A new method for evaluating the variation of heavy load and vehicle composition on asphalt pavement

Abstract

This invention belongs to the field of road engineering technology, and specifically provides a novel method for evaluating the variation patterns of heavy load and vehicle composition in asphalt pavements. It collects continuous axle load data and original vehicle type data for vehicles of types 2 to 11 on multiple asphalt pavement sections with axle load collection devices deployed in the same observation period. Based on the equivalent axle load conversion coefficient for each type of vehicle, it calculates the cumulative number of equivalent standard axle loads (ESAL) within the statistical period. Based on the traffic vehicle conversion coefficient, it calculates the cumulative equivalent traffic volume (pcu) of all vehicles of types 2 to 11 within the statistical period. Using pcu as the independent variable and ESAL and the total number of natural vehicles as dependent variables, it simultaneously constructs two sets of correlation models using linear regression. By simultaneously constructing the ESAL-pcu and natural vehicle number-pcu correlation models and performing coupled correlation analysis, it achieves a comprehensive evaluation of the variation patterns of heavy load and vehicle composition, providing a precise decision-making basis for maintenance and heavy load control.

Description

Technical Field

[0001] This invention relates to the field of road engineering technology, specifically a new method for evaluating the variation of heavy load and vehicle composition in asphalt pavement. Background Technology

[0002] In the highway infrastructure system, asphalt pavement is widely used in highway construction at all levels due to its advantages of convenient construction and driving comfort. With the continuous expansion of socio-economic activities, highway traffic volume has shown a significant growth trend, with the proportion of heavy-duty vehicles constantly increasing. These heavy-duty vehicles, with axle loads far exceeding design standards, exert repeated rolling and dynamic impacts on the pavement during operation, leading to accelerated deterioration of the asphalt pavement structure, causing early-stage defects such as rutting, cracking, and fatigue damage. This shortens the service life of the pavement, increases maintenance costs, and poses a potential threat to driving safety. The cumulative standard axle load exposure (ESAL) is a core parameter for quantifying the cumulative effect of pavement load and is directly related to the pavement fatigue life and performance degradation process. Equivalent traffic volume (pcu), through standardized conversion, unifies traffic flows of different vehicle types and axle loads into equivalent vehicle units, becoming a key basis for highway design and maintenance planning. In-depth exploration of the intrinsic relationship between ESAL and pcu has important practical value for accurately assessing the current service status of the pavement, predicting the evolution trend of defects, and optimizing the allocation of maintenance resources.

[0003] However, existing research on heavy-duty traffic often focuses on improving the ESAL calculation model or analyzing the single impact of overloading on pavement mechanical performance. It fails to achieve a simultaneous and comprehensive evaluation of the vehicle composition and load states (empty load, standard axle load, first-level overload, second-level overload, and severe overload) of vehicles ranging from 2 to 11 types (covering two-axle small motor vehicles to eleven-axle heavy semi-trailers and container trucks). Furthermore, in maintenance decision-making practice, equivalent traffic volume (pcu) or the natural number of trucks is often used directly to replace ESAL estimation. Due to a lack of systematic understanding of the correlation between the two, this substitution method is prone to significant bias when the relationship between ESAL and pcu is unclear, leading to a disconnect between maintenance plans and actual needs, affecting resource utilization efficiency and long-term pavement performance.

[0004] Therefore, we propose a new method to evaluate the variation of heavy load and vehicle composition on asphalt pavement to solve the above problems. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides a novel method for evaluating the variation patterns of heavy load and vehicle composition on asphalt pavements. This method can simultaneously evaluate the variation patterns of heavy load and vehicle composition on asphalt pavements, providing accurate decision-making basis for maintenance and heavy load control, and solving the problems mentioned in the background art.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention specifically adopts the following technical solution:

[0009] A new method for evaluating the variation of heavy load and vehicle composition on asphalt pavement includes the following steps:

[0010] S1. Collect continuous axle load data and original vehicle type data of vehicles of types 2 to 11 within the same observation period on asphalt road sections with multiple axle load collection devices deployed in the collection area.

[0011] S2. Based on the equivalent design axle load conversion factor for various types of vehicles, calculate the cumulative number of equivalent standard axle load applications (ESAL) within the statistical period.

[0012] S3. Based on the traffic survey vehicle conversion factor, calculate the cumulative equivalent traffic volume (pcu) of all vehicles of types 2 to 11 within the statistical period;

[0013] S4. Using pcu as the independent variable and ESAL and total number of natural vehicles as dependent variables, two sets of correlation models are constructed simultaneously using linear regression.

[0014] S5. Based on the correlation coefficients of the two sets of models, conduct coupled correlation analysis to evaluate the changing patterns of heavy load and vehicle composition on road sections, and output maintenance and heavy load control decision schemes.

[0015] Furthermore, in step S1, the axle load acquisition device is a dynamic weighing WIM device; the acquired raw data includes the number of natural vehicles, vehicle type secondary classification, axle type, axle load, and axle load interval distribution; the observation period can be selected as daily, weekly, quarterly, or annual, with the annual period being preferred for the overall evaluation of the regional road network; the acquired data is synchronously bound to the storage of road section pavement structure, service life, and historical defects.

[0016] Furthermore, The formula for calculating the equivalent design axle load conversion factor for vehicle type A is:

[0017]

[0018] in: for Equivalent design axle load conversion factor for vehicle class; for Class of vehicles Average number of axes for each type of shaft; for Class of vehicles Type of shaft in the first Equivalent design axle load conversion factor for axle load range; for Class of vehicles Type of shaft in the first Axle load distribution coefficient for axle load range;

[0019] ESAL is based on the formula Calculations show that Within the statistical period The total number of vehicles in each category; the axle load range is divided into unloaded, standard axle load, first-level overload, second-level overload, and severe overload ranges, and the ESAL contribution rate of overloaded vehicles is calculated separately.

[0020] Furthermore, the conversion factor standards for vehicles in step S3 are as follows: conversion factor 1 for small and medium-sized passenger cars and small trucks; conversion factor 1.5 for large passenger cars and medium-sized trucks; conversion factor 3 for large trucks; conversion factor 4 for extra-large trucks and container trucks; when special engineering vehicles exist in the area, additional conversion factors can be added to complete the PCU conversion.

[0021] Furthermore, step S4 includes data preprocessing sub-steps: removing outlier data from equipment, completing missing detection data, and standardizing the statistical period length for each road segment; independently fitting linear regression models for each road segment to obtain... , Two sets of models, and solve for the definite coefficients respectively. , As a basis for determining relevance.

[0022] Furthermore, the correlation determination threshold in step S5: Highly correlated The correlation is moderate. Low correlation is defined as a negative value, and negative values ​​represent negative correlation. Based on the two sets of correlation coefficients, four coupling conditions are defined: dual-high positive correlation, dual-high negative correlation, ESAL-pcu high negative correlation and natural vehicle number-pcu high positive correlation, and ESAL-pcu high positive correlation and natural vehicle number-pcu moderate negative correlation. A scatter plot is generated with pcu as the horizontal axis and ESAL and natural vehicle number as the two vertical axes, respectively.

[0023] Furthermore, in step S5, the traffic load levels are divided into three levels: medium, heavy, and extremely heavy, based on the ESAL of each road segment, and the maintenance priority is determined in conjunction with the coupled working conditions: road segments with extremely heavy loads and negative correlation coefficients are designated as first-level priority maintenance road segments; road segments with heavy loads and positive correlation coefficients in both heights are designated as second-level routine maintenance road segments; and road segments with medium loads and stable traffic flow structure are designated as third-level preventive maintenance road segments. Simultaneously, dynamic weighing enforcement, time-limited passage for trucks, and graded heavy load control schemes for overload diversion are output.

[0024] Furthermore, it also includes step S6, a cross-time series longitudinal comparison step: retrieving multi-year ESAL, pcu, and natural vehicle count data for the road segment, updating the regression model and correlation coefficients year by year, and longitudinally analyzing the evolution of heavy traffic flow patterns; outputting a comprehensive regional road network assessment report, including load level proportions, overload ESAL contribution rate, and maintenance fund allocation calculation data, to support road reconstruction and expansion and structural optimization design.

[0025] Furthermore, the vehicles of types 2 to 11 cover two-axle small motor vehicles to eleven-axle heavy semi-trailers and container trucks, excluding non-motorized vehicles and temporary construction machinery; the data collection process distinguishes between passenger and freight vehicles, and separately counts the proportion of ESAL (Equivalent Axle Load) of freight vehicles in the total equivalent axle load.

[0026] Furthermore, the entire method is integrated into the intelligent pavement load analysis terminal, which has built-in modular functions for WIM data docking, ESAL batch calculation, PCU conversion, regression fitting, and maintenance decision output, realizing fully automated calculation and report export.

[0027] (III) Beneficial Effects

[0028] Compared with existing technologies, this invention provides a new method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement, which has the following beneficial effects:

[0029] 1. Intuitively identify the characteristics of load-bearing asphalt pavement: This invention uses a dual correlation analysis of ESAL and PCU, and natural vehicle number and PCU, which can analyze the traffic flow structure, heavy load mixing rate and actual load rate of asphalt pavement load from two dimensions: heavy load and vehicle type. This is helpful to understand the real cause of road damage and propose targeted maintenance strategies.

[0030] 2. Improve the accuracy of maintenance fund allocation: By analyzing multiple road sections in the same area, this invention can identify which road sections can use the change in pcuU to estimate the change in ESAL, and which road sections cannot use only the change in pcuU to estimate the change in ESAL, thereby improving the accuracy of allocating maintenance funds based on traffic volume.

[0031] 3. Evaluating the axle load variation pattern of the same road at different times: This invention can also realize the continuous variation pattern of road traffic axle load through the dual correlation combination analysis of ESAL and pcu, and natural vehicle number and pcu.

[0032] By simultaneously constructing two sets of correlation models, ESAL-pcu and natural vehicle number-pcu, and performing coupled correlation analysis, a comprehensive evaluation of the changing patterns of heavy load and vehicle composition can be achieved. This approach can simultaneously evaluate the changing patterns of heavy load and vehicle composition on asphalt pavement, providing accurate decision-making basis for maintenance and heavy load control. Attached Figure Description

[0033] Figure 1 This is a general flowchart of the overall method of the present invention;

[0034] Figure 2 This is a schematic diagram illustrating that ESAL and PCU, as well as the number of natural vehicles and PCU, are highly positively correlated in this embodiment of the invention.

[0035] Figure 3 This is a schematic diagram illustrating that ESAL and pcu, as well as the number of natural vehicles and pcu, are highly negatively correlated in this embodiment of the invention.

[0036] Figure 4 This is a schematic diagram illustrating that ESAL and pcu are highly negatively correlated, while the number of natural vehicles and pcu are highly positively correlated in this embodiment of the invention.

[0037] Figure 5 This is a schematic diagram illustrating that ESAL and pcu are highly positively correlated, while the number of natural vehicles and pcu are moderately negatively correlated in this embodiment of the invention.

[0038] Figure 6 This is a classification flowchart for the PCU equivalent traffic volume conversion of the present invention;

[0039] Figure 7 This is the logic diagram for the dual-model coupling correlation determination and maintenance grading of the present invention;

[0040] Figure 8 This is a modular system architecture diagram of the smart terminal of the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Example

[0043] Traditional studies on heavy traffic on asphalt pavements primarily focus on calculating the cumulative standard axle load (ESAL) and the impact of overloading on pavement mechanical properties, but fail to simultaneously assess the composition of vehicles of types 2 to 11 and their unloaded and overloaded conditions. Furthermore, in pavement maintenance planning and decision-making, equivalent traffic volume (pcu) or the natural number of freight vehicles is often used directly to substitute for ESAL. Without a clear understanding of the correlation between the two, this substitution method is prone to introducing significant errors.

[0044] In this regard, such as Figure 1-8 As shown, this application proposes a new method for evaluating the variation of heavy load and vehicle composition in asphalt pavement, including the following steps:

[0045] S1. Collect continuous axle load data and original vehicle type data of vehicles of types 2 to 11 within the same observation period on asphalt road sections with multiple axle load collection devices deployed in the collection area.

[0046] S2. Based on the equivalent design axle load conversion factor for various types of vehicles, calculate the cumulative number of equivalent standard axle load applications (ESAL) within the statistical period.

[0047] S3. Based on the traffic survey vehicle conversion factor, calculate the cumulative equivalent traffic volume (pcu) of all vehicles of types 2 to 11 within the statistical period;

[0048] S4. Using pcu as the independent variable and ESAL and total number of natural vehicles as dependent variables, two sets of correlation models are constructed simultaneously using linear regression.

[0049] S5. Based on the correlation coefficients of the two sets of models, conduct coupled correlation analysis to evaluate the changing patterns of heavy load and vehicle composition on road sections, and output maintenance and heavy load control decision schemes.

[0050] Specifically, this embodiment provides a new method for evaluating the variation patterns of heavy loads and vehicle composition on asphalt pavements.

[0051] First, in step S1, data acquisition is performed. This data acquisition can be achieved in various ways. For example, through manual observation and recording, vehicles on the road section can be visually classified and axle counts can be performed, and portable weighing equipment can be used for sampling weighing to obtain axle load data. As another approach, fixed or mobile sensors, such as piezoelectric sensors and bending plate sensors, can be deployed, which can sense pressure changes when vehicles pass by, and thus infer axle load information. Raw vehicle model data can be obtained by manually recording license plates, obtaining vehicle model photos, or through vehicle counters. The observation period can be set according to actual needs.

[0052] Secondly, in step S2, based on the equivalent design axle load conversion factors for various types of vehicles, the cumulative number of equivalent standard axle load applications (ESAL) within the statistical period is calculated. After obtaining the axle load data and vehicle model data, the calculation of ESAL involves assigning equivalent design axle load conversion factors to different axle types, axle loads, and vehicle models according to empirical formulas or preset damage models.

[0053] Further, in step S3, the cumulative equivalent traffic volume (pcu) of all vehicles of types 2 to 11 within the statistical period is calculated based on the traffic flow conversion factor. The calculation of pcu requires the traffic flow conversion factor. These factors are typically specified by research or standards in the field of traffic engineering, based on the degree of impact of different vehicle types on road capacity.

[0054] Subsequently, in step S4, using pcu as the independent variable and ESAL and total number of natural vehicles as dependent variables, two sets of correlation models are constructed simultaneously using linear regression. The construction of these correlation models is crucial for analyzing the relationships between traffic parameters. First, the ESAL and pcu data calculated in steps S2 and S3, along with the total number of natural vehicles in the original data, are processed. Then, a linear regression method can be selected using statistical analysis software or the statistical module of a programming language. For example, by manually inputting data points, drawing a scatter plot, and visually fitting a straight line, or by using the least squares principle, the linear equation between ESAL and pcu can be calculated. ) and the linear equation between the total number of natural vehicles and the PCU ( These models describe the expected trends in ESAL and the total number of natural vehicles given a given pcu value.

[0055] Finally, in step S5, a coupled correlation analysis is conducted based on the correlation coefficients of the two sets of models to evaluate the changing patterns of heavy loads and vehicle composition on the road segment, and to output maintenance and heavy load control decision-making schemes. After constructing two sets of linear regression models, the goodness of fit of these models needs to be evaluated, usually by calculating correlation coefficients (such as the coefficient of determination R²). The correlation coefficient reflects the degree to which the independent variable explains the change in the dependent variable. For example, the strength and direction of the correlation between ESAL and PCU, and between the total number of natural vehicles and PCU, can be determined based on the R² value output by statistical software. Based on these correlation analysis results, the changing patterns of heavy loads and vehicle composition on the road segment can be judged. For example, if ESAL and PCU show a positive correlation, it indicates that an increase in traffic volume will lead to an increase in the heavy load effect. Based on these judgments, preliminary maintenance recommendations (such as regular inspections and local repairs) and heavy load control measures (such as setting up weight limit signs and strengthening patrols) can be formulated by experts based on their experience.

[0056] This application systematically collects vehicle axle load and vehicle type data across multiple road sections, and calculates ESAL and pcu based on equivalent design axle load conversion factors and traffic flow vehicle conversion factors. By constructing linear regression models of ESAL and pcu, and of the total number of natural vehicles and pcu, and conducting coupled correlation analysis, it can deeply reveal the inherent laws governing heavy loads and changes in vehicle composition on asphalt pavements. This provides accurate theoretical basis and technical support for highway engineering design optimization, maintenance decisions, and traffic control, extending highway service life, reducing maintenance costs, and improving road network traffic safety and stability.

[0057] This application further proposes that in step S1, the axle load acquisition device is a dynamic weighing WIM device; the acquired raw data includes the number of natural vehicles, vehicle type secondary classification, axle type, axle load, and axle load interval distribution; the observation period can be selected as daily, weekly, quarterly, or annual, with the annual period being preferred for the overall evaluation of the regional road network; the acquired data is synchronously bound to the storage of road section pavement structure, service life, and historical defects.

[0058] Specifically, Dynamic Weighing (WIM) equipment is a device that can continuously and in real time measure parameters such as axle load, total weight, and wheelbase of a vehicle while it is not stopped or traveling at low speed. Compared to static weighing, WIM equipment has advantages such as being non-contact, not interrupting traffic, having a large data acquisition capacity, and being highly efficient, and can accurately reflect the dynamic changes in road traffic load. WIM equipment typically consists of weighing sensors (such as piezoelectric quartz sensors, bending plate sensors, etc.) buried under the road surface, vehicle detectors (such as induction coils, laser detectors, etc.), and a data acquisition and processing unit. The sensors are responsible for sensing pressure changes as vehicles pass by, the detectors are used to identify vehicle type, speed, and wheelbase, and the data processing unit processes the raw signals to calculate accurate axle load and total weight data.

[0059] The collected raw data includes the number of natural vehicles, vehicle type secondary classification, axle type, axle load, and axle load interval distribution. The number of natural vehicles refers to the total number of vehicles actually passing through a specific cross-section within a given observation period, serving as a fundamental indicator for measuring traffic flow. Vehicle type secondary classification provides a more detailed classification of vehicles, such as further subdividing trucks into two-axle trucks, three-axle trucks, and semi-trailers, helping to accurately identify the load effect of different vehicle types on the road surface. Axle type refers to the number of axles and axle group configuration of a vehicle, such as single-axle, double-axle, and triple-axle configurations; different axle types result in different stress distributions on the road surface. Axle load is the weight borne by a single axle and is a key parameter for calculating the Equivalent Standard Axle Load (ESAL). Axle load interval distribution involves statistically analyzing axle load data according to preset intervals (such as unloaded, standard axle load, first-level overload, second-level overload, and severe overload), which can intuitively reflect the proportion and degree of overloaded vehicles, providing a basis for overload management. These detailed raw data provide accurate input for subsequent calculations of equivalent standard axle load (ESAL) and equivalent traffic volume (pcu), ensuring the accuracy of the evaluation results and enabling in-depth analysis of the impact of different vehicle types and load levels on the road surface.

[0060] By employing the aforementioned technical solution and utilizing dynamic weighing (WIM) equipment for data acquisition, continuous, efficient, and non-contact acquisition of vehicle axle load data can be achieved, significantly improving the efficiency and accuracy of data collection. Simultaneously, the collected raw data not only includes the number of vehicles but also details the vehicle type, axle type, axle load, and axle load range distribution, providing ample and refined data support for subsequent accurate calculations of the equivalent standard axle load (ESAL) and equivalent traffic volume (pcu), thus ensuring the reliability of the evaluation of heavy load and vehicle composition variation patterns. Furthermore, flexible observation period settings, particularly prioritizing annual periods for overall regional road network evaluation, help capture long-term trends and avoid interference from short-term fluctuations in the evaluation results. More importantly, synchronously binding and storing the collected data with the road segment's pavement structure, service life, and historical damage attributes establishes a direct correlation between traffic load data and pavement performance data. This provides a foundation for in-depth analysis of the evolution mechanism of pavement damage under load, enabling more scientific and precise formulation of maintenance and heavy load control decisions, effectively improving the refinement level and effectiveness of pavement maintenance management.

[0061] This application further proposes the following formula for calculating the equivalent design axle load conversion factor for vehicles of category m: .in: for Equivalent design axle load conversion factor for vehicle class; for Class of vehicles Average number of axes for each type of shaft; for Class of vehicles Type of shaft in the first Equivalent design axle load conversion factor for axle load range; for Class of vehicles Type of shaft in the first Axle load distribution coefficient for axle load range;

[0062] ESAL is based on the formula Calculations show that Within the statistical period The total number of vehicles in each category; the axle load range is divided into unloaded, standard axle load, first-level overload, second-level overload, and severe overload ranges, and the ESAL contribution rate of overloaded vehicles is calculated separately.

[0063] The formula for calculating the equivalent design axle load conversion factor is used to accurately calculate specific... Equivalent design axle load conversion factor for vehicle type ; This is a key parameter for measuring the ability of different types of vehicles to damage road structures, and the accuracy of its calculation directly affects the reliability of the Equivalent Standard Axle Load Accumulated Action Count (ESAL). This formula comprehensively considers... The number of axle types for vehicle types, the equivalent design axle load conversion factor for each axle type in different axle load ranges, and the axle load distribution factor enable a refined quantification of vehicle damage effects. It represents This coefficient represents the number of equivalent standard axle loads exerted on the road surface by a vehicle of a certain type during a single passage. A higher value indicates a greater ability of that type of vehicle to damage the road surface. This coefficient is determined based on factors such as the vehicle's structural characteristics, load distribution, and road surface response characteristics. express The average number of axle types of class i in vehicles. This parameter reflects the characteristics of the vehicle's axle configuration and is an important component in calculating the total damage effect. It is aimed at Class of vehicles For a specific axle load class within a certain j-th level axle load range, a conversion factor is used to convert its axle load to a standard axle load. This factor takes into account the nonlinear effect of axle load on road surface damage; that is, the greater the axle load, the more exponentially the damage to the road surface increases. By subdividing the axle load range and assigning different conversion factors, the destructive force of vehicles on the road surface under different load conditions can be reflected more accurately. express Class of vehicles The frequency or proportion of axle type occurring within the j-th axle load range. This coefficient reflects the distribution of different load states of a vehicle during actual operation. By statistically analyzing a large amount of vehicle axle load data, the distribution coefficients for different axle load ranges can be obtained, thereby enabling... The calculation more closely reflects the characteristics of actual traffic flow. ESAL is calculated using the formula... The formula is used to calculate the cumulative equivalent standard axle load (ESAL) on the road surface caused by all vehicle types within a statistical period. ESAL is a core indicator for evaluating road fatigue damage and designing road structure thickness. This is achieved by converting the equivalent design axle load of each vehicle type using a conversion factor. Compared with the total number of natural vehicles in the statistical period Multiplying and summing allows for a comprehensive and accurate quantification of the traffic load borne by the road surface. This refers to the period within a specific statistical period that falls under the category of The total number of vehicles actually passing through the road is directly derived from the original vehicle model data collected in step S1 and is one of the fundamental data for calculating ESAL (Emergency Response Level). Axle load intervals are divided into unloaded, standard axle load, first-level overload, second-level overload, and severe overload intervals. Subdividing axle load into multiple intervals, including unloaded, standard axle load, and different degrees of overload, allows for a more precise assessment of the damage to the road surface caused by vehicles under different load conditions. This classification method highlights the severity of road surface damage caused by overloaded vehicles, providing data support for subsequent overload control. Separately calculating the ESAL contribution rate of overloaded vehicles means that the proportion of damage caused by overloaded vehicles can be clearly identified within the total ESAL. This measure is significant for understanding the causes of early road surface defects, assessing the harm of overloading behavior, and developing targeted heavy-load control strategies. In this way, the shortening effect of overloading on road surface life can be quantified, thus providing a basis for decision-making by law enforcement agencies.

[0064] This application proposes a novel method for evaluating the variation patterns of heavy loads and vehicle composition on asphalt pavements. In some of the aforementioned embodiments, a method is proposed to calculate the cumulative equivalent traffic volume (PCU) of all vehicle types 2 to 11 within a statistical period based on traffic flow conversion factors. However, in practical applications, the lack of a unified and clear standard for traffic flow conversion factors may lead to inaccurate assessments of the contribution of different vehicle types to traffic volume, thereby affecting the accuracy of PCU calculations and the comparability of data analysis between different road segments. This results in deviations in subsequent correlation model construction and the evaluation results of the variation patterns of heavy loads and vehicle composition.

[0065] This application further proposes that in step S3, the conversion factor standard for dispatch vehicles is clearly defined as follows: conversion factor 1 for small and medium-sized passenger cars and small trucks; conversion factor 1.5 for large passenger cars and medium-sized trucks; conversion factor 3 for large trucks; conversion factor 4 for extra-large trucks and container trucks; and when there are special engineering vehicles in the area, additional conversion factors for new vehicle types can be added to complete the PCU conversion.

[0066] Specifically, the traffic survey vehicle conversion factor standard refers to the factor used in traffic surveys to convert different types of vehicles into a unified standard vehicle (usually a small passenger car). Its purpose is to eliminate differences in road occupancy and capacity impact among different vehicle types, thereby obtaining a comprehensive equivalent traffic volume (PCU) that reflects the actual traffic flow load. This standard provides a unified basis for subsequent PCU calculations, ensuring the standardization of data processing and the accuracy of results. Small passenger cars and small trucks are typically smaller in size and have a relatively smaller impact on road space occupation and capacity; therefore, their conversion factor is set to 1 as the baseline vehicle type. This means that one small passenger car or small truck is considered a standard equivalent unit. Large buses and medium-sized trucks are larger than small passenger cars and small trucks, and their impact on road space occupation and capacity increases accordingly; therefore, their conversion factor is set to 1.5 to reflect their greater impact on traffic flow. Large trucks, with their larger size and heavier loads, significantly impact road capacity. Their conversion factor is set to 3, reflecting their effect equivalent to three standard trucks in traffic flow. Extra-large trucks and container trucks are among the largest and heaviest vehicle types on the road, exhibiting the most significant traffic congestion and road resource occupation. Therefore, their conversion factor is set to 4 to accurately reflect their substantial impact on traffic volume. Furthermore, considering the possibility of non-standard or special-purpose engineering vehicles in different areas, whose size, weight, and driving characteristics may differ significantly from conventional vehicles, this method allows for the addition of corresponding conversion factors for these special engineering vehicles based on actual conditions. This enables the PCU conversion to cover all actual traffic flows, improving the method's applicability and flexibility.

[0067] This application further proposes that step S4 includes a data preprocessing sub-step: removing outlier data from equipment, completing missing detection data, and standardizing the statistical period length for each road segment; independently fitting linear regression models for each road segment to obtain... , Two sets of models, and solve for the definite coefficients respectively. , As a basis for determining relevance.

[0068] The data preprocessing steps described above effectively identify and remove outliers from the original data, fill in missing data, and standardize the statistical period length for each road segment, thereby significantly improving the quality and consistency of the original data used for model construction. Based on this, linear regression models are independently fitted to each road segment, and the coefficient of determination is calculated. and This ensures that the constructed correlation models between ESAL and PCU, and between the total number of natural vehicles and PCU, can more accurately and reliably reflect the actual traffic characteristics and heavy load patterns of each road segment. This not only avoids model bias caused by data quality issues, but also provides a solid data and model foundation for subsequent coupled correlation analysis based on correlation coefficients. This makes the evaluation of the heavy load and vehicle composition change patterns of road segments more accurate, and provides a more convincing scientific basis for the formulation of maintenance and heavy load control decision-making schemes.

[0069] This application further proposes a clear determination of correlation in the aforementioned coupling correlation analysis. Specifically, the correlation determination threshold is set as follows: when the coefficient of determination... When, it is judged as highly correlated; when When, it is determined to be moderately correlated; when A value of 0.5 indicates a low correlation; a negative value indicates a negative correlation. This threshold defines the criteria for quantifying the degree of correlation in a linear regression model. Coefficient of Determination This is an indicator that measures the goodness of fit of a model. The closer its value is to 1, the stronger the model's explanatory power of the data, and the higher the degree to which the independent variables explain the dependent variable. By setting... The threshold values ​​categorize correlation into three levels: high correlation, moderate correlation, and low correlation, providing a quantitative basis for subsequent coupled-condition analysis. Negative values ​​are directly identified as negative correlations, indicating an inverse trend between the independent and dependent variables. These threshold settings provide a unified and objective standard for interpreting model fitting results, avoiding the ambiguity of subjective judgment.

[0070] Based on this, four coupling conditions are classified according to two sets of correlation coefficients: highly positive correlation, highly negative correlation, highly negative correlation between ESAL and pcu and highly positive correlation between the number of natural vehicles and pcu, and highly positive correlation between ESAL and pcu and moderately negative correlation between the number of natural vehicles and pcu. This technical feature further refines the interaction mode between heavy load and vehicle composition based on the correlation coefficients of the two sets of correlation models: ESAL and pcu, and the total number of natural vehicles and pcu. By combining the judgment results of different correlation levels (highly positive correlation, highly negative correlation, and moderately negative correlation), four typical coupling conditions can be identified.

[0071] Simultaneously, a scatter plot is generated with the horizontal axis representing PCU and the two vertical axes corresponding to ESAL and the total number of natural vehicles, respectively. This technical feature aims to visually demonstrate the relationship between PCU and ESAL, and between PCU and the total number of natural vehicles. The scatter plot clearly presents the distribution trend of the original data, while the fitted curves intuitively reflect the fitting effect and correlation of the linear regression model. The dual vertical axis design allows for the simultaneous display of the trends of ESAL and the total number of natural vehicles with PCU in the same chart, facilitating comparative analysis. This graphical output not only enhances the readability and comprehensibility of the analysis results but also provides decision-makers with intuitive judgment criteria, helping them quickly grasp the changing patterns of road segment loads and vehicle composition.

[0072] By introducing a clear correlation threshold, this application can quantitatively evaluate the goodness of fit of the linear regression model, avoiding the ambiguity of subjective judgment and making the analysis of the changing patterns of heavy load and vehicle composition more objective and accurate. Based on this, according to the correlation coefficients of the two sets of correlation models, four specific coupling conditions are further divided. This elevates the understanding of the complex interaction between road section heavy load and vehicle composition from a single correlation judgment to multi-dimensional, refined pattern recognition. Simultaneously, the generated scatter plot visually displays the relationship between PCU, ESAL, and the total number of natural vehicles, making the complex statistical analysis results easier to understand and grasp, greatly improving decision-making efficiency and the relevance of solutions. This comprehensive analysis and visualization method makes the evaluation of the changing patterns of heavy load and vehicle composition on asphalt pavements more comprehensive and in-depth, providing strong support for formulating scientific and reasonable maintenance strategies and heavy load control measures.

[0073] This application further proposes that in step S5, the traffic load levels are divided into three levels: medium, heavy, and extremely heavy, based on the road segment ESAL, and the maintenance priority is determined in combination with the coupled working conditions: road segments with extremely heavy loads and negative correlation coefficients are designated as first-level priority maintenance road segments; road segments with heavy loads and positive correlation between the two heights are designated as second-level routine maintenance road segments; and road segments with medium loads and stable traffic flow structure are designated as third-level preventive maintenance road segments; and dynamic weighing enforcement, time-limited passage for trucks, and graded heavy load control schemes for overload diversion are output simultaneously.

[0074] Specifically, ESAL (Equivalent Standard Axle Load Cumulative Actions) is a key indicator for measuring the cumulative effect of traffic load on a road surface. By setting specific ESAL thresholds, the traffic load intensity of a road segment can be classified into three levels: medium, heavy, and extremely heavy. For example, based on industry standards or empirical data, road segments with ESAL values ​​below a certain threshold can be classified as medium loads, those between two thresholds as heavy loads, and those above the highest threshold as extremely heavy loads. This classification method provides a basic load intensity assessment for subsequent maintenance decisions. Coupled conditions refer to the comprehensive judgment result of the changes in road segment heavy load and vehicle composition obtained through correlation analysis of ESAL and PCU, and the total number of natural vehicles and PCU. Combining traffic load levels with coupled conditions allows for a more comprehensive assessment of road segment maintenance needs. For example, even for heavy load road segments, the coupling relationship between heavy load and traffic volume changes differs, and the urgency and strategies for maintenance should also differ. This combination makes the classification of maintenance priorities more scientific and refined.

[0075] When a road segment is subjected to extremely heavy loads, it indicates that its pavement structure is under tremendous stress. If a negative correlation coefficient is also present (e.g., a high negative correlation between ESAL and pcu), this may mean that the pavement's heavy load effect decreases with increased traffic volume, or that the vehicle composition has changed in a way that is unfavorable to the pavement's load-bearing capacity. This anomaly often indicates that there may be deep-seated problems or structural damage to the pavement, requiring immediate intervention. Therefore, it should be designated as a first-priority maintenance segment to ensure the highest priority for resource allocation and emergency treatment. For road segments subjected to heavy loads, if ESAL and pcu, as well as the total number of natural vehicles and pcu, are both highly positively correlated, it indicates that both the pavement's heavy load effect and the total number of vehicles increase significantly with increasing traffic volume. This is a typical and predictable pattern of load increase due to traffic volume growth. Although such road segments have large loads, the change pattern is relatively clear, and they can be included in the routine maintenance plan and managed as second-priority maintenance segments. When the traffic load on a road segment is at a moderate level, the pressure on the pavement structure is relatively small. If the traffic flow structure remains stable, meaning the changes in heavy loads and vehicle composition are not drastic or have a low correlation, it indicates that the road surface condition is relatively good and the trend of change is controllable. For these road sections, preventive maintenance measures are mainly adopted to delay the degradation of road surface performance and extend its service life, and they are classified as Level III preventive maintenance road sections.

[0076] In addition to prioritizing maintenance, this method also generates specific heavy-load control plans based on the load level and coupled operating conditions of road sections. Dynamic weighing enforcement controls heavy loads at the source by monitoring and penalizing overloaded vehicles in real time; time-limited passage for trucks reduces the instantaneous impact on the road surface by restricting the passage of trucks in specific time periods or areas; and overload diversion guides overloaded vehicles to other road sections or requires them to unload, thereby reducing the load on the target road section. These tiered and differentiated control plans provide traffic management departments with direct and actionable decision-making basis, enabling precise management of heavy-load vehicles.

[0077] This application further proposes a novel method for evaluating the changing patterns of heavy loads and vehicle composition on asphalt pavements, which includes step S6, a longitudinal comparison across time series. This step aims to extend the assessment from a single time point to a dynamic analysis across multiple time points. By introducing a time dimension, the understanding of changes in pavement heavy loads and vehicle composition shifts from static to dynamic, thereby revealing their evolutionary patterns. Specifically, this step first retrieves multi-year data on ESAL, PCU, and natural vehicle counts for the road segment. This means establishing a historical database to continuously accumulate and archive road segment ESAL, PCU, and natural vehicle counts for each year or multiple observation periods (e.g., quarterly, annual). This data can be acquired and stored using continuously operating axle load acquisition equipment (e.g., dynamic weighing WIM equipment). The data retrieval process can be automated and efficiently managed through a database interface or data warehouse.

[0078] Based on this, the method updates the regression model and correlation coefficients annually. Given the retrieved multi-year data, the steps of constructing the association model and performing coupled correlation analysis need to be re-executed for each year (or each observation period). This means that new data will be generated each year. and The model and its corresponding coefficient of determination This update process can be periodic, for example, once a year at the end of the year, to ensure the timeliness and accuracy of the model.

[0079] Subsequently, a longitudinal analysis of the evolution of heavy traffic flow patterns is conducted. This involves performing time series analysis on the updated model and correlation coefficients. For example, trend graphs of ESAL, PCU, and natural vehicle count over time can be plotted to observe the regression coefficients. , and correlation coefficient , By analyzing changes over different years, we can identify whether the trend of heavy-load traffic is increasing or decreasing, whether the vehicle composition has changed, and how the relationship between these changes and traffic volume has evolved. This helps predict future trends and provides forward-looking information for decision-making.

[0080] Simultaneously, this method also outputs a comprehensive regional road network assessment report. This report summarizes and presents the results of the longitudinal analysis, aiming to provide decision-makers with comprehensive and systematic information. The report includes load class proportions, which, based on defined load classes (such as medium, heavy, and extra heavy), statistically analyzes the proportion of road segments with different load classes within the region and their changes over time, for example, showing whether the proportion of extra heavy load road segments increases or decreases year by year. The report also includes the overload ESAL contribution rate, which statistically analyzes the contribution ratio of different vehicle types or different axle load ranges to the total ESAL and its changes over time, helping to identify major overload sources and their evolution. Furthermore, the report provides maintenance funding allocation calculation data. Based on the road segment's load class, heavy load evolution patterns, coupled working conditions, and maintenance priorities, combined with basic attributes such as pavement structure, service life, and damage history, it estimates the maintenance funds required for different road segments and provides recommendations for fund allocation. This may involve establishing a fund allocation model that considers various factors such as road segment importance, damage severity, and traffic volume. Ultimately, this comprehensive assessment report supports road reconstruction and expansion as well as structural optimization design. Its data and analysis results provide a scientific basis for road management departments to conduct long-term road reconstruction and expansion planning, structural design optimization, and the application of new materials. For example, if the heavy load trend on a certain road section continues to rise, it may be necessary to consider a stronger road structure design or to carry out reconstruction and expansion ahead of schedule.

[0081] By introducing a cross-time series longitudinal comparison step, this application can retrieve multi-year data on ESAL, PCU, and natural vehicle count for road segments, and update the regression model and correlation coefficients annually. This dynamic analysis method transforms the understanding of the evolution of heavy traffic flow patterns from a static assessment at a single point in time to a dynamic insight into continuous time series. By observing the changes in these key indicators and their interrelationships over time, the growth trend of road heavy load, the direction of changes in vehicle composition structure, and the potential impact of these changes on road performance can be accurately identified. Furthermore, by outputting a comprehensive regional road network assessment report that includes load level proportions, overload ESAL contribution rates, and maintenance fund allocation calculation data, this application provides road management departments with a comprehensive, systematic, and forward-looking decision-making basis. This not only supports more accurate and efficient allocation of maintenance funds, ensuring that limited resources are invested in the most needed road segments, but also provides scientific support for road reconstruction and expansion planning and structural optimization design, thereby effectively extending road service life, improving the overall service level of the road network, and reducing long-term operation and maintenance costs.

[0082] This application further proposes to clearly define the types of vehicles involved (2 to 11), specifically covering two-axle small motor vehicles to eleven-axle heavy semi-trailers and container trucks, and excluding non-motorized vehicles and temporary construction machinery; at the same time, passenger and freight vehicles are distinguished during the data collection process, and the proportion of ESAL of freight vehicles in the total equivalent axle load is counted separately.

[0083] This technical solution aims to clarify the scope of vehicles involved in evaluating the changes in heavy loads and vehicle composition on asphalt pavements. By limiting the analysis to vehicles of types 2 to 11, covering motor vehicles that significantly load the pavement structure, from two-axle small motor vehicles to eleven-axle heavy semi-trailers and container trucks, the effectiveness and relevance of data collection are ensured. Simultaneously, non-motorized vehicles and temporary construction machinery are explicitly excluded, avoiding interference from loads with minimal or occasional impact on the long-term performance of the pavement, thereby improving data processing efficiency and the accuracy of evaluation results. In practice, dynamic weighing WIM equipment can be used to identify the number and type of axles of vehicles, and vehicle classification algorithms can be used for further filtering to ensure that only data from vehicles meeting the criteria are included in subsequent analysis.

[0084] Furthermore, this technical solution aims to conduct a refined analysis of the load contributions of vehicles of different types. Freight vehicles, typically with higher axle loads and more frequent heavy-load behavior, are a major factor contributing to fatigue damage in asphalt pavement structures. By clearly distinguishing between passenger and freight vehicles during data acquisition and processing, and separately calculating the proportion of the equivalent standard axle load cumulative action number (ESAL) of freight vehicles to the total ESAL, the source and impact of heavy loads can be revealed more deeply. This helps to accurately identify the main driving factors of pavement damage, providing data support for developing targeted heavy-load control measures and maintenance strategies. For example, in the data processing module, logical judgments can be made based on vehicle type classification information (such as buses, trucks, semi-trailers, etc.) to extract freight vehicle data separately, calculate its ESAL, and then compare it with the total ESAL of all vehicles to determine its contribution rate.

[0085] By precisely defining vehicle types 2 to 11 and excluding non-motorized vehicles and temporary construction machinery, this method ensures that the collected and analyzed traffic load data more accurately focuses on vehicle types that significantly impact asphalt pavement structure. This avoids interference from irrelevant data and improves the accuracy of equivalent axle load calculation. Furthermore, by distinguishing between passenger and freight vehicles during data collection and separately calculating the proportion of freight vehicle ESAL (Equivalent Axle Load) in the total equivalent axle load, this method can deeply analyze the actual contribution of heavy-load vehicles to pavement damage and accurately identify the main sources of heavy loads. This not only enhances the refinement of the evaluation of pavement heavy load and vehicle composition changes but also provides a solid data foundation for developing more targeted maintenance strategies and heavy load control plans, making decision-making more scientific and effective.

[0086] In some of the embodiments described above in this application, a method for evaluating the variation law of heavy load and vehicle composition of asphalt pavement is proposed. This method involves multiple steps, including data acquisition, equivalent axle load calculation, traffic volume conversion, model construction, and coupling correlation analysis. However, in practical applications, to achieve continuous monitoring and dynamic evaluation of multiple road sections in a regional road network and to output maintenance and heavy load control decision-making schemes in a timely manner, a large amount of manpower is required for data processing, calculation, and analysis. Furthermore, this method is easily affected by human factors, leading to low efficiency and delayed response, making it difficult to meet the real-time and accuracy requirements of modern road maintenance management.

[0087] In response, this application further proposes to integrate the entire method into a pavement load intelligent analysis terminal. This terminal has built-in modular functions for WIM data docking, ESAL batch calculation, pcu conversion, regression fitting, and maintenance decision output, realizing fully automated calculation and report export.

[0088] Specifically, the entire method is integrated into a road load intelligent analysis terminal. This means that all steps of the aforementioned evaluation method, from raw data acquisition to the output of the final decision scheme, are integrated into a dedicated, intelligent hardware and software system. This terminal can be an industrial-grade computer system with sufficient computing power and storage space, or an embedded device, designed to automate the management and execution of complex analysis processes. The terminal has a built-in WIM data interface, enabling data communication and interaction with dynamic weighing WIM equipment. Specifically, the terminal is equipped with corresponding communication interfaces and data parsing modules, capable of automatically receiving, identifying, and processing real-time or near-real-time vehicle axle load data and vehicle model raw data from WIM equipment, ensuring the accuracy and timeliness of data transmission. The ESAL batch calculation function enables the terminal to efficiently calculate the ESAL (Equivalent Standard Axle Load Accumulated Action Times). This module incorporates the calculation formula for the equivalent design axle load conversion factor and the ESAL accumulation formula, enabling automated processing of large-scale, continuous vehicle axle load data, quickly obtaining the ESAL value within a specified statistical period, and allowing for separate calculation of the ESAL contribution rate of overloaded vehicles as needed. The PCU conversion function enables the terminal to convert the collected natural numbers of various types of vehicles into Cumulative Equivalent Traffic Volume (PCU) based on preset traffic survey vehicle conversion factors. This module automatically converts different vehicle types (such as small and medium-sized passenger cars, large passenger cars, and various types of trucks) according to their corresponding conversion factors, ensuring the accuracy and standardization of PCU calculation. The regression fitting function enables the terminal to automatically construct linear regression models. This module can automatically perform linear regression analysis with PCU as the independent variable and ESAL and the total number of natural vehicles as dependent variables, constructing a PCU model. and Two sets of correlation models were used, and the coefficients of determination were calculated simultaneously. and This provides a basis for subsequent correlation determination. The modular function of maintenance decision output enables the terminal to automatically generate maintenance and heavy load control decision schemes based on the analysis results. This functional module, based on the results of coupled correlation analysis and traffic load level classification standards, has built-in decision logic and rules, which can intelligently determine the maintenance priority of road segments (such as first-level priority maintenance, second-level routine maintenance, and third-level preventive maintenance), and output specific heavy load control measures (such as dynamic weighing enforcement, time-limited passage for trucks, and overload diversion). Its modular design facilitates updates and expansions according to actual needs. Finally, it achieves fully automated calculation and report export, which means that from WIM data access, ESAL and pcu calculation, regression model fitting, coupled correlation analysis, to the generation of the final maintenance decision scheme, the entire process can be completed automatically within the terminal without manual intervention. At the same time, the terminal also provides a report export function, which can output the analysis results, decision suggestions, and related charts (such as scatter plots) in a standardized format, making it easy for users to view, archive, and share.

[0089] The following example will provide a more detailed explanation of the above technical solution:

[0090] A regional road network management department, A, is responsible for managing multiple asphalt road sections. The department faces the challenge that, with the continuous increase in traffic volume and the rising proportion of heavy-duty vehicles, existing methods cannot simultaneously assess changes in the composition of different vehicle types, overloading situations, and the impact of these factors on the road surface load (ESAL), leading to significant errors in developing maintenance and heavy-duty control strategies. To address this issue, management department A has decided to adopt this method.

[0091] First, in step S1, management department A deployed dynamic weighing (WIM) devices on multiple asphalt road sections within its jurisdiction. These devices continuously collected axle load data and raw vehicle type data for vehicles of types 2 to 11 during the annual observation period. The collected raw data included detailed information such as the number of vehicles, secondary vehicle type classification (e.g., passenger cars, large trucks, extra-large trucks), axle type, axle load, and axle load range distribution (including empty, standard axle load, first-level overload, second-level overload, and severe overload). Simultaneously, the collected data was linked and stored with basic attributes of the corresponding road section, such as pavement structure, service life, and history of road defects. Unlike traditional methods that only count total traffic volume or the number of trucks, this method uses WIM devices to obtain more refined axle load and vehicle type classification data and can distinguish overloading situations, laying the foundation for accurate subsequent analysis.

[0092] Next, in step S2, management department A uses the collected data to calculate the cumulative number of equivalent standard axle load applications (ESAL) within the statistical period based on the equivalent design axle load conversion factors for various types of vehicles. Specifically, for For vehicles of this type, the equivalent design axle load conversion factor Through formula Calculation, where for Equivalent design axle load conversion factor for vehicle class; for Class of vehicles Average number of axes for each type of shaft; for Class of vehicles Type of shaft in the first Equivalent design axle load conversion factor for axle load range; for Class of vehicles Type of shaft in the first The axle load distribution coefficient for the axle load range. ESAL is determined by the formula... The calculation yielded, where Within the statistical period The total number of vehicles in this category. This method specifically calculates the ESAL contribution rate of overloaded vehicles, which differs from existing methods that use empirical coefficients or simplified models to calculate ESAL, ignoring the effects of different axle types, axle load ranges, and overloading conditions. This significantly improves the accuracy of ESAL calculation.

[0093] Subsequently, in step S3, management department A calculates the cumulative equivalent traffic volume (pcu) of all vehicles of types 2 to 11 within the statistical period based on the preset traffic conversion factor standards. For example, the conversion factor for small and medium-sized passenger cars and small trucks is 1; the conversion factor for large passenger cars and medium-sized trucks is 1.5; the conversion factor for large trucks is 3; and the conversion factor for extra-large trucks and container trucks is 4. If special engineering vehicles exist in the area, additional conversion factors are added to complete the pcu conversion. This detailed classification of vehicle types and corresponding conversion factors makes the pcu more representative than traditional methods that use rough classifications or uniform coefficients.

[0094] In step S4, before constructing the model, management department A first preprocesses the collected data, including removing outlier data from abnormal equipment, supplementing missing detection data, and standardizing the statistical period length for each road segment. Then, using pcu as the independent variable, and ESAL and the total number of natural vehicles (… Using ( ) as the dependent variable, two sets of correlation models were constructed simultaneously using linear regression. For each road segment, the model was independently fitted to obtain: and At the same time, the coefficients of determination for these two sets of models were calculated separately. and This serves as the basis for subsequent correlation determination. Unlike existing methods that typically analyze ESAL or pcu independently, this method constructs two sets of regression models simultaneously, providing a quantitative foundation for subsequent coupled correlation analysis.

[0095] Finally, in step S5, management department A calculates the coefficient of determination based on the two sets of models. and Coupling correlation analysis was conducted. The correlation threshold was set as follows: Highly correlated The correlation is moderate. A low correlation indicates a low correlation, and a negative value indicates a negative correlation. Based on the two sets of correlation coefficients, the system classifies road segments into four coupled load conditions. For example, there is a dual-high positive correlation (ESAL increases significantly with increasing pcu, and the number of natural vehicles also increases significantly with increasing pcu), or a high negative correlation between ESAL and pcu and a high positive correlation between the number of natural vehicles and pcu (when pcu increases, the heavy load effect decreases, but the total number of vehicles increases, which may mean a shift towards lighter vehicle structures). The system generates a scatter plot with pcu on the horizontal axis and ESAL and the number of natural vehicles on the vertical axes, respectively, to visually display the analysis results. Based on the ESAL value of the road segment, the system classifies it into three traffic load levels: medium, heavy, and extremely heavy. Based on coupled operating conditions, the system prioritizes maintenance: for example, road sections with extremely heavy loads and negative correlation coefficients (potentially indicating a decrease in heavy-load vehicles but an increase in total traffic volume, or improved efficiency of heavy-load vehicles) are classified as Level 1 priority maintenance sections; road sections with heavy loads and positive correlation coefficients (significant simultaneous increases in both heavy loads and total traffic volume) are classified as Level 2 routine maintenance sections; and road sections with moderate loads and stable traffic flow structures are classified as Level 3 preventive maintenance sections. Simultaneously, the system outputs specific heavy-load control decision-making schemes, such as dynamic weighing enforcement, time-limited passage for trucks, and overload diversion and classification. Unlike existing maintenance decisions, which often rely on single indicators and lack a deep understanding of the complex coupling relationship between heavy loads and vehicle composition, this method, through coupled correlation analysis and multi-dimensional load level classification, can more comprehensively assess road section conditions and provide more targeted maintenance and control solutions.

[0096] Furthermore, this method includes a cross-time-series longitudinal comparison in step S6. Management department A retrieves multi-year data on ESAL, PCU, and natural vehicle count for the road segment. The system updates the regression model and correlation coefficients annually, conducting longitudinal analysis to reveal the evolution of heavy traffic patterns. Finally, the system outputs a comprehensive regional road network assessment report, which includes load level proportions, ESAL contribution rate of overload, and maintenance fund allocation calculation data, providing data support for road reconstruction and expansion and structural optimization design. This enables management departments to dynamically adjust strategies and achieve longer-term planning, compensating for the shortcomings of traditional methods in capturing long-term trends.

[0097] Throughout the implementation process, vehicle types 2 to 11 covered everything from two-axle small motor vehicles to eleven-axle heavy semi-trailers and container trucks, excluding non-motorized vehicles and temporary construction machinery. The data collection process differentiated between passenger and freight vehicles, and separately calculated the proportion of ESAL (Equivalent Axle Load) of freight vehicles in the total equivalent axle load to more accurately assess the impact of heavy freight loads. The entire methodology was integrated into a pavement load intelligent analysis terminal, which incorporates modular functions such as WIM data integration, batch ESAL calculation, PCU conversion, regression fitting, and maintenance decision output, achieving fully automated calculation and report export, thus improving efficiency and practicality.

[0098] Using the above methods, regional road network management department A can clearly understand the correlation between heavy loads and vehicle composition on various road sections within its network, identify road sections with high heavy load risks and abnormal vehicle structure changes, and formulate scientific and precise maintenance strategies and heavy load control measures accordingly. This effectively extends the service life of asphalt pavements, reduces maintenance costs, and improves the safety and stability of road network traffic. This solves the technical problems of traditional methods that cannot simultaneously evaluate vehicle composition and overloading, and the errors caused by directly substituting ESAL and PCU without understanding their correlation.

[0099] Example 1: like Figure 1 As shown, this invention provides a new method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement, comprising the following steps:

[0100] 1) Select road segments and collect data: Select multiple road segments in the region. Heavy traffic road segments of asphalt roads in the same region can be selected as the analysis objects. Collect detailed data of these road segments from 2022 to 2024. Each set of data includes the number of natural vehicles, axle type, axle load of each axle, and vehicle type for the same observation period.

[0101] 2) Calculate the equivalent standard axle load cumulative number of times (ESAL): Calculate the equivalent design axle load conversion factor for different vehicle models, multiply it by the corresponding number of natural vehicles, and obtain the ESAL data. The results are shown in Table 1.

[0102] Table 1. ESAL, PCU, and Natural Vehicle Count at Site 1, 2022-2024

[0103]

[0104] 3) Calculate the cumulative equivalent traffic volume of all vehicles of type 2 to 11 in each statistical period based on the traffic survey vehicle conversion factor.

[0105] 4) Simultaneously construct correlation models between ESAL and PCU, and between natural vehicle count and PCU: Using PCU as the independent variable and cumulative ESAL and natural vehicle count as dependent variables, fit the data samples to establish quantitative relationship models between PCU and cumulative standard axle load, and between PCU and natural vehicle count for each asphalt road. See [link to relevant documentation]. Figures 2-5 .

[0106] 5) Correlation analysis and maintenance decision output: Calculate the correlation coefficient R based on the fitting results. 2 The strength of the correlation between the two factors was determined. The variation patterns of heavy loads and vehicle composition on each road surface were evaluated using the dual correlation coefficients between PCU and cumulative standard axle load, and between PCU and natural vehicle count. As shown in Table 2, based on the road surface damage, recommendations for road segment maintenance priorities and reference schemes for heavy traffic control were derived.

[0107] Table 2. Patterns of Heavy Load and Vehicle Composition under Different Modes

[0108]

[0109] Step 2) calculates the equivalent design axle load conversion factor for different vehicle models using the following formula:

[0110]

[0111] In the formula, for Equivalent design axle load conversion factor for vehicle type. for In the category of vehicles The average number of axes for each type of shaft for In the category of vehicles Type of shaft in Equivalent design axle load conversion factor for axle load range. for In the category of vehicles Type of shaft in Axle load distribution coefficient for the grade axle load range.

[0112] In step 3), based on Table 1, the number of natural vehicles, and vehicle types, the cumulative equivalent traffic volume of all vehicles of types 2 to 11 detected at the axle load stations on each road within a certain period is obtained. The detection cycle is in years, as shown in Table 2.

[0113] In step 4), a linear regression method is used to simultaneously construct correlation models between ESAL and PCU, and between natural vehicle count and PCU, as shown in the figure. Figures 2-5 .

[0114] In step 5), based on the bicorrelation coefficients of ESAL and PCU, and natural vehicle count and PCU for 15 road segments, it can be found that the change in ESAL cannot be directly estimated using the change in PCU. Therefore, road design and maintenance should not rely solely on PCU to judge pavement damage caused by axle load, nor should maintenance funds be allocated based on PCU. 60% of road segments showed a high positive correlation, while 40% showed a low to medium correlation, a moderate negative correlation, or even a strong negative correlation. This indicates that traffic flow structure, heavy-load mixing rate, and actual load rate are the three core factors disrupting the correlation.

[0115] As illustrated above with reference to the accompanying drawings, a novel method for evaluating the variation patterns of heavy loads and vehicle composition in asphalt pavements, according to the present invention, has been described by way of example. However, those skilled in the art should understand that various modifications can be made to the analytical method proposed in the present invention without departing from the scope of the invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

[0116] In summary, the analytical method provided in this application has at least the following effects or advantages:

[0117] 1. Intuitively identify the characteristics of load-bearing asphalt pavement. This invention employs a dual correlation analysis combining ESAL and PCU, and natural vehicle count and PCU, to analyze the traffic flow structure, heavy load mixing rate, and actual load rate of asphalt pavement load from two dimensions: heavy load and vehicle type. This helps to understand the true causes of road damage and propose targeted maintenance strategies.

[0118] 2. Improve the accuracy of maintenance fund allocation. This invention analyzes multiple road segments in the same area to identify which road segments can use changes in pcuU to estimate changes in ESAL, and which road segments cannot use only changes in pcuU to estimate changes in ESAL, thereby improving the accuracy of allocating maintenance funds based on traffic volume.

[0119] 3. Evaluate the axle load variation pattern of the same road at different times. This invention can also realize the continuous variation pattern of road traffic axle load through a combination analysis of the bicorrelation between ESAL and PCU, and between natural vehicle number and PCU.

[0120] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A new method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement, characterized in that: Includes the following steps: S1. Collect continuous axle load data and original vehicle type data of vehicles of types 2 to 11 within the same observation period on asphalt road sections with multiple axle load collection devices deployed in the collection area. S2. Based on the equivalent design axle load conversion factor for various types of vehicles, calculate the cumulative number of equivalent standard axle load applications (ESAL) within the statistical period. S3. Based on the traffic survey vehicle conversion factor, calculate the cumulative equivalent traffic volume (pcu) of all vehicles of types 2 to 11 within the statistical period; S4. Using pcu as the independent variable and ESAL and total number of natural vehicles as dependent variables, two sets of correlation models are constructed simultaneously using linear regression. S5. Based on the correlation coefficients of the two sets of models, conduct coupled correlation analysis to evaluate the changing patterns of heavy load and vehicle composition on road sections, and output maintenance and heavy load control decision schemes.

2. The new method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement according to claim 1, characterized in that: In step S1, the axle load acquisition device is a dynamic weighing WIM device; the acquired raw data includes the number of natural vehicles, vehicle type secondary classification, axle type, axle load, and axle load range distribution; the observation period can be selected as daily, weekly, quarterly, or annual, with the annual period being preferred for the overall evaluation of the regional road network; the acquired data is synchronously bound to the storage of road section pavement structure, service life, and historical defects.

3. A novel method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement according to claim 1, characterized in that: The formula for calculating the equivalent design axle load conversion factor for vehicle type A is: ;in: for Equivalent design axle load conversion factor for vehicle class; for Class of vehicles Average number of axes for each type of shaft; for Class of vehicles Type of shaft in the first Equivalent design axle load conversion factor for axle load range; for Class of vehicles Type of shaft in the first Axle load distribution coefficient for axle load range; ESAL is based on the formula Calculations show that Within the statistical period The total number of vehicles in each category; the axle load range is divided into unloaded, standard axle load, first-level overload, second-level overload, and severe overload ranges, and the ESAL contribution rate of overloaded vehicles is calculated separately.

4. A novel method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement according to claim 1, characterized in that: The conversion factor standards for vehicles in step S3 are as follows: conversion factor 1 for small and medium-sized passenger cars and small trucks; conversion factor 1.5 for large passenger cars and medium-sized trucks; conversion factor 3 for large trucks; conversion factor 4 for extra-large trucks and container trucks; when there are special engineering vehicles in the area, additional conversion factors can be added to complete the PCU conversion.

5. A novel method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement according to claim 1, characterized in that: Step S4 includes data preprocessing sub-steps: removing outlier data from equipment, completing missing detection data, and standardizing the statistical period length for each road segment; independently fitting linear regression models for each road segment to obtain... , Two sets of models, and solve for the definite coefficients respectively. , As a basis for determining relevance.

6. A novel method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement according to claim 1, characterized in that: The correlation determination threshold in step S5: Highly correlated The correlation is moderate. Low correlation is defined as a negative value, and negative values ​​represent negative correlation. Based on the two sets of correlation coefficients, four coupling conditions are defined: dual-high positive correlation, dual-high negative correlation, ESAL-pcu high negative correlation and natural vehicle number-pcu high positive correlation, and ESAL-pcu high positive correlation and natural vehicle number-pcu moderate negative correlation. A scatter plot is generated with pcu as the horizontal axis and ESAL and natural vehicle number as the two vertical axes, respectively.

7. A novel method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement according to claim 1, characterized in that: In step S5, the traffic load levels are divided into three levels: medium, heavy, and extremely heavy according to the ESAL of the road segment, and the maintenance priority is determined in combination with the coupled working conditions: road segments with extremely heavy load and negative correlation coefficient are first-level priority maintenance road segments; road segments with heavy load and positive correlation between the two heights are second-level routine maintenance road segments. Road sections with medium loads and stable traffic flow are designated as Level III preventive maintenance sections; dynamic weighing enforcement, time-limited passage for trucks, and graded heavy load control schemes for overload diversion are implemented simultaneously.

8. A novel method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement according to claim 1, characterized in that: It also includes step S6, cross-time series longitudinal comparison step: retrieve multi-year ESAL, pcu, and natural vehicle count data of road segments, update the regression model and correlation coefficients year by year, and longitudinally analyze the evolution of traffic flow heavy load; Output a comprehensive assessment report of the regional road network, including load level proportions, ESAL contribution rate of overload, and maintenance fund allocation calculation data, to support road reconstruction and expansion and structural optimization design.

9. A novel method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement according to claim 1, characterized in that: The types of vehicles 2 to 11 cover two-axle small motor vehicles to eleven-axle heavy semi-trailers and container trucks, excluding non-motorized vehicles and temporary construction machinery; the data collection process distinguishes between passenger and freight vehicles, and separately counts the proportion of ESAL (Equivalent Axle Load) of freight vehicles in the total equivalent axle load.

10. A novel method for evaluating the variation law of heavy load and vehicle composition in asphalt pavement according to claim 1, characterized in that: The entire method is integrated into the intelligent pavement load analysis terminal. The terminal has built-in modular functions such as WIM data docking, ESAL batch calculation, pcu conversion, regression fitting, and maintenance decision output, realizing fully automated calculation and report export.

Images