Carbon emission evaluation system based on machine learning

Abstract

The invention discloses a carbon emission evaluation system based on machine learning, and relates to the technical field of pet insurance management, and the system comprises a data processing module which is used for collecting real-time traffic data and outputting real-time carbon emission evaluation, and a microscopic behavior prediction module which is used for predicting the real-time carbon emission evaluation. The microscopic behavior prediction module is used for analyzing an individual travel mode according to the data of the data processing module; the interaction analysis module is used for integrating the data collected by the data processing module; a planning module; the accuracy of carbon emission evaluation can be improved, the current traffic condition and the real-time change of individual behaviors can be reflected, the system can predict the traffic behavior change and the potential influence of the traffic behavior change on carbon emission in combination with microscopic behavior prediction and multi-factor interaction analysis, and a scientific basis is provided for urban planning and policy making; real-time monitoring and prediction results provided by the system can help traffic managers to optimize the traffic flow.

Description

Technical Field

[0001] The present invention relates to the technical field of pet insurance management, and more specifically, to a carbon emission evaluation system based on machine learning. Background Art

[0002] Climate change is a major hot issue that the international community is currently paying attention to. The series of consequences it triggers will not only have a significant impact on the global environment and ecology, but also have a profound impact on the production, consumption and lifestyle of human society. As society gradually pays more attention to environmental protection, carbon emission assessment, as an environmental management system, can conduct qualitative or quantitative evaluation of transportation carbon emissions. With the development of social economy and the improvement of people's living standards, the overall level of motorization is still improving, which means that the use of high-carbon emission vehicles such as cars may increase. In the process of urbanization, changes in urban spatial structure will also affect travel modes and distances, and thus affect carbon emissions. The Chinese patent application number 202310172198.0 discloses a carbon market carbon emission data quality evaluation method, which relates to the technical field of carbon emission data monitoring and evaluation. The method includes: receiving evaluation requests from multiple evaluation objects, selecting multiple evaluation indicators for the carbon emission data quality of the evaluation object; using the Bendeford law to measure the index value of each evaluation indicator of each evaluation object; according to the index value of each evaluation indicator of each evaluation object, using the linear weighted sum method, TOPSIS approximation ideal solution method and VIKOR compromise method to calculate the integrated value of the evaluation index of the carbon emission data quality of each evaluation object; performing gray statistical evaluation and grading on each carbon emission location to obtain the quality grade of the carbon emission data quality of each carbon emission location; However, in actual use, urban traffic conditions change in real time, and it is difficult for the system to respond quickly and provide immediate carbon emissions assessments. It is necessary to go deep into the behavioral patterns at the individual and vehicle levels, which is very difficult in data collection and analysis. Factors such as urbanization, socio-economic development, personal behavior, and policy changes are constantly changing and intertwined, which have a compound impact on carbon emissions. Existing assessment and prediction methods are difficult to accurately predict the carbon emission dynamics caused by changes in micro-traffic behavior. Summary of the invention

[0003] To solve the above problems, the present invention provides a carbon emission evaluation system based on machine learning.

[0004] A carbon emission evaluation system based on machine learning includes: a data processing module, the data processing module is used to collect real-time traffic data and output real-time carbon emission evaluation, specifically: Set an evaluation area in advance; according to the formula , calculate and obtain the daily traffic carbon emissions within the area , where n is the type of activity level, is the data of the ith activity level, is the emission factor corresponding to the ith activity level data; A micro-behavior prediction module, which is used to analyze individual travel patterns according to the data from the data processing module, and feed the prediction results back to the data processing module to obtain a preliminary predicted carbon emission evaluation; An interactive analysis module, the interactive analysis module is used to integrate the data collected by the data processing module, analyze the impact on traffic carbon emissions based on the integrated data collected by the data processing module, and output a carbon emission impact analysis; The planning module is used to perform long-term trend analysis based on the output data of the above three modules and output future carbon emission forecasts and planning suggestions.

[0005] Preferably, the n is 2, corresponding to, Vehicle mileage and fuel consumption ; The first correlation factor and the second related factor , where the first correlation factor The specific way to obtain is as follows: According to the formula , calculate and obtain the first correlation factor ,in is the consumption of the ith fossil fuel, in millions.

[0006] Preferably, the specific working mode of the micro-behavior prediction module is as follows: Get time features , spatial characteristics , Vehicle characteristics and user behavior characteristics ; According to the formula , get the prediction results ,in Represents the jth feature, specifically the time feature , spatial characteristics , Vehicle characteristics and user behavior characteristics Any of the above, m represents the number of features, and m is 4, is the decision threshold corresponding to the feature, Indicates the predicted category under current conditions; Prediction results The percentage of people's travel modes in the area tomorrow; The predicted results Feedback to the data processing module.

[0007] Preferably, the specific working method of the data processing module after obtaining the prediction result of the micro-behavior prediction module is as follows: Get the percentage of people's travel mode in the area on that day; According to the prediction results The difference between the percentage of the travel mode of the people in the area tomorrow and the percentage of the travel mode of the people in the area today is calculated to obtain the motor vehicle travel ratio ; According to the formula , get the predicted traffic carbon emissions value of the area ,in is the preset proportional coefficient; The predicted carbon emissions from transportation in this area Transmitted to the interaction analysis module.

[0008] Preferably, the interaction analysis module works as follows: According to the formula , calculate and obtain the impact value of carbon emissions in the area ,in It is The regression coefficients of the influencing factors are It is The actual value of the influencing factor, is the number of influencing factors; According to the formula , calculate and obtain the adjusted forecast value of tomorrow's traffic carbon emissions .

[0009] Preferably, the number of k is 3, corresponding to for: Population Density and collect the average distance between residence and work ; Public transport service indicators .

[0010] Preferably, the working method of the interaction analysis module also includes: According to the formula , calculate and obtain the optimized predicted traffic carbon emissions in the area tomorrow ,in is the influence coefficient of weather factors on the choice of travel mode, It is the weather impact index.

[0011] Preferably, the planning module includes a prediction unit and a planning unit, and the specific working mode of the prediction unit is as follows: According to the formula , calculate and obtain the predicted traffic carbon emissions of the area for the day after tomorrow ,in is the rate of change based on the time trend, is the rate of change based on season; Based on the obtained forecast of traffic carbon emissions in the area for the next day The predicted traffic carbon emissions for each day of the month are calculated by this method; Set a threshold for traffic carbon emissions in advance, compare the predicted traffic carbon emissions for each day of the month with the threshold, and obtain the number of days above the threshold. , the number of days below the threshold Transfer to the planning unit.

[0012] Preferably, the specific working mode of the planning unit is to obtain the existing traffic carbon emission planning scheme, as follows: Including transportation structure, encouragement of new energy vehicles and energy efficiency of transportation vehicles; If the number of days above the threshold If the pollution level is less than 2, it is marked as low pollution. If the pollution level is above the threshold If the pollution level is greater than 2 and less than 5, it is marked as moderate pollution. If it is greater than 5, it is marked as high pollution; Adjust transportation carbon emission planning schemes based on pollution results.

[0013] Preferably, the specific working method of the planning unit also includes: If it is low pollution, maintain the current transportation carbon emission planning scheme; If the pollution is moderate, strengthen the promotion policy of new energy vehicles, increase subsidies, expand the scope of tax incentives, implement stricter energy efficiency standards, encourage the use of low-carbon transportation, optimize the transportation structure, and increase investment in public transportation and non-motorized transportation; If the pollution is high, emergency measures will be implemented, such as restricting the use of high-emission vehicles, optimizing the traffic signal system to reduce congestion, expanding incentive policies for new energy vehicles, which may include car purchase subsidies, free parking, free charging, etc., increasing investment in public transportation systems, and improving service frequency and coverage.

[0014] Beneficial effects: It can increase the accuracy of carbon emission assessment, reflect the real-time changes of current traffic conditions and individual behaviors, and combine micro-behavior prediction and multi-factor interaction analysis. The system can predict changes in traffic behavior and its potential impact on carbon emissions, providing a scientific basis for urban planning and policy making. The real-time monitoring and prediction results provided by the system can help traffic managers optimize traffic flow, reduce congestion, and thus reduce carbon emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of the management system of the present invention. DETAILED DESCRIPTION

[0016] Application scenarios: Climate change is a major hot issue that the international community is currently concerned about. The series of consequences it triggers will not only have a significant impact on the global environment and ecology, but also have a profound impact on human production, consumption and lifestyle. As society gradually pays more attention to environmental protection, carbon emission assessment, as an environmental management system, can conduct qualitative or quantitative evaluation of transportation carbon emissions. With the development of social economy and the improvement of people's living standards, the overall level of motorization is still improving, which means that the use of high-carbon emission vehicles such as cars may increase. In the process of urbanization, changes in urban spatial structure will also affect travel modes and distances, and thus affect carbon emissions; However, in actual use, urban traffic conditions change in real time, and it is difficult for the system to respond quickly and provide immediate carbon emissions assessments. It is necessary to go deep into the behavioral patterns at the individual and vehicle levels, which is very difficult in data collection and analysis. Factors such as urbanization, socio-economic development, personal behavior, and policy changes are constantly changing and intertwined, which have a compound impact on carbon emissions. Existing assessment and prediction methods are difficult to accurately predict the carbon emission dynamics caused by changes in micro-traffic behavior.

[0017] like Figure 1 As shown: A carbon emission evaluation system based on machine learning includes: a data processing module, the data processing module is used to collect real-time traffic data and output real-time carbon emission evaluation, specifically: An evaluation area is set in advance; it should be noted that the area is obtained according to the user's division, and can be a fixed area or a fixed area within a city; According to the formula , calculate and obtain the daily traffic carbon emissions within the area , where n is the type of activity level, is the data of the ith activity level, is the emission factor corresponding to the ith activity level data; it should be noted that, in this embodiment, the activity level data reflects the scale and intensity of a specific activity, such as the mileage of a vehicle or the amount of fuel consumed, which are directly related to the amount of activity where emissions occur. The emission factor is a conversion coefficient used to convert activity level data into emissions, representing the greenhouse gas emissions generated per unit of activity (such as per kilometer traveled, per liter of fuel consumed); It should be noted that, in this embodiment, the collected traffic data includes vehicle flow and speed: data collected by traffic monitoring equipment is used to analyze traffic congestion; Traffic accidents and repair events: Real-time reporting of traffic accidents and road repair information that affect traffic flow and speed; Public transportation usage: passenger volume, operating frequency and route information of public transportation such as buses and subways; Vehicle type and fuel type: different types of vehicles (e.g. private cars, trucks, public transport) and the type of fuel they use (e.g. petrol, diesel, electricity); Transportation infrastructure usage: such as usage and maintenance records of roads, bridges, and tunnels; It should also be noted that providing instant carbon emission assessment helps city managers respond quickly to changes in traffic conditions. Through the analysis of real-time data, traffic signal control can be optimized to reduce congestion and emissions; A micro-behavior prediction module, which is used to analyze individual travel patterns according to the data from the data processing module, and feed the prediction results back to the data processing module to obtain a preliminary predicted carbon emission evaluation; It should be noted that a deep understanding of individual and group travel decisions can provide support for the formulation of targeted transportation policies, predict changes in transportation demand, and provide data support for transportation planning and infrastructure construction; An interactive analysis module, the interactive analysis module is used to integrate the data collected by the data processing module, analyze the impact on traffic carbon emissions based on the integrated data collected by the data processing module, and output a carbon emission impact analysis; It should be noted that revealing how different factors jointly affect carbon emissions can provide a scientific basis for integrated transportation and environmental policies, assess the potential impact of policy changes on carbon emissions, and help policymakers intervene effectively; Planning module, which is used to perform long-term trend analysis based on the output data of the above three modules, and output future carbon emission forecasts and planning suggestions. It should be noted that it provides forecast support for long-term transportation planning and carbon emission reduction strategies; It should be noted that the above modules can increase the accuracy of carbon emission assessment, reflect the real-time changes of current traffic conditions and individual behaviors, and combine micro-behavior prediction and multi-factor interaction analysis. The system can predict changes in traffic behavior and their potential impact on carbon emissions, providing a scientific basis for urban planning and policy formulation. The real-time monitoring and prediction results provided by the system can help traffic managers optimize traffic flow, reduce congestion, and thus reduce carbon emissions. As an optional embodiment: the n is 2, correspondingly, Vehicle mileage and fuel consumption ; It should be noted that it can be collected through the vehicle's odometer reading or GPS tracking system. For public transportation vehicles, it can be obtained through the vehicle management system. For private cars, driving data can be recorded through the on-board diagnostic system (OBD) device or mobile application; Fuel consumption data can be collected through fuel card transaction records, gas station sales records or vehicle fuel consumption monitoring systems. For corporate fleets, an internal management system can be established to regularly record the fuel consumption and mileage of each vehicle. The above information is collected on the same day; The first correlation factor and the second related factor , where the first correlation factor The specific way to obtain is as follows: According to the formula , calculate and obtain the first correlation factor ,in is the consumption of the fossil fuel in the ith unit, in million kilojoules, is the carbon dioxide emission factor of the i-th fossil fuel, in tons of carbon dioxide / million kilojoules. It should be noted that the consumption of the i-th fossil fuel is calculated by The analysis shows that in this embodiment, the main types are fuel oil, gasoline, kerosene, diesel and natural gas. The carbon dioxide emission factor of the corresponding fossil fuel in the ith position can be calculated by multiplying the carbon content per unit calorific value of the fossil fuel by the carbon oxidation rate and then by obtain; The carbon oxidation rates of fuel oil, gasoline, kerosene, diesel and natural gas are all about 98%, and the carbon content per unit calorific value of fuel oil, gasoline, kerosene, diesel and natural gas are 0.0211tC / GJ, 0.0189tC / GJ, 0.0196tC / GJ, 0.0202tC / GJ and 0.01532tC / GJ respectively; It should also be noted that the second correlation factor It can be obtained by directly measuring the actual vehicle test data through equipment such as the Portable Emission Measurement System (PEMS).

[0018] As an optional embodiment: the specific working mode of the micro-behavior prediction module is as follows: Get time features , spatial characteristics , Vehicle characteristics and user behavior characteristics It should be noted that the above characteristics are obtained by collecting data such as vehicle type, mileage, usage frequency, route selection, timestamp, weather conditions, traffic rules changes, etc. through the data processing module, among which the time characteristics is the time feature of the day, and the spatial feature The characteristics of the vehicle are the routes in the area, the vehicle characteristics are the model and age of the vehicle, and the user behavior characteristics are driving habits and travel frequency. According to the formula , get the prediction results ,in Represents the jth feature, specifically the time feature , spatial characteristics , Vehicle characteristics and user behavior characteristics Any of the above, m represents the number of features, and m is 4, is the decision threshold corresponding to the feature, Indicates the predicted category under the current conditions; it should be noted that Represents a logical OR operation, meaning either is true, then select the corresponding As a prediction result, the prediction result represents the category judgment made by the model based on the input features and the learned decision rules. In this embodiment, if the decision tree is used to predict the mode of transportation (such as driving, taking public transportation, riding a bicycle, etc.), the prediction result will be to determine the most likely mode of transportation to be selected based on the input features (such as time, weather, distance, etc.); The above is the decision tree algorithm in machine learning. It is a machine learning algorithm that imitates the human decision-making process. It predicts the value of the target variable by learning decision rules from data. The decision tree consists of nodes and branches, where each internal node represents a judgment on a feature, each branch represents the result of the judgment, and the leaf node represents the final decision or prediction result. A probability value is output for each possible category, indicating the probability of choosing this travel mode; Prediction results is the percentage of the travel modes of the people in the area tomorrow. It should be noted that the above scheme can obtain the predicted percentage change of the travel modes of the people in the area tomorrow, which is conducive to the subsequent feedback adjustment of carbon emissions. The predicted results Feedback to the data processing module.

[0019] As an optional embodiment: the specific working method of the data processing module after obtaining the prediction result of the micro-behavior prediction module is as follows: The percentage of the travel modes of the people in the area on that day is obtained; it should be noted that the data processing module first needs to collect the travel data of the people in the area, which may include traffic flow monitoring, public transportation use records, questionnaires, mobile device tracking, etc.; According to the prediction results The difference between the percentage of the travel mode of the people in the area tomorrow and the percentage of the travel mode of the people in the area today is calculated to obtain the motor vehicle travel ratio ; It should be noted that the difference is calculated by subtracting the ratio of motor vehicle travel among the travel modes of the people in the area on the current day from the ratio of motor vehicle travel among the travel modes of the people in the area tomorrow; According to the formula , get the predicted traffic carbon emissions value of the area ,in is the preset proportional coefficient; The predicted carbon emissions from transportation in this area Transmitted to the interaction analysis module.

[0020] As an optional embodiment: the interactive analysis module works as follows: It should be noted that in actual use, traffic carbon emissions will also be affected by various complex factors in the region. This technical solution can comprehensively consider and quantify the factors affecting carbon emissions; accurate impact assessment can help policymakers formulate more effective traffic and environmental policies to reduce carbon emissions; According to the formula , calculate and obtain the impact value of carbon emissions in the area ,in It is The regression coefficients of the influencing factors are It is The actual value of the influencing factor, is the number of influencing factors; it should be noted that, in this embodiment, Indicates that the corresponding independent variable changes by one unit, which is the average change in transportation carbon emissions; According to the formula , calculate and obtain the adjusted forecast value of tomorrow's traffic carbon emissions It should be noted that the predicted carbon emissions from transportation in this area are In this embodiment, it is expressed as the predicted traffic carbon emissions when all influencing factors are 0. Through the above calculation, the predicted value of tomorrow's traffic carbon emissions after considering the impact in the region can be obtained. , which can be more in line with the actual results.

[0021] As an optional embodiment: the number of k is 3, corresponding to for: Population Density and collect the average distance between residence and work It should be noted that population density The population density data for the region is in units of 100 people per square kilometer, and the average distance between residence and work is collected. The unit is hundred meters, and the above data are all obtained through the data processing unit; Public transport service indicators It should be noted that, in this embodiment, it is obtained by integrating the public transportation coverage rate and service quality index of the staff, which is specifically an integer from 1 to 10; It should also be noted that the corresponding Also for , and , In this embodiment, the least square method can be used to obtain the regression coefficient by minimizing the sum of squares of the error terms, and the obtained value can be , and .

[0022] As an optional embodiment: the working method of the interactive analysis module also includes: It should be noted that the traffic carbon emissions in the area will also be affected by the weather. In bad weather, many people will choose to travel by private car or taxi, which will also affect the traffic carbon emissions. This technical solution can comprehensively consider the above influences to give a specific predicted traffic carbon emissions; According to the formula , calculate and obtain the optimized predicted traffic carbon emissions in the area tomorrow ,in is the influence coefficient of weather factors on the choice of travel mode, is the weather impact index. It should be noted that, in this embodiment, the impact coefficient The weather impact index is obtained through historical data evaluation. In this embodiment, the value is 1.023. It is obtained by staff based on tomorrow's weather forecast evaluation, and the specific value range is 0.5 to 1.5, and is 1.204 in this embodiment.

[0023] As an optional embodiment: the planning module includes a prediction unit and a planning unit, and the specific working mode of the prediction unit is as follows: According to the formula , calculate and obtain the predicted traffic carbon emissions of the area for the day after tomorrow ,in is the rate of change based on the time trend, is the seasonal change rate; it should be noted that and It can be calculated through historical data. In this embodiment, the values ​​are 1.023 and 0.945 respectively. It should also be noted that the predicted traffic carbon emissions for the next day are obtained by integrating the change rate of time and season through the data of the current day and tomorrow. ; Based on the obtained forecast of traffic carbon emissions in the area for the next day The predicted traffic carbon emissions for each day of the month are calculated by the method; it should be noted that the calculations are performed in sequence according to the above scheme; Set a threshold for traffic carbon emissions in advance, compare the predicted traffic carbon emissions for each day of the month with the threshold, and obtain the number of days above the threshold. , the number of days below the threshold Transfer to the planning unit.

[0024] As an optional embodiment: the specific working mode of the planning unit is to obtain the existing traffic carbon emission planning scheme, which is as follows: Including transportation structure, encouragement of new energy vehicles and energy efficiency of transportation vehicles; If the number of days above the threshold If the pollution level is less than 2, it is marked as low pollution. If the pollution level is above the threshold If the pollution level is greater than 2 and less than 5, it is marked as moderate pollution. If it is greater than 5, it is marked as high pollution; Adjust transportation carbon emission planning schemes based on pollution results.

[0025] As an optional embodiment: the specific working method of the planning unit also includes: If it is low pollution, maintain the current transportation carbon emission planning scheme; If the pollution is moderate, strengthen the promotion policy of new energy vehicles, increase subsidies, expand the scope of tax incentives, implement stricter energy efficiency standards, encourage the use of low-carbon transportation, optimize the transportation structure, and increase investment in public transportation and non-motorized transportation; If the pollution is high, emergency measures will be implemented, such as restricting the use of high-emission vehicles, optimizing the traffic signal system to reduce congestion, expanding incentive policies for new energy vehicles, which may include car purchase subsidies, free parking, free charging, etc., increasing investment in public transportation systems, and improving service frequency and coverage.

[0026] How it works It can increase the accuracy of carbon emission assessment, reflect the real-time changes in current traffic conditions and individual behaviors, and combine micro-behavior prediction and multi-factor interaction analysis. The system can predict changes in traffic behavior and its potential impact on carbon emissions, providing a scientific basis for urban planning and policy making. The real-time monitoring and prediction results provided by the system can help traffic managers optimize traffic flow, reduce congestion, and thus reduce carbon emissions.

[0027] The above are only preferred implementations of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical staff in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of this template.

Claims

1. The carbon emission evaluation system based on machine learning is characterized by: include: The data processing module is used to collect real-time traffic data and output real-time carbon emission evaluation, specifically: Set an evaluation area in advance; according to the formula , calculate and obtain the daily traffic carbon emissions within the area , where n is the type of activity level, is the data of the ith activity level, is the emission factor corresponding to the ith activity level data; A micro-behavior prediction module, which is used to analyze individual travel patterns according to the data from the data processing module, and feed the prediction results back to the data processing module to obtain a preliminary predicted carbon emission evaluation; An interactive analysis module, the interactive analysis module is used to integrate the data collected by the data processing module, analyze the impact on traffic carbon emissions based on the integrated data collected by the data processing module, and output a carbon emission impact analysis; The planning module is used to perform long-term trend analysis based on the output data of the above three modules and output future carbon emission forecasts and planning suggestions.

2. The carbon emission evaluation system based on machine learning according to claim 1 is characterized in that: The n is 2, corresponding to, Vehicle mileage and fuel consumption ; The first correlation factor and the second related factor , where the first correlation factor The specific way to obtain is as follows: According to the formula , calculate and obtain the first correlation factor ,in is the consumption of the fossil fuel in the ith unit, in million kilojoules, is the carbon dioxide emission factor of the ith fossil fuel, in tons of carbon dioxide per million kilojoules.

3. The carbon emission evaluation system based on machine learning according to claim 2 is characterized in that: The specific working mode of the micro-behavior prediction module is as follows: Get time features , spatial characteristics , Vehicle characteristics and user behavior characteristics ; According to the formula , get the prediction results ,in Represents the jth feature, specifically the time feature , spatial characteristics , Vehicle characteristics and user behavior characteristics Any of the above, m represents the number of features, and m is 4, is the decision threshold corresponding to the feature, Indicates the predicted category under current conditions; Prediction results The percentage of people's travel modes in the area tomorrow; The predicted results Feedback to the data processing module.

4. The carbon emission evaluation system based on machine learning according to claim 3 is characterized in that: The specific working method of the data processing module after obtaining the prediction result of the micro-behavior prediction module is as follows: Get the percentage of people's travel mode in the area on that day; According to the prediction results The difference between the percentage of the travel mode of the people in the area tomorrow and the percentage of the travel mode of the people in the area today is calculated to obtain the motor vehicle travel ratio ; According to the formula , get the predicted traffic carbon emissions value of the area ,in is the preset proportional coefficient; The predicted carbon emissions from transportation in this area Transmitted to the interaction analysis module.

5. The carbon emission assessment system based on machine learning according to claim 1 is characterized in that: The interaction analysis module works as follows: According to the formula , calculate and obtain the impact value of carbon emissions in the area ,in It is The regression coefficients of the influencing factors are It is The actual value of the influencing factor, is the number of influencing factors; According to the formula , calculate and obtain the adjusted forecast value of tomorrow's traffic carbon emissions .

6. The carbon emission assessment system based on machine learning according to claim 5 is characterized in that: Said The number of is 3, corresponding to is: population density and collect the average distance between residence and work ; Public transport service indicators .

7. The carbon emission assessment system based on machine learning according to claim 6 is characterized in that: The working method of the interaction analysis module also includes: According to the formula , calculate and obtain the optimized predicted traffic carbon emissions in the area tomorrow ,in is the influence coefficient of weather factors on the choice of travel mode, It is the weather impact index.

8. The carbon emission assessment system based on machine learning according to claim 5, characterized in that: The planning module includes a prediction unit and a planning unit. The specific working mode of the prediction unit is as follows: According to the formula , calculate and obtain the predicted traffic carbon emissions of the area for the day after tomorrow ,in is the rate of change based on the time trend, is the rate of change based on season; Based on the obtained forecast of traffic carbon emissions in the area for the next day The predicted traffic carbon emissions for each day of the month are calculated by this method; Set a threshold for traffic carbon emissions in advance, compare the predicted traffic carbon emissions for each day of the month with the threshold, and obtain the number of days above the threshold. , the number of days below the threshold Transfer to the planning unit.

9. The carbon emission assessment system based on machine learning according to claim 8, characterized in that: The specific working method of the planning unit is to obtain the existing traffic carbon emission planning scheme, as follows: Including transportation structure, encouragement of new energy vehicles and energy efficiency of transportation vehicles; If the number of days above the threshold If the pollution level is less than 2, it is marked as low pollution. If the pollution level is above the threshold If the pollution level is greater than 2 and less than 5, it is marked as moderate pollution. If it is greater than 5, it is marked as high pollution; Adjust transportation carbon emission planning schemes based on pollution results.

10. The carbon emission assessment system based on machine learning according to claim 9, characterized in that: The specific working method of the planning unit also includes: If it is low pollution, maintain the current transportation carbon emission planning scheme; If the pollution is moderate, strengthen the promotion policy of new energy vehicles, increase subsidies, expand the scope of tax incentives, implement stricter energy efficiency standards, encourage the use of low-carbon transportation, optimize the transportation structure, and increase investment in public transportation and non-motorized transportation; If the pollution is high, emergency measures will be implemented, such as restricting the use of high-emission vehicles, optimizing the traffic signal system to reduce congestion, expanding incentive policies for new energy vehicles, which may include car purchase subsidies, free parking, free charging, etc., increasing investment in public transportation systems, and improving service frequency and coverage.

Images