A method for calculating atmospheric pollutant concentration based on urban traffic simulation

Abstract

The present invention discloses a method for calculating the concentration of atmospheric pollutants based on urban traffic simulation. First, field road network data is obtained, and a two-dimensional map is constructed according to the obtained data; urban motor vehicle data is obtained and deployed on the generated map for traffic simulation; the motor vehicle simulation data of the traffic simulation and the average emission factors of local urban vehicles are obtained, the road emission line sources are calculated, the pollutant concentrations of the roads are calculated using a diffusion model, a correlation model between the pollutant concentration and the urban air quality is obtained through classification model training, and the pollutant concentrations of each road in the city are estimated using the model with real vehicle data. The present invention can calculate the concentration of atmospheric pollutants generated due to motor vehicle exhaust emissions under specific traffic conditions through a traffic simulation system. And it can guide the adjustment of traffic plans according to the atmospheric pollutant concentrations obtained from the simulation to maximize the reduction of the atmospheric pollution degree caused by motor vehicle exhaust emissions.

Description

Technical Field

[0001] The present invention relates to the fields of environmental protection and traffic management, and in particular to a method for calculating atmospheric pollutant concentration based on urban traffic simulation. Background Art

[0002] The main components of motor vehicle exhaust emissions are CO, HC, NO x , SO2, PM, photochemical smog, etc. At present, motor vehicle exhaust emissions have become the main local atmospheric pollution sources in large and medium-sized cities. Motor vehicle emissions of pollutants have become one of the main sources of urban air pollution. The contribution ratio of mobile sources to fine particulate matter (PM 2.5 ) in local emission sources in each place is as Figure 1 shown, and since the pollutants generated by motor vehicles are ultra-low altitude emissions, the pollutant sharing rate of motor vehicles will be higher for urban roads.

[0003] Currently, most of the detection schemes for atmospheric pollutant concentration in the industry are to detect the current pollutant concentration in real time through some atmospheric pollutant detection devices. The advantage of real-time detection is that the data is relatively accurate, but it is very difficult to estimate the pollutant concentration at future moments. Moreover, the pollutants detected by the detection device include various sources such as industrial waste gas, domestic cooking fume, and waste gas emitted by motor vehicles. It is still relatively difficult to refine to a certain pollution source such as motor vehicle exhaust emissions. Summary of the Invention

[0004] The purpose of the present invention is to propose a method for calculating atmospheric pollutant concentration based on urban traffic simulation in view of the deficiencies of the prior art. It is mainly to alleviate the environmental problems brought about by the increasingly crowded urban traffic in recent years, use the urban traffic simulation system to reflect the pollutant concentration caused by motor vehicle emissions at a macroscopic level, and use the air quality detection data for processing to further improve the data accuracy and achieve the idea of predicting the future environment. System simulation is a third research method in addition to theoretical derivation and experiments. Traffic simulation aims to realize the imaginary scheme of urban road management under the computer simulation in the case of the uncertainty of theoretical derivation and the excessive experimental cost or the inability to conduct experiments at all due to the complexity of traffic conditions, so as to make the most correct adjustment or optimization in real traffic management, greatly simplifying the experimental steps and saving the experimental cost.

[0005] The purpose of the present invention is achieved through the following technical solutions: A method for calculating atmospheric pollutant concentration based on urban traffic simulation, the specific steps are as follows:

[0006] Step (1): Obtain field road network data; the road network data mainly includes map data and traffic signal data;

[0007] Step (2): Construct a two-dimensional map based on the data obtained in step (1); simulate the real road using a three-layer road network structure of lanes, edges, and roads, and simulate real intersections using junctions;

[0008] Step (3): Obtain urban motor vehicle data, including motor vehicle information data and motor vehicle flow data;

[0009] Step (4): Based on the traffic simulation system, deploy the data obtained in step (3) on the map generated in step (2) for traffic simulation. The data inputs of the traffic simulation system include user demand parameters, the generated road network, traffic signal data, the departure and destination data of each motor vehicle. The path of each trip is obtained through the Dijkstra algorithm based on the starting point and ending point of each trip, and finally, the path is mapped to the road network through the fuzzy logic algorithm; define a car-following model and a lane-changing model to control the behavior of motor vehicles; at the same time, define a timer task that repeats every several milliseconds to calculate and generate a simulation frame to keep the simulation going. In each simulation frame, it is necessary to update the status of traffic signals, generate the vehicles that appear in the current frame, destroy the vehicles that need to disappear in the current frame, obtain the longitude and latitude coordinates, speed, acceleration, driver intention, driver behavior, distance to the intersection, and distance to the front and rear vehicles of each vehicle in the current simulation frame, calculate the longitude and latitude coordinates of each vehicle in the next frame of simulation, and infer the driver intention and driver behavior of the next frame based on the distance to the intersection and the distance to the front and rear vehicles. Finally, refresh the vehicle status and display the vehicles on the road network according to the longitude and latitude coordinates of the vehicles.

[0010] Step (5): Obtain the number of vehicles, the average speed of each vehicle, the acceleration of each vehicle, the fuel type of each vehicle, and the road length data on different roads in the traffic simulation system in step (4);

[0011] Step (6): Obtain the average emission factor of local urban vehicles. Specifically, look up the pollutant emission rate limits of vehicles in different national standards, calculate the average emission factor according to the proportion of local motor vehicles in different standards, or look up the fleet average emission factor provided by the local traffic police, compare the two factors, select one as the benchmark according to human needs, and then use the total correction factor CF to correct the emission factor. The correction formula is as follows:

[0012]

[0013] EF′ w = EF w × CF

[0014] In the formula: is the average speed correction factor, is the temperature correction factor, is the humidity correction factor, is the altitude correction factor, is the fuel correction factor, is the deterioration correction factor, is the load correction factor; EF w is the emission factor of emission type w, EF′ w is the corrected emission factor of emission type w;

[0015] Step (7): Calculate the line source strength of road emission line sources according to the data in Step (5) and Step (6). The formula for the line source strength is as follows.

[0016] Q iw = q i × l × EF′ iw

[0017]

[0018] In the formula: Q iw is the line source strength of vehicle emission type w of a certain line source road type i; q i is the traffic flow of vehicles of type i on the road; l is the road length; EF′ iw is the emission factor of vehicle emission type w of type i; Q w is the total line source strength of vehicle emission type w on a certain line source road, that is, the pollutant emission per unit time on a certain road; n is the number of motor vehicle types;

[0019] Step (8): According to the results in Step (7), use the Gaussian diffusion model to estimate the air pollution concentration generated by motor vehicle exhaust emissions. The formula is as follows:

[0020]

[0021] In the formula: C w (x,y,z) is the pollutant concentration at coordinates (x,y,z); H is the effective height of the exhaust stack; U is the average wind speed at the outlet of the exhaust stack; y is the coordinate perpendicular to the X-axis on the horizontal plane; Z is the coordinate in the vertical direction (ground elevation); δ y , δ z are the diffusion parameters in the horizontal transverse and vertical directions;

[0022] Step (9): Obtain the data of local environmental monitoring points and microwave detectors around the road to obtain the urban air quality detection data.

[0023] Step (10): Obtain the correlation model between pollutant concentration and urban air quality through classification model training based on the data in Step (8) and Step (9), and obtain the pollutant concentration corrected based on the urban air quality detection data according to the model.

[0024] Step (11): Estimate the pollutant concentration of each road in the city using the model obtained in Step (10) with the actual statistical or simulated vehicle data.

[0025] Further, in Step (1), the map data includes road data and intersection data; the road data includes road length, road speed limit, number of road lanes, road attributes, road width, and road facility data; the intersection data includes the number of approach lanes at the intersection, the number of departure lanes at the intersection, and the road data connected to the intersection.

[0026] Further, in Step (1), the traffic signal data includes traffic signal groups and traffic signal cycle data.

[0027] Further, in Step (2), the section Edge is divided into two types: Internal Edge and Normal Edge. Normal Edge is the section between intersections; Internal Edge refers to the driving route inside the intersection. The data included in the section Edge are: section speed limit, section attributes, section length, section width, included lanes, and included devices;

[0028] The lane Lane is a smaller unit than the section. One lane cannot span two sections and can only be arranged side by side inside the section. The data included are: lane speed limit, lane length, lane width, the section to which the lane belongs, and included devices;

[0029] The road Road is a specific concept, at a higher level than Edge, and is reflected as the name of the road;

[0030] Further, in Step (2), the intersection Junction is the point where two or more directed roads intersect and is used to connect adjacent edges; the data included are: the coordinates of the intersection center point, the sections entering the intersection, and the sections leaving the intersection.

[0031] Further, the traffic signal TrafficLight at the intersection belongs to the devices included in the section, and the data included are: signal cycle and the lanes controlled by the signal.

[0032] Further, in step (3), the motor vehicle information data includes: motor vehicle type, maximum acceleration of the motor vehicle, maximum speed of the motor vehicle, maximum deceleration of the motor vehicle, length of the motor vehicle, minimum distance at rest of the motor vehicle, driver reaction time, driver personality, and motor vehicle distance to the destination data;

[0033] The motor vehicle flow data includes: total traffic volume in each time period, standard vehicle traffic volume, average point traffic volume, morning and evening peak periods, travel volume and regression volume of traffic zones, and origin and destination data of each vehicle.

[0034] Further, in step (4), the car-following model is the Intelligent Driver Model (IDM), which describes that a rational driver, on the premise of maintaining a safe distance from the vehicle in front during driving and when the environment permits, will increase the speed to a certain feasible range;

[0035] Further, in step (4), the lane-changing model is divided into mandatory lane-changing and non-mandatory lane-changing. Mandatory lane-changing is the lane-changing that must be completed, and non-mandatory lane-changing is the lane-changing that needs to reach the destination quickly; the two types of lane-changing are mainly mandatory lane-changing, and non-mandatory lane-changing is only considered when there is no requirement for mandatory lane-changing.

[0036] Further, in step (10), a correlation model is established between the obtained urban air quality detection data and the calculated concentrations of various pollutants, and the Pearson correlation coefficient method or a classification algorithm in machine learning is used for correlation analysis, so that the results obtained through traffic simulation are more in line with the actual situation.

[0037] The beneficial effects of the present invention: The present invention proposes a method for calculating the concentration of atmospheric pollutants based on urban traffic simulation, which can set the corresponding pollutant emissions according to the attributes such as speed, acceleration, vehicle type, and displacement of the traffic simulation system, and can predict the concentration of atmospheric pollutants generated by motor vehicle exhaust emissions under specific traffic conditions through the traffic simulation system. And it can guide the adjustment of traffic plans according to the concentration of atmospheric pollutants obtained from the simulation to maximize the reduction of air pollution caused by motor vehicle exhaust emissions. Description of the Drawings

[0038] Figure 1 It is a schematic diagram of the contribution of mobile sources to fine particulate matter in local emission sources in various places;

[0039] Figure 2 It is a flowchart of the method of the present invention;

[0040] Figure 3 It is a schematic diagram of the three-layer structure of the road network;

[0041] Figure 4This is a traffic simulation effect diagram. DETAILED DESCRIPTION

[0042] The specific implementation modes of the present invention are further described in detail below with reference to the accompanying drawings.

[0043] The present invention provides a method for calculating the concentration of atmospheric pollutants based on urban traffic simulation, such as Figure 2 As shown, the specific process is as follows:

[0044] Step (1): Obtain actual road network data. Road network data mainly includes map data and traffic light data.

[0045] Map data includes: road data and intersection data.

[0046] Road data includes: road length, road speed limit, number of road lanes, road attributes, road width, and road facilities;

[0047] Intersection data includes: the number of lanes at the intersection entrance, the number of lanes at the intersection exit, and the roads connected to the intersection.

[0048] Traffic light data includes: traffic light group and traffic light cycle.

[0049] Step (2): Construct a two-dimensional map based on the data obtained in step (1).

[0050] A three-layer road network structure of lanes, edges and roads is used to simulate real roads, and junctions are used to simulate real intersections.

[0051] There are two types of road edge: Internal Edge and Normal Edge. Normal Edge (hereinafter referred to as Normal) is the road section in the usual sense, that is, the road section between intersections; while Internal Edge (hereinafter referred to as Internal) specifically refers to the driving route inside the intersection. The two have a lot in common. The only difference is that Normal can contain multiple lanes, while Internal has only one lane. This is done to simplify the processing logic inside the intersection. The data included are: road section speed limit, road section attributes, road section length, road section width, included lanes, and included equipment.

[0052] Lane is a smaller unit than road section. A lane cannot cross two road sections and can only be arranged in parallel inside the road section. The difference between Normal and Internal at the Edge level is unified at the Lane level. The data included are: lane speed limit, lane length, lane width, road section to which the lane belongs, and the equipment included.

[0053] A Road is a concrete concept, hierarchically above an Edge, and is a concept similar to a complete road such as "Renmin Road" and "Jiefang Road".

[0054] A Junction is a point where two or more directed roads intersect and is used to connect adjacent edges. The data it contains are: the coordinates of the center point of the junction, the road segments entering the junction, and the road segments leaving the junction. The three-layer structure diagram of the road network is as Figure 3 shown.

[0055] The traffic lights at a junction belong to the equipment included in a road segment. The data it contains are: the signal cycle and the lanes controlled by the signal lights.

[0056] Step (3): Obtain urban motor vehicle data.

[0057] Motor vehicle data includes: motor vehicle information data and motor vehicle flow data.

[0058] Motor vehicle information data includes: motor vehicle type, maximum acceleration of the motor vehicle, maximum speed of the motor vehicle, maximum deceleration of the motor vehicle, length of the motor vehicle, minimum spacing when the motor vehicle is stationary, driver reaction time, driver personality, distance of the motor vehicle from the destination;

[0059] Motor vehicle flow data includes: total traffic volume, standard vehicle traffic volume, average point traffic volume, morning and evening peak hours, trip volume and regression volume of traffic zones, and the departure and destination of each vehicle.

[0060] Step (4): Deploy the data obtained in step (3) on the map generated in step (2) for traffic simulation.

[0061] Building a traffic simulation system requires data input, including user demand parameters, generated road network, traffic signal data, departure and destination of each motor vehicle, and then generating a path through the departure and destination. The specific implementation steps are: obtaining the path of each trip through the Dijkstra algorithm based on the starting point and ending point of each trip, and finally mapping this path to the road network through the fuzzy logic algorithm.

[0062] Define a timer task that repeats the calculation to generate a simulation frame every certain number of milliseconds (set by user parameters) to keep the simulation running continuously. In each simulation frame, it is necessary to update the status of traffic lights, generate the vehicles that appear in the current frame, destroy the vehicles that need to disappear in the current frame, obtain the longitude and latitude coordinates, speed, acceleration, driver intention, driver behavior, distance to the intersection, distance to the vehicle in front and behind, etc. of each vehicle in the current simulation frame, calculate the longitude and latitude coordinates of each vehicle in the next frame of the simulation, and infer the driver intention and driver behavior of the next frame based on the distance to the intersection and the distance to the vehicle in front and behind, etc. Finally, refresh the vehicle status and transfer the vehicle data to the front-end page according to the longitude and latitude coordinates of the vehicle. The simulation effect diagram is as shown in Figure 4 shown.

[0063] Define two types of models, namely Car Following Models (CFM) and Lane Change Models (LCM), to control the behavior of motor vehicles.

[0064] The car-following model includes: free driving speed, following driving speed, speed approaching the destination, speed when the vehicle brakes, vehicle safety distance, vehicle observation distance, etc. The Intelligent Driver Model (IDM) is a classic dynamic car-following model, which describes that a rational driver will increase the speed to a certain feasible range when maintaining a safe distance from the vehicle in front and when the environment permits during the driving process.

[0065] There is currently no unified standard for the lane change model. In the present invention, the lane change model is divided into mandatory lane change and non-mandatory lane change. Mandatory lane change is a lane change that must be completed, otherwise an abnormality will occur in the subsequent simulation; non-mandatory lane change is a lane change that needs to reach the destination quickly. The two types of lane changes are mainly mandatory lane changes, and non-mandatory lane changes are only considered when there is no requirement for mandatory lane changes.

[0066] Step (5): Obtain the motor vehicle simulation data of the traffic simulation in step (4).

[0067] By calculation, obtain the number of vehicles on different roads in the simulation system, the average speed of each vehicle, the acceleration of each vehicle, the fuel type of each vehicle, the road length, etc. These data of pollutant concentration have a certain impact on the types of pollutants emitted by vehicles, the speed of pollutant emission, and the total amount of pollutant emission. For example, the main emissions of diesel vehicles are NO xand PM; the main emissions of gasoline vehicles are CO and HC. The emissions generated by motor vehicles when driving at low speeds are higher than those when driving at high speeds. The number of road vehicles indicates the degree of road congestion, and the more congested the road, the more emissions will be generated, and so on.

[0068] Step (6): Obtain the average emission factors of local urban vehicles.

[0069] Consult the pollutant emission rate limits of vehicles under different national standards, calculate the average emission factors according to the proportion of local motor vehicles in different standards, or obtain the average emission factors of the vehicle fleet provided by the local traffic police. Compare the two factors and select one as the benchmark according to human needs. Then, through the quantification of temperature, humidity, average speed, altitude, and emission deterioration (due to increased driving mileage and engine carbon deposition), sulfur emissions, and diesel vehicle load estimation, use the total correction factor CF to correct the emission factor EF. The correction formula is as follows.

[0070]

[0071] In the formula: CF is the product of each correction factor, is the average speed correction factor; is the temperature correction factor; is the humidity correction factor; is the altitude correction factor; is the fuel correction factor; is the deterioration correction factor; is the load correction factor.

[0072] EF′ w = EF w × CF

[0073] In the formula: EF w is the emission factor of emission type w, CF is the total correction factor, and EF′ w is the corrected emission factor of emission type w.

[0074] Step (7): Calculate the line source strength of road emission sources based on the data in Steps (5) and (6).

[0075] According to the urban motor vehicle emission air pollution measurement method issued by the State Environmental Protection Administration, calculate the data obtained from the traffic simulation system and the emission factor benchmark to obtain the line source strength of road pollution sources. The line source strength calculation formula is as follows.

[0076] Q iw = q i × l × EF′ iw

[0077] Where: Q iw is the line source strength of vehicle emission pollutant type w for a certain line source road type i, g / h; q i is the traffic flow of vehicles of type i on the road, vehicles / h; l is the road length, km; EF′ iw is the emission factor of vehicle emission pollutant type w for vehicles of type i, g / km;

[0078]

[0079] Where: Q w is the total line source strength of pollutant type w emitted on a certain line source road, g / h; Q iw is the line source strength of vehicle emission pollutant type w for a certain line source road type i; n is the number of motor vehicle types.

[0080] Step (8): According to the results in step (7), use appropriate meteorological data and diffusion model to calculate the pollutant concentration of the road.

[0081] The diffusion model can estimate the air pollution concentration generated by motor vehicles from motor vehicle exhaust emissions. The present invention uses the Gaussian diffusion model to estimate the concentration of each pollutant. The parameters that need to be provided are: emission factor reference, wind speed, atmospheric diffusion parameters, effective source height, etc. The formula is shown as follows.

[0082]

[0083] Where: C w (x, y, z) is the pollutant concentration at the point (x, y, z), mg / m 3 ; Q w is the line source strength, the emission amount of pollutants per unit time, mg / s; H is the effective height of the exhaust stack, m; U is the average wind speed at the outlet of the exhaust stack, m; y is the coordinate perpendicular to the X-axis on the horizontal plane, m; Z is the coordinate in the vertical direction (ground elevation), m; δ y 、δ z are the diffusion parameters in the horizontal transverse and vertical directions, m.

[0084] Step (9): Obtain the urban air quality detection data, specifically: obtain the data of local environmental detection points and microwave detectors around the road.

[0085] Step (10): Through classification model training based on the data in step (8) and step (9), obtain the correlation model between pollutant concentration and urban air quality, and obtain the pollutant concentration corrected based on the urban air quality detection data according to the model.

[0086] The obtained urban air quality detection data is correlated with the pollutant concentrations calculated by the present invention to establish a correlation model. The Pearson correlation coefficient method or the classification algorithm in machine learning can be used for correlation analysis, so that the results obtained through traffic simulation are more in line with the actual situation.

[0087] The Pearson correlation coefficient method is mainly used to analyze ordinal data. Based on the data nature of the pollutant concentrations calculated by the present invention, the Pearson correlation coefficient can be tried as a research parameter, and its formula is shown as follows:

[0088]

[0089] In the formula: X is the factor affecting the air pollutant concentration, and the possible influencing factors are: temperature, humidity, pressure, wind level, average speed of vehicles, total number of vehicles. Y is the air pollutant concentration. are the weighted averages of X and Y respectively.

[0090] The method of machine learning training can also be used to correct the results. Since there are many data influencing factors in the present invention, the principal component analysis (PCA) method can be used to reduce the dimension of the data, and then the support vector machine model is used for classification training.

[0091] Step (11): Use the model obtained in step (10) to estimate the pollutant concentrations of each road in the city with the real or simulated vehicle data, and present them in the traffic simulation system to display the pollutant concentrations of the roads in real time.

[0092] Example:

[0093] The present invention takes Liuzhou City as the research object to explain the process of the present invention as follows:

[0094] In step (1), satellite maps and map data provided by the Liuzhou Municipal Department are used to draw the map, and all road data and traffic signal data are stored in xml format.

[0095] In step (2), the data obtained in step (1) is used for modeling to implement a two-dimensional map of the three-layer structure of the Lane, Edge, and Road road network in the main urban area of Liuzhou, and the intersection traffic lights are configured as road devices.

[0096] In step (3), the data provided in the Liuzhou traffic special investigation report is used to establish traffic zones, and different travel volumes and regression volumes are assigned to each traffic zone.

[0097] Step (4) Deploy the data obtained in step (3) on the generated map in step (2), and start the simulation system to perform simulation for a certain period of time.

[0098] Step (5) Obtain the data during the entire simulation process, such as the length of each lane, the number of vehicles in each lane, the average speed of vehicles in each lane, etc.

[0099] Step (6) Directly obtain the average emission factors under different standards. The average emission factors of the National IV vehicle fleet are shown in Table 1.

[0100] Table 1 Average Emission Factors of the National IV Vehicle Fleet (g / km)

[0101]

[0102]

[0103] Step (7) Calculate the pollutant line source strengths of different vehicle types in the road respectively, and add up the calculated results to obtain the pollutant line source strengths generated by all motor vehicles on the entire road.

[0104] Step (8) Calculate the pollutant concentration of the road according to the result in step (7).

[0105] Step (9) Obtain the urban air quality detection data published by the local ecological environment department.

[0106] Step (10) According to the data in step (8) and step (9), obtain the correlation model between the pollutant concentration and the urban air quality by the Pearson correlation coefficient method, and display the calculated pollutant concentration on the simulation interface according to the model, so that the entire visualization method has better accuracy.

[0107] Step (11) Estimate the pollutant concentrations of each road in Liuzhou using the model obtained in step (10) with real or simulated vehicle data, and classify the road pollutant concentrations, and present them in the traffic simulation system related to Liuzhou.

[0108] The above embodiments are used to explain the present invention, rather than limit the present invention. Any modifications and changes made within the spirit and scope of the claims of the present invention fall within the protection scope of the present invention.

Claims

1. A method for calculating atmospheric pollutant concentration based on urban traffic simulation, characterized in that, The specific steps are as follows: Step (1): Obtain the field road network data; the road network data includes map data and traffic signal data; Step (2): Construct a two-dimensional map based on the data obtained in step (1); Adopt a three-layer road network structure of lane, edge, and road to simulate the real road, and adopt a junction to simulate the real intersection; Step (3): Obtain urban motor vehicle data, including motor vehicle information data and motor vehicle flow data; Step (4): Based on the traffic simulation system, deploy the data obtained in step (3) on the map generated in step (2) for traffic simulation. The data input of the traffic simulation system includes user demand parameters, the generated road network, traffic signal data, the departure and destination data of each motor vehicle. The path of each trip is obtained through the Dijkstra algorithm based on the starting point and ending point of each trip, and finally, the path is mapped to the road network through the fuzzy logic algorithm; define a car-following model and a lane-changing model to control the behavior of motor vehicles; at the same time, define a timer task to repeat the calculation every several milliseconds to generate a simulation frame to keep the simulation going. In each simulation frame, it is necessary to update the status of traffic signals, generate the vehicles that appear in the current frame, destroy the vehicles that need to disappear in the current frame, obtain the latitude and longitude coordinates, speed, acceleration, driver intention, driver behavior, distance to the intersection, and distance to the front and rear vehicles of each vehicle in the current simulation frame, calculate the latitude and longitude coordinates of each vehicle in the next frame of simulation, and infer the driver intention and driver behavior of the next frame based on the distance to the intersection and the distance to the front and rear vehicles data. Finally, refresh the vehicle status and display the vehicle on the road network according to the latitude and longitude coordinates of the vehicle; Step (5): Obtain the number of vehicles, the average speed of each vehicle, the acceleration of each vehicle, the fuel type of each vehicle, and the road length data on different roads in the traffic simulation system in step (4); Step (6): Obtain the average emission factor of local urban vehicles. Specifically, look up the pollutant emission rate limits of vehicles in different national standards, calculate the average emission factor according to the proportion of local motor vehicle numbers in different standards, or look up the fleet average emission factor provided by the local traffic police, compare the two factors, select one as the benchmark according to human needs, and then use the total correction factor CF to correct the emission factor. The correction formula is as follows: EF′ w = EF w × CF Wherein: is the average speed correction factor, is the temperature correction factor, is the humidity correction factor, is the altitude correction factor, is the fuel correction factor, is the deterioration correction factor, is the load correction factor; EF w is the emission factor for the emission type w, EF′ w is the corrected emission factor for the emission type w; Step (7): Calculate the line source strength of road emissions according to the data in steps (5) and (6). The line source strength calculation formula is as follows: Q iw = q i × l × EF′ iw Where: Q iw is the line source strength of vehicle emissions of pollutant type w for a line source road of type i; q i is the traffic flow of vehicles of type i on the road; l is the road length; EF′ iw is the emission factor of vehicle emissions of pollutant type w for vehicles of type i; Q w is the total line source strength of vehicle emissions of pollutant type w on a certain line source road, that is, the emission amount of pollutants per unit time on a certain road; n is the number of motor vehicle types; Step (8): According to the result in step (7), use the Gaussian diffusion model to estimate the air pollution concentration generated by motor vehicle exhaust emissions. The formula is as follows: Where: C w (x, y, z) is the pollutant concentration at the coordinates (x, y, z); H is the effective height of the exhaust stack; U is the average wind speed at the outlet of the exhaust stack; y is the coordinate perpendicular to the X-axis on the horizontal plane; Z is the coordinate in the vertical direction (ground elevation); δ y , δ z are the diffusion parameters in the horizontal transverse and vertical directions; Step (9): Obtain the data of local environmental monitoring points and microwave detectors around the road to obtain urban air quality detection data; Step (10): Obtain the correlation model between pollutant concentration and urban air quality through training with the data in Step (8) and Step (9), and obtain the pollutant concentration corrected based on the urban air quality detection data according to the model. Step (11): Estimate the pollutant concentration of each road in the city using the model obtained in Step (10) for the actual statistical or simulated vehicle data.

2. The method for calculating the concentration of atmospheric pollutants based on urban traffic simulation according to claim 1, wherein, In Step (1), the map data includes road data and intersection data; the road data includes road length, road speed limit, number of road lanes, road attributes, road width, and road facility data; the intersection data includes the number of approach lanes at the intersection, the number of departure lanes at the intersection, and the road data connected to the intersection.

3. A method for calculating the concentration of air pollutants based on urban traffic simulation according to claim 1, characterized in that, In Step (1), the traffic signal data includes traffic signal groups and traffic signal cycle data.

4. A method for calculating atmospheric pollutant concentration based on urban traffic simulation according to claim 1, characterized in that, In Step (2), the road segment Edge is divided into two types: Internal Edge and Normal Edge. Normal Edge is the road segment between intersections; Internal Edge refers to the driving route inside the intersection. The data included in the road segment Edge are: segment speed limit, segment attributes, segment length, segment width, included lanes, and included devices. The lane Lane is a smaller unit than the road segment. One lane cannot span two road segments and can only be arranged side by side inside the road segment. The data included are: lane speed limit, lane length, lane width, the road segment to which the lane belongs, and included devices. The road Road is a specific concept, at a higher level than Edge, and is reflected as the name of the road.

5. A method for calculating atmospheric pollutant concentration based on urban traffic simulation according to claim 4, characterized in that In Step (2), the intersection Junction is the point where two or more directed roads intersect, used to connect adjacent edges; the data included are: the coordinates of the intersection center point, the road segments entering the intersection, and the road segments leaving the intersection.

6. The method for calculating the concentration of air pollutants based on urban traffic simulation according to claim 5, wherein The traffic signal TrafficLight at the intersection belongs to the devices included in the road segment, and the data included are: signal cycle and the lanes controlled by the signal.

7. A method for calculating the concentration of atmospheric pollutants based on urban traffic simulation according to claim 1, characterized in that, In Step (3), the motor vehicle information data includes: motor vehicle type, maximum acceleration of the motor vehicle, maximum speed of the motor vehicle, maximum deceleration of the motor vehicle, motor vehicle length, minimum distance when the motor vehicle is stationary, driver reaction time, driver personality, and motor vehicle distance to the destination data. The motor vehicle flow data includes: the total traffic volume in each time period, the standard vehicle traffic volume, the average point traffic volume, the morning and evening peak periods, the travel volume and regression volume of traffic zones, and the departure place and destination data of each vehicle.

8. A method for calculating atmospheric pollutant concentration based on urban traffic simulation according to claim 1, characterized in that In Step (4), the car-following model is the Intelligent Driver Model (IDM), which describes that a rational driver, on the premise of maintaining a safe distance from the vehicle in front during driving and when the environment permits, will increase the speed to a certain feasible range.

9. A method for calculating the concentration of atmospheric pollutants based on urban traffic simulation according to claim 1, characterized in that, In step (4), the lane-changing model is divided into mandatory lane-changing and non-mandatory lane-changing. Mandatory lane-changing is the lane-changing that must be completed, and non-mandatory lane-changing is the lane-changing that needs to reach the destination quickly. The two types of lane-changing are mainly mandatory lane-changing, and non-mandatory lane-changing is only considered when there is no requirement for mandatory lane-changing.

10. A method for calculating atmospheric pollutant concentration based on urban traffic simulation according to claim 1, characterized in that, In step (10), a correlation model is established between the obtained urban air quality detection data and the calculated concentrations of various pollutants. The Pearson correlation coefficient method or a classification algorithm in machine learning is used for correlation analysis, so that the results obtained through traffic simulation are more in line with the real situation.

Images