Truck emission exposure measuring and tracing method, system and device based on resident activity distribution and medium

Abstract

The invention discloses a method, a system, equipment and a medium for calculating and tracing truck emission exposure based on resident activity distribution, and belongs to the technical field of environment and traffic, and the method comprises the steps: building a microscopic air diffusion model based on the emission data of a vehicle, and simulating the concentration distribution of each pollutant after diffusion; capturing activity tracks of residents, considering activity characteristics of urban residents, combining pollutant concentration data with crowd activity distribution information data, and calculating dynamic and static exposure levels of resident pollutants in unit grids within a preset time; and establishing a risk assessment model, assessing and analyzing influence factors of pollutant exposure and an action mechanism of the influence factors, and executing a space treatment strategy. According to the method, the influence of truck emission on resident health is accurately measured and traced by combining a microscopic air diffusion model and mobile phone signaling data. According to the method, the activity track of residents is captured by simulating pollutant concentration distribution, and the dynamic and static exposure levels of the residents are calculated.

Description

Technical Field

[0001] The present invention relates to the field of environment and transportation technology, and in particular to a method, system, equipment and medium for measuring and tracing truck emission exposure based on resident activity distribution. Background Art

[0002] Vehicle emissions are a significant source of atmospheric pollutants and have consistently received significant attention and attention from urban environmental monitoring departments. With the deepening globalization of logistics and the rise of e-commerce, freight demand in major cities continues to grow, significantly increasing the intensity and scope of freight vehicle activity. Diesel combustion in trucks produces significant amounts of NOx, PM, and other conventionally harmful pollutants. Truck emissions, through photochemical reactions, can easily cause smog and significantly impact the human respiratory and cardiovascular systems, becoming a key contributor to major illnesses such as asthma, lung cancer, and heart disease. While heavy-duty vehicle emission standards are being continuously improved, the long-standing accumulation of vehicles with low emission standards, coupled with significant limitations in the application scenarios of new energy heavy-duty truck technologies, means that freight vehicles will continue to be one of the most significant emissions sources in the road transport sector for some time to come.

[0003] The impact of road freight is further magnified in megacities, where demand for freight transport and resident travel is intense and the interactions between these activities are complex. First, due to the distribution of freight demand, truck emissions within cities are highly spatially uneven. Heavy truck emissions typically peak between midday and afternoon, and last for a long time. Furthermore, due to urban land use characteristics, emissions are concentrated in key freight nodes such as ports, freight train stations, major intercity freight corridors, and important industrial areas. To improve productivity, large freight facilities are clustered together, creating negative externalities that impact the surrounding ecological environment and rental prices. Second, population mobility within large cities is high, and residents balance work conditions, economic circumstances, and lifestyle needs to develop different work-residence patterns, resulting in diverse activity patterns. Finally, driven by freight demand, trucks crisscross the city, intersecting with residents' movements. Especially during traffic congestion, trucks idling at high speeds generate high emissions. Pollutants emitted by trucks diffuse from the roadside to residential areas, where they can be inhaled and pose health risks. The social costs of freight transport would need to be reassessed if the public health costs of exposure to pollutants emitted by trucks were taken into account.

[0004] When assessing exposure to truck emissions, it's important to consider residents' activities. In the past, due to the difficulty of quantifying residents' actual activities, static exposure values ​​were typically calculated based on their location, resulting in discrepancies with actual data. However, with the advancement of mobile communications technology, it has become possible to extract precise information on the distribution of people's activities from mobile phone data. Calculating dynamic exposure values ​​that closely approximate actual data represents an improvement and enhancement to existing research assessing the public health impacts of truck emissions.

[0005] Therefore, it is particularly important to provide a method for measuring and tracing truck emission exposure that takes into account the distribution of residents' activities. This aims to reduce truck emissions and resident exposure in major corridors and key areas without affecting cargo transportation demand. This will become a focus of attention for relevant departments in major cities, especially analyzing the exposure levels of truck emission-related pollutants from the perspective of temporal and spatial differentiation, exploring the influencing factors and differences in the dynamic and static exposure values ​​of truck emissions from a spatial perspective, and finally proposing practical and effective spatial governance strategies in combination with urban transportation and land use planning, providing an important foundation for supporting urban sustainable development. Summary of the Invention

[0006] To address these technical issues, a method for estimating and tracing truck emissions exposure based on resident activity distribution was proposed. This method involves building a microscopic air diffusion model based on vehicle emission data to simulate the concentration distribution of each pollutant after diffusion.

[0007] Capturing residents' activity trajectories, considering the activity characteristics of urban residents, combining pollutant concentration data with population activity distribution information data, and calculating the dynamic and static exposure levels of residents to pollutants in the unit grid within a preset time;

[0008] Establish risk assessment models, evaluate and analyze the influencing factors and mechanisms of pollutant exposure, and implement spatial governance strategies.

[0009] As a preferred solution of the truck emission exposure measurement and source tracing method based on resident activity distribution described in the present invention, the simulation of the concentration distribution of each pollutant after diffusion includes:

[0010] The research space is divided into grid cells, each grid cell represents a spatial unit, and the pollutant emissions are diffused;

[0011] The Gaussian atmospheric diffusion model is used to establish a Gaussian plume model, simplifying the emission source to the center of the grid;

[0012] Input the emission source intensity per hour and calculate the pollutant concentration at any point downwind.

[0013] As a preferred solution of the truck emission exposure measurement and source tracing method based on resident activity distribution described in the present invention, the simulation of the concentration distribution of each pollutant after diffusion also includes:

[0014] Taking environmental factors into consideration, for the research space, the concentration of each grid is the superposition of the Gaussian diffusion concentrations of all grids as independent emission sources.

[0015] As a preferred solution of the method for truck emission exposure measurement and tracing based on resident activity distribution described in the present invention, the method of capturing resident activity trajectories includes:

[0016] Based on user anonymous signaling data and population weights, obtain full crowd activity information;

[0017] Divide user status into stay status and travel status;

[0018] Match the location information between each stop to the road network to form a travel record;

[0019] Long-term observation of users' most frequent locations at night and during weekdays to identify their home and work locations;

[0020] The crowd activity information is aggregated and processed in grid space units to obtain detailed crowd activity distribution information every hour.

[0021] As a preferred solution of the truck emission exposure measurement and source tracing method based on resident activity distribution described in the present invention, the calculation of the dynamic and static exposure levels of residents' pollutants in the unit grid within a preset time includes:

[0022] The individual exposure value is calculated based on the product of the individual's residence time in the grid and the grid pollutant concentration during that time period;

[0023] Divide each individual's exposure value into each grid area according to location characteristics, and collect grid-based exposure values;

[0024] The time dimension is introduced to calculate the dynamic exposure value based on the grid, that is, the product of the individual's activity time in the grid during the time period and the grid pollutant concentration;

[0025] The average value was used to assess the exposure level of individuals who had been active in a certain area to take into account the phenomenon of population clustering in urban functional areas.

[0026] As a preferred solution of the truck emission exposure measurement and tracing method based on resident activity distribution described in the present invention, wherein: the establishment of the risk assessment model includes:

[0027] Use OLS regression method to analyze the linear relationship between variables and build a regression model;

[0028] Establish a spatial econometric model, consider the impact of spatial proximity on the relationship between observations, and introduce spatial lag terms and spatial error terms;

[0029] A random forest model was established to rank the importance of the regression results and evaluate the impact of various influencing factors on the average exposure value of individuals in the region.

[0030] As a preferred solution of the truck emission exposure measurement and tracing method based on resident activity distribution described in the present invention, the variables include truck emission factors and resident activity patterns.

[0031] As a preferred solution of the truck emission exposure measurement and traceability system based on resident activity distribution described in the present invention, it is characterized by including: an emission analysis module for building a microscopic air diffusion model based on vehicle emission data to simulate the concentration distribution of each pollutant after diffusion;

[0032] The exposure analysis module is used to capture the activity trajectories of urban residents, consider the activity characteristics of urban residents, combine pollutant concentration data with population activity distribution information data, and calculate the dynamic and static exposure levels of residents in the unit grid within a preset time;

[0033] The spatial governance module is used to establish risk assessment models, evaluate and analyze the influencing factors and mechanisms of pollutant exposure, and implement spatial governance strategies.

[0034] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method for measuring and tracing truck emission exposure based on resident activity distribution are implemented.

[0035] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for measuring and tracing truck emission exposure based on resident activity distribution.

[0036] The beneficial effects of this invention include accurately measuring and tracing the impact of truck emissions on resident health by combining microscopic air diffusion models with mobile phone signaling data. This method simulates pollutant concentration distributions, captures resident movement trajectories, and calculates both dynamic and static exposure levels. Furthermore, multiple models are developed to assess the factors influencing pollutant exposure and their mechanisms of action, and to propose measures to reduce truck pollutant exposure and protect public health. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 This is an overall flow chart of a method for estimating and tracing truck emissions exposure based on resident activity distribution, provided in accordance with one embodiment of the present invention.

[0039] Figure 2 This is a dynamic and static difference map of truck emissions in Shanghai (regional overall exposure) that provides a method for calculating and tracing truck emission exposure based on resident activity distribution, which is provided in one embodiment of the present invention.

[0040] Figure 3 A dynamic and static difference map of truck emissions in Shanghai (weighted average exposure) is provided for a truck emission exposure measurement and traceability method based on resident activity distribution, which is an embodiment of the present invention.

[0041] Figure 4 A schematic diagram of population exposure calculation for an existing method for truck emission exposure measurement and tracing based on resident activity distribution provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0042] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0043] Example 1, with reference to Figure 1 , which is the first embodiment of the present invention, provides a method for estimating and tracing truck emission exposure based on resident activity distribution, including:

[0044] like Figure 1As shown, combined with detailed emission data based on road freight vehicles, a microscopic air diffusion model is built to simulate the concentration distribution of each pollutant after diffusion; secondly, mobile phone signaling data is integrated to accurately capture the residents' activity trajectory, fully consider the activity characteristics of urban residents, and combine pollutant concentration data with residents' activity data to calculate the dynamic and static exposure levels of residents' pollutants in a 24-hour unit grid; on this basis, linear models, spatial econometric models, and machine learning models are established to evaluate and analyze the influencing factors of pollutant exposure and their mechanisms of action; finally, based on the spatial governance strategy for truck emission control, specific measures and suggestions are proposed to reduce truck pollutant exposure and further protect residents' public health. This invention provides a useful reference for the formulation of new vehicle traffic control and land planning guidance policies, provides a more systematic and comprehensive research perspective for urban traffic pollution control, and is an important direction that meets the actual interests of residents and achieves the goal of healthy city development.

[0045] Step 1: Based on emission data from road freight vehicles, a microscopic air diffusion model is constructed to simulate the concentration distribution of each pollutant after diffusion;

[0046] Step 2: Integrate mobile phone signaling data to accurately capture residents' activity trajectories, fully consider the activity characteristics of urban residents, combine pollutant concentration data with population activity distribution information data, and calculate the dynamic and static exposure levels of residents to pollutants in a 24-hour unit grid;

[0047] Step 3: Establish linear models, spatial econometric models, and machine learning models to evaluate and analyze the influencing factors and mechanisms of pollutant exposure;

[0048] Step 4: Based on the spatial governance strategy for truck emission control, propose specific measures and suggestions to reduce exposure to truck pollutants and further protect residents' public health.

[0049] In step 1, the method for processing emission data includes the following steps:

[0050] a1. Divide the study space into grid cells, where each grid cell represents a spatial unit, and perform diffusion processing on pollutant emissions;

[0051] a2. Use the Gaussian atmospheric diffusion model to establish a Gaussian plume model, simplifying the emission source to the center of the grid;

[0052] a3. Input the hourly emission source intensity and calculate the pollutant concentration at any point downwind.

[0053] In the emission data processing process of step one, environmental factors such as atmospheric stability, aerosol characteristics, wind speed, wind direction and wind type are taken into account in the actual calculation; for the research space, the concentration of each grid is the superposition of the Gaussian diffusion concentrations of all grids as independent emission sources.

[0054] In a preferred embodiment of the present invention, the grid is used as the spatial unit to perform diffusion processing on pollutant emissions. In the process of atmospheric environmental impact assessment, atmospheric diffusion calculations are usually based on the Gaussian atmospheric diffusion model. The Gaussian model is a simple and practical atmospheric diffusion model. In a uniform, steady turbulent atmosphere, it is assumed that pollutants exhibit a Gaussian (bell-shaped) distribution of diffusion in the atmosphere. Based on a series of meteorological, topographical, emission source characteristics and atmospheric boundary layer factors, the propagation trajectory and concentration distribution of emission source pollutants in the atmosphere are estimated. A Gaussian plume model is established using the Gaussian atmospheric diffusion model, the emission source is simplified to the center of the grid, the emission source intensity is input every hour, and the pollutant concentration at any point C (x, y, z) downwind is:

[0055]

[0056] Where C represents the concentration of pollutants at the spatial point (x, y, z), in mg / m 3 ;q represents the emission source intensity, unit is mg / s; σ y , σ z It represents the diffusion coefficient in the horizontal and vertical directions, which is related to meteorological conditions and terrain; μ represents the average wind speed, in m / s; x represents the distance from the spatial point to the source on the wind direction axis; y represents the distance from the spatial point to the source in the direction perpendicular to the wind direction axis; z represents the height of the spatial point.

[0057] For the study space, the concentration of each grid is the superposition of the Gaussian diffusion concentrations of all grids as independent emission sources. In the actual calculation, the main environmental factors considered include atmospheric stability, aerosol characteristics, wind speed, wind direction and wind type.

[0058] It should be noted that the difficulty in processing emissions data lies in deducing the emission and diffusion processes of air pollutants. This requires selecting appropriate temporal and spatial granularity, and striking a balance between the difficulty of data acquisition, the complexity of data processing, and the accuracy of the calculation results. Current methods include measurement based on field measurements, estimation based on statistical models, and modeling and simulation based on emission sources.

[0059] In an optional embodiment of the present invention, when the exposure value is measured by actual measurement, the subjects are usually invited to use portable wearable devices to record the pollutant concentration information of the subject's environment. In smaller-scale, specific-site studies, especially in laboratory environments or exposure assessments of individual blocks, equipment measurement has a strong advantage. However, the number of measurement equipment is limited and expensive, requiring high economic investment and professional technical support. On-site measurements of subjects are time-consuming and labor-intensive, and the data samples that can be obtained are limited by the number of samples. There are deviations in the representativeness and universality of the results. Therefore, when the spatial scale of the study is expanded to the city or regional level, the organization and implementation of the actual measurement experiment are highly complex, and the demand for resources increases significantly. Equipment measurement becomes an unrealistic option, and traditional actual measurement methods are no longer applicable.

[0060] When estimating exposure values ​​based on statistical models, concentration data from air monitoring stations in the study area are generally collected and interpolated to obtain concentration values ​​for the entire space. This method has good interpretability in regional pollution studies over large spatial scales and long time spans. However, according to the definition of exposure values, methods based on statistical models completely ignore the factor of exposure duration when measuring exposure values ​​and cannot fully reflect the exposure level.

[0061] In an optional embodiment of the present invention, various diffusion models developed in the field of atmospheric science can be used for modeling and simulation based on emission sources. Examples include the NAQPMS (Nested Air-Quality Prediction Modeling System) model developed by the Institute of Atmospheric Physics, Chinese Academy of Sciences, which is applicable to comprehensive regional scales; the CMAQ (Community Multiscale Air Quality Model), a community multiscale air quality model that uses a three-dimensional Euler model; the ADMS (Advanced Dispersion Modeling System), the AERMOD (AMS / EPA Regulatory Model), the CALPUFF model, and the RLINE model, which is applicable to finite line sources. This method returns to the model foundation of various diffusion models, namely the Gaussian diffusion model. Compared with field-based measurements and statistical model-based estimates, this method reduces economic costs and improves accuracy. Compared with various complex diffusion models, this method adopts an open street setting and does not consider urban morphological factors such as building obstruction, thereby maximizing the simulation spatial scale and computational efficiency.

[0062] In step 2, the method for processing crowd activity distribution information includes the following steps:

[0063] b1. Obtaining full-scale population activity information based on anonymous mobile phone user signaling data and population weights provided by operators;

[0064] b2. User status is divided into stay status and travel status, where stay status is defined as staying in the same location for more than 30 minutes;

[0065] b3. Match the location information between each stop to the road network to form a travel record;

[0066] b4. Long-term observation of the user's most frequent locations at night and during weekdays to identify the user's home and workplace;

[0067] b5. Collect and process crowd activity information in grid space units to obtain detailed crowd activity distribution information every hour.

[0068] It should be noted that the dynamic and static differences in regional total exposure include dynamic and static differences in the distribution of population activities, and are closely related to the number of people in the grid. Figure 2 The figure shows the dynamic and static differences in the total exposure of truck emissions in Shanghai. The red grid indicates the overestimated residential area where the static exposure is higher than the dynamic exposure, and the blue grid indicates the underestimated employment / mobility area where the static exposure is lower than the dynamic exposure.

[0069] In the core area within the outer ring road, an east-west belt-shaped undervalued agglomeration zone has formed, namely area ④, with Lujiazui as the center, covering Huangpu District, Jing'an District, the Lujiazui area of ​​Pudong New Area, and the Zhangjiang Science City area of ​​Pudong New Area. Combined with the economic development of Shanghai, these areas have generated a large number of jobs and commercial activities due to high-intensity economic agglomeration, and have extremely high economic vitality. However, due to the high rent and land prices in these areas and the relatively small residential space, employees often need to live in other areas and use public transportation to commute. The separation of work and residence leads to an underestimation of the static exposure value.

[0070] At the junction of the Outer Ring Expressway, there are two distinct statically overestimated grid groups: Area 1 and Area 2. Area 1 is located in Baoshan District, encompassing the Gucun and Gongkang sections, and is home to numerous residential communities. Area 2 is primarily located in Minhang District, also home to numerous densely populated residential communities, such as the Xinzhuang community. Therefore, the static exposure response based on residential locations exhibits a significant overestimation.

[0071] Regions ③ and ⑥ are home to the Songjiang Industrial Zone and the Jiading Automobile Industrial Base, where manufacturing jobs are concentrated but the permanent population is sparse. Therefore, the static exposure will be underestimated. However, due to the large spatial scope, the degree of underestimation will not be very obvious.

[0072] Area ⑤ is the Hongqiao hub area. As a purely mobile node with an average daily passenger flow of over 1 million, there is a large number of mobile people, but no residential population. Therefore, there is a very obvious and concentrated static underestimation phenomenon.

[0073] Furthermore, the dynamic and static differences based on the population-weighted average exposure are due to the time-weighted mechanism rather than the difference in population distribution. Static exposure assumes that residents stay in the grid for 24 hours, and the time in each grid is 86400s, that is, 3600s*24. The static weighted average exposure is the reflection of the grid concentration value level, while the dynamic individual grid residence time per hour adopts the actual residence time average, which must be less than 3600s. As a result, the static exposure estimate of all grids must be higher than the dynamic exposure, such as Figure 3 As shown in the figure, the degree of deviation is positively correlated with the spatial gradient of emission concentration, and is particularly significant in high-concentration areas around major freight arteries.

[0074] In a preferred embodiment of the present invention, crowd activity distribution information processing is based on anonymous signaling data of mobile phone users provided by operators, and obtains relatively accurate full-scale crowd activity information according to the population weights of mobile phone users provided by mobile phone operators. Mobile phone user status can be divided into stay status and travel status. The situation where a user stays at the same location for more than 30 minutes is defined as stay status. The location information between each stay point is matched to the road network to form a travel record, and the user is defined as travel status during this period. The most frequent residence locations of mobile phone users at night (8 pm to 8 am the next day) and weekdays (8 am to 8 pm) are observed for a long time to identify the user's home and workplace. This information is recorded and processed in grid space units to obtain detailed crowd activity distribution information every hour.

[0075] In step 3, the method for calculating the dynamic exposure value of the spatial area includes the following steps:

[0076] c1. Calculation of individual exposure value: Calculate the individual exposure value by multiplying the individual's residence time within the grid by the pollutant concentration within that time period;

[0077] c2. Grid-based exposure values: Divide each individual's exposure value into grid areas according to location characteristics and aggregate the grid-based exposure values.

[0078] c3. Grid-based dynamic exposure value: Introducing the time dimension, the grid-based dynamic exposure value is calculated as the product of the individual's activity time in the grid during the time period and the grid pollutant concentration;

[0079] c4. Average exposure value: Use the average value to assess the exposure level of individuals who have been active in a certain area to take into account the phenomenon of population concentration in urban functional areas.

[0080] In a preferred embodiment of the present invention, the factors affecting the dynamic and static exposure values ​​of the spatial region include:

[0081] d1. Truck emissions: Consider factors related to freight activities, such as the spatial distribution of freight facilities, the location of public freight hubs, and the distribution of transportation networks;

[0082] d2. Resident activity patterns: This includes the spatial distribution of residences and employment, the spatial distribution of commercial entities, the proportion of resident types, and commuting distance characteristics;

[0083] d3.OLS regression analysis: Use the OLS regression method to gain a preliminary understanding of the linear relationship between variables and construct a regression model;

[0084] d4. Spatial econometric model: Establish a spatial econometric model, consider the impact of spatial proximity on the relationship between observations, and introduce spatial lag terms and spatial error terms;

[0085] d5. Random forest model: A random forest model was established to rank the importance of regression results and evaluate the impact of various influencing factors on the average exposure value of individuals in the region.

[0086] In a preferred embodiment of the present invention, when evaluating the impact of truck emissions on public health, since this study wants to understand the public health impact of truck emissions from a spatial perspective, the exposure amount is calculated based on the grid. Traditional exposure value calculations usually use individuals as the calculation unit, and calculate the individual exposure value through Formula 1. However, in order to better understand the impact of truck emissions on public health in urban areas, this study has made adjustments to the conventional individual-based exposure value calculation, extending the calculation of exposure values ​​from the individual level to the spatial level, and dividing the exposure value of each individual into corresponding grid areas according to their activity location. Then, the exposure values ​​of all individuals are aggregated in each grid to obtain a grid-based exposure value, that is, the aggregate value of the product of the residence time of each individual in the grid and the concentration of the grid pollutant in that time period, expressed as,

[0087]

[0088] Among them, E i,dymamic represents the dynamic exposure of individual i, s·mg / m 3 , E k,dynamic represents the dynamic exposure of grid k, s·mg / m 3 , c k Indicates the pollutant concentration of grid k, mg / m 3 , T ik represents the activity time of individual i in grid k, s.

[0089] When the mobility of people in the city and the impact of time changes on exposure are further considered and the time dimension is introduced, the grid-based dynamic exposure is obtained, which is expressed as:

[0090]

[0091] Among them, E k,dynamic represents the dynamic exposure of grid k, s·mg / m 3 , c kt Represents the pollutant concentration of grid k in time period, mg / m 3 , T ikt represents the activity time of individual i in grid k during time period t, s, and N represents the total number of people with activity records in grid k.

[0092] E k The activity time of all people with activity tracks in the grid is recorded, which actually contains the density information of people. Generally speaking, due to the different urban functions assumed by different locations in the city during land use planning, there is a clustering phenomenon in residence and employment. For Shanghai, due to its high intensity of economic activities and population mobility in the city center, the clustering of crowd activities is a very obvious rule. The exposure in a certain area may be particularly high due to the gathering of a large number of people, but since the total exposure level is the accumulation of all individual exposure values ​​in the area, it does not actually pose a significant health threat to each individual. Therefore, in order to more accurately assess the exposure level of individuals in a specific area, it is more meaningful to use the average value for analysis, which is expressed as,

[0093]

[0094] in, represents the weighted average dynamic exposure of grid k, s·mg / m 3 , c kt Represents the pollutant concentration of grid k in time period, mg / m 3 , T ilt represents the activity time of individual i in grid k during time period τ, s, and N represents the total number of people with activity records in grid k.

[0095] When calculating the static exposure value, it is assumed that all residents remain stationary at their residence throughout the day, and only the average concentration of NOx emitted by trucks throughout the day is considered. Therefore, the static regional total exposure and weighted average exposure are expressed as follows:

[0096]

[0097] Among them, E k,static represents the total static exposure in the area of ​​grid k, s·mg / m 3 , represents the weighted average static exposure of grid k, s·mg / m 3 , N 0k represents the number of residents in grid k, T 0k Indicates the duration of each hour, 3600s.

[0098] It should be noted that there are currently methods that simply consider the dilution rate of pollutant concentration and the average breathing rate of the population to derive the inhalation factor and then calculate the population exposure, such as Figure 4 As shown, while this method also comprehensively considers both population distribution and pollutant concentration, it completely simplifies the natural diffusion of pollutants. Concentration decays proportionally over time, ignoring the impact of the emission source on the surrounding space. When calculating exposure using "crowd density," the dimensions considered are similar to those of the present invention, but per capita exposure generally better reflects the impact on individuals. Therefore, this embodiment proposes the use of average values.

[0099] Example 2, which is the second embodiment of the present invention, provides a method for calculating and tracing truck emission exposure based on resident activity distribution.

[0100] For individuals, dynamic and static exposure values ​​are derived from the concentration of truck emissions within each individual's activity area. By shifting perspective and clustering individuals based on their spatial location, we can incorporate the attributes of that spatial region, providing a more spatial perspective on the core influencing factors and mechanisms of truck emission exposure.

[0101] Specifically, two factors need to be considered: truck emissions and resident activity patterns. NOx concentrations generated by truck emissions are primarily related to freight activity in the area, including factors influencing freight demand and convenience, such as the spatial distribution of freight facilities, the location of public freight hubs, and the distribution of truck-accessible transportation networks. Resident activity patterns primarily involve the spatial distribution of residences and employment, the spatial distribution of highly active commercial entities, the proportion of resident types (workers, minors, and the elderly) in the area, and the commuting distance characteristics of workers in the area, as shown in Table 1.

[0102] Table 1. Variable description.

[0103]

[0104]

[0105] The present invention first uses the OLS regression method to preliminarily understand the linear relationship between variables, and constructs regression models with dynamic and static spatial individual average exposure as dependent variables. Taking spatial dynamic average exposure as an example, it is expressed as:

[0106]

[0107] To address the OLS regression's inability to account for spatial dependence and autocorrelation, a spatial econometric model was developed to further consider the impact of spatial proximity on the relationships between observations. By incorporating spatial structure into the spatial weight matrix, the spatial characteristics of truck emission exposure levels were more accurately captured. Based on the results of the LM (Lagrange Multiplier) test, the inclusion of spatial lag and spatial error terms in the model was considered, expressed as:

[0108] Spatial lag model:

[0109] Spatial error model:

[0110] Among them, X is the sample matrix of the independent variable, β is the independent variable coefficient, W is the spatial weight matrix, and ∈ is the random error term.

[0111] On this basis, to further understand the importance of each attribute feature of a spatial unit, a random forest model was established to rank the importance of regression results. Random forest is an ensemble learning algorithm based on decision trees. It uses a bagging (bootstrap aggregation) strategy to aggregate the predictions of all decision trees to form a final prediction. It can effectively handle nonlinear problems and is a commonly used classification, regression, and anomaly detection method. The random forest method can handle a large number of features and data and provides feature importance assessment, which is very valuable for understanding data and model prediction results.

[0112] A validation model based on a random forest was constructed, selecting 80% of the samples as a training set and 20% as a test set. The accuracy test results primarily included two metrics: MAE (mean absolute error) and RMSE (root mean square error). Based on these metrics, variable importance analysis of the model results was used to measure the impact of various influencing factors on the average exposure values ​​of individuals in the region.

[0113] In an optional embodiment of the present invention, existing methods generally use traditional machine learning models, such as support vector machines, to verify the accuracy of pollutant exposure simulation. Support vector regression (SVR) maps data to a high-dimensional space through a kernel function (such as an RBF kernel), finds the optimal hyperplane to fit the nonlinear relationship, and uses cross-validation to optimize hyperparameters (such as the penalty coefficient C and the kernel width γ). However, the computational complexity is high, the interpretability is poor, and it is sensitive to the data distribution. When the sample size is large, such as city-level data, the training time will increase significantly, and it is difficult to directly analyze the importance of variables, which is not conducive to policy making or exposure attribution analysis. Moreover, if the scale of the input variables is large, additional standardization processing is required.

[0114] The method of the present invention (random forest) has nonlinear modeling capabilities, can automatically capture complex relationships between variables, is robust and interpretable, is insensitive to outliers and missing data, can perform variable importance analysis to quantify the impact of various factors on exposure values, and compared with SVR, random forest is easier to parallelize and is suitable for large-scale data sets.

[0115] The beneficial effects of this preferred technical solution are: it improves the accuracy of truck emission exposure assessment and makes the assessment results closer to the actual situation by considering the actual distribution of residents' activities; it provides a systematic and comprehensive research perspective for urban traffic pollution control, and proposes effective spatial governance strategies based on urban traffic and land use planning; by reducing truck pollutant exposure, it further protects residents' public health and achieves the goal of healthy city development; it provides useful reference for the formulation of new vehicle traffic control and land planning guidance policies, and promotes urban sustainable development.

[0116] Example 3 is the third embodiment of the present invention, which differs from the first two embodiments in that:

[0117] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0118] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0119] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0120] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, it can be implemented using a combination of any of the following technologies known in the art: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0121] Example 4, the fourth embodiment of the present invention, provides a truck emission exposure measurement and traceability system based on resident activity distribution, including:

[0122] The emission analysis module is used to build a microscopic air diffusion model based on vehicle emission data to simulate the concentration distribution of each pollutant after diffusion;

[0123] The exposure analysis module is used to capture the activity trajectories of urban residents, consider the activity characteristics of urban residents, combine pollutant concentration data with population activity distribution information data, and calculate the dynamic and static exposure levels of residents in the unit grid within a preset time;

[0124] The spatial governance module is used to establish risk assessment models, evaluate and analyze the influencing factors and mechanisms of pollutant exposure, and implement spatial governance strategies.

[0125] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for estimating and tracing truck emissions exposure based on resident activity distribution, characterized by: include, Based on vehicle emission data, a microscopic air diffusion model is built to simulate the concentration distribution of each pollutant after diffusion; Capturing residents' activity trajectories, considering the activity characteristics of urban residents, combining pollutant concentration data with population activity distribution information data, and calculating the dynamic and static exposure levels of residents to pollutants in the unit grid within a preset time; Establish risk assessment models, evaluate and analyze the influencing factors and mechanisms of pollutant exposure, and implement spatial governance strategies.

2. The method for estimating and tracing truck emissions exposure based on resident activity distribution according to claim 1, characterized in that: The simulation of the concentration distribution of each pollutant after diffusion includes: The research space is divided into grid cells, each grid cell represents a spatial unit, and the pollutant emissions are diffused; The Gaussian atmospheric diffusion model is used to establish a Gaussian plume model, simplifying the emission source to the center of the grid; Input the emission source intensity per hour and calculate the pollutant concentration at any point downwind.

3. The method for estimating and tracing truck emissions exposure based on resident activity distribution according to claim 2, characterized in that: The simulation of the concentration distribution of each pollutant after diffusion also includes: Taking environmental factors into consideration, for the research space, the concentration of each grid is the superposition of the Gaussian diffusion concentrations of all grids as independent emission sources.

4. The method for estimating and tracing truck emissions exposure based on resident activity distribution according to claim 3, characterized in that: The capturing of residents' activity trajectories includes: Based on user anonymous signaling data and population weights, obtain full crowd activity information; Divide user status into stay status and travel status; Match the location information between each stop to the road network to form a travel record; Long-term observation of users' most frequent locations at night and during weekdays to identify their home and work locations; The crowd activity information is aggregated and processed in grid space units to obtain detailed crowd activity distribution information every hour.

5. The method for estimating and tracing truck emissions exposure based on resident activity distribution according to claim 4, characterized in that: The calculation of the dynamic and static exposure levels of residents of the unit grid within the preset time includes: The individual exposure value is calculated based on the product of the individual's residence time in the grid and the grid pollutant concentration during that time period; Divide each individual's exposure value into each grid area according to location characteristics, and collect grid-based exposure values; The time dimension is introduced to calculate the dynamic exposure value based on the grid, that is, the product of the individual's activity time in the grid during the time period and the grid pollutant concentration; The average value was used to assess the exposure level of individuals who had been active in a certain area to take into account the phenomenon of population clustering in urban functional areas.

6. The method for estimating and tracing truck emissions exposure based on resident activity distribution according to claim 5, characterized in that: The establishment of the risk assessment model includes: Use OLS regression method to analyze the linear relationship between variables and build a regression model; Establish a spatial econometric model, consider the impact of spatial proximity on the relationship between observations, and introduce spatial lag terms and spatial error terms; A random forest model was established to rank the importance of the regression results and evaluate the impact of various influencing factors on the average exposure value of individuals in the region.

7. The method for estimating and tracing truck emissions exposure based on resident activity distribution according to claim 6, characterized in that: The variables include truck emission factors and residents' activity patterns.

8. A truck emission exposure measurement and tracing system based on resident activity distribution, applying a truck emission exposure measurement and tracing method based on resident activity distribution as described in any one of claims 1 to 7, characterized in that: include: The emission analysis module is used to build a microscopic air diffusion model based on vehicle emission data to simulate the concentration distribution of each pollutant after diffusion; The exposure analysis module is used to capture the activity trajectories of urban residents, consider the activity characteristics of urban residents, combine pollutant concentration data with population activity distribution information data, and calculate the dynamic and static exposure levels of residents in the unit grid within a preset time; The spatial governance module is used to establish risk assessment models, evaluate and analyze the influencing factors and mechanisms of pollutant exposure, and implement spatial governance strategies.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the truck emission exposure measurement and tracing method based on resident activity distribution according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the truck emission exposure measurement and tracing method based on resident activity distribution according to any one of claims 1 to 7 are implemented.

Images