A method for preparing a dynamic emission inventory of urban road fugitive dust based on road dust load walk-by monitoring

Abstract

The application provides a preparation method of a city road dust dynamic emission inventory based on road dust load walk monitoring, which comprises the following steps: performing walk monitoring on road dust by establishing a vehicle-mounted particulate matter monitoring device calibration model and a road dust load model, and preparing a city road dust inventory according to monitoring data. Through the two models established by the application, the humidity of different regions and the processing of abnormal data in the walk are considered, and the obtained result is more accurate than that of a traditional method. Moreover, the preparation of the city road dust inventory is simpler, the updating speed is faster, the spatial resolution is higher, and the result can be specific to each road in the city. The application can solve the shortcomings of the traditional method, such as complex inventory preparation, high labor cost, low updating frequency, inability to distinguish the emission characteristics of time and space, and only annual overall city road dust emission inventory.

Description

Technical Field

[0001] This invention relates to the field of air pollution control technology, and in particular to a method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load. Background Technology

[0002] In recent years, air pollution has become increasingly severe. Haze pollution, caused by increased concentrations of atmospheric particulate matter, not only reduces visibility and causes significant inconvenience to people's lives and work, but also directly impacts human health. The main sources of atmospheric particulate matter include industrial emissions, vehicle exhaust emissions, biomass combustion, and road dust emissions. Studies indicate that with the increasing number of vehicles in cities and the rapid pace of road construction, road dust has become a major source of urban atmospheric particulate matter.

[0003] Therefore, rapid monitoring of road environmental pollutants is fundamental to refined urban air pollution control. However, urban roads are widely distributed, making it difficult to monitor them using fixed-point air quality monitoring stations. Currently, the main methods for monitoring road dust include the dustfall method and the AP-42 method. The dustfall method has a long sampling cycle, is easily affected by meteorological factors, and is prone to mixing with atmospheric particulate matter from other sources in the dustfall tank, leading to an overestimation of road dust emissions. The AP-42 method requires sampling at certain intervals on actual road surfaces, which is time-consuming, labor-intensive, and costly. Furthermore, the parameters in its emission coefficients are determined based on US test results, which are not suitable for my country's national conditions.

[0004] Furthermore, with the rise of sensor technology, vehicle-mounted mobile monitoring technology for road dust load has emerged. Based on a vehicle-mounted particulate matter detector, it obtains the environmental background concentration through a sampling port on the vehicle roof and the concentration of road dust emitted by vehicles through sampling ports at the tires under the vehicle. The road dust load is then calculated using empirical formulas. The advantages of mobile monitoring of road dust load are: ① rapid detection and simple operation; ② continuous monitoring of roads. However, current research both domestically and internationally has explored very little about the data quality and application forms of mobile monitoring of road dust load. Most cities are simply applying it blindly, lacking scientifically sound methods and evaluation. The large amounts of data monitored are only analyzed superficially, lacking in-depth research and analysis. It can only be used to detect the relative distribution characteristics of urban road dust and cannot be used to calculate road dust emissions or compile road dust emission inventories.

[0005] Road dust emission inventories are a crucial foundation for government departments to formulate various control plans and conduct environmental management. Domestic road dust emission inventory studies are generally categorized by road type, and the emission coefficients use the nationally standardized "Technical Guidelines for Compiling Particulate Matter Emission Inventories from Dust Sources." However, significant differences may exist in parameters such as traffic volume, average vehicle weight, and dust load among different road types and sampling seasons within cities. Using the nationally standardized coefficients results in inaccurate calculations and lacks dynamic updating capabilities, only providing data from the previous year.

[0006] Therefore, the method proposed in this invention improves the accuracy of road dust load obtained by mobile monitoring; and based on the improved mobile monitoring method for road dust load, a new method for preparing dynamic emission inventories of urban road dust is proposed, expanding the application scope of the mobile monitoring method for road dust load. Summary of the Invention

[0007] This invention provides a method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load. The results obtained by this invention are more accurate than those of traditional methods, and the preparation of the urban road dust inventory is simpler, faster, and has higher spatial resolution. It can be specific to each road in the city, which can solve the shortcomings of traditional methods, such as complex inventory preparation, high labor costs, low update frequency, inability to distinguish emission characteristics in time and space, and the inability to produce an overall urban emission inventory.

[0008] The specific technical solution is as follows: A method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load includes: (1) Use standard air station particulate matter monitoring equipment and mobile vehicle particulate matter monitoring equipment to monitor the concentration of atmospheric particulate matter at the same location, and obtain atmospheric particulate matter concentration data and meteorological data; (2) Using the data from step (1) as input, construct a particulate matter concentration calibration model suitable for mobile vehicle-mounted particulate matter monitoring equipment under different meteorological conditions; (3) Use mobile monitoring vehicles to monitor various roads in the city, obtain raw particulate matter concentration data and mobile speed data, and input the obtained raw particulate matter concentration data into the particulate matter concentration calibration model in step (2) to obtain calibrated particulate matter concentration data. (4) Using the calibrated particulate matter concentration data and driving speed corresponding to each road in the city obtained in step (3), as well as the measured road dust load data, as input, a road dust load model is constructed; (5) Use a mobile monitoring vehicle to conduct real-time mobile monitoring of the urban roads under test, obtain the original particulate matter concentration data and meteorological data of the urban roads under test, input them into the particulate matter concentration calibration model obtained in step (2), obtain the calibrated particulate matter concentration data, and then input them into the road dust load model obtained in step (4) to obtain the road dust load corresponding to the urban roads under test. (6) Collect road dust load data of the urban roads to be tested, calculate road dust data, and establish a dynamic emission inventory of urban road dust.

[0009] Further, in step (1), the atmospheric particulate matter concentration data includes: PM 2.5 Particulate matter concentration data, PM 10 Particulate matter concentration data; the meteorological data includes: temperature data, humidity data, wind speed data, and atmospheric pressure data.

[0010] Furthermore, the particulate matter concentration data and meteorological data in step (1) are divided into a training set and a test set. The training set is used to input and construct a particulate matter concentration calibration model suitable for mobile vehicle-mounted particulate matter monitoring equipment under different meteorological conditions. The test set is used to test and verify the improvement of the constructed calibration model.

[0011] Further, in step (2), the method for constructing the particulate matter concentration calibration model includes: (2-1) Using the meteorological data obtained in step (1) as input, the key meteorological factors are identified by using a decision tree model analysis; (2-2) Using the key meteorological factors obtained in step (2-1) as classification criteria, the K-prototype model is used to perform cluster analysis on the atmospheric particulate matter concentration data obtained in step (1) to obtain a classification set of particulate matter concentration data based on different numerical ranges of key meteorological factors. (2-3) Fit the classification set corresponding to each numerical interval of the key meteorological factors to obtain the calibration coefficient between the particulate matter concentration on the mobile monitoring vehicle and the particulate matter concentration at the standard air station. Based on the calibration coefficient, establish the calibration formula to obtain the calibration model of the particulate matter concentration on the mobile monitoring vehicle.

[0012] Further, in step (2-1), the method for identifying key meteorological factors is as follows: using the particulate matter data obtained from a standard air station particulate matter monitoring device as a standard, the key meteorological factors affecting the deviation of the particulate matter data obtained from the vehicle-mounted particulate matter monitoring device are obtained. The meteorological data is used as the independent variable, and the vehicle-mounted PM2.5 data is then analyzed. 2.5 Particulate matter concentration values ​​and PM 2.5 The difference in particulate matter concentration values ​​between standard stations, vehicle-mounted PM2.5 concentrations 10 Particulate matter concentration values ​​and PM 10The difference in particulate matter concentration values ​​at standard stations is used as the dependent variable for decision tree modeling, outputting a decision number structure diagram and a feature weight table to derive key factors; In step (2-1), the key meteorological factor identified is relative humidity; in step (2-2), the different numerical ranges of relative humidity are divided into three ranges: low, medium, and high, which are 10%-68%, 69%-85%, and 86%-100%, respectively. Furthermore, in step (2-3), the calibration formula is as follows: Lower range: y = x + a; (1) In equation (1), x represents the vehicle-mounted PM2.5 concentration obtained by the vehicle-mounted particulate matter concentration monitoring equipment. 2.5 or PM 10 Particulate matter concentration value; y represents the calibrated vehicle PM2.5 concentration value. 2.5 or PM 10 Particulate matter concentration value; 'a' is the calibration coefficient; Middle section: y=x+b RH (2) In equation (2), RH is the relative humidity value; b is the calibration coefficient; and x represents the PM2.5 concentration obtained by the mobile vehicle-mounted particulate matter concentration monitoring equipment. 2.5 or PM 10 Particulate matter concentration value; y represents the calibrated vehicle PM corresponding to x. 2.5 or PM 10 Particulate matter concentration values; High range: y1=x1+e1×RH 3 +e2×RH 2 +e3×RH+e4;(3) In equation (3), RH represents the relative humidity; x1 represents the onboard PM2.5 concentration obtained by the mobile vehicle particulate matter concentration monitoring equipment. 2.5 Particulate matter concentration or PM 10 Particulate matter concentration value; y1 represents the calibrated vehicle PM corresponding to x1. 2.5 Particulate matter concentration or PM 10 Particulate matter concentration values; e1, e2, e3, and e4 are calibration coefficients; Furthermore, in step (2), the improved calibration model is tested and verified using test set data. By comparing the relative deviation values ​​of the data without partitioning and the data with partitioning, the relative deviation value of the data without partitioning is 13.16%-23.27%, while the relative deviation value of the data with partitioning is only 7.28%. It is believed that the new calibration model has improved the data calibration.

[0013] Furthermore, in step (4), the method for determining the measured data of road dust load is as follows: (4-1) Select testing points for each urban road; (4-2) The detection points were measured using the vacuum dust collection method to obtain the measured data of road dust load; The training set for the road dust load model includes: measured road dust load data, calibrated particulate matter concentration data and travel speed obtained by a mobile monitoring vehicle, all three data points being obtained at the same detection point.

[0014] Furthermore, the training set of the road dust load model also includes an estimation dataset; The method for obtaining the estimation dataset is as follows: (4-3) Establish a correlation formula between the original particulate matter concentration data and the navigation speed data; (4-4) Based on the correlation formula, obtain the original particulate matter concentration data corresponding to different cruising speeds to obtain the original particulate matter concentration estimation dataset; (4-5) Based on the particulate matter concentration calibration model, the original particulate matter concentration estimation dataset is transformed into a calibrated particulate matter concentration estimation dataset. (4-6) The measured data of road dust load obtained at the same detection point and the calibrated particulate matter concentration estimation dataset are combined to obtain the estimation dataset.

[0015] y1=b1×V c1 (4) In equation (4), V is the driving speed; b1 and c1 are coefficients; and y1 is the estimated vehicle-mounted PM. 2.5 or PM 10 Particulate matter concentration values; Furthermore, in step (4), the formula for the road dust load model is: SL = a5 × (m × y) PM2.5 +n×y PM10 ) / (V g (5) In equation (5), SL is the road dust load value; y PM2.5 The vehicle-mounted PM after calibration using the aforementioned calibration formula 2.5 Particulate matter concentration value; y PM10 For vehicle-mounted PM after calibration using the calibration formula 10 Particulate matter concentration value; V is the cruise speed value; a5 is a constant; m is PM2.5 concentration. 2.5 Weighting coefficients for particulate matter concentration numerical indicators; n represents PM2.5 concentration. 10 Weighting coefficients for particulate matter concentration numerical indicators; g is the velocity-related constant; the formula is as follows: g=m×c1PM10 +n×c1 PM2.5 (6) In equation (6), m represents PM. 2.5 Weighting coefficients for particulate matter concentration numerical indicators; n represents PM2.5 concentration. 10 Weighting coefficient for particulate matter concentration numerical index; c1 PM2.5 For PM in equation (4) 2.5 The corresponding c1 value; c1 PM10 For PM in equation (4) 10 The corresponding c1 value; Furthermore, the obtained models are evaluated for their merits using a single indicator, vehicle-mounted PM. 2.5 or PM 10 The formulas for numerical particulate matter concentration, the coefficient of determination for comprehensive index formulas, and the relative deviation value for calculating road dust accumulation data are used to evaluate the model's performance. The specific formulas are as follows: Single vehicle PM 2.5 or PM 10 Formula for particulate matter concentration: SL = a6 × y2 / (V g1 (7) In equation (7), SL is the road dust accumulation value, and y2 is the calibrated vehicle-mounted PM2.5 concentration. 2.5 or PM 10 The particulate matter concentration value, V is the value of the cruise speed, a6 is a constant, obtained by fitting a power function; g1 is the velocity-related constant, g1 is the value of c1 in formula (4); The formula for the comprehensive index is shown in equation (5); Furthermore, in step (5), the original particulate matter concentration data is processed by removing abnormal data before being input into the particulate matter concentration calibration model; The abnormal data includes data from the mobile monitoring vehicle in stationary, accelerating, and turning states; the operating status of the mobile monitoring vehicle is obtained using the following formula: Parking status: (8) Acceleration state: (9) Turning position: >10 (10) T represents the driving state, k represents the number of seconds, when T is 0 in equation (8) it represents the parking state, when T is 0 in equation (9) it represents the parking state. The acceleration state is represented by T > 10 in equation (10), which represents the turning state; X in equations (8) and (10) represents the turning state. k Let X be the longitude data at the k-th second. k+1 For the longitude data at the (k+1)th second, Y k For the latitude data at the k-th second, Y k+1 For the latitude data at the (k+1)th second; V in equation (9)k V represents the driving speed data at the k-th second. k+1 For the driving speed data at the (k+1)th second, t k For the time data at the kth second, Y k+1 This is the time data for the (k+1)th second.

[0016] Furthermore, in step (6), the steps for establishing the dynamic emission inventory of urban road dust are as follows: (6-1) Based on the road dust load values ​​calculated above and the latitude and longitude data obtained by the positioning system, use software to draw a road dust load map for each road in the city; (6-2) Using the road dust load map of each road in the city obtained in step (6-1), and in combination with the requirements for inventory compilation, construct an urban road dust emission inventory; (6-3) Display the dust emission situation of various roads in the city through the platform.

[0017] When establishing a dynamic emission inventory of dust from urban roads, GIS software is used to draw high-resolution road dust maps, and in accordance with the requirements of the "Technical Guidelines for Compiling Particulate Emission Inventories of Dust Sources (Trial)" (Announcement No. 92 of the Ministry of Environmental Protection in 2014), a high-resolution emission inventory of dust from urban roads is constructed. The road dust inventory equals the sum of all road dust emissions in the surveyed area; the "Technical Guidelines for Compiling Particulate Matter Emission Inventories from Dust Sources" provides the following formula for calculating road dust emissions: (11) In equation (11), W Ri Total emissions of road dust, unit: tons / year (t / a); K i SL represents the particle size multiplier of the generated dust; SL is the road dust load, in grams per square meter (g / m²). 2 W represents the average vehicle weight, in tons (t). The average vehicle weight indicates the average weight of all vehicles passing through the road. L R N represents the length of each road, in km; R Average traffic volume on the road, unit: vehicles / year (vehicles / year).

[0018] In this invention It provides the total emissions of road dust sources for each road on a weekly or monthly basis, rather than the overall annual emissions data, thus offering higher temporal resolution. K i The measured particle size multiplier (i.e., the PM2.5 value calibrated after measurement by the mobile monitoring vehicle equipment). 2.5 Concentration data / PM 10(Concentration data), not the reference data recommended in Table 5 of the technical guidelines; SL is the dust load data of each road after actual measurement and calibration, not the nationally unified reference coefficient, and has higher spatial resolution.

[0019] Finally, GIS software was used to display high-resolution data on urban road dust emissions.

[0020] Compared with the prior art, the present invention has the following beneficial effects: (1) The method provided by the present invention conducts mobile monitoring of road dust by establishing a calibration model of vehicle-mounted particulate matter monitoring equipment and a model of vehicle-mounted particulate matter monitoring equipment concentration and road dust load, and prepares an inventory of urban road dust based on the monitoring data; by establishing two models, the humidity of different regions and the processing of abnormal data in mobile monitoring are considered, and the results obtained are more accurate than those of traditional methods.

[0021] (2) This invention proposes a calibration algorithm model applicable to vehicle-mounted particulate matter detectors, which improves the accuracy of obtaining road dust load through mobile monitoring. Based on the improved mobile monitoring method for road dust load, a new method for preparing dynamic emission inventory of urban road dust is proposed, which expands the application scope of the mobile monitoring method for road dust load.

[0022] (3) The method provided by the present invention is faster and more frequently updated than existing methods in preparing the list of urban road dust, and can accurately obtain different road sections in each road in the city. It can solve the shortcomings of traditional methods, such as slow list preparation, low update frequency and only applicable to the whole city.

[0023] (4) Due to the relative humidity, research on road dust collection equipment is generally focused on the dry northern regions, and there is very little research and calibration work for the high humidity in the south. Therefore, the calibration algorithm of this invention is more suitable for the high humidity in the south. Attached Figure Description

[0024] Figure 1 This is a flowchart of the present invention.

[0025] Figure 2 The calibration flowchart is for an application example.

[0026] Figure 3 The diagram shown is a decision tree structure diagram for application purposes. The content within the boxes in the diagram does not affect the present invention.

[0027] Figure 4 This is a plot showing the percentage of cluster analysis results for application examples; where cluster_1 (relative humidity 86%-100%), cluster_2 (relative humidity 10%-68%), and cluster_3 (relative humidity 69%-85%). Detailed Implementation

[0028] The present invention will be further described below with reference to specific embodiments. The following are only specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto.

[0029] Example A method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load includes: (1) Use standard air station particulate matter monitoring equipment and mobile vehicle particulate matter monitoring equipment to monitor the concentration of atmospheric particulate matter at the same location, and obtain atmospheric particulate matter concentration data and meteorological data; Atmospheric particulate matter concentration data include: PM2.5 2.5 Particulate matter concentration data, PM 10 Particulate matter concentration data; the meteorological data includes: temperature data, humidity data, wind speed data, and atmospheric pressure data. The same location means that the mobile particulate matter monitoring equipment is placed in the same area as the standard air station particulate matter monitoring equipment for data monitoring. The particulate matter concentration data and meteorological data are divided into training set and test set. The training set is used to input and construct a particulate matter concentration calibration model suitable for the mobile particulate matter monitoring equipment under different meteorological conditions. The test set is used to test and verify the improvement of the constructed calibration model.

[0030] (2) Using the data from step (1) as input, construct a particulate matter concentration calibration model suitable for mobile vehicle-mounted particulate matter monitoring equipment under different meteorological conditions; Using the meteorological data obtained in step (1) as input, a decision tree model is used for analysis. Taking the particulate matter data obtained from a standard air station particulate matter monitoring device as the standard, the key meteorological factors affecting the deviation of the particulate matter data obtained from the vehicle-mounted particulate matter monitoring device are identified. Meteorological data are used as independent variables, and the vehicle-mounted PM2.5 data is analyzed separately. 2.5 Particulate matter concentration values ​​and PM 2.5 The difference in particulate matter concentration values ​​between standard stations, vehicle-mounted PM2.5 concentrations 10 Particulate matter concentration values ​​and PM 10 The difference in particulate matter concentration values ​​at standard stations is used as the dependent variable for decision tree modeling, outputting a decision number structure diagram and a feature weight table to derive key factors; Using the obtained key meteorological factors as classification criteria, the K-prototype model is used to perform cluster analysis on the atmospheric particulate matter concentration data obtained in step (1) to obtain a classification set of particulate matter concentration data based on different numerical ranges of key meteorological factors. For each numerical range of key meteorological factors, the corresponding classification set is fitted with linear regression equation, quadratic curve equation and nonlinear regression equation respectively to obtain the corresponding fitting formula. The correlation between the vehicle-mounted particulate matter concentration and the particulate matter concentration of the standard air station is established to obtain the calibration model of the vehicle-mounted particulate matter monitoring equipment. The key meteorological factor identified was relative humidity; the different numerical ranges of the key meteorological factor were divided into three ranges: low, medium, and high, namely: 10%-68%, 69%-85%, and 86%-100%, respectively; the calibration formula is as follows: Lower range: y = x + a; (1) In equation (1), x represents the vehicle-mounted PM2.5 concentration obtained by the vehicle-mounted particulate matter concentration monitoring equipment. 2.5 or PM 10 Particulate matter concentration value; y represents the calibrated vehicle PM2.5 concentration value. 2.5 or PM 10 Particulate matter concentration value; 'a' is the calibration coefficient; Middle section: y=x+b RH (2) In equation (2), RH is the relative humidity value; b is the calibration coefficient; and x represents the PM2.5 concentration obtained by the mobile vehicle-mounted particulate matter concentration monitoring equipment. 2.5 or PM 10 Particulate matter concentration value; y represents the calibrated vehicle PM corresponding to x. 2.5 or PM 10 Particulate matter concentration values; High range: y1=x1+e1×RH 3 +e2×RH 2 +e3×RH+e4;(3) In equation (3), RH represents the relative humidity; x1 represents the onboard PM2.5 concentration obtained by the mobile vehicle particulate matter concentration monitoring equipment. 2.5 Particulate matter concentration or PM 10 Particulate matter concentration value; y1 represents the calibrated vehicle PM corresponding to x1. 2.5 Particulate matter concentration or PM 10 Particulate matter concentration values; e1, e2, e3, and e4 are calibration coefficients; The improved calibration model was tested and verified using test set data. The relative deviation values ​​of the non-partitioned and partitioned fitting data were compared to verify the improvement. The relative deviation value of the non-partitioned data was 13.16%-23.27%, while the relative deviation value of the partitioned fitting data was only 7.28%. It is believed that the new calibration model has improved the data calibration.

[0031] (3) Use mobile monitoring vehicles to monitor various roads in the city, obtain raw particulate matter concentration data and mobile speed data, and input the obtained raw particulate matter concentration data into the particulate matter concentration calibration model in step (2) to obtain calibrated particulate matter concentration data. The present invention can use the mobile monitoring vehicle disclosed in the utility model patent with authorization announcement number CN 218726431 U for mobile monitoring.

[0032] The raw particulate matter concentration data is uncalibrated and processed data obtained from the vehicle-mounted particulate matter monitoring equipment, while the travel speed data is the speed data that corresponds one-to-one with the vehicle-mounted particulate matter monitoring equipment.

[0033] (4) Using the calibrated particulate matter concentration data and driving speed corresponding to each road in the city obtained in step (3), as well as the measured road dust load data, as input, a road dust load model is constructed; The method for determining the measured data of road dust load is as follows: (4-1) Select testing points for each urban road; (4-2) The vacuum dust collection method was used to measure the detection points and obtain the measured data of road dust load; The training set for the road dust load model includes: measured road dust load data, calibrated particulate matter concentration data and travel speed obtained by a mobile monitoring vehicle, all three data points being obtained at the same detection point.

[0034] The training set for the road dust load model also includes the estimation dataset; The method for obtaining the estimation dataset is as follows: (4-3) Establish a correlation formula between the original particulate matter concentration data and the navigation speed data; (4-4) Based on the correlation formula, the original particulate matter concentration data corresponding to different cruising speeds are obtained, and the original particulate matter concentration estimation dataset is obtained. (4-5) Based on the particulate matter concentration calibration model, the original particulate matter concentration estimation dataset is transformed into a calibrated particulate matter concentration estimation dataset. (4-6) The measured data of road dust load obtained at the same detection point and the calibrated particulate matter concentration estimation dataset are combined to obtain the estimation dataset.

[0035] The correlation formula in step (4-3) is: y1=b1×V c1 (4) In equation (4), V is the driving speed; b1 and c1 are coefficients, and y1 is the estimated vehicle-mounted PM. 2.5 or PM 10 Particulate matter concentration values.

[0036] In step (4-3), dust pollution tests are conducted at various vehicle speeds, with multiple tests performed at each speed on the same road section. A dataset is created and grouped, with each test result at different speeds forming a group. Vehicle particulate matter concentration data is then matched to each group based on speed, and PM2.5 concentration is calculated by fitting a power function. 2.5 Vehicle particulate matter concentration, PM 10 Formula for vehicle-mounted particulate matter concentration.

[0037] In step (4), the formula for the road dust load model is: SL = a5 × (m × y) PM2.5 +n×y PM10 ) / (V g (5) In equation (5), SL is the road dust load value; y PM2.5 The vehicle-mounted PM after calibration using the aforementioned calibration formula 2.5 Particulate matter concentration value; y PM10 For vehicle-mounted PM after calibration using the calibration formula 10 Particulate matter concentration value; V is the cruise speed value; a5 is a constant; m is PM2.5 concentration. 2.5 Weighting coefficients for particulate matter concentration numerical indicators; n represents PM2.5 concentration. 10 Weighting coefficients for particulate matter concentration numerical indicators; g is the velocity-related constant; the formula is as follows: g=m×c1 PM10 +n×c1 PM2.5 (6) In equation (6), m is PM 2.5 The weighting coefficient for particulate matter concentration numerical index, where n is the PM2.5 concentration. 10 Weighting coefficient of particulate matter concentration numerical index, c1 PM2.5 For PM in equation (4) 2.5 The corresponding c1 value, c1 PM10 For PM in equation (4) 10 The corresponding c1 value.

[0038] (5) Use a mobile monitoring vehicle to conduct real-time mobile monitoring of the urban roads under test, obtain the original particulate matter concentration data and meteorological data of the urban roads under test, input them into the particulate matter concentration calibration model obtained in step (2), obtain the calibrated particulate matter concentration data, and then input the measured original particulate matter concentration data into the road dust load model obtained in step (4) to obtain the road dust load corresponding to the urban roads under test. The raw particulate matter concentration data is processed by removing outlier data before being input into the particulate matter concentration calibration model; The abnormal data includes data from the mobile monitoring vehicle in stationary, accelerating, and turning states; the operating status of the mobile monitoring vehicle is obtained using the following formula: Parking status: (7) Acceleration state: (8) Turning position: >10 (9) T represents the driving state, k represents the number of seconds, when T is 0 in equation (7) it represents the parking state, when T is 0 in equation (8) it represents the parking state. Represents the acceleration state; when T > 10 in equation (9), it represents the turning state; X in equations (7) and (9) k Let X be the longitude data at the k-th second. k+1 For the longitude data at the (k+1)th second, Y k For the latitude data at the k-th second, Y k+1 For the latitude data at the (k+1)th second; V in equation (8) k V represents the driving speed data at the k-th second. k+1 For the driving speed data at the (k+1)th second, t k For the time data at the kth second, Y k+1 This is the time data for the (k+1)th second.

[0039] (6) Collect road dust load data of the urban roads to be tested, calculate road dust data, and establish a dynamic emission inventory of urban road dust.

[0040] The steps for establishing a dynamic emission inventory of urban road dust are as follows: (6-1) Based on the road dust load values ​​calculated above and the latitude and longitude data obtained by the positioning system, use software to draw a road dust load map for each road in the city; (6-2) Using the road dust load map of each road in the city obtained in step (6-1), and in combination with the requirements for inventory compilation, construct an urban road dust emission inventory; (6-3) Display the dust emission situation of various roads in the city through the platform.

[0041] The aforementioned urban roads can refer to any road or any type of road, such as expressways, arterial roads, and secondary arterial roads. When establishing a dynamic dust emission inventory for urban roads, GIS software is used to create a high-resolution road dust map. Furthermore, in accordance with the requirements of the "Technical Guidelines for Compiling Particulate Emission Inventories of Dust Sources (Trial)" (Ministry of Environmental Protection Announcement No. 92 of 2014), a high-resolution urban road dust emission inventory is constructed. The road dust inventory equals the sum of all road dust emissions in the surveyed area; the "Technical Guidelines for Compiling Particulate Matter Emission Inventories from Dust Sources" provides the following formula for calculating road dust emissions: (10) In equation (10), W Ri Total emissions of road dust, unit: tons / year (t / a); K i SL represents the particle size multiplier of the generated dust; SL is the road dust load, in grams per square meter (g / m²). 2 W represents the average vehicle weight, in tons (t). The average vehicle weight indicates the average weight of all vehicles passing through the road. L R N represents the length of each road, in km; R Average traffic volume on the road, unit: vehicles / year (vehicles / year).

[0042] In this invention It provides the total emissions of road dust sources for each road on a weekly or monthly basis, rather than the overall annual emissions data, thus offering higher temporal resolution. K i The measured particle size multiplier (i.e., the PM2.5 value calibrated after measurement by the mobile monitoring vehicle equipment). 2.5 Concentration data / PM 10 (Concentration data), not the reference data recommended in Table 5 of the technical guidelines; SL is the dust load data of each road after actual measurement and calibration, not the nationally unified reference coefficient, and has higher spatial resolution.

[0043] Finally, GIS software was used to display high-resolution data on urban road dust emissions.

[0044] Application examples The method described in this application was applied to the preparation of a dynamic emission inventory of urban road dust in Jiaxing City based on mobile monitoring of road dust load.

[0045] A method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load includes: (1) Use standard air station particulate matter monitoring equipment and mobile vehicle particulate matter monitoring equipment to monitor the concentration of atmospheric particulate matter at the same location, and obtain atmospheric particulate matter concentration data and meteorological data; Atmospheric particulate matter concentration data include: PM2.5 2.5 Particulate matter concentration data, PM 10 Particulate matter concentration data; the meteorological data includes: temperature data, humidity data, wind speed data, and atmospheric pressure data. The same location means that the mobile particulate matter monitoring equipment is placed in the same area as the standard air station particulate matter monitoring equipment for data monitoring. The particulate matter concentration data and meteorological data are divided into training set and test set. The training set is used to input and construct a particulate matter concentration calibration model suitable for the mobile particulate matter monitoring equipment under different meteorological conditions. The test set is used to test and verify the improvement of the constructed calibration model.

[0046] (2) Using the data from step (1) as input, construct a particulate matter concentration calibration model suitable for mobile vehicle-mounted particulate matter monitoring equipment under different meteorological conditions; The calibration procedure is attached. Figure 2 The method for constructing a particulate matter concentration calibration model is as follows: Using the meteorological data obtained in step (1) as input, a decision tree model (model from the Python machine learning library Scikit-learn) is used for analysis. Particulate matter data obtained from a standard air station particulate matter monitoring device is used as the standard to identify the key meteorological factors affecting the deviation of particulate matter data obtained from the vehicle-mounted particulate matter monitoring device. Meteorological data is used as the independent variable, and the vehicle-mounted PM2.5 data is analyzed separately. 2.5 Particulate matter concentration values ​​and PM 2.5 The difference in particulate matter concentration values ​​between standard stations, vehicle-mounted PM2.5 concentrations 10 Particulate matter concentration values ​​and PM 10 The difference in particulate matter concentration values ​​at standard stations was used as the dependent variable for decision tree modeling. The output decision tree structure diagram and feature weight table were used to identify key factors. Relative humidity accounted for 99.55% of the weight, making it the highest-weighted feature and playing a crucial role in model construction. Temperature accounted for 0.24%, atmospheric pressure for 0.19%, and wind speed for 0.01%. Therefore, relative humidity was selected as the key factor. The parameter settings for the decision tree model are shown in Table 1, and the decision tree structure diagram is shown in Appendix 1. Figure 3 ; Table 1 - Model Summary Table

[0047] The characteristic weights of each factor are shown in Table 2. Table 2 Feature Weight Values

[0048] Using the obtained key meteorological factors as classification criteria, the K-prototype model (model source: TheSPSSAU project (2023). SPSSAU. (Version 23.0) [Online Application Software]. Retrieved from https: / / www.spssau.com) was used to perform cluster analysis on the atmospheric particulate matter concentration data obtained in step (1), resulting in a classification set of particulate matter concentration data based on different numerical ranges of key meteorological factors. The different numerical ranges of key meteorological factors were divided into three ranges: low, medium, and high, namely: 10%-68%, 69%-85%, and 86%-100%, respectively; with proportions of 65.93%, 17.58%, and 16.48%, respectively. The cluster analysis results are shown in Table 3, and the percentage of cluster analysis results is shown in the attached chart. Figure 4 Table 3 - Cluster Analysis Data Results

[0049] For each numerical interval of key meteorological factors, the corresponding classification set was fitted using linear regression equations, quadratic curve equations, and nonlinear regression equations respectively (all models are from The SPSSAU project (2023). SPSSAU. (Version 23.0) [Online Application Software]. Retrieved from https: / / www.spssau.com), and the corresponding fitting formulas were obtained. The correlation between the particulate matter concentration on the mobile monitoring vehicle and the particulate matter concentration at the standard air station was established, and the calibration model of the mobile particulate matter monitoring equipment was obtained. Lower range: y = x + 2; coefficient of determination R 2 =0.97 (I) In equation (I), x represents the vehicle-mounted PM2.5 concentration obtained by the mobile vehicle-mounted particulate matter concentration monitoring equipment. 2.5 Particulate matter concentration value; y represents the calibrated vehicle PM corresponding to x. 2.5 Particulate matter concentration values; y = x + 3.5; coefficient of determination R 2 =0.95 (II) In equation (II), x represents the vehicle-mounted PM2.5 concentration obtained by the mobile vehicle-mounted particulate matter concentration monitoring equipment. 10 Particulate matter concentration value; y represents the calibrated vehicle PM corresponding to x. 10 Particulate matter concentration values; Mid-range: Three models were used: linear regression equation, quadratic curve equation, and nonlinear regression equation. The determination coefficients R of the three formulas were compared. 2 , and thus PM 2.5 Linear regression equation for particulate matter concentration numerical deviation, coefficient of determination R 2 =0.677; PM 2.5 The quadratic regression equation for the numerical deviation of particulate matter concentration, with a coefficient of determination R. 2 =0.797; PM 2.5 Nonlinear regression equation for numerical deviation of particulate matter concentration, coefficient of determination R2 =0.90; PM 10 Linear regression equation for particulate matter concentration numerical deviation, coefficient of determination R 2 =0.653; PM 10 The quadratic regression equation for the numerical deviation of particulate matter concentration, with a coefficient of determination R. 2 =0.738; PM 10 Nonlinear regression equation for numerical deviation of particulate matter concentration, coefficient of determination R 2 =0.893; The fitting results are shown in the table below: Table 4 PM 2.5 Linear regression analysis results of particulate matter concentration

[0050] Table 5 PM 10 Linear regression analysis results of particulate matter concentration

[0051] Table 6 PM 2.5 Summary of quadratic regression models for particulate matter concentration values

[0052] Table 7 PM 10 Summary of quadratic regression models for particulate matter concentration values

[0053] Table 8 PM 2.5 Summary of nonlinear regression models for particulate matter concentration

[0054] Table 9 PM 10 Summary of nonlinear regression models for particulate matter concentration

[0055] The coefficient of determination R of the nonlinear regression equation for the relative humidity range of 69%-85% 2 At its highest level, a nonlinear regression equation is chosen to derive PM. 2.5 Vehicle particulate matter concentration, PM 10 Calibration formula for vehicle-mounted particulate matter concentration values; PM is derived from a nonlinear regression equation. 2.5 Vehicle particulate matter concentration, PM 10 Calibration formula for vehicle particulate matter concentration values: y1=x1+0.0576eRH (III) In equation (III), RH represents the relative humidity, and x1 represents the onboard PM2.5 concentration obtained by the mobile vehicle-mounted particulate matter concentration monitoring equipment. 2.5 The particulate matter concentration value, y1 represents the calibrated vehicle PM corresponding to x1. 2.5 Particulate matter concentration value, coefficient of determination R 2 =0.90; y² = x² + 0.0773e RH (IV) In equation (IV), RH represents the relative humidity, and x2 represents the onboard PM2.5 concentration obtained by the mobile vehicle-mounted particulate matter concentration monitoring equipment. 10 The particulate matter concentration value, y2 represents the calibrated PM2.5 concentration corresponding to x2. 10 Particulate matter concentration value, coefficient of determination R 2 =0.89; High interval: Multiple linear regression equations and nonlinear regression equations were used, and the determination coefficients R of the two formulas were compared. 2 , and thus Table 10 PM 2.5 Summary of nonlinear regression models for particulate matter concentration

[0056] Table 11 PM 10 Summary of nonlinear regression models for particulate matter concentration

[0057] Table 12 PM 2.5 Summary of multiple linear regression models for particulate matter concentration

[0058] Table 13 PM 10 Summary of multiple linear regression models for particulate matter concentration

[0059] The coefficient of determination R of the multiple linear regression equation for relative humidity in the range of 86%-100% 2 At its highest, PM was obtained through a multiple linear regression equation. 2.5 Vehicle particulate matter concentration, PM 10 Calibration formula for vehicle particulate matter concentration values: y1=x1+14564RH 3 -41578RH 2 +39573RH-12524; (V) In formula (V), RH represents the relative humidity; x1 represents the PM2.5 concentration obtained by the vehicle-mounted particulate matter concentration monitoring equipment. 2.5 The particulate matter concentration value, y1 represents the calibrated vehicle PM corresponding to x1. 2.5 Particulate matter concentration value, coefficient of determination R 2 =0.93; y2=x2+14664RH 3 -41768RH 2 +37583RH-15527; (VI) In equation (VI), RH represents the relative humidity; x2 represents the PM2.5 concentration obtained by the vehicle-mounted particulate matter concentration monitoring equipment. 10 The particulate matter concentration value, y2 represents the calibrated vehicle PM2.5 concentration value corresponding to x2. 10 Particulate matter concentration value, coefficient of determination R 2 =0.91.

[0060] The improvement of the model was verified by comparing the relative deviation values ​​of the non-partitioned and partitioned data. The final relative deviation values ​​for the non-partitioned data were 13.16%-23.27%, while the relative deviation value for the partitioned data was 7.28%. The model evaluation results are shown in the table below: Table 14 Relative Deviation

[0061] (3) Use mobile monitoring vehicles to monitor various roads in the city, obtain raw particulate matter concentration data and mobile speed data, and input the obtained raw particulate matter concentration data into the particulate matter concentration calibration model in step (2) to obtain calibrated particulate matter concentration data. The raw particulate matter concentration data is the unprocessed data obtained by the vehicle-mounted particulate matter monitoring equipment, and the travel speed data is the speed data that corresponds one-to-one with the vehicle-mounted particulate matter monitoring equipment.

[0062] (4) Using the calibrated particulate matter concentration data and driving speed corresponding to each road in the city obtained in step (3), as well as the measured road dust load data, as input, a road dust load model is constructed; The method for determining the measured data of road dust load is as follows: (4-1) Select testing points for each urban road; (4-2) The vacuum dust collection method was used to measure the detection points and obtain the measured data of road dust load; The training set for the road dust load model includes: measured road dust load data, calibrated particulate matter concentration data and travel speed obtained by a mobile monitoring vehicle, all three data points being obtained at the same detection point.

[0063] (4-3) Establish a correlation formula between the original particulate matter concentration data and the navigation speed data; The correlation formula is: y1=0.5011×V 1.67 (VII) In equation (VII), V represents the driving speed, and y1 is the vehicle-mounted PM value estimated by the formula. 2.5 Particulate matter concentration value, coefficient of determination R 2 =0.91; y1=0.2193×V 2.23 (VIII) In equation (VIII), V represents the driving speed, and y1 is the vehicle-mounted PM value estimated by the formula. 10 Particulate matter concentration value, coefficient of determination R 2 =0.93.

[0064] Dust emissions tests were conducted at various vehicle speeds, with multiple tests performed at each speed on the same road section. A dataset was created and grouped, with each test result at different speeds grouped together. Vehicle particulate matter concentration data was then matched to speed within each group, and PM2.5 concentration was calculated using a power function fit. 2.5 Vehicle particulate matter concentration, PM 10 Formula for vehicle-mounted particulate matter concentration; Dust pollution tests were conducted at various vehicle speeds: 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, and 80 km / h. Each speed was tested four times on the same road segment. A dataset was created and grouped, with each test result at different speeds forming one group, for a total of 60 groups. PM2.5 concentration was calculated by fitting a power function. 2.5 Vehicle particulate matter concentration, PM 10 Formula for vehicle-mounted particulate matter concentration.

[0065] (4-4) Based on the correlation formula, the original particulate matter concentration data corresponding to different cruising speeds are obtained, and the original particulate matter concentration estimation dataset is obtained. (4-5) Based on the particulate matter concentration calibration model, the original particulate matter concentration estimation dataset is transformed into a calibrated particulate matter concentration estimation dataset. (4-6) The measured data of road dust load obtained at the same detection point and the calibrated particulate matter concentration estimation dataset are combined to obtain the estimation dataset.

[0066] PM is calculated using the entropy method (algorithm model source: The SPSSAU project (2023). SPSSAU. (Version 23.0) [Online Application Software]. Retrieved from https: / / www.spssau.com). 2.5 Particulate matter concentration and PM 10 The weighting of particulate matter concentration as a numerical index for road dust load yields the following results: PM 2.5 The particulate matter concentration index has a weight of 32.35%, PM 10 The particulate matter concentration numerical index has a weight of 67.65%, and the specific formula is obtained by fitting a power function. The formula for the road dust load model is: SL = 2.13 × (0.32 × y) PM2.51 +0.67×y PM10 ) / (V 1.83 (IX) In equation (IX), SL represents the road dust accumulation value, and y PM2.5 The vehicle-mounted PM after calibration using the aforementioned calibration formula 2.5 Particulate matter concentration value, y PM10 For vehicle-mounted PM after calibration using the calibration formula 10 Particulate matter concentration value, V is the sailing speed, and the coefficient of determination R 2 =0.91; See Table 15 for specific weight calculation results. Table 15 Weight Calculation Results

[0067] The obtained models were evaluated for their merits using a single indicator, vehicle-mounted PM. 2.5 or PM 10 The formulas for numerical particulate matter concentration, the coefficient of determination for comprehensive index formulas, and the relative deviation value for calculating road dust accumulation data are used to evaluate the model's performance. The specific formulas are as follows: Single vehicle PM 2.5 Formula for particulate matter concentration: SL = 4 × y² / (V 1.67 (X) In equation (X), SL is the road dust accumulation value, and y2 is the calibrated vehicle-mounted PM2.5 concentration. 2.5 Particulate matter concentration value, V is the sailing speed; Single vehicle PM 10 Formula for particulate matter concentration: SL = 15 × y² / (V 2.23 (XI) In equation (XI), SL is the road dust accumulation value, and y2 is the calibrated vehicle-mounted PM2.5 concentration. 10 Particulate matter concentration value, V is the sailing speed; Comprehensive index formula: SL = 2.13 × (0.32 × y) PM2.5 +0.67×y PM10 ) / (V 1.83 (XII) In formula (XII), SL is the road dust accumulation value, and y PM2.5 The vehicle-mounted PM after calibration using the aforementioned calibration formula 2.5 Particulate matter concentration value, y PM10 For vehicle-mounted PM after calibration using the calibration formula 10 Particulate matter concentration value, V is the sailing speed; Single vehicle PM 2.5 Formula for particulate matter concentration numerical index: Coefficient of determination R 2 =0.79, relative deviation 19.43%; Single vehicle PM 10 Formula for particulate matter concentration numerical index: Coefficient of determination R 2 =0.83, relative deviation 17.35%; Comprehensive index formula: Coefficient of determination R 2 =0.91, relative deviation 8.31%; The evaluation results of the road dust load model are shown in Table 16. Table 16 Evaluation Results of Road Dust Load Model

[0068] The results show that the index formula is effective, with a high coefficient of determination and a small relative deviation.

[0069] (5) Use a mobile monitoring vehicle to conduct real-time mobile monitoring of the urban roads under test, obtain the original particulate matter concentration data and meteorological data of the urban roads under test, input them into the particulate matter concentration calibration model obtained in step (2), obtain the calibrated particulate matter concentration data, and then input them into the road dust load model obtained in step (4) to obtain the road dust load corresponding to the urban roads under test. The raw particulate matter concentration data is processed by removing outlier data before being input into the particulate matter concentration calibration model; The abnormal data includes data from the mobile monitoring vehicle in stationary, accelerating, and turning states; the operating status of the mobile monitoring vehicle is obtained using the following formula: Parking status: (XIII) Acceleration state: (XIV) Turning position: >10 (XV) T represents the driving state, k represents the number of seconds, when T is 0 in equation (XIII) it represents the parking state, when T is 0 in equation (XIV) it represents the parking state. Represents the acceleration state; when T > 10 in equation (XV), it represents the turning state; in equations (XIII) and (XV), X... k Let X be the longitude data at the k-th second. k+1 For the longitude data at the (k+1)th second, Y k For the latitude data at the k-th second, Y k+1 V represents the latitude data at the (k+1)th second; in equation (XIV)... k V represents the driving speed data at the k-th second. k+1 For the driving speed data at the (k+1)th second, t k For the time data at the kth second, Y k+1 This is the time data for the (k+1)th second.

[0070] (6) Collect road dust load data of the urban roads to be tested, calculate road dust data, and establish a dynamic emission inventory of urban road dust.

[0071] The steps for establishing a dynamic emission inventory of urban road dust are as follows: (6-1) Based on the road dust load values ​​calculated above and the latitude and longitude data obtained by the positioning system, use software to draw a road dust load map for each road in the city; (6-2) Using the road dust load map of each road in the city obtained in step (6-1), and in combination with the requirements for inventory compilation, construct an urban road dust emission inventory; (6-3) Display the dust emission situation of various roads in the city through the platform.

[0072] When establishing a dynamic emission inventory of dust from urban roads, GIS software is used to draw high-resolution road dust maps, and in accordance with the requirements of the "Technical Guidelines for Compiling Particulate Emission Inventories of Dust Sources (Trial)" (Announcement No. 92 of the Ministry of Environmental Protection in 2014), a high-resolution emission inventory of dust from urban roads is constructed. The road dust inventory equals the sum of all road dust emissions in the surveyed area; the "Technical Guidelines for Compiling Particulate Matter Emission Inventories from Dust Sources" provides the following formula for calculating road dust emissions: (XVI) In formula (XVI), W Ri Total emissions of road dust, unit: tons / year (t / a); K i SL represents the particle size multiplier of the generated dust; SL is the road dust load, in grams per square meter (g / m²). 2W represents the average vehicle weight, in tons (t). The average vehicle weight indicates the average weight of all vehicles passing through the road. L R N represents the length of each road, in km; R Average traffic volume on the road, unit: vehicles / year (vehicles / year).

[0073] In this invention It provides the total emissions of road dust sources for each road on a weekly or monthly basis, rather than the overall annual emissions data, thus offering higher temporal resolution. K i The measured particle size multiplier (i.e., the PM2.5 value calibrated after measurement by the mobile monitoring vehicle equipment). 2.5 Concentration data / PM 10 (Concentration data), not the reference data recommended in Table 5 of the technical guidelines; SL is the dust load data of each road after actual measurement and calibration, not the nationally unified reference coefficient, and has higher spatial resolution.

[0074] Finally, GIS software was used to display high-resolution data on urban road dust emissions.

Claims

1. A method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load, characterized in that, include: (1) Use standard air station particulate matter monitoring equipment and mobile vehicle particulate matter monitoring equipment to monitor the concentration of atmospheric particulate matter at the same location, and obtain atmospheric particulate matter concentration data and meteorological data; (2) Using the data from step (1) as input, construct a particulate matter concentration calibration model suitable for mobile vehicle-mounted particulate matter monitoring equipment under different meteorological conditions; (3) Use mobile monitoring vehicles to monitor various roads in the city, obtain raw particulate matter concentration data and mobile speed data, and input the obtained raw particulate matter concentration data into the particulate matter concentration calibration model in step (2) to obtain calibrated particulate matter concentration data. (4) Using the calibrated particulate matter concentration data and driving speed corresponding to each road in the city obtained in step (3), as well as the measured road dust load data, as input, a road dust load model is constructed; (5) Use a mobile monitoring vehicle to conduct real-time mobile monitoring of the urban roads under test, obtain the original particulate matter concentration data and meteorological data of the urban roads under test, input them into the particulate matter concentration calibration model obtained in step (2), obtain the calibrated particulate matter concentration data, and then input them into the road dust load model obtained in step (4) to obtain the road dust load corresponding to the urban roads under test. (6) Collect road dust load data of the urban roads to be tested, calculate road dust data, and establish a dynamic emission inventory of urban roads dust.

2. The method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load as described in claim 1, characterized in that, In step (1), the atmospheric particulate matter concentration data includes: PM 2.5 Particulate matter concentration data, PM 10 Particulate matter concentration data; the meteorological data includes: temperature data, humidity data, wind speed data, and atmospheric pressure data.

3. The method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load as described in claim 1, characterized in that, In step (2), the method for constructing the particulate matter concentration calibration model includes: (2-1) Using the meteorological data obtained in step (1) as input, the key meteorological factors are identified by using a decision tree model analysis; (2-2) Using the key meteorological factors obtained in step (2-1) as classification criteria, the K-prototype model is used to perform cluster analysis on the atmospheric particulate matter concentration data obtained in step (1) to obtain a classification set of particulate matter concentration data based on different numerical ranges of key meteorological factors. (2-3) Fit the classification set corresponding to each numerical range of key meteorological factors to obtain the calibration coefficient between the mobile monitoring vehicle particulate matter concentration and the standard air station particulate matter concentration. Based on the calibration coefficient, establish the calibration formula to obtain the calibration model of the mobile monitoring vehicle particulate matter concentration.

4. The method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load as described in claim 3, characterized in that, In step (2-1), the key meteorological factor identified is relative humidity; in step (2-2), the different numerical ranges of relative humidity are divided into three ranges: low, medium, and high, which are 10%-68%, 69%-85%, and 86%-100%, respectively. In steps (2-3), the calibration formula is as follows: Lower range: y = x + a; (1) In equation (1), x represents the PM2.5 concentration obtained by the vehicle-mounted particulate matter concentration monitoring equipment. 2.5 or PM 10 The particulate matter concentration value, y represents the calibrated vehicle PM2.5 concentration. 2.5 or PM 10 Particulate matter concentration value, where 'a' is the calibration factor; Middle section: y=x+b RH ;(2) In equation (2), RH is the relative humidity value, b is the calibration coefficient, and x represents the PM2.5 concentration obtained by the mobile vehicle-mounted particulate matter concentration monitoring equipment. 2.5 or PM 10 The particulate matter concentration value, y represents the calibrated vehicle PM2.5 concentration value corresponding to x. 2.5 or PM 10 Particulate matter concentration values; High range: y1=x1+e1×RH 3 +e2×RH 2 +e3×RH+e4;(3) In equation (3), RH represents the relative humidity; x1 represents the onboard PM2.5 concentration obtained by the mobile vehicle particulate matter concentration monitoring equipment. 2.5 Particulate matter concentration or PM 10 The particulate matter concentration value, y1 represents the calibrated vehicle PM corresponding to x1. 2.5 Particulate matter concentration or PM 10 The particulate matter concentration values, e1, e2, e3, and e4 are calibration coefficients.

5. The method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load as described in claim 4, characterized in that, In step (4), the method for determining the measured data of road dust load is as follows: (4-1) Select testing points for each urban road; (4-2) The detection points were measured using the vacuum dust collection method to obtain the measured data of road dust load; The training set for the road dust load model includes: measured road dust load data, calibrated particulate matter concentration data and travel speed obtained by a mobile monitoring vehicle, all three data being obtained at the same detection point.

6. The method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load as described in claim 5, characterized in that, The training set for the road dust load model also includes an estimation dataset; The method for obtaining the estimation dataset is as follows: (4-3) Establish a correlation formula between the original particulate matter concentration data and the navigation speed data; (4-4) Based on the correlation formula, obtain the original particulate matter concentration data corresponding to different cruising speeds to obtain the original particulate matter concentration estimation dataset; (4-5) Based on the particulate matter concentration calibration model, the original particulate matter concentration estimation dataset is transformed into a calibrated particulate matter concentration estimation dataset. (4-6) The measured data of road dust load obtained at the same detection point and the calibrated particulate matter concentration estimation dataset are combined to obtain the estimation dataset.

7. The method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load as described in claim 6, characterized in that, The correlation formula mentioned in step (4-3) is: y1=b1×V c1 (4) In equation (4), V is the driving speed; b1 and c1 are coefficients; and y1 is the estimated vehicle-mounted PM. 2.5 or PM 10 Particulate matter concentration values.

8. The method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load as described in claim 7, characterized in that, In step (4), the formula for the road dust load model is: SL=a5×(m×y PM2.5 +n×y PM10 ) / (V g )(5) In equation (5), SL is the road dust load value; y PM2.5 The vehicle-mounted PM after calibration using the aforementioned calibration formula 2.5 Particulate matter concentration value, y PM10 For vehicle-mounted PM after calibration using the calibration formula 10 Particulate matter concentration value; V is the cruise speed value; a5 is a constant; m is PM2.5 concentration. 2.5 The weighting coefficient for particulate matter concentration numerical index, where n is the PM2.5 concentration. 10 Weighting coefficients for particulate matter concentration numerical indicators; g is a velocity-related constant; The formula for the velocity-related constant is as follows: g=m×c1 PM10 +n×c1 PM2.5 (6) In equation (6), m is PM 2.5 Particulate matter concentration weighting coefficient, where n is the PM2.5 concentration. 10 Particulate matter concentration weighting coefficient, c1 PM2.5 For PM in equation (4) 2.5 The corresponding c1 value, c1 PM10 For PM in equation (4) 10 The corresponding c1 value.

9. The method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load as described in claim 1, characterized in that, In step (5), the original particulate matter concentration data is processed by removing abnormal data before being input into the particulate matter concentration calibration model; The abnormal data includes data from the mobile monitoring vehicle in stationary, accelerating, and turning states; the operating status of the mobile monitoring vehicle is obtained using the following formula: Parking status: (7) Acceleration state: (8) Turning position: >10 (9) T represents the driving state, k represents the number of seconds, when T is 0 in equation (7) it represents the parking state, when T is 0 in equation (8) it represents the parking state. Represents the acceleration state; when T > 10 in equation (9), it represents the turning state; X in equations (7) and (9) k Let X be the longitude data at the k-th second. k+1 For the longitude data at the (k+1)th second, Y k For the latitude data at the k-th second, Y k+1 For the latitude data at the (k+1)th second; V in equation (8) k V represents the driving speed data at the k-th second. k+1 For the driving speed data at the (k+1)th second, t k For the time data at the kth second, Y k+1 This is the time data for the (k+1)th second.

10. The method for preparing a dynamic emission inventory of urban road dust based on mobile monitoring of road dust load as described in claim 1, characterized in that, In step (6), the steps for establishing the dynamic emission inventory of urban road dust are as follows: (6-1) Based on the calculated road dust load and the latitude and longitude data obtained from the positioning system, use software to draw a road dust load map for each road in the city; (6-2) Using the road dust load map of each road in the city obtained in step (6-1), and in combination with the requirements for inventory compilation, construct an urban road dust emission inventory; (6-3) Display the dust emission situation of various roads in the city through the platform.

Images