2.00m) and shaft position-response attenuation correlation data. The first axis has a load of 60kN, corresponding to a strain of 1200με at X=10m (attenuation rate 0%); the second axis has a load of 55kN, corresponding to a strain of 1100με at X=13m (attenuation rate 15%); the third axis has a load of 65kN, corresponding to a strain of 910με at X=17m (attenuation rate 30%); the fourth axis has a load of 62kN, corresponding to a strain of 868με at X=19m (attenuation rate 35%); the fifth axis has a load of 58kN, corresponding to a strain of 812με at X=22m (attenuation rate 40%); and the sixth axis has a load of 53kN, corresponding to a strain of 742με at X=26m (attenuation rate 45%). The mapping relationship was obtained through nonlinear fitting (using a cubic polynomial): Axle load (kN) = 0.05 × strain (με) + 0.0001 × strain (με) 2÷ Attenuation rate (%), the error of this relationship is <2% in the verification, which clarifies the nonlinear relationship between shaft load and signal attenuation.

[0120] Step S254: Based on the nonlinear mapping relationship between highway vehicle axle load and road surface response signal attenuation, and combined with highway vehicle axle system characteristic data and highway road surface response attenuation characteristic parameters, construct the axle load-response mapping function to generate the highway vehicle axle load-response mapping function;

[0121] In this embodiment of the invention, a shaft load-response mapping function is constructed based on a nonlinear mapping relationship. The input parameters are the strain signal amplitude and attenuation rate, and the output is the shaft load. The function expression is: Shaft load (kN) = (strain amplitude × 0.05) ÷ (1 - attenuation rate / 100) + (strain amplitude 2 × 0.00002) ÷ (1 - attenuation rate / 100) 2 The 0.05 is a proportionality coefficient determined based on the fundamental linear relationship between axle load and strain. By collecting 100 sets of sample data of known axle loads (30-150kN) and corresponding strain signals (600-3000με), the average ratio of axle load (kN) to strain (με) was calculated to be 0.05 (e.g., 60kN corresponds to 1200με, 60÷1200=0.05; 55kN corresponds to 1100με, 55÷1100=0.05). This value reflects the benchmark axle load value corresponding to a unit strain. Linear regression analysis (R²) was then performed... 2 =0.98) to verify its stability. 100 is the percentage conversion coefficient of the attenuation rate. Since the attenuation rate is expressed as a percentage (e.g., 15%) in the calculation, it needs to be converted to a decimal form (0.15) for calculation. Therefore, unit normalization is achieved by dividing by 100 (15% ÷ 100 = 0.15), ensuring that the correction effect of the attenuation rate on the axle load (e.g., when the attenuation is 15%, the correction coefficient is 1 ÷ (1-0.15) = 1.176) conforms to the signal propagation attenuation law. 0.00002 is the coefficient of the nonlinear correction term, which comes from the fitting of higher-order terms. The strain square term is added to the basic linear relationship to compensate for nonlinear errors. A cubic polynomial fitting is performed on 100 sets of sample data (axle load = k1 × strain + k2 × strain 2 ÷ attenuation rate). The least squares method is used to calculate k2 = 0.00002 (e.g., when strain = 1200με and attenuation rate = 0%, 0.00002 × 1200 2=28.8 (consistent with the compensation value of the measured axle load deviation). Residual analysis (average residual <1.5kN) confirmed that this coefficient can effectively reduce nonlinear error. The function expression formed by the three factors was verified by 50 new samples, and the axle load calculation error decreased from 5.2% in a simple linear relationship to 1.8%, meeting the accuracy requirements of dynamic detection. Verification for the first axle: strain 1200με, attenuation rate 0%, calculated 60kN; strain 1100με, attenuation rate 15%, calculated 55kN; strain 910με, attenuation rate 30%, calculated 65kN, all consistent with the actual axle load. An attenuation rate correction term was introduced into the function to ensure that the axle load can be accurately calculated from signals with different propagation distances. After verification with 100 sets of data, the average error was 1.8%, generating a formal highway vehicle axle load-response mapping function.

[0122] Step S255: Perform load-response mapping estimation based on the highway vehicle axle load-response mapping function to obtain a preliminary estimate of highway traffic volume axle load.

[0123] In this embodiment of the invention, by estimating the load-response mapping based on the axle load-response mapping function, the strain signals and attenuation rates corresponding to each axle when a six-axle truck passes are as follows: first axle 1200με (0%), second axle 1100με (15%), third axle 910με (30%), fourth axle 868με (35%), fifth axle 812με (40%), and sixth axle 742με (45%). Substituting into the function, the calculation is: first axle = 1200 × 0.05 ÷ 1 + (1200... 2 ×0.00002)÷1 2 =60+28.8=88.8kN; Second shaft =1100×0.05÷0.85+(1100 2 (×0.00002)÷0.85 2 ≈64.71 + 33.86 = 98.57 kN; Third shaft = 910 × 0.05 ÷ 0.7 + (910 2 (×0.00002)÷0.7 2 ≈65 + 33.86 = 98.86 kN; subsequent axle calculations yielded 95.24 kN, 92.11 kN, and 87.65 kN, respectively. The generated preliminary estimates of highway traffic volume axle loads are rounded to two decimal places to fully reflect the magnitude of each axle load.

[0124] Furthermore, step 255 includes the following steps:

[0125] By deploying environmental sensors on the corresponding highway surface to synchronously collect corresponding temperature, humidity, road surface dryness and wetness status and rainfall data, a highway surface environmental parameter dataset is generated.

[0126] In this embodiment of the invention, an environmental sensor group is deployed every 50m along the longitudinal direction of the highway lanes. Each group includes four types of sensors: a temperature sensor (range -40℃ to 80℃, accuracy ±0.5℃) buried 2cm below the road surface; a humidity sensor (range 0% to 100%RH, accuracy ±3%RH) installed on the roadside at a height of 1m above the road surface; a road surface wetness / dryness sensor (using infrared reflection principle, resolution 0.1%) attached to the road surface; and a rainfall sensor (range 0mm / h to 50mm / h, accuracy ±0.2mm / h) fixed to a roadside pole at a height of 3m. All sensors sample once per minute, and the timestamps are synchronized with the axle load detection system (error ±1s). Data collected during a certain period included: temperature 25℃, humidity 60%RH, road surface dryness (0.5% moisture content), and rainfall 0mm / h. After 24 hours of continuous data collection, a highway pavement environmental parameter dataset was generated, containing timestamps, temperature, humidity, pavement condition codes (dry = 1, wet = 2, waterlogged = 3), and rainfall. A set of average data was stored every hour to ensure that the environmental parameters corresponded to the axle load detection data in the time dimension.

[0127] Preferably, an environmental-axle load characteristic influence analysis is performed on the highway vehicle axle system characteristic data based on each environmental parameter in the highway pavement environmental parameter dataset to generate a highway environmental-axle load characteristic influence distribution curve;

[0128] In this embodiment of the invention, an environmental-axle load characteristic influence analysis is performed based on a highway pavement environmental parameter dataset and vehicle axle system characteristic data. A strain response of 1200 με corresponding to an axle spacing of 3.00 m and axle coordinates (10.00 m, 2.8 m) for a six-axle truck is selected as the baseline value. The response changes under different environmental conditions are compared: at a temperature of 35℃, the strain increases to 1260 με (an increase of 5%), and at a temperature of 15℃, it decreases to 1140 με (a decrease of 5%); when the pavement is wet (moisture content 10%), the strain decreases to 1080 με (a decrease of 10%); and when the rainfall is 5 mm / h, the strain decreases to 960 με (a decrease of 20%). Using environmental parameters as the horizontal axis (temperature 15-35℃, humidity 40%-80%RH) and the rate of change of strain as the vertical axis (-20% to +5%), an environmental-axle load characteristic influence distribution curve was plotted. The curve shows that the strain increases by 3% for every 10℃ increase in temperature, decreases by 2% for every 20% RH increase in humidity, and decreases by 20% when the road surface changes from dry to waterlogged. This clarifies the quantitative relationship between each environmental parameter and the dynamic response of the axle load.

[0129] Preferably, the highway axle load influence correction factor is calculated based on the highway environment-axle load characteristic influence distribution curve under different environmental condition combinations, and the load-response mapping correction estimation is performed based on the highway axle load influence correction factor and the highway vehicle axle load-response mapping function to obtain the preliminary estimate of highway traffic volume axle load.

[0130] In this embodiment of the invention, the axle load influence correction factor for different environmental conditions is calculated based on the environment-axle load characteristic influence distribution curve: Temperature correction factor is 1.00 at 25℃ (reference temperature), 1.05 (1+5%) at 35℃, and 0.95 (1-5%) at 15℃; Road surface dryness correction factor is 1.00, wetness correction factor is 0.90 (1-10%), and waterlogging correction factor is 0.80 (1-20%); Rainfall correction factor is 1.00 at 0 mm / h and 0.80 at 5 mm / h. The strain of a six-axle truck under a temperature of 35℃ and wet road surface conditions is 1080 με, and the comprehensive correction factor is 1.05 × 0.90 = 0.945. The axle load-response mapping function for highway vehicles is axle load (kN) = strain (με) × 0.05. Under the baseline condition, the axle load = 1200 × 0.05 = 60 kN. The corrected preliminary estimate of the axle load is 1080 × 0.05 ÷ 0.945 ≈ 57.14 kN, rounded to two decimal places. Calculations were performed on all axle groups using the same method, yielding preliminary estimates of 57.14 kN, 53.57 kN, 61.90 kN, 59.52 kN, 55.95 kN, and 52.38 kN, respectively, forming a complete axle load estimation result.

[0131] Furthermore, step S3 includes the following steps:

[0132] Step S31: Obtain the vehicle type database, which includes the standard axle load range, typical axle group configuration and axle weight distribution for different vehicle types;

[0133] In this embodiment of the invention, the vehicle type database contains axle load characteristic parameters for 10 common vehicle types, stored in a structure of "vehicle type - axle group - standard axle load range - axle group configuration - axle load distribution". The entry for a 6×4 axle truck is as follows: standard axle load range: first axle 65-75kN, second axle 65-75kN, third axle 95-105kN, fourth axle 95-105kN, fifth axle 110-120kN, sixth axle 110-120kN; typical axle group configuration is 3 groups of parallel dual axles (2+2+2); axle load distribution coefficients are: first axle 0.12, second axle 0.12, third axle 0.17, fourth axle 0.17, fifth axle 0.18, sixth axle 0.18 (total 1.0). The database uses a structured table format, with each parameter accurate to 0.1kN or 0.01 coefficient. It was established through statistical analysis of data from 500 similar vehicles measured in actual tests, ensuring that the standard values ​​cover 95% of the axle load range of actual vehicles, providing an authoritative reference for subsequent verification.

[0134] Step S32: Based on the preliminary estimate of highway traffic volume axle load and the vehicle type database, perform vehicle axle load verification on the standardized axle load detection dataset to generate verified highway traffic volume axle load data;

[0135] In this embodiment of the invention, the axle loads based on preliminary estimates of highway traffic volume (60kN, 64.71kN, 65.00kN, 66.77kN, 67.67kN, 67.45kN) are verified against a vehicle type database. The first axle load of 60kN is lower than the lower limit of the standard range of 65kN, with a calculation deviation of 5kN; the second axle load of 64.71kN is close to the lower limit of 65kN, with a deviation of 0.29kN; the third to sixth axles are all lower than the lower limit of the standard range, with deviations of 30-47.53kN. Using the axle group balancing correction method, the axle load difference of the parallel double axle groups is forcibly constrained to ≤2kN: the third and fourth axles are corrected to 98kN and 102kN (both within the range of 95-105kN), and the fifth and sixth axles are corrected to 113kN and 117kN (both within the range of 110-120kN); the first and second axles remain at their original values ​​(deviation within the allowable error of ±5kN). The verified highway traffic volume axle load data are generated as follows: 60kN, 64.71kN, 98kN, 102kN, 113kN, and 117kN. All values ​​are rounded to two decimal places, which conforms to the database standard and retains a reasonable error range.

[0136] Step S33: Perform time-segmented distribution statistics on the verified highway traffic volume axle load data according to the preset time period, so as to calculate the number of vehicles, total axle load and proportion corresponding to different axle load intervals in each time period, so as to obtain the highway time-segmented axle load distribution data.

[0137] In this embodiment of the invention, the verified axle load data is statistically analyzed by time period according to a preset time period cycle (morning peak 7:00-9:00, off-peak 9:00-17:00, evening peak 17:00-19:00). During the morning peak, data from 100 six-axle trucks are analyzed: 200 axles are in the 50-70kN range (first and second axles), with a total axle load of (60+64.71)×100=12471kN, accounting for 22%; there is no data for the 70-100kN range; and 400 axles are in the 100-120kN range (third to sixth axles), with a total axle load of (98+102+113+117)×100=43000kN, accounting for 78%. During off-peak hours, 80 trucks: total axle load in the 50-70kN range is 9976.8kN, accounting for 22%; total axle load in the 100-120kN range is 34400kN, accounting for 78%. During evening peak hours, 120 trucks: total axle load in the 50-70kN range is 14965.2kN, accounting for 22%; total axle load in the 100-120kN range is 51600kN, accounting for 78%. The generated highway axle load distribution data by time period includes the number of axles, total load, and percentage for each time period, accurately reflecting the axle load distribution pattern in different time periods.

[0138] Step S34: Obtain the structural parameters of the highway pavement, including the thickness, material type, compressive strength, flexural tensile strength and fatigue characteristics of each pavement structural layer;

[0139] In this embodiment of the invention, the obtained highway pavement structure parameters were determined through core drilling and laboratory testing: the surface layer is asphalt concrete, consisting of three layers with a total thickness of 18cm (4cm top layer, 6cm middle layer, and 8cm bottom layer), a compressive strength of 4.5MPa at 25℃, a flexural tensile strength of 1.0MPa at 15℃, and a fatigue life coefficient of 5.0×10⁻⁶. 6 Secondary; the base layer is cement-stabilized crushed stone, 36cm thick, with a 7-day compressive strength of 3.0MPa, a flexural tensile strength of 0.8MPa, and a fatigue life coefficient of 3.0×10⁻⁶. 6 The subbase consists of graded crushed stone, 20cm thick, with a compressive strength of 1.5MPa. All parameters were averaged after three parallel tests, with thickness measurements accurate to 0.1cm and strength tests accurate to 0.1MPa. Fatigue characteristics were determined through indoor small beam bending fatigue tests (loading frequency 10Hz, stress ratio 0.5) to ensure that the data accurately reflects the pavement structure performance.

[0140] Step S35: Based on the highway pavement structure parameters and the highway time-segmented axle load distribution data, conduct a pavement axle load cumulative damage assessment to obtain the highway pavement axle load cumulative damage value.

[0141] In this embodiment of the invention, cumulative damage is assessed based on pavement structure parameters and time-segmented axle load distribution data. Using Miner's theory, the equivalent axle load during the morning peak is 354.65 times, and the surface layer damage is calculated as 354.65 / 5.0 × 10⁻⁶. 6 =7.09×10 -5 Root layer damage = 354.65 / 3.0 × 10 6 =1.18×10 -4 Peak equivalent shaft rotations: 210.32; surface layer damage: 4.21 × 10⁻⁶. -5 7.01×10⁻⁶ basal layer damage -5 During the evening rush hour, the equivalent number of axle strikes was 380.17, and the surface layer damage was 7.60 × 10⁻⁶. -5 1.27 × 10⁻⁶ basal layer damage -4 Cumulative damage over time: Total surface layer damage = 7.09 × 10⁻⁶ -5 +4.21×10 -5 +7.60×10 -5 =1.89×10 -4 Total damage at the grassroots level = 1.18 × 10 -4 +7.01×10 -5 +1.27×10 -4 =3.15×10 -4 The summation yields a cumulative axle load damage value of 5.04 × 10⁻⁶ for the highway pavement. -4 This value is used to assess the remaining life of the road surface. Maintenance and repair are required when the cumulative damage reaches 1.0.

[0142] Furthermore, step S32 includes the following steps:

[0143] The standard axle load parameters for matching the corresponding vehicle type axle group are obtained from the vehicle type database, and the deviation is calculated by comparing the preliminary estimated value of highway traffic volume axle load with the standard axle load parameters to obtain the highway traffic volume axle load deviation rate.

[0144] In this embodiment of the invention, the standard axle load parameters for a 6×4 axle truck are retrieved from the vehicle type database: 70kN for the first axle, 70kN for the second axle, 100kN for the third axle, 100kN for the fourth axle, 115kN for the fifth axle, and 115kN for the sixth axle. The preliminary estimated axle load values ​​for highway traffic volume (60kN, 64.71kN, 65.00kN, 66.77kN, 67.67kN, 67.45kN) are compared with the standard axle load parameters to calculate the deviation. The deviation rates for the first axle are calculated as follows: (60-70) / 70 × 100% = -14.29%, the second axle as (64.71-70) / 70 × 100% = -7.56%, the third axle as (65.00-100) / 100 × 100% = -35.00%, the fourth axle as (66.77-100) / 100 × 100% = -33.23%, the fifth axle as (67.67-115) / 115 × 100% = -41.16%, and the sixth axle as (67.45-115) / 115 × 100% = -41.35%. All deviation rates are rounded to two decimal places to generate the highway traffic volume axle load deviation rate, clearly indicating the degree of deviation between the estimated value and the standard value for each axle.

[0145] Preferably, the axle load deviation rate of highway traffic volume is compared and judged with the preset axle load deviation range to filter out the standardized axle load detection dataset corresponding to the highway traffic volume axle load deviation rate exceeding the preset axle load deviation range, so as to obtain highway axle load data with deviation rate not meeting the requirements.

[0146] In this embodiment of the invention, the deviation rate of each shaft is compared with the preset axle load deviation range of ±20% to determine the deviation. The deviation rate of the first shaft is -14.29% (within the range), the deviation rate of the second shaft is -7.56% (within the range), the deviation rate of the third shaft is -35.00% (out of the range), the deviation rate of the fourth shaft is -33.23% (out of the range), the deviation rate of the fifth shaft is -41.16% (out of the range), and the deviation rate of the sixth shaft is -41.35% (out of the range). Axle load data with deviation rates exceeding the preset range were filtered out: third axle 65.00kN, fourth axle 66.77kN, fifth axle 67.67kN, and sixth axle 67.45kN. The corresponding standardized axle load test datasets include strain signals (910με, 868με, 812με, 742με), axle coordinates (17.00m, 19.00m, 22.00m, 26.00m), and attenuation rates (30%, 35%, 40%, 45%) for these axles. This yielded highway axle load data with non-compliant deviation rates, providing a clear target for subsequent corrections.

[0147] Preferably, the wheelbase and tire quantity constraints corresponding to the vehicle type are obtained, and the highway axle load data with non-compliant deviation rates are constrained, corrected, adjusted, and verified based on the wheelbase and tire quantity constraints corresponding to the vehicle type. The verified highway axle load data is then integrated with the highway axle load data with compliant deviation rates to generate verified highway traffic volume axle load data.

[0148] In this embodiment of the invention, the wheelbase constraints (distance between adjacent axle groups 3.00-5.00m) and tire quantity constraints (two tires per axle group, load distribution coefficient 1.0) of a 6×4 axle truck are obtained. Axle load data with non-compliant deviation rates are corrected: the third and fourth axles are parallel axles with a standard axle load of 100kN each. Based on the dual-axle load balance constraint, the corrected load is 98kN for the third axle and 102kN for the fourth axle (total 200kN, consistent with the standard total of 200kN); the fifth and sixth axles are parallel axles with a standard axle load of 115kN each. After correction, the load is 113kN for the fifth axle and 117kN for the sixth axle (total 230kN, consistent with the standard total of 230kN). The corrected data were verified as follows: third axle 98kN (deviation rate -2.00%), fourth axle 102kN (deviation rate +2.00%), fifth axle 113kN (deviation rate -1.74%), and sixth axle 117kN (deviation rate +1.74%). All of these were within the preset range. The corrected data were then integrated with the first axle 60kN and the second axle 64.71kN, which also met the deviation rate requirements, to generate verified highway traffic volume axle load data. This ensured that all axle loads met both vehicle structural constraints and were within the allowable deviation range.

[0149] Furthermore, step S35 includes the following steps:

[0150] Step S351: Based on the material type and fatigue characteristic parameters in the highway pavement structure parameters, perform axle load conversion matching analysis to determine the equivalent axle number conversion coefficients corresponding to different axle loads and generate a highway fatigue axle load equivalent conversion table.

[0151] In this embodiment of the invention, based on highway pavement structural parameters (asphalt concrete surface layer, compressive strength 4.5 MPa, fatigue life coefficient 5.0 × 10⁻⁶), the following method is used: 6 Secondary; cement-stabilized crushed stone base course, compressive strength 3.0 MPa, fatigue life coefficient 3.0 × 10⁻⁶. 6 (Time) Axle load conversion matching analysis was performed, using the modified AASHTO axle load conversion formula. A 100kN single-axle dual-wheel set was used as the standard axle load to calculate the equivalent axle load conversion factor for different axle loads: The generated highway fatigue axle load equivalent conversion table is divided into axle load ranges (50-70kN, 70-100kN, 100-120kN), and the conversion coefficients corresponding to each range are clearly defined to ensure the accurate quantitative relationship between axle load and equivalent axle load.

[0152] Step S352: Based on the highway fatigue axle load equivalent conversion table and the highway time-sharing axle load distribution data, calculate the equivalent axle number corresponding to different axle load intervals in each time period to generate highway time-sharing equivalent axle number data.

[0153] In this embodiment of the invention, the equivalent axle loads for each time period are calculated based on a highway fatigue axle load equivalent conversion table combined with time-segmented axle load distribution data (morning peak 7:00-9:00, off-peak 9:00-17:00, evening peak 17:00-19:00). During the morning peak period: 60kN axles appear 120 times, 70kN axles appear 150 times, 98kN axles appear 80 times, 102kN axles appear 70 times, 113kN axles appear 50 times, and 117kN axles appear 40 times. Calculate the equivalent number of axle loads for each axle load range: 60kN axle = 120 × 0.1296 = 15.55 times, 70kN axle = 150 × 0.2401 = 36.02 times, 98kN axle = 80 × 0.8852 = 70.82 times, 102kN axle = 70 × 1.0824 = 75.77 times, 113kN axle = 50 × 1.6305 = 81.53 times, 117kN axle = 40 × 1.8739 = 74.96 times. Total equivalent number of axle loads = 15.55 + 36.02 + 70.82 + 75.77 + 81.53 + 74.96 = 354.65 times. The total equivalent axle load during off-peak hours is 210.32 times, and during evening peak hours it is 380.17 times. Time-based equivalent axle load data for highways is generated and two decimal places are retained to reflect the differences in the fatigue impact of axle load on the road surface at different times.

[0154] Step S353: Construct a pavement axle load damage assessment model based on the thickness, compressive strength and flexural tensile strength of each pavement structural layer in the pavement structural parameters of the highway, and input the equivalent axle data of the highway in different time periods into the pavement axle load damage assessment model to calculate the cumulative damage corresponding to each pavement structural layer according to the time series, and calculate the cumulative damage summation based on the cumulative damage corresponding to each pavement structural layer to obtain the cumulative damage value of the highway pavement axle load.

[0155] In this embodiment of the invention, a pavement axle load damage assessment model is constructed based on pavement structural parameters (surface layer thickness 18cm, compressive strength 4.5MPa, flexural tensile strength 1.0MPa; base layer thickness 36cm, compressive strength 3.0MPa, flexural tensile strength 0.8MPa). The model adopts Miner's linear cumulative damage theory, with the damage coefficient = equivalent axle load / fatigue life coefficient. The equivalent axle load data for different time periods is input into the model: morning peak surface layer damage = 354.65 / 5.0 × 10⁻¹⁰ 6 =7.09×10 -5 The damage amount to the basal layer = 354.65 / 3.0 × 10 6 =1.18×10 -4 Damage to the flat-peak surface layer = 210.32 / 5.0 × 10 6 =4.21×10 -5 Damage to the basal layer = 210.32 / 3.0 × 10 6 =7.01×10 -5 Surface damage during evening peak hours = 380.17 / 5.0 × 10 6 =7.60×10 -5 Damage to the basal layer = 380.17 / 3.0 × 10 6 =1.27×10 -4 Cumulative damage over time: Total surface layer damage = 7.09 × 10⁻⁶ -5 +4.21×10 -5 +7.60×10 -5 =1.89×10 -4 Total damage to the primary layer = 1.18 × 10 -4 +7.01×10 -5 +1.27×10 -4 =3.15×10 -4 The cumulative damage value of the highway pavement under axle load, obtained by summing the cumulative damage, is 1.89 × 10⁻⁶. -4 +3.15×10 -4 =5.04×10 -4 This value is used to assess the degree of fatigue damage to the road surface.

[0156] Furthermore, step S4 includes the following steps:

[0157] Step S41: Obtain the number and proportion of highway vehicle axle loads and the cumulative axle load damage value of the highway pavement in different axle load intervals within each time period by using the highway time-segmented axle load distribution data and the highway pavement axle load cumulative damage value.

[0158] In this embodiment of the invention, information for each time period is extracted from the time-segmented axle load distribution data and the cumulative axle load damage value of the highway: During the morning peak from 7:00 to 9:00, there are 200 axle loads in the 50-70kN range, accounting for 22%, with a corresponding cumulative road damage value of 1.89 × 10⁻⁶. -5 (Top layer) + 3.15 × 10 -5 (Basic level) = 5.04 × 10 -5 The number of axle loads in the 100-120kN range was 400, accounting for 78%, with a corresponding damage value of 1.70×10⁻⁶. -4 (Top layer) + 8.65×10 -5 (Basic level) = 2.565 × 10 -4 During the off-peak hours of 9:00-17:00, 160 axle loads were in the 50-70kN range, accounting for 22% of the total, with a damage value of 1.43×10⁻⁶. -5 +2.42×10 -5 =3.85×10 -5 The number of axle loads in the 100-120kN range was 320, accounting for 78%, with a damage value of 1.31×10⁻⁶. -4 +6.68×10 -5 =1.978×10 -4 During the evening rush hour from 5:00 PM to 7:00 PM, 240 axle loads were in the 50-70kN range, accounting for 22% of the total, with a damage value of 2.15 × 10⁻⁶. -5 +3.63×10 -5 =5.78×10 -5 The number of axle loads in the 100-120kN range was 480, accounting for 78%, with a damage value of 2.05×10⁻⁶. -4 +1.03×10 -4 =3.08×10 -4 All data are categorized and organized by axle load interval and time period, clearly defining the correspondence between the number and proportion of axle loads in each interval and the damage value.

[0159] Step S42: Based on the number and proportion of axle loads of highway vehicles in different axle load intervals within each time period and the cumulative axle load damage value borne by the road surface, perform axle load-damage correlation analysis to analyze and identify the correlation between vehicle axle load distribution and cumulative damage within each time period, and generate highway vehicle axle load-cumulative damage correlation features.

[0160] In this embodiment of the invention, axle load-damage correlation analysis is performed based on data from different time periods to calculate the damage contribution of different axle load ranges: the 100-120kN range during the early peak accounts for 78%, contributing 2.565×10 -4 Damage value (total damage 3.069 × 10⁻⁶) -4(83.6%); the 50-70kN range accounted for 22%, contributing 5.04×10 -5 (16.4%). The peak load contribution from the 100-120kN range was 1.978×10. -4 (Total 2.363×10) -4 (83.7%), with a contribution of 3.85 × 10 in the 50-70 kN range. -5 (16.3%). The evening peak hour contribution from the 100-120kN range was 3.08 × 10⁻⁶. -4 (Total 3.658×10) -4 (84.2%), with the 50-70kN range contributing 5.78×10. -5 (15.8%). A correlation curve was plotted, with the horizontal axis representing the axle load range and the vertical axis representing the damage percentage. The curve showed that the damage contribution in the 100-120kN range was stable at 83%-85%. The correlation characteristics of axle load and cumulative damage on highways were generated: the larger the axle load and the higher the percentage, the more significant the contribution to pavement damage. Moreover, the damage rate during the evening peak was 1.2 times higher than that during the morning peak.

[0161] Step S43: Obtain real-time axle load detection data of the highway, and perform dynamic high-precision detection and correction of real-time traffic volume axle load based on the highway vehicle axle load-cumulative damage correlation characteristics to generate a dynamic detection report of highway traffic volume axle load.

[0162] In this embodiment of the invention, real-time axle load detection data of highways is obtained: at a certain moment, the number of axle loads in the 100-120kN range suddenly increases by 500 (25% higher than the historical average for the same period), and the number in the 50-70kN range increases by 200. Based on the axle load-cumulative damage correlation characteristics, the real-time data is corrected: the 100-120kN range is corrected using a peak-hour damage coefficient of 1.2, and the damage value is adjusted from the theoretical 2.565 × 10⁻⁶. -4 ×1.25=3.206×10 -4 Adjusted to 3.206×10 -4 ×1.2=3.847×10 -4 The 50-70kN range is corrected at 22%, with a damage value of 5.04×10⁻⁶. -5 The values ​​remain unchanged. The generated dynamic detection report includes the number of real-time axle loads (700), the percentage of each range (71.4% for 100-120kN, 28.6% for 50-70kN), and the corrected damage value of 4.351×10⁻⁶. -4 The deviation rate from the historical average (+25%) ensures that the data reflects the current actual damage trend.

[0163] Step S44: Based on the cumulative axle load damage value of the road surface corresponding to the dynamic detection report of axle load of highway traffic volume, compare and evaluate the road surface damage with the preset road surface design service life threshold to obtain the current damage level of the highway road surface.

[0164] In this embodiment of the invention, by setting the pavement design service life threshold to a cumulative damage value of 1.0 (corresponding to a design service life of 15 years), the current cumulative axle load damage value of the highway pavement is 5.04 × 10⁻⁶. -4 (Daily). Calculation of current damage level = Daily damage value × 365 days × Years of use (3 years) = 5.04 × 10 -4 ×365×3≈0.552. Compared to a threshold of 1.0, the current damage level is 55.2%, with surface layer damage accounting for 37.5% (0.189 / 0.504) and base layer damage accounting for 62.5% (0.315 / 0.504). The assessment results show that the base layer damage is progressing faster and requires close monitoring. The remaining lifespan is calculated as (1.0-0.552)÷(5.04×10). -4 (×365)≈2.4 years, providing a time basis for maintenance planning.

[0165] Step S45: Based on the current damage level of the highway pavement and combined with the dynamic detection report of highway traffic volume and axle load, formulate corresponding vehicle axle load limits and maintenance timing plans, and generate a highway pavement axle load maintenance plan that includes maintenance process, maintenance scope and implementation priority.

[0166] In this embodiment of the invention, based on the current damage level of 55.2% and the dynamic detection report, vehicle axle load restrictions are established: 30% of vehicles with axle loads of 100-120kN are restricted during evening peak hours (17:00-19:00), and the speed limit is reduced from 100km / h to 80km / h (reducing the impact coefficient by 15%). The maintenance plan includes: emulsified asphalt grouting for the base layer (process parameters: asphalt dosage 3.5kg / m³). 2 The application temperature is 160℃, covering the area from K10+000 to K15+000 (the section with concentrated damage). The surface layer undergoes micro-surfacing treatment (aggregate particle size 3-5mm, asphalt-aggregate ratio 6.0%), covering the entire inspection section. Implementation priority: base course maintenance (within 1 month) > axle load limitation (immediate implementation) > surface course maintenance (within 3 months). The plan clearly defines the material usage, construction temperature, and acceptance standards for each process, ensuring that the base course damage rate is reduced by 40% and the surface course service life is extended by 2 years after maintenance.

[0167] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for dynamic high-precision detection of axle load in highway traffic volume, characterized in that, Includes the following steps: Step S1: Collect dynamic response signals and vehicle traffic characteristic data corresponding to the highway road surface through a multi-type sensor array to obtain the original axle load detection data of highway traffic volume; perform spatiotemporal synchronization and noise suppression processing on the original axle load detection data of highway traffic volume to generate a standardized axle load detection dataset. Step S2: Based on the standardized axle load detection dataset, identify vehicle axle type and locate axle position to extract axle group distribution features and axle spacing parameters, and generate highway vehicle axle system feature data; Load-response mapping estimation is performed based on the characteristic data of highway vehicle axle system and the attenuation characteristics corresponding to the road surface response signal to obtain a preliminary estimate of highway traffic volume axle load. Step S3: Obtain the vehicle type database, and perform vehicle axle load verification and time-segmented distribution statistics on the standardized axle load detection dataset based on the preliminary estimate of highway traffic volume axle load and the vehicle type database to obtain highway time-segmented axle load distribution data; obtain highway pavement structure parameters, and perform pavement axle load cumulative damage assessment based on highway pavement structure parameters combined with highway time-segmented axle load distribution data to obtain highway pavement axle load cumulative damage value. Step S4: Perform dynamic high-precision detection based on the time-segmented axle load distribution data of the expressway and the cumulative axle load damage value of the expressway pavement to generate a dynamic detection report of expressway traffic volume axle load; formulate a corresponding expressway pavement axle load maintenance plan based on the dynamic detection report of expressway traffic volume axle load.

2. The method for dynamic high-precision detection of axle load in highway traffic volume according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: By deploying strain sensors and acceleration sensors at different depths on the highway surface, and deploying a multi-type sensor array consisting of laser profile sensors and video recognition equipment on the side of the highway, the strain sensors collect the dynamic response signal corresponding to the road surface strain when a vehicle passes, the acceleration sensors collect the dynamic response signal corresponding to the road surface vibration, the laser profile sensors collect the vehicle length and wheelbase geometric parameters, and the video recognition equipment collects the vehicle type and travel trajectory, the original axle load detection data of highway traffic volume is obtained. Step S12: Based on the time synchronization module corresponding to the multi-type sensor array, the acquisition timestamps of each sensor are unified to the same clock reference, and the error is controlled within the preset range to generate multi-source data time synchronization parameters. Step S13: Establish the coordinate mapping relationship of each sensor detection point according to the spatial layout of each sensor in the multi-type sensor array, and associate the dynamic response signal in the original axle load detection data of highway traffic volume with the vehicle traffic characteristic data to the same coordinate mapping relationship to generate multi-source data spatial calibration parameters. Step S14: Based on the time synchronization parameters and spatial calibration parameters of multi-source data, the original axle load detection data of highway traffic volume is spatiotemporally synchronized and integrated to generate a spatiotemporally aligned detection dataset of highway axle load containing time and space. Step S15: Perform noise suppression and standardization on the highway axle load spatiotemporal alignment detection dataset. Wavelet threshold denoising is used to remove high-frequency interference noise and standardize the data to generate a standardized axle load detection dataset.

3. The method for dynamic high-precision detection of axle load in highway traffic volume according to claim 2, characterized in that, Step S2 includes the following steps: Step S21: Perform peak value statistics on the corresponding dynamic response signals in the standardized axle load detection dataset to statistically analyze the peak occurrence time, peak amplitude, and peak signal spectrum of each peak value of the highway axle load dynamic signal, and generate a highway axle load signal peak value sequence. Step S22: Perform signal peak feature analysis based on the peak sequence of highway axle load signal to obtain the peak time interval, amplitude change rate and spectral energy distribution corresponding to the dynamic signal of highway axle load; Step S23: Based on the peak time interval, amplitude change rate, and spectral energy distribution of the dynamic signal of highway axle load, and combined with the corresponding vehicle type and traffic trajectory in the standardized axle load detection dataset, perform vehicle axle type identification analysis to identify and distinguish vehicle axle type groups corresponding to single axle, parallel dual axle, and parallel triple axle, and obtain the highway traffic volume axle type group classification results. Step S24: Based on the classification results of highway traffic volume axle type groups and combined with the peak time interval and amplitude change rate corresponding to the highway axle load dynamic signal, perform vehicle axle position positioning statistics on the standardized axle load detection dataset to generate highway vehicle axle system feature data. Step S25: Based on the axle system characteristic data of highway vehicles and combined with the attenuation characteristics of the road surface response signal, perform load-response mapping estimation to obtain a preliminary estimate of the axle load of highway traffic volume.

4. The method for dynamic high-precision detection of axle load in highway traffic volume according to claim 3, characterized in that, Step S24 includes the following steps: Step S241: Obtain the corresponding highway vehicle speed by matching the vehicle type and travel trajectory within the standardized axle load detection dataset; Step S242: Based on the peak time interval and amplitude change rate corresponding to the dynamic signal of highway axle load, and combined with the speed of highway vehicles, the vehicle axle spacing is converted to obtain the initial parameters of highway vehicle axle spacing; Step S243: Based on the vehicle length and wheelbase geometric parameters corresponding to the standardized axle load detection dataset, perform axle spacing comparison and correction calculation on the initial parameters of highway vehicle axle spacing to obtain highway axle spacing comparison and correction parameters; Step S244: Based on the classification results of axle type groups of highway traffic volume and combined with the highway vehicle speed, calculate the axle position of each axle group to locate the corresponding highway pavement coordinates and generate a dataset of axle position coordinates for each axle group; perform axle group distribution statistics based on the dataset of axle position coordinates for each axle group to analyze and extract the number and distribution of axles in each axle group, and obtain the distribution characteristics of each axle group of the highway. Step S245: Based on the highway axle spacing comparison correction parameters and the distribution characteristics of each axle group on the highway, integrate the vehicle axle system features to generate highway vehicle axle system feature data containing the number of axles, axle position and axle spacing of each axle group.

5. The method for dynamic high-precision detection of axle load in highway traffic volume according to claim 3, characterized in that, Step S25 includes the following steps: Step S251: Perform signal attenuation characteristic analysis on the pavement response signals corresponding to pavement strain and pavement vibration in the standardized axle load test dataset to calculate the amplitude attenuation rate and phase shift of the pavement response signal at different propagation distances, and generate highway pavement response attenuation characteristic parameters. Step S252: Based on the axle coordinate information in the highway vehicle axle system feature data, determine the road surface response signal acquisition points corresponding to each axle group, and perform axle position-attenuation correlation analysis based on the road surface response signal acquisition points corresponding to each axle group and the highway road surface response attenuation feature parameters to obtain highway vehicle axle position-response attenuation correlation data. Step S253: Based on the number of axles and axle spacing in the axle system characteristic data of highway vehicles and combined with the response attenuation characteristic parameters corresponding to the axle position-response attenuation correlation data of highway vehicles, perform load-response nonlinear mapping analysis to obtain the nonlinear mapping relationship between highway vehicle axle load and road surface response signal attenuation. Step S254: Based on the nonlinear mapping relationship between highway vehicle axle load and road surface response signal attenuation, and combined with highway vehicle axle system characteristic data and highway road surface response attenuation characteristic parameters, construct the axle load-response mapping function to generate the highway vehicle axle load-response mapping function; Step S255: Perform load-response mapping estimation based on the highway vehicle axle load-response mapping function to obtain a preliminary estimate of highway traffic volume axle load.

6. The method for dynamic high-precision detection of axle load in highway traffic volume according to claim 5, characterized in that, Step 255 includes the following steps: By deploying environmental sensors on the corresponding highway surface to synchronously collect corresponding temperature, humidity, road surface dryness and wetness status and rainfall data, a highway surface environmental parameter dataset is generated. Based on the environmental parameters in the highway pavement environmental parameter dataset, an environmental-axle load characteristic influence analysis is performed on the highway vehicle axle system characteristic data to generate the highway environmental-axle load characteristic influence distribution curve. Based on the distribution curve of the influence of highway environment-axle load characteristics, the corresponding highway axle load influence correction factor under different environmental condition combinations is calculated. Based on the highway axle load influence correction factor and the highway vehicle axle load-response mapping function, the load-response mapping correction estimation is performed to obtain the preliminary estimate of highway traffic volume axle load.

7. The method for dynamic high-precision detection of axle load in highway traffic volume according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Obtain the vehicle type database, which includes the standard axle load range, typical axle group configuration and axle weight distribution for different vehicle types; Step S32: Based on the preliminary estimate of highway traffic volume axle load and the vehicle type database, perform vehicle axle load verification on the standardized axle load detection dataset to generate verified highway traffic volume axle load data; Step S33: Perform time-segmented distribution statistics on the verified highway traffic volume axle load data according to the preset time period, so as to calculate the number of vehicles, total axle load and proportion corresponding to different axle load intervals in each time period, so as to obtain the highway time-segmented axle load distribution data. Step S34: Obtain the structural parameters of the highway pavement, including the thickness, material type, compressive strength, flexural tensile strength and fatigue characteristics of each pavement structural layer; Step S35: Based on the highway pavement structure parameters and the highway time-segmented axle load distribution data, conduct a pavement axle load cumulative damage assessment to obtain the highway pavement axle load cumulative damage value.

8. The method for dynamic high-precision detection of axle load in highway traffic volume according to claim 7, characterized in that, Step S32 includes the following steps: The standard axle load parameters for matching the corresponding vehicle type axle group are obtained from the vehicle type database, and the deviation is calculated by comparing the preliminary estimated value of highway traffic volume axle load with the standard axle load parameters to obtain the highway traffic volume axle load deviation rate. The standard axle load test dataset corresponding to the axle load deviation rate of highway traffic volume is selected by comparing and judging between the axle load deviation rate of highway traffic volume and the preset axle load deviation range, so as to obtain the highway axle load data with deviation rate that does not meet the requirements. Obtain the wheelbase and tire quantity constraints corresponding to the vehicle type, and perform constraint correction and adjustment verification on the highway axle load data with non-compliant deviation rates based on the wheelbase and tire quantity constraints corresponding to the vehicle type. Then, integrate the verified highway axle load data with the highway axle load data with compliant deviation rates to generate verified highway traffic volume axle load data.

9. The method for dynamic high-precision detection of axle load in highway traffic volume according to claim 7, characterized in that, Step S35 includes the following steps: Step S351: Based on the material type and fatigue characteristic parameters in the highway pavement structure parameters, perform axle load conversion matching analysis to determine the equivalent axle number conversion coefficients corresponding to different axle loads and generate a highway fatigue axle load equivalent conversion table. Step S352: Based on the highway fatigue axle load equivalent conversion table and the highway time-sharing axle load distribution data, calculate the equivalent axle number corresponding to different axle load intervals in each time period to generate highway time-sharing equivalent axle number data. Step S353: Construct a pavement axle load damage assessment model based on the thickness, compressive strength and flexural tensile strength of each pavement structural layer in the pavement structural parameters of the highway, and input the equivalent axle data of the highway in different time periods into the pavement axle load damage assessment model to calculate the cumulative damage corresponding to each pavement structural layer according to the time series, and calculate the cumulative damage summation based on the cumulative damage corresponding to each pavement structural layer to obtain the cumulative damage value of the highway pavement axle load.

10. The method for dynamic high-precision detection of axle load in highway traffic volume according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Obtain the number and proportion of highway vehicle axle loads and the cumulative axle load damage value of the highway pavement in different axle load intervals within each time period by using the highway time-segmented axle load distribution data and the highway pavement axle load cumulative damage value. Step S42: Based on the number and proportion of axle loads of highway vehicles in different axle load intervals within each time period and the cumulative axle load damage value borne by the road surface, perform axle load-damage correlation analysis to analyze and identify the correlation between vehicle axle load distribution and cumulative damage within each time period, and generate highway vehicle axle load-cumulative damage correlation features. Step S43: Obtain real-time axle load detection data of the highway, and perform dynamic high-precision detection and correction of real-time traffic volume axle load based on the highway vehicle axle load-cumulative damage correlation characteristics to generate a dynamic detection report of highway traffic volume axle load. Step S44: Based on the cumulative axle load damage value of the road surface corresponding to the dynamic detection report of axle load of highway traffic volume, compare and evaluate the road surface damage with the preset road surface design service life threshold to obtain the current damage level of the highway road surface. Step S45: Based on the current damage level of the highway pavement and combined with the dynamic detection report of highway traffic volume and axle load, formulate corresponding vehicle axle load limits and maintenance timing plans, and generate a highway pavement axle load maintenance plan that includes maintenance process, maintenance scope and implementation priority.

Images