Highway network traffic flow carbon emission estimation method fusing multi-source data

Abstract

The invention discloses a highway network traffic flow carbon emission estimation method fusing multi-source data. The method comprises the following steps: S1, carrying out multi-source data fusion and target road section traffic flow database construction; s2, classifying vehicles on the target road section of the expressway; s3, investigating and surveying and calibrating vehicle energy consumption of different types and different fuels; s4, obtaining energy consumption correction coefficients at different operation speeds through testing; s5, energy consumption correction coefficients under different road conditions are obtained through test training, and energy consumption correction coefficients under different road conditions are obtained through LightGBM training; and S6, estimating the carbon emission of the target road section of the expressway in specific time. According to the method, the highway target road section traffic flow database is constructed, the information data of the vehicle type, the fuel type, the interval speed and the road attribute of the target road section are extracted, and the highway road section estimation model is constructed and trained; and the accuracy and the reliability of the carbon emission calculation of each road section of the expressway network reach the quick estimation of an ideal level.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic emission, and in particular to a method for estimating carbon emission of highway network traffic flow by integrating multi-source data. Background Art

[0002] With the rapid economic development, carbon dioxide emissions continue to increase, and the greenhouse effect continues to intensify. As a key industry for energy consumption, the transportation industry's carbon emissions account for about 10% of the country's total carbon emissions. Carbon emissions in the transportation industry have become a global concern. In the transportation industry, carbon emissions from road transportation account for 74% of the total carbon emissions in the transportation industry. Among them, highway transportation, as an important part of road transportation, has a dominant position in carbon emissions in road transportation. Therefore, it is necessary to focus on highway transportation emission reduction.

[0003] In order to promote emission reduction in the highway transportation industry, we should first improve the carbon emission statistical monitoring system and calculate the carbon emissions in the highway transportation field. Some studies focus on the carbon emissions of the entire industry and adopt a "top-down" total estimation method to obtain the carbon emissions of the entire industry based on the overall energy consumption data of the industry. The statistical caliber of this study is relatively coarse and cannot specifically reflect the carbon emissions of highway network sections and nodes; in addition, some studies have considered integrating multi-source data and conducting carbon emission studies on highway networks for vehicles. They mainly adopt a "bottom-up" approach to calculate energy consumption using various types of transportation vehicles as energy users, but focus on the calculation of carbon emissions of target sections of highways. Most of them consider the vehicle type or vehicle fuel type data of the integrated traffic flow, without considering the changes in fuel consumption and emissions caused by road conditions and speed of traffic flow during actual driving on the highway, which will lead to inaccurate estimated data.

[0004] In view of this, this paper proposes a highway network traffic carbon emission estimation method that integrates multi-source data to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to provide a method for estimating carbon emissions from highway network traffic flow by integrating multi-source data to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for estimating carbon emissions of highway network traffic flow by integrating multi-source data, the estimation method comprising the following steps:

[0007] S1, multi-source data fusion and target road section traffic database construction, collecting gantry data, toll data and highway section attribute data in the road network model of the highway network toll collection system as basic data for fusion analysis;

[0008] S2, classifying vehicles on the target section of the expressway, wherein the classification is based on the target section vehicle information table, and the vehicles passing through the target section of the expressway are classified according to vehicle type and fuel;

[0009] S3, conduct research and calibrate vehicle energy consumption by vehicle type and fuel;

[0010] S4, obtain the energy consumption correction coefficient at different operating speeds through testing;

[0011] S5, obtain the energy consumption correction coefficient under different road conditions through test training, and use Light GBM training to obtain the energy consumption correction coefficient under different road conditions;

[0012] S6, estimating the carbon emissions of a target highway section at a specific time, by obtaining the road characteristics and weather characteristics data of the target highway section at a specific time period, and based on the traffic database of the target highway section at a specific time period, estimating the carbon emissions of the target highway section at a specific time period.

[0013] Preferably, the construction of the target road section traffic flow database includes the following steps:

[0014] S101, based on the road network model, using the road network topology and the location information of the gantry and toll station, segment the sections of the expressway network;

[0015] S102, based on the segmented road sections, matching the road attribute data of the target road section of the expressway, wherein the road attribute data includes slope, slope length, line type and road section attributes, and obtaining the target road section attribute information table;

[0016] S103, based on the highway network toll collection data, identifying the entry and exit node sequence and entry and exit time sequence of the highway traffic flow;

[0017] S104, based on the road network topology and the location information of the gantries and toll stations, the complete trajectory of the moving vehicles is traced through the entry and exit sequence of the high-speed traffic flow, the traffic flow information table of the target section is obtained, and the traffic flow database of the target section is constructed by associating the road section and the license plate.

[0018] Preferably, the road section attributes include an interchange section, a merging area, a diverging area, a tunnel section and a bridge section.

[0019] Preferably, the entry and exit node sequence is expressed as:

[0020] {S 1 ,M 1 ,M 2 ,…,M k ,S 2}(1);

[0021] Among them, S1 , S 2 Indicates the name of the entrance and exit of a high-speed trip of the vehicle, M k represents the kth gantry that the vehicle passes through during a high-speed trip;

[0022] The in-and-out time series is expressed as:

[0023] {T s1 ,T m1 ,T m2 ,…,T mk ,T s2}(2);

[0024] Among them, T s1 , T s2 Indicates the time of passing through the entrance and exit, T mk Indicates the time when the vehicle passes through the kth gantry.

[0025] Preferably, the classification of vehicles on the target section is based on the vehicle type field, passenger cars are divided into 1-4 categories according to the number of seats, trucks are divided into 1-6 categories according to the number of axles, special operation vehicles on toll roads are included as trucks, and based on the license plate number, license plate color, and license plate digit fields, the fuel types of vehicles passing through the target section of the highway are divided into pure electric, non-pure electric, and traditional energy.

[0026] Preferably, the calibration of the vehicle energy consumption is to obtain the theoretical energy consumption value per 100 kilometers of the vehicle's factory model, calculate the theoretical average energy consumption per 100 kilometers of vehicles by model and fuel, investigate the average actual energy consumption per 100 kilometers of vehicles by model and fuel type, and calibrate the energy consumption correction coefficient by model and fuel.

[0027] Preferably, the energy consumption correction coefficient at different operating speeds is obtained by testing the fuel consumption of vehicles of different fuel types at different average speed ranges, and calibrating the vehicle energy consumption speed correction coefficient.

[0028] Preferably, the test of the energy consumption correction coefficient under different road conditions includes the following steps:

[0029] S501, obtaining training data, and testing and obtaining the energy consumption of the vehicle under different road characteristics and weather characteristics;

[0030] S502, calculating energy consumption correction coefficient data, calculating the theoretical carbon emissions of the vehicle according to the vehicle model mileage, interval running speed, and carbon emission factor based on the obtained average energy consumption per 100 kilometers of the vehicle by vehicle model and fuel, the energy consumption correction coefficients of the vehicle model and fuel, and the energy consumption correction coefficients of different running speeds, and the actual carbon emissions are compared with the theoretical carbon emissions to obtain energy consumption correction coefficient training data;

[0031] S503, constructing a characteristic vector of influencing factors, extracting road characteristic data of the test section and weather characteristic data of the test period, recoding according to the data type, and constructing road characteristics and weather characteristics;

[0032] S504, input the feature vectors of the m samples into the Light GBM for training, establish the relationship between the feature vector and the energy consumption correction coefficient, and obtain the optimal parameters of the Light GBM model through k-fold cross validation, which are used to characterize the mapping relationship between the feature vector and the energy consumption correction coefficient.

[0033] Preferably, the training of the feature vector includes:

[0034] The first step is to extract the feature vector set of the sample, expressed as:

[0035]

[0036] Among them, x m represents the feature vector of the mth sample, and n represents the number of feature vectors;

[0037] In the second step, the feature vectors of m samples are input into Light GBM for training to establish the relationship between road conditions and energy consumption, which is expressed as:

[0038]

[0039] Among them, Y m represents the energy consumption correction factor of the mth sample, f(θ 1 ,θ 2 ,...,θ n ) indicates that the optimal parameters of the Light GBM model are obtained through k-fold cross validation, which is used to characterize the mapping relationship from road conditions to energy consumption, that is, the energy consumption correction coefficient under different road conditions.

[0040] Preferably, the estimation of carbon emissions includes the following steps:

[0041] S601, extracting the length of the target expressway section and the road characteristic data of the section;

[0042] S602, obtaining weather characteristic data of a target section of a highway during a specific time period;

[0043] S603, calculating the average running speed data of the traffic flow on the target section of the expressway, calibrating the average running speed of the traffic flow between the node sequences according to the traffic flow entry and exit node sequence and the entry and exit time sequence, finding the sequence interval closest to the section, and using the average running speed of the traffic flow in the sequence interval to calibrate it as the average running speed of the vehicles on the target section;

[0044] S604, based on the traffic database, estimate the carbon emissions of the target section of the highway during a specific time period, expressed as:

[0045]

[0046] Among them, Q s represents the carbon emissions of the target highway section, L s Indicates the mileage of the target section of the expressway. represents the theoretical average energy consumption per 100 kilometers of the i-th type of vehicle and the j-th type of fuel, represents the energy consumption correction factor of the i-th type of vehicle and the j-th type of fuel, f v,j It represents the energy consumption speed correction coefficient of the j-th fuel vehicle in the v-th speed interval, N s,i,j,v It represents the number of vehicles of type i, type j, fuel and speed interval v that pass through the target section of the highway at a specific time period. represents the carbon emission factor of the jth energy source, f(θ 1 ,θ 2 ,...,θ n ) represents the energy consumption correction coefficient under the road conditions and weather conditions of the target section of the highway.

[0047] Technical effects and advantages of the present invention:

[0048] Based on the problems of coarse statistical caliber and inability to focus on road sections in existing methods, the present invention proposes a method for estimating carbon emissions from highway network traffic that integrates multi-source data. By constructing a traffic database for target highway sections, extracting information data on vehicle type, fuel type, interval speed and road attributes of the target sections, and constructing and training a highway section estimation model, carbon emissions from each section of the highway network can be rapidly estimated with ideal accuracy and reliability. The data used in the implementation of the present invention are easy to obtain, and the samples cover a wide range of time and space. The method is data-driven, the algorithm is simple and feasible, and the method has strong universality. Carbon emissions from highway sections and node traffic can be rapidly estimated at an ideal level. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a framework diagram of the implementation steps of the estimation method of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] The present invention provides Figure 1 A highway network traffic carbon emission estimation method integrating multi-source data is shown, which includes the following steps:

[0052] S1, multi-source data fusion and target road section traffic database construction, collecting gantry data, toll data and highway section attribute data in the road network model of the highway network toll collection system as basic data for fusion analysis;

[0053] Specifically, the construction of the target road section traffic flow database includes the following steps:

[0054] S101, based on the road network model, using the road network topology and the location information of the gantry and toll station, segment the sections of the expressway network;

[0055] S102, based on the segmented road sections, matching the road attribute data of the target road section of the expressway, the road attribute data including slope, slope length, line type and road section attributes, and obtaining the target road section attribute information table, the information table is shown in Table 1 below;

[0056] Table 1 Example table of target road segment attribute data

[0057]

[0058]

[0059] Furthermore, the road section attributes include interchange sections, merging areas, diverging areas, tunnel sections, and bridge sections.

[0060] S103, based on the highway network toll collection data, identifying the entry and exit node sequence and entry and exit time sequence of the highway traffic flow;

[0061] Furthermore, the entry and exit node sequence is expressed as:

[0062] {S 1 ,M 1 ,M 2 ,…,M k ,S 2}(1);

[0063] Among them, S 1 , S 2 Indicates the name of the entrance and exit of a high-speed trip of the vehicle, M k represents the kth gantry that the vehicle passes through during a high-speed trip;

[0064] The in-and-out time series is expressed as:

[0065] {T s1 ,T m1 ,Tm2 ,…,T mk ,T s2}(2);

[0066] Among them, T s1 , T s2 Indicates the time of passing through the entrance and exit, T mk Indicates the time when the vehicle passes through the kth gantry.

[0067] S104, based on the road network topology and the location information of the gantries and toll stations, the complete trajectory of the moving vehicles is traced through the entry and exit sequence of the high-speed traffic flow, and the traffic flow information table of the target section is obtained, as shown in Table 2 below, and the traffic flow database of the target section is constructed by association based on the road section and the license plate.

[0068] Table 2 Target road section traffic information table

[0069]

[0070] S2, classifying vehicles on the target section of the expressway, wherein the classification is based on the target section vehicle information table, and the vehicles passing through the target section of the expressway are classified according to vehicle type and fuel, wherein the vehicle type classification table and the fuel classification table are shown in Table 3 and Table 4 below;

[0071] Table 3 Vehicle classification standards

[0072]

[0073] Table 4 Fuel classification standards

[0074] Specifically, the classification of vehicles on the target section is based on the vehicle type field. Passenger buses are divided into categories 1-4 according to the number of seats, trucks are divided into categories 1-6 according to the number of axles, and special-purpose vehicles on toll roads are included as trucks. In addition, based on the license plate number, license plate color, and license plate digits fields, the fuel types of vehicles passing through the target section of the highway are divided into pure electric, non-pure electric, and traditional energy.

[0075] S3, conduct research and calibrate vehicle energy consumption by vehicle type and fuel;

[0076] Specifically, the calibration of vehicle energy consumption is to obtain the theoretical energy consumption per 100 kilometers of the vehicle's factory model, calculate the theoretical average energy consumption per 100 kilometers of vehicles based on different models and fuel types, investigate the average actual energy consumption per 100 kilometers of vehicles based on different models and fuel types, and calibrate the energy consumption correction coefficient based on different models and fuel types.

[0077] S4, obtain the energy consumption correction coefficient at different operating speeds through testing;

[0078] Specifically, the energy consumption correction coefficient at different operating speeds is obtained by testing the fuel consumption of vehicles of different fuel types in different average speed ranges, and calibrating the vehicle energy consumption speed correction coefficient, as shown in Table 5 below.

[0079] Table 5. Energy consumption speed correction coefficients of vehicles on expressway network

[0080]

[0081] S5, obtain the energy consumption correction coefficient under different road conditions through test training, and use Light GBM training to obtain the energy consumption correction coefficient under different road conditions;

[0082] Specifically, the test of the energy consumption correction coefficient under different road conditions includes the following steps:

[0083] S501, obtaining training data, and testing and obtaining the energy consumption of the vehicle under different road characteristics and weather characteristics;

[0084] S502, calculating energy consumption correction coefficient data, calculating the theoretical carbon emissions of the vehicle according to the vehicle model mileage, interval running speed, and carbon emission factor based on the obtained average energy consumption per 100 kilometers of the vehicle by vehicle model and fuel, the energy consumption correction coefficients of the vehicle model and fuel, and the energy consumption correction coefficients of different running speeds, and the actual carbon emissions are compared with the theoretical carbon emissions to obtain energy consumption correction coefficient training data;

[0085] S503, constructing a feature vector of influencing factors, extracting road feature data of the test section and weather feature data of the test period, recoding them according to the data type, constructing road features and weather features, and constructing feature vectors thereof as shown in Table 6 below;

[0086] Table 6 Feature vector construction table

[0087]

[0088]

[0089] S504, input the feature vectors of the m samples into the Light GBM for training, establish the relationship between the feature vector and the energy consumption correction coefficient, and obtain the optimal parameters of the Light GBM model through k-fold cross validation, which are used to characterize the mapping relationship between the feature vector and the energy consumption correction coefficient.

[0090] Furthermore, the training of feature vectors includes:

[0091] The first step is to extract the feature vector set of the sample, expressed as:

[0092]

[0093] Among them, x m represents the feature vector of the mth sample, and n represents the number of feature vectors;

[0094] In the second step, the feature vectors of m samples are input into Light GBM for training to establish the relationship between road conditions and energy consumption, which is expressed as:

[0095]

[0096] Among them, Y m represents the energy consumption correction factor of the mth sample, f(θ 1 ,θ 2 ,...,θ n ) indicates that the optimal parameters of the Light GBM model are obtained through k-fold cross validation, which is used to characterize the mapping relationship from road conditions to energy consumption, that is, the energy consumption correction coefficient under different road conditions.

[0097] S6, estimating the carbon emissions of a target highway section at a specific time, by obtaining the road characteristics and weather characteristics data of the target highway section at a specific time period, and based on the traffic database of the target highway section at a specific time period, estimating the carbon emissions of the target highway section at a specific time period.

[0098] Specifically, the estimation of carbon emissions includes the following steps:

[0099] S601, extracting the length of the target expressway section and the road characteristic data of the section;

[0100] S602, obtaining weather characteristic data of a target section of a highway during a specific time period;

[0101] S603, calculating the average running speed data of the traffic flow on the target section of the expressway, calibrating the average running speed of the traffic flow between the node sequences according to the traffic flow entry and exit node sequence and the entry and exit time sequence, finding the sequence interval closest to the section, and using the average running speed of the traffic flow in the sequence interval to calibrate it as the average running speed of the vehicles on the target section;

[0102] S604, based on the traffic database, estimate the carbon emissions of the target section of the highway during a specific time period, expressed as:

[0103]

[0104] Among them, Q s represents the carbon emissions of the target highway section, L s Indicates the mileage of the target section of the expressway. represents the theoretical average energy consumption per 100 kilometers of the i-th type of vehicle and the j-th type of fuel, represents the energy consumption correction factor of the i-th type of vehicle and the j-th type of fuel, fv,j It represents the energy consumption speed correction coefficient of the j-th fuel vehicle in the v-th speed interval, N s,i,j,v It represents the number of vehicles of type i, type j, fuel and speed interval v that pass through the target section of the highway at a specific time period. represents the carbon emission factor of the jth energy source, f(θ 1 ,θ 2 ,...,θ n ) represents the energy consumption correction coefficient under the road conditions and weather conditions of the target section of the highway.

[0105] By constructing a traffic database for target highway sections, extracting traffic information data on target sections at specific times, as well as traffic information data on vehicle type, fuel type, and interval speed, a highway section carbon emission estimation model is constructed and trained to achieve rapid estimation of carbon emissions for each section of the highway network with an ideal level of accuracy and reliability.

[0106] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A highway network traffic carbon emission estimation method integrating multi-source data, characterized in that: The estimation method includes the following steps: S1, multi-source data fusion and target road section traffic database construction, collecting gantry data, toll data and highway section attribute data in the road network model of the highway network toll collection system as basic data for fusion analysis; S2, classifying vehicles on the target section of the expressway, wherein the classification is based on the target section vehicle information table, and the vehicles passing through the target section of the expressway are classified according to vehicle type and fuel; S3, conduct research and calibrate the vehicle energy consumption of different models and fuels; S4, obtain the energy consumption correction coefficient at different operating speeds through testing; S5, obtain the energy consumption correction coefficient under different road conditions through test training, and use Light GBM training to obtain the energy consumption correction coefficient under different road conditions; S6, estimating the carbon emissions of a target highway section at a specific time, by obtaining the road characteristics and weather characteristics data of the target highway section at a specific time period, and based on the traffic database of the target highway section at a specific time period, estimating the carbon emissions of the target highway section at a specific time period.

2. According to claim 1, a highway network traffic flow carbon emission estimation method integrating multi-source data is characterized in that: The construction of the target road section traffic flow database includes the following steps: S101, based on the road network model, using the road network topology and the location information of the gantry and toll station, segment the sections of the expressway network; S102, based on the segmented road sections, matching the road attribute data of the target road section of the expressway, wherein the road attribute data includes slope, slope length, line type and road section attributes, and obtaining the target road section attribute information table; S103, based on the highway network toll collection data, identifying the entry and exit node sequence and entry and exit time sequence of the highway traffic flow; S104, based on the road network topology and the location information of the gantries and toll stations, the complete trajectory of the moving vehicles is traced through the entry and exit sequence of the high-speed traffic flow, the traffic flow information table of the target section is obtained, and the traffic flow database of the target section is constructed by associating the road section and the license plate.

3. The method for estimating carbon emissions from highway network traffic flow by integrating multi-source data according to claim 2 is characterized in that: The road section attributes include an interchange section, a merging area, a diverging area, a tunnel section, and a bridge section.

4. According to claim 2, a highway network traffic carbon emission estimation method integrating multi-source data is characterized in that: The in-and-out node sequence is expressed as: {S1,M1,M2,…,M k ,S2}(1); Among them, S1 and S2 represent the names of the entrance and exit of a high-speed trip of the vehicle, and M k represents the kth gantry that the vehicle passes through during a high-speed trip; The in-and-out time series is expressed as: {T s1 ,T m1 ,T m2 ,…,T mk ,T s2 }(2); Among them, T s1 , T s2 Indicates the time of passing through the entrance and exit, T mk Indicates the time when the vehicle passes through the kth gantry.

5. The method for estimating carbon emissions of highway network traffic flow by integrating multi-source data according to claim 1 is characterized in that: The classification of vehicles on the target section is based on the vehicle type field. Passenger cars are divided into categories 1-4 according to the number of seats, trucks are divided into categories 1-6 according to the number of axles, and special-purpose vehicles on toll roads are included as trucks. In addition, based on the license plate number, license plate color, and license plate number of digits fields, the fuel types of vehicles passing through the target section of the highway are divided into pure electric, non-pure electric, and traditional energy.

6. The method for estimating carbon emissions of highway network traffic flow by integrating multi-source data according to claim 1 is characterized in that: The calibration of the vehicle energy consumption is to obtain the theoretical energy consumption value per 100 kilometers of the vehicle's factory model, calculate the theoretical average energy consumption per 100 kilometers of vehicles by model and fuel, investigate the average actual energy consumption per 100 kilometers of vehicles by model and fuel type, and calibrate the energy consumption correction coefficient by model and fuel.

7. The method for estimating carbon emissions of highway network traffic flow by integrating multi-source data according to claim 1 is characterized in that: The energy consumption correction coefficient at different operating speeds is obtained by testing the fuel consumption of vehicles of different fuel types in different average speed ranges, and calibrating the vehicle energy consumption speed correction coefficient.

8. The method for estimating carbon emissions of highway network traffic flow by integrating multi-source data according to claim 1 is characterized in that: The test of the energy consumption correction coefficient under different road conditions comprises the following steps: S501, obtaining training data, and testing and obtaining the energy consumption of the vehicle under different road characteristics and weather characteristics; S502, calculating energy consumption correction coefficient data, calculating the theoretical carbon emissions of the vehicle according to the vehicle model mileage, interval running speed, and carbon emission factor based on the obtained average energy consumption per 100 kilometers of the vehicle by vehicle model and fuel, the energy consumption correction coefficients of the vehicle model and fuel, and the energy consumption correction coefficients of different running speeds, and the actual carbon emissions are compared with the theoretical carbon emissions to obtain energy consumption correction coefficient training data; S503, constructing a characteristic vector of influencing factors, extracting road characteristic data of the test section and weather characteristic data of the test period, recoding according to the data type, and constructing road characteristics and weather characteristics; S504, input the feature vectors of the m samples into the Light GBM for training, establish the relationship between the feature vector and the energy consumption correction coefficient, and obtain the optimal parameters of the Light GBM model through k-fold cross validation, which are used to characterize the mapping relationship between the feature vector and the energy consumption correction coefficient.

9. The method for estimating carbon emissions of highway network traffic flow by integrating multi-source data according to claim 8 is characterized in that: The training of the feature vector includes: The first step is to extract the feature vector set of the sample, expressed as: Among them, x m represents the feature vector of the mth sample, and n represents the number of feature vectors; In the second step, the feature vectors of m samples are input into Light GBM for training to establish the relationship between road conditions and energy consumption, which is expressed as: Among them, Y m represents the energy consumption correction factor of the mth sample, f(θ1,θ2,...,θ n ) indicates that the optimal parameters of the Light GBM model are obtained through k-fold cross validation, which is used to characterize the mapping relationship from road conditions to energy consumption, that is, the energy consumption correction coefficient under different road conditions.

10. The method for estimating carbon emissions of highway network traffic flow by integrating multi-source data according to claim 1, characterized in that: The estimation of carbon emissions includes the following steps: S601, extracting the length of the target expressway section and the road characteristic data of the section; S602, obtaining weather characteristic data of a target section of a highway during a specific time period; S603, calculating the average running speed data of the traffic flow on the target section of the expressway, calibrating the average running speed of the traffic flow between the node sequences according to the traffic flow entry and exit node sequence and the entry and exit time sequence, finding the sequence interval closest to the section, and using the average running speed of the traffic flow in the sequence interval to calibrate it as the average running speed of the vehicles on the target section; S604, based on the traffic database, estimate the carbon emissions of the target section of the highway during a specific time period, expressed as: Among them, Q s represents the carbon emission of the target highway section, L s Indicates the mileage of the target section of the expressway. represents the theoretical average energy consumption per 100 kilometers of the i-th type of vehicle and the j-th type of fuel, represents the energy consumption correction factor of the i-th type of vehicle and the j-th type of fuel, f v,j It represents the energy consumption speed correction coefficient of the j-th fuel vehicle in the v-th speed interval, N s,i,j,v It represents the number of vehicles of type i, type j, fuel and speed interval v that pass through the target section of the highway at a specific time period. represents the carbon emission factor of the jth energy source, f(θ1,θ2,...,θ n ) represents the energy consumption correction coefficient under the road conditions and weather conditions of the target section of the highway.

Images