Road carbon emission evaluation method and system

Abstract

The invention discloses a road carbon emission evaluation method and system, and relates to the technical field of carbon emission evaluation, and the method comprises the following steps: S1, collecting the traffic flow data and environment data of a target road section in real time through a roadside sensing array; s2, constructing a road body carbon emission calculation model; s3, establishing a maintenance activity carbon emission compensation model; s4, constructing a driving behavior carbon emission correction model; s5, establishing a climate accumulation effect correlation model; and S6, generating a dynamic emission coefficient matrix, and establishing a carbon emission coefficient dynamic correction mechanism of different generations of vehicles. According to the method, the road body carbon emission calculation model is constructed, the unit mileage road maintenance carbon emission factor is dynamically generated, oxygenolysis carbon emission of the asphalt concrete material is considered, the influence of the road surface flatness on the vehicle rolling resistance is also taken into account, and therefore comprehensive and accurate accounting of the road carbon emission is achieved, and the calculation efficiency is improved. And the accuracy of the evaluation result is obviously improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of carbon emission assessment, in particular to a highway carbon emission assessment method and system. BACKGROUND

[0002] The precise quantification of highway traffic carbon emissions has become a key technical bottleneck for green transportation development. The existing carbon emission assessment methods generally lack key elements, resulting in systematic deviations between the assessment results and the actual emissions, which are manifested in the following aspects: (1) Road carbon emission accounting is missing: traditional methods only focus on vehicle emissions, do not account for the oxidation and decomposition of asphalt concrete materials during the service period, and ignore the increase in vehicle rolling resistance caused by road roughness degradation. Studies have shown that CO2 emissions from asphalt oxidation can account for 12-18% of the total life cycle carbon emissions of the road, and an increase in the International Roughness Index (IRI) of 1 m / km will result in a 2.3% increase in fuel consumption.

[0003] (2) Dynamic impact modeling of maintenance activities is insufficient: existing technologies use static path assignment models to compensate for carbon emissions during construction road closures, without dynamically calculating the incremental bypass path based on real-time traffic flow characteristics. Actual monitoring data shows that the carbon emissions of bypass paths during construction can increase by 27-35% compared to normal paths, but the existing model error is as high as ±22%.

[0004] (3) Simplification of non-linear effects of driving behavior: Microscopic driving operations such as rapid acceleration and rapid braking are simplified as linear correction coefficients, which cannot reflect the non-linear relationship between acceleration and energy consumption in actual working conditions. Experiments have shown that frequent rapid acceleration (>0.3 m / s²) can increase single-car carbon emissions by 18-25%, but the average treatment results in a correction error of more than ±15%.

[0005] (4) Fragmented verification system: existing assessment methods rely on a single data source for verification, lack a multi-modal data fusion mechanism, and cannot identify local model failure problems. Studies have shown that a verification system that does not fuse laser scanning and thermal imaging data can have a 30% error in predicting road oxidation rates, and the abnormal condition detection rate can be as high as 45%. SUMMARY

[0006] The present application aims to provide a highway carbon emission assessment method and system to solve the problems raised in the background art.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solution: a highway carbon emission assessment method and system, comprising the following steps: S1, collecting traffic flow data and environmental data of the target road section in real time through a roadside perception array, the traffic flow data including vehicle density and vehicle type distribution, and the environmental data including temperature, humidity and sunshine intensity; S2, constructing a road ontology carbon emission calculation model, dynamically generating a unit mileage road maintenance carbon emission factor based on the oxidation decomposition characteristics of the pavement material and the pavement state parameters; S3, establishing a maintenance activity carbon emission compensation model, when a road construction event is detected, dynamically modeling and compensating the additional carbon emissions generated by the detour path according to the real-time traffic flow distribution characteristics; S4, constructing a driving behavior carbon emission correction model, acquiring vehicle micro-operation data through a preset driving behavior parameter acquisition device, and generating a nonlinear carbon emission correction coefficient based on a machine learning algorithm; S5, establishing a climate cumulative effect correlation model, calculating the cumulative effective temperature equivalent value according to historical climate data, and simultaneously establishing a dynamic correlation between the tire performance degradation function and the air conditioner energy consumption growth; S6, generating a dynamic emission coefficient matrix, correlating the corresponding emission standards based on the vehicle year of manufacture, and establishing a dynamic correction mechanism for the carbon emission coefficients of different generations of vehicles.

[0008] Preferably, in step S1: The roadside perception array is composed of a microwave radar (working frequency band 76-77GHz), a video acquisition device (resolution ≥1280×960) and a laser scanning device (scanning frequency ≥50Hz), and a multi-source data space-time alignment is performed through an edge computing gateway to generate a timestamp-synchronized traffic flow data matrix; The environmental data fusion adopts a hierarchical calibration mechanism: the ground meteorological station data and the MODIS land surface temperature product are fused by the following formula: Wherein, α is the satellite data weight coefficient, β is the meteorological station data weight coefficient, γ is the elevation correction coefficient, Δh is the elevation difference between the road section and the meteorological station (unit: meters).

[0009] Preferably, in step S2: The oxidation rate function model of asphalt concrete material satisfies the formula: Wherein, is the oxidation rate at time t, is the initial oxidation rate (0.15-0.25), k is the temperature influence factor (0.021-0.035 / ℃), TI is the temperature time integral value (℃·h); The rolling resistance correction coefficient μ satisfies: Wherein, Cf is the reference rolling resistance coefficient, IRI is the International Roughness Index (m / km), β1=0.15, and β2=0.08.

[0010] Preferably, the detection method of the construction road closure event in S3 is: The cone bucket placement form is identified through the roadside perception array, and when the distance between the cone buckets is detected to be ≤20 m for 5 consecutive frames, it is determined to be a construction zone. Synchronous access to the traffic management platform to obtain construction license record information for double verification; Dynamically update the road closure period data to the high-precision map service interface; The incremental calculation of the detour path uses an improved Dijkstra algorithm, and the weight function includes: where L is the path length (km), T is the travel time (min), is the unit mileage carbon emission factor (kg / km), and λ1-λ3 are dynamic weight coefficients (∑λ=1); The detour carbon emission compensation value is calculated by the following formula: where is the i-type vehicle flow (vehicles / h), and D is the road closure duration (h) Preferably, step S4 specifically includes the following steps: S41, driving behavior data acquisition and preprocessing, real-time acquisition of driving operation multi-source heterogeneous data through a vehicle-mounted data interface, high-precision acquisition and cleaning of acceleration, braking, and steering signals, establishment of a spatiotemporally aligned driving behavior time series dataset, and identification of typical abnormal operation modes through an event marking mechanism; S42, feature engineering and model construction, based on multi-dimensional driving behavior feature extraction, construction of a feature vector that fuses time-frequency domain characteristics and driving environment coupling relationships, adoption of a hierarchical model architecture combining pre-training and online learning to realize nonlinear relationship modeling of driving behavior and carbon emissions, and deployment of a dynamic tuning mechanism to maintain model adaptability; S43, dynamic application and optimization of correction coefficients, real-time generation and scene adaptive compensation of correction coefficients through an embedded inference engine, establishment of a closed-loop optimization system combined with multi-source verification data, and formation of a sustainable optimization system including abnormality early warning, behavior feedback, and model iteration.

[0011] Preferably, in step S5: The cumulative effective temperature equivalent value ETTI is calculated to satisfy: where is the road surface temperature at time t (℃), is the tire softening threshold temperature (45℃), and Δt is the monitoring interval (0.5 h); Air conditioning load dynamic correlation model output air conditioning energy consumption correction coefficient: Wherein γ = 0.32, δ = 0.78 Preferably, step S6 specifically comprises the following steps: S61, vehicle generation data correlation and standardization, establish multi-source data acquisition and standard matching mechanism, accurately correlate emission standard version through vehicle characteristic information, construct structured database containing generation attribute, and implement data quality verification process; S62, multi-dimensional dynamic matrix construction and verification, based on vehicle technology generation evolution law, construct dynamic emission coefficient matrix integrating time, space and vehicle type dimensions, embed transition period coefficient calculation method, and establish multi-level verification system to guarantee data reliability; S63, real-time collaborative correction and closed-loop optimization, deploy flexible correction framework of multi-source real-time data collaboration, integrate vehicle degradation compensation and use feedback mechanism, form dynamic evolution capability and closed-loop verification link of emission coefficient.

[0012] Preferably, S7, a multi-modal data fusion verification mechanism is established, specifically comprising: S71, synchronously collect tail gas telemetry data and model calculation results within the evaluation period, realize dynamic error compensation through Kalman filtering algorithm; trigger model parameter recalibration for road sections with deviation exceeding ± 15% for 3 consecutive times; establish carbon emission intensity confidence interval, and mark data points exceeding the interval range as abnormal working conditions; S72, deploy cross-validation module: compare the actual road surface texture depth measured by laser scanner with the predicted value of oxidation rate model, and trigger material parameter update when the difference is > 20%; obtain tire ground temperature distribution through unmanned aerial vehicle thermal imager to verify the local adaptability of climate cumulative effect model; S73, generate verification report and sign on chain: output auditable daily log containing original data hash value, model version and average error rate per day; automatically trigger data quality anomaly warning through smart contract.

[0013] A highway carbon emission evaluation system, comprising: A roadside perception array module composed of microwave radar, video acquisition device and laser scanning equipment, configured on both sides of the target road section, for real-time collection of traffic flow data including vehicle density and vehicle type distribution, and state data of road surface temperature and texture depth; An edge computing gateway connected to the roadside perception array, with a multi-source data space-time alignment module built-in, for time stamp synchronization processing of traffic flow data and generation of traffic flow matrix, while fusing ground meteorological station and satellite remote sensing data to calibrate environmental parameters; A dynamic modeling server for dynamic modeling of carbon emissions; The hierarchical calibration module of the edge computing gateway fuses satellite temperature data, weather station temperature data and elevation correction values by using dynamic weight coefficients to generate calibrated environmental parameters.

[0014] Preferably, the dynamic modeling server comprises: The road ontology computing engine dynamically calculates the road maintenance carbon emission factor based on the oxidation rate characteristics of asphalt concrete materials and the pavement smoothness index; The repair compensation processor is configured to detect construction road closure events and call the path planning algorithm to calculate the carbon emission increment generated by the detour path; The driving behavior analysis unit processes vehicle acceleration and braking micro-operation data through a machine learning model to output a nonlinear carbon emission correction coefficient; The climate effect modeling library establishes a dynamic correlation model between tire performance degradation and air conditioner energy consumption growth based on historical climate data; The intergenerational compensation database associates vehicle manufacturing year and emission standards to generate a dynamic emission coefficient matrix; The verification terminal integrates tail gas telemetry equipment, road scanners and thermal imagers to compare measured data with model calculation results and achieve auditable storage of evaluation logs through a blockchain node.

[0015] Technical effects and advantages of the present application: The present application dynamically generates a unit mileage road maintenance carbon emission factor by constructing a road ontology carbon emission calculation model, not only considering the oxidation and decomposition carbon emission of asphalt concrete materials, but also taking into account the influence of pavement smoothness on vehicle rolling resistance, thereby achieving comprehensive and accurate accounting of highway carbon emissions, significantly improving the accuracy of the evaluation results; The present application establishes a repair activity carbon emission compensation model to dynamically calculate the additional carbon emissions generated by the detour path using real-time traffic flow data, effectively solving the evaluation error problem caused by the static path allocation model in the prior art, making the carbon emission evaluation more close to the actual working condition, and improving the accuracy and practicality of the evaluation; The present application constructs a driving behavior carbon emission correction model and generates a nonlinear carbon emission correction coefficient using a machine learning algorithm, which can finely depict the nonlinear influence of driving behavior on carbon emission, significantly improving the accuracy of the evaluation results and providing strong support for developing more scientific emission reduction strategies; The present application establishes a multi-modal data fusion verification mechanism during the evaluation process, which synchronously collects multi-source information such as tail gas telemetry data, laser scanning data and thermal imaging data to achieve comprehensive verification and dynamic error compensation of the model calculation results. This mechanism effectively improves the reliability and stability of the evaluation results, providing a solid guarantee for the accurate quantification of highway carbon emissions. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 This is a schematic diagram of the evaluation method framework of the present invention. Detailed Implementation

[0017] 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.

[0018] This invention provides, for example Figure 1 The method for assessing highway carbon emissions includes the following steps: S1. Real-time collection of traffic flow data and environmental data for the target road segment using a roadside sensing array. Traffic flow data includes vehicle density and vehicle type distribution, while environmental data includes temperature, humidity, and solar radiation intensity. S2. Construction of a road carbon emission calculation model. Based on the oxidative decomposition characteristics of pavement materials and pavement condition parameters, a dynamic carbon emission factor per unit mileage for road maintenance is generated. S3. Establishment of a carbon emission compensation model for maintenance activities. When road construction events are detected, the additional carbon emissions generated by detour routes are dynamically modeled and compensated based on real-time traffic flow distribution characteristics. S4. Construction of a driving behavior carbon emission correction model. Vehicle micro-operation data is acquired through a preset driving behavior parameter acquisition device, and a nonlinear carbon emission correction coefficient is generated based on a machine learning algorithm. S5. Establishment of a climate cumulative effect correlation model. The cumulative effective temperature equivalent value is calculated based on historical climate data, and a dynamic correlation between tire performance degradation function and air conditioning energy consumption growth is established simultaneously. S6. Generation of a dynamic emission coefficient matrix. Based on the vehicle's manufacturing year and corresponding emission standards, a dynamic correction mechanism for carbon emission coefficients of different generations of vehicles is established.

[0019] In step S1: the roadside sensing array consists of a microwave radar (operating frequency band 76-77GHz), a video acquisition device (resolution ≥1280×960), and a laser scanning device (scanning frequency ≥50Hz). Multi-source data is spatiotemporally aligned via an edge computing gateway to generate a traffic flow data matrix with synchronized timestamps. Environmental data fusion employs a hierarchical calibration mechanism: ground weather station data and MODIS surface temperature products are fused using the following formula: Where α is the satellite data weighting coefficient, β is the meteorological station data weighting coefficient, γ is the elevation correction coefficient, and Δh is the elevation difference between the road segment and the meteorological station. (Unit: meters).

[0020] Preferably, in step S2: The oxidation rate function model for asphalt concrete materials satisfies the following formula: wherein is the oxidation rate at time t, is the initial oxidation rate (0.15-0.25), k is the temperature influence factor (0.021-0.035 / ℃), TI is the temperature time integral value (℃·h); The rolling resistance correction coefficient μ is calculated to satisfy: wherein is the reference rolling resistance coefficient, IRI is the international roughness index (m / km), β1=0.15, β2=0.08.

[0021] Specifically, by constructing a road ontology carbon emission calculation model, a unit mileage road maintenance carbon emission factor is dynamically generated, not only considering the oxidation decomposition carbon emission of asphalt concrete material, but also taking into account the influence of pavement roughness on vehicle rolling resistance, so as to realize the comprehensive and accurate accounting of highway carbon emissions, and significantly improve the accuracy of the evaluation results.

[0022] wherein, the detection method of the construction closure event in S3 is: the road side perception array identifies the cone bucket placement form, when the distance between the cone buckets is ≤20m for 5 consecutive frames, it is determined to be a construction zone; the construction license record information is obtained from the traffic management platform for double verification; the closure period data is dynamically updated to the high-precision map service interface; the improved Dijkstra algorithm is used for incremental calculation of the detour path, and the weight function includes: wherein L is the path length (km), T is the travel time (min), is the unit mileage carbon emission factor (kg / km), λ1-λ3 are dynamic weight coefficients (∑λ=1); The detour carbon emission compensation value is calculated by the following formula: wherein is the i-type vehicle flow (vehicles / h), D is the closure duration (h) Specifically, by establishing a maintenance activity carbon emission compensation model, the additional carbon emissions generated by the detour path are dynamically calculated using real-time traffic flow data, effectively solving the evaluation error problem caused by the static path allocation model in the prior art, making the carbon emission evaluation more close to the actual working condition, and improving the accuracy and practicality of the evaluation.

[0023] In step S4, the following steps are included: S41, driving behavior data collection and preprocessing, real-time acquisition of driving operation multi-source heterogeneous data through a vehicle data interface, high-precision collection and cleaning of acceleration, braking and steering signals, establishment of a driving behavior time series dataset aligned in time and space, and identification of typical abnormal operation modes through an event marking mechanism; S42, feature engineering and model construction, based on multi-dimensional driving behavior feature extraction, construction of a feature vector that fuses time-frequency domain characteristics and driving environment coupling relationship, adoption of a hierarchical model architecture combining pre-training and online learning to realize nonlinear relationship modeling of driving behavior and carbon emissions, and deployment of a dynamic tuning mechanism to maintain model adaptability; S43, dynamic application and optimization of correction coefficients, real-time generation and scene adaptive compensation of correction coefficients through an embedded inference engine, establishment of a closed-loop optimization system combined with multi-source verification data, formation of a sustainable optimization system including abnormal warning, behavior feedback and model iteration, through the construction of a driving behavior carbon emission correction model, the use of machine learning algorithms to generate nonlinear carbon emission correction coefficients, which can finely depict the nonlinear influence of driving behavior on carbon emissions, significantly improving the accuracy of the evaluation results and providing strong support for developing more scientific emission reduction strategies.

[0024] In step S5, the cumulative effective temperature equivalent value ETTI is calculated as follows: Wherein is the road surface temperature (℃) at time t, is the tire softening threshold temperature (45℃), and Δt is the monitoring interval (0.5h); The air conditioning load dynamic correlation model outputs the air conditioning energy consumption correction coefficient: Wherein γ=0.32, δ=0.78 In step S6, the following steps are included: S61, vehicle intergenerational data correlation and standardization, establishment of a multi-source data collection and standard matching mechanism, accurate correlation of emission standard versions through vehicle feature information, construction of a structured database containing intergenerational attributes, and implementation of a data quality verification process; S62, multi-dimensional dynamic matrix construction and verification, based on the evolution law of vehicle technology generations, construction of a dynamic emission coefficient matrix that fuses time, space and vehicle type dimensions, embedding of a transition period coefficient calculation method, and establishment of a multi-level verification system to ensure data reliability; S63, real-time collaborative correction and closed-loop optimization, deployment of an elastic correction framework for multi-source real-time data collaboration, integration of vehicle degradation compensation and usage feedback mechanisms, formation of dynamic evolution capabilities and closed-loop verification links for emission coefficients.

[0025] S7, a multi-modal data fusion verification mechanism is established, specifically comprising: S71, synchronously collecting tail gas telemetry data and model calculation results within an evaluation period, and realizing dynamic error compensation through Kalman filtering algorithm; triggering model parameter re-calibration for road sections with deviation exceeding ±15% for three consecutive times; establishing a carbon emission intensity confidence interval, and marking data points exceeding the interval range as abnormal working conditions; S72, deploying a cross-validation module: comparing the actual road surface texture depth measured by a laser scanner with the predicted value of the oxidation rate model, and triggering material parameter updating when the difference is >20%; obtaining tire ground temperature distribution through a UAV thermal imager to verify the local adaptability of the climate cumulative effect model; S73, generating a verification report and signing a chain: outputting an auditable daily log containing the hash value of the original data, the model version, and the average error rate per day; automatically triggering data quality anomaly warning through a smart contract; a multi-modal data fusion verification mechanism is established during the evaluation process, multi-source information such as tail gas telemetry data, laser scanning data, and thermal imaging data is synchronously collected, and comprehensive verification and dynamic error compensation of model calculation results are realized. This mechanism effectively improves the reliability and stability of the evaluation results, providing a solid guarantee for the accurate quantification of highway carbon emissions.

[0026] A highway carbon emission evaluation system, comprising: a roadside perception array module composed of a microwave radar, a video acquisition device, and a laser scanning device, configured on both sides of a target road section, for real-time collection of traffic flow data including vehicle density and vehicle distribution, and state data of road surface temperature and texture depth; an edge computing gateway connected to the roadside perception array, with a multi-source data space-time alignment module built-in, for timestamp synchronization processing of the traffic flow data and generation of a traffic flow matrix, while fusing ground meteorological station and satellite remote sensing data to calibrate environmental parameters; a dynamic modeling server for dynamic modeling of carbon emissions; a hierarchical calibration module of the edge computing gateway, which adopts dynamic weight coefficients to fuse satellite temperature data, meteorological station temperature data, and elevation correction values to generate calibrated environmental parameters.

[0027] The dynamic modeling server comprises: a road ontology computing engine that dynamically calculates road maintenance carbon emission factors based on the oxidation rate characteristics of asphalt concrete materials and road roughness indicators; a repair compensation processor configured to detect construction road closure events and call a path planning algorithm to calculate the carbon emission increment generated by the detour path; a driving behavior analysis unit that processes vehicle acceleration and braking micro-operation data through a machine learning model to output a non-linear carbon emission correction coefficient; a climate effect modeling library that establishes a dynamic correlation model of tire performance degradation and air conditioner energy consumption growth based on historical climate data; a generation compensation database that associates vehicle manufacturing year and emission standards to generate a dynamic emission coefficient matrix; and a verification terminal integrating tail gas telemetry equipment, a road scanner, and a thermal imager for comparing measured data with model calculation results, and realizing auditable storage of evaluation logs through a blockchain node.

[0028] Finally, it should be noted that the above is only the preferred embodiments of the present application, and is not intended to limit the present application, although the foregoing embodiments of the present application are described in detail, for those skilled in the art, it can still be modified to the technical solutions described in the foregoing embodiments, or equivalent replacement of some of the technical features, any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application, should be included in the protection scope of the present application.

Claims

1. A method for assessing carbon emissions from highways, characterized in that, Includes the following steps: S1. Real-time collection of traffic flow data and environmental data of the target road segment through roadside sensing array. Traffic flow data includes vehicle density and vehicle type distribution, and environmental data includes temperature, humidity and solar radiation intensity. S2. Construct a carbon emission calculation model for the road itself, and dynamically generate a carbon emission factor for road maintenance per unit mileage based on the oxidative decomposition characteristics of road materials and road condition parameters. S3. Establish a carbon emission compensation model for maintenance activities. When a road construction event is detected, dynamically model and compensate for the additional carbon emissions generated by the detour route based on the real-time traffic flow distribution characteristics. S4. Construct a driving behavior carbon emission correction model. Obtain vehicle micro-operation data through a preset driving behavior parameter acquisition device, and generate nonlinear carbon emission correction coefficients based on machine learning algorithms. S5. Establish a climate cumulative effect correlation model, calculate the cumulative effective temperature equivalent value based on historical climate data, and simultaneously establish a dynamic correlation between tire performance degradation function and air conditioning energy consumption growth. S6. Generate a dynamic emission coefficient matrix and establish a dynamic correction mechanism for carbon emission coefficients of different generations of vehicles based on the vehicle's manufacturing year and corresponding emission standards.

2. The method for assessing highway carbon emissions according to claim 1, characterized in that, In step S1: The roadside sensing array consists of microwave radar, video acquisition devices and laser scanning equipment. It performs spatiotemporal alignment of multi-source data through an edge computing gateway to generate a traffic flow data matrix with synchronized timestamps. Environmental data fusion employs a hierarchical calibration mechanism: ground weather station data and MODIS surface temperature products are fused using the following formula: Where α is the satellite data weighting coefficient, β is the meteorological station data weighting coefficient, γ is the elevation correction coefficient, and Δh is the elevation difference between the road section and the meteorological station.

3. A method for assessing highway carbon emissions according to claim 1, characterized in that, In step S2: The oxidation rate function model for asphalt concrete materials satisfies the following formula: in Let be the oxidation rate at time t. Where is the initial oxidation rate, k is the temperature influence factor, and TI is the temperature-time integral value; The calculation of the rolling resistance correction factor μ satisfies: in The reference rolling resistance coefficient is β1 = 0.15, and IRI is the International Roughness Index, with β1 = 0.15 and β2 = 0.

08.

4. A method for assessing highway carbon emissions according to claim 1, characterized in that, The detection method for road closure events during construction in S3 is as follows: The placement pattern of traffic cones is identified by a roadside sensing array. When the spacing between traffic cones is ≤20m for 5 consecutive frames, it is determined to be a construction zone. Simultaneously access the traffic management platform to obtain construction permit filing information for dual verification; Dynamically update road closure time data to the high-precision map service interface; The detour path increment calculation uses a modified Dijkstra algorithm, and the weight function includes: Where L is the path length and T is the travel time. λ1-λ3 is the carbon emission factor per unit mileage, and λ1-λ3 are dynamic weighting coefficients. The carbon emission compensation value for detours is calculated using the following formula: where Let i represent the traffic flow rate and D represent the road closure duration.

5. A method for assessing highway carbon emissions according to claim 1, characterized in that, Step S4 specifically includes the following steps: S41. Driving behavior data acquisition and preprocessing: real-time acquisition of multi-source heterogeneous driving operation data through the vehicle data interface; high-precision acquisition and cleaning of acceleration, braking and steering signals; establishment of a spatiotemporally aligned driving behavior time series dataset; and identification of typical abnormal operation patterns through an event tagging mechanism. S42. Feature Engineering and Model Building: Based on multi-dimensional driving behavior feature extraction, a feature vector that integrates time-frequency domain characteristics and the coupling relationship with the driving environment is constructed. A hierarchical model architecture combining pre-training and online learning is adopted to realize the nonlinear relationship modeling between driving behavior and carbon emissions, and a dynamic tuning mechanism is deployed to maintain model adaptability. S43. Dynamic application and optimization of correction coefficients: Real-time generation of correction coefficients and scene adaptive compensation are achieved through an embedded inference engine. A closed-loop optimization system is established by combining multi-source verification data to form a sustainable optimization system that includes anomaly warning, behavior feedback and model iteration.

6. A method for assessing highway carbon emissions according to claim 1, characterized in that, In step S5: The cumulative effective temperature equivalent (ETTI) value calculation satisfies: in Let t be the road surface temperature. The tire softening threshold temperature is Δt, where Δt is the monitoring interval. The air conditioning load dynamic correlation model outputs the air conditioning energy consumption correction coefficient: Where γ=0.32 and δ=0.

78.

7. A method for assessing highway carbon emissions according to claim 1, characterized in that, Step S6 specifically includes the following steps: S61. Vehicle generational data association and standardization: Establish a multi-source data collection and standard matching mechanism, accurately associate emission standard versions with vehicle characteristic information, construct a structured database containing generational attributes, and implement a data quality verification process. S62. Multidimensional dynamic matrix construction and verification: Based on the generational evolution of vehicle technology, a dynamic emission coefficient matrix integrating time, space, and vehicle model dimensions is constructed, a transition period coefficient calculation method is embedded, and a multi-level verification system is established to ensure data reliability. S63. Real-time collaborative correction and closed-loop optimization: Deploy a flexible correction framework based on multi-source real-time data collaboration, integrate vehicle degradation compensation and usage feedback mechanisms, and form a dynamic evolution capability for emission coefficients and a closed-loop verification link.

8. A method for assessing highway carbon emissions according to claim 1, characterized in that, Also includes: S7. Establish a multimodal data fusion verification mechanism, specifically including: S71. During the evaluation period, exhaust gas telemetry data and model calculation results are collected synchronously, and dynamic error compensation is achieved through the Kalman filter algorithm. For road sections with deviations exceeding ±15% for three consecutive times, model parameter recalibration is triggered; a confidence interval for carbon emission intensity is established, and data points exceeding the interval are marked as abnormal operating conditions; S72. Deploy the cross-validation module: Compare the measured road surface texture depth with the oxidation rate predicted by the model using a laser scanner. When the difference is greater than 20%, trigger the material parameter update; obtain the tire ground contact temperature distribution using a UAV thermal imager to verify the local adaptability of the climate cumulative effect model. S73. Generate verification reports and sign them on the blockchain: Output auditable logs daily containing the original data hash value, model version, and average error rate; automatically trigger data quality anomaly warnings through smart contracts.

9. A highway carbon emission assessment system, characterized in that, include: The roadside sensing array module consists of microwave radar, video acquisition device and laser scanning equipment. It is configured on both sides of the target road section to collect traffic flow data including vehicle density and vehicle type distribution, as well as status data such as road surface temperature and texture depth in real time. The edge computing gateway connects to the roadside sensing array and has a built-in multi-source data spatiotemporal alignment module for timestamp synchronization of traffic flow data and generation of traffic flow matrix. It also integrates ground weather station and satellite remote sensing data to calibrate environmental parameters. Dynamic modeling server, used for dynamic modeling of carbon emissions; The hierarchical calibration module of the edge computing gateway uses dynamic weighting coefficients to fuse satellite temperature data, meteorological station temperature data, and elevation correction values ​​to generate calibrated environmental parameters.

10. A highway carbon emission assessment system according to claim 9, characterized in that, Dynamic modeling server, including: The road body calculation engine dynamically calculates the carbon emission factor of road maintenance based on the oxidation rate characteristics of asphalt concrete materials and the pavement smoothness index. The maintenance compensation processor is configured to detect road closure events and invoke a path planning algorithm to calculate the carbon emission increment generated by the detour path. The driving behavior analysis unit processes vehicle acceleration and braking micro-operation data through machine learning models and outputs a nonlinear carbon emission correction coefficient. Climate effect modeling library, which establishes a dynamic correlation model between tire performance degradation and air conditioning energy consumption growth based on historical climate data; An intergenerational compensation database is used to link vehicle manufacturing year with emission standards to generate a dynamic emission coefficient matrix. The verification terminal integrates exhaust gas telemetry equipment, road scanner, and thermal imager to compare measured data with model calculation results, and uses blockchain nodes to achieve auditable storage of evaluation logs.

Images