CN111259539A

CN111259539A - Simulation method and system for road PM2.5 refined pollution distribution and computer storage medium

Info

Publication number: CN111259539A
Application number: CN202010038518.XA
Authority: CN
Inventors: 周素红; 林荣平; 刘凯; 宋洁
Original assignee: Sun Yat Sen University
Current assignee: Sun Yat Sen University
Priority date: 2020-01-14
Filing date: 2020-01-14
Publication date: 2020-06-09
Anticipated expiration: 2040-01-14
Also published as: CN111259539B

Abstract

The invention discloses a simulation method and a system for road PM2.5 refined pollution distribution and a computer storage medium. The simulation method comprises the following steps: s1, acquiring PM2.5 sampling data in the target area, wherein the PM2.5 sampling data comprise position information; s2, acquiring an environment data variable corresponding to the PM2.5 sampling data; s3, inputting PM2.5 sampling data and environment data variables into a multiple collinearity detection model, and eliminating common variables and non-significant variables in the environment data variables; s4, establishing a causal relationship model of the environmental data variable and the PM2.5 sampling data after the insignificant variable is removed in S3, and checking the causal relationship model to obtain an optimal relationship model of the PM2.5 sampling data and the environmental data variable; s5, setting points in a road network of a target area, extracting environment data variables of the set points, inputting the environment data variables into the optimal relation model of S4, and calculating PM2.5 simulation data of the set points; and S6, interpolating the PM2.5 simulation data of each set point of S5 in the target area to generate the PM2.5 simulation distribution of the road.

Description

Simulation method and system for road PM2.5 refined pollution distribution and computer storage medium

Technical Field

The invention belongs to the technical field of regional pollution early warning and prevention and control, and particularly relates to a simulation method and system for road PM2.5 refined pollution distribution and a computer storage medium.

Background

PM2.5 refers to particulate matter, also known as fine particulate matter, having an aerodynamic equivalent diameter of 2.5 microns or less. Because of its easy inhalation, long-term residue and easy adhesion with toxic substances, it is seriously harmful to human health. The human body can induce the death rate and the disease incidence rate due to the long-term or short-term exposure to the PM2.5 air environment with high pollution level, so the method has important practical significance for the simulation and early warning of the regional PM2.5 pollutant distribution.

In the existing method for revealing regional PM2.5 distribution pattern, the main technical idea is to construct a regression model by using the causal relationship between the regional established environment (such as land utilization property, road traffic flow, population density and the like) and PM2.5, and the method specifically comprises the following steps:

1) collecting monitoring data of fixed stations in a research area and related built-up environment data in the range of the monitoring data;

2) establishing a regression model based on the monitoring data and the constructed environment data, quantifying causal connection and action degree between the monitoring data and the constructed environment data, and obtaining an optimal relation factor combination according to a model result;

3) dividing an equidistant grid with a certain unit scale for a research area, and extracting built environment data which is obviously associated with PM2.5 in the grid based on a model result;

4) the relational model is applied throughout the study grid to thereby calculate the overall spatio-temporal pattern of the region PM 2.5.

In the process of implementing the invention, the applicant finds that the method for the PM2.5 distribution pattern in the existing region has the following problems:

in the past, a linear regression method is mostly adopted in a model fitting method, the nonlinear relation and the spatial non-stationarity among variables are not considered, and the accuracy of the relation between urban air pollutants and constructed environmental data is influenced.

Disclosure of Invention

The invention aims to solve the technical problem that the accuracy of the traditional research method needs to be improved, and provides a simulation method and a simulation system for road PM2.5 refined pollution distribution and a computer storage medium.

In order to solve the problems, the invention is realized according to the following technical scheme:

the invention relates to a simulation method for road PM2.5 refined pollution distribution, which comprises the following steps:

s1, acquiring PM2.5 sampling data in the target area, wherein the PM2.5 sampling data comprise position information;

s2, acquiring an environment data variable corresponding to the PM2.5 sampling data;

s3, inputting PM2.5 sampling data and environment data variables into a multiple collinearity detection model, and eliminating common variables and non-significant variables in the environment data variables;

s4, establishing a causal relationship model of the environmental data variable and the PM2.5 sampling data after the common variable and the non-significant variable are removed in S3, and checking the causal relationship model to obtain an optimal relationship model of the PM2.5 sampling data and the environmental data variable;

s5, setting points in a road network of a target area, extracting environment data variables of the set points, inputting the environment data variables into the optimal relation model of S4, and calculating PM2.5 simulation data of the set points;

and S6, interpolating the PM2.5 simulation data of each set point of S5 in the target area.

Preferably, the S6 further includes:

and according to the PM2.5 simulation data of each set point of S5, performing interpolation calculation on the PM2.5 simulation data of each set point in the target area to obtain the fine simulation distribution of the road PM2.5 of the target area.

Preferably, the S5 sets points in the road network of the target area, and the distance between adjacent set points is 50-150 m.

Preferably, the S1 further includes preprocessing the homogeneity data of the PM2.5 sample data.

Preferably, the homogeneity data preprocessing specifically includes the following steps:

s11, performing homogeneity data division on PM2.5 sampling data by using a half-variation function to obtain a plurality of sub-sampling data sets;

s12, calculating the data median of each sub-sampling data set;

and S13, taking the data median of all the sub-sampling data sets of S12 as new PM2.5 sampling data.

Preferably, the S2 specifically includes the following steps:

s21, generating a plurality of corresponding buffer areas according to the PM2.5 sampling data;

and S22, extracting the built environment data in the buffer area to obtain environment data variables.

7. The method for simulating road PM2.5 refined pollution distribution according to claim 1, wherein the method comprises the following steps:

and the PM2.5 sampling data of S1 is acquired by acquiring equipment in a road moving or fixed monitoring mode in a target area.

The invention also provides a simulation system for road PM2.5 refined pollution distribution, which comprises:

the system comprises a sampling data module, a data processing module and a data processing module, wherein the sampling data module is used for acquiring PM2.5 sampling data in a target area, and the PM2.5 sampling data comprise position information;

the environment variable module is used for acquiring an environment data variable corresponding to PM2.5 sampling data;

the variable eliminating module is used for inputting the PM2.5 sampling data and the environmental data variable into the multiple collinearity detection model and eliminating the collinearity variable and the non-significant variable in the environmental data variable;

the model establishing module is used for establishing a relation model of the environmental data variable and the PM2.5 sampling data after the common variable and the non-significant variable are removed, and checking the relation model to obtain an optimal relation model of the PM2.5 sampling data and the environmental variable data;

the road network point setting module is used for setting points in a road network of a target area, extracting environment data variables of the set points, inputting the environment data variables into the optimal relation model, and calculating PM2.5 simulation data of the set points;

and the simulation distribution module is used for interpolating the PM2.5 simulation data of each set point in the target area to generate the PM2.5 simulation distribution of the road.

Further, the analog distribution module further includes:

and the interpolation unit is used for carrying out interpolation calculation on the PM2.5 simulation data of each set point in the target area according to the PM2.5 simulation data of each set point to obtain the fine simulation distribution of the road PM2.5 of the target area.

The invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by the processor, implements the steps of the simulation method as described above.

Compared with the prior art, the invention has the beneficial effects that:

1. in terms of model results, the method of establishing a model based on the quantitative causal rule improves the model accuracy compared to the conventional LUR multiple regression model.

2. The method can be applied to similar regions or can be used for comparing models in a plurality of regions, and a relation model with higher universality can be obtained; the method is also more suitable for a PM2.5 pollutant distribution map of all roads with a larger range and more refinement, and the daily pollution early warning effect of the corresponding region is better.

Drawings

Embodiments of the invention are described in further detail below with reference to the attached drawing figures, wherein:

FIG. 1 is a first schematic flow chart of a simulation method for road PM2.5 fine pollution distribution according to the present invention;

FIG. 2 is a schematic flow chart diagram II of a simulation method for road PM2.5 refined pollution distribution according to the present invention;

FIG. 3 is a schematic diagram of a regional road network node of the target region of the present invention;

FIG. 4 is a schematic representation of the full-area PM2.5 spatio-temporal distribution of the target area of the present invention.

Detailed Description

The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.

It is to be understood that, in the present application, "at least one" means one or more, "a plurality" means two or more. "and/or" for describing an association relationship of associated objects, indicating that there may be three relationships, e.g., "a and/or B" may indicate: only A, only B and both A and B are present, wherein A and B may be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of single item(s) or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", wherein a, b, c may be single or plural.

"first," "second," and the like are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. Furthermore, the terms "comprising" and "having," as well as any variations thereof, are intended to cover a non-exclusive inclusion, such as a list of modules or elements. The system, product, or device is not necessarily limited to those modules or elements explicitly listed, but may include other modules or elements not explicitly listed or inherent to such system, module, or element.

It is noted that, in the present application, words such as "exemplary" or "for example" are used to mean exemplary, illustrative, or descriptive. Any embodiment or design described herein as "exemplary" or "e.g.," is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, use of the word "exemplary" or "such as" is intended to present concepts related in a concrete fashion.

Supplementary description of the prior art:

based on the introduction of the prior art method for disclosing the distribution pattern of PM2.5 in a region, the applicant has found that the following main problems still exist in the prior art:

1. the source of the PM2.5 data is monitored by a national control air quality monitoring site. However, the applicant researches and discovers that the monitoring sites for monitoring the pollutants are fixed sites, the distribution of the monitoring sites is sparse, and the distribution number of the monitoring sites in the whole city is only single digit. Based on the conditions, data analysis is mostly used for medium and macro scales, and simultaneously, the space-time granularity of the calculation result of the corresponding relation model is too large, so that the environmental change characteristics of local space scales cannot be revealed, and the accuracy of causal connection judgment between urban air pollutants and the built environment is influenced.

2. The problem of overlarge scale of the unit for extracting the environment grid due to the fact that the whole research area range needs to be covered to obtain the space-time distribution pattern of the whole PM2.5 exists, the generated result may overestimate or underestimate the model explanatory power, further deviation of the calculation result occurs, and the environment exposure process and the influence degree in daily life of residents in the city cannot be finely disclosed and explained.

Before describing the embodiments of the present application, first, the following description is made on the manner of acquiring the PM2.5 sampling data of the embodiments of the present application:

currently, in the existing research methods, the source of PM2.5 data is monitored by the national control air quality monitoring site. However, the applicant researches and discovers that the monitoring sites for monitoring the pollutants are fixed sites, the distribution of the monitoring sites is sparse, and the distribution number of the monitoring sites in the whole city is only single digit.

Therefore, the PM2.5 sampling data can be acquired by a road-following movable acquisition or fixed monitoring acquisition method, specifically, for the PM2.5 data of the target area, a movable method (such as low-speed riding) is adopted to perform movable acquisition along the road or fixed equipment is used to acquire road PM2.5 with different typical traffic levels on the road. Different from the existing monitoring station acquisition, the two acquisition modes of the invention acquire air pollutant monitoring data with high space-time resolution. And the collected PM2.5 data all contain corresponding spatial position coordinate points.

For the mobile method, the used equipment is portable monitoring equipment, and the slow-speed moving tool moving along the road can be a bicycle, an electric vehicle and the like which keep running at a constant speed. PM2.5 sampling data of all roads in the target area are acquired by using a mobile tool and wearing a portable monitoring device to drive along the roads. For a fixed method, PM2.5 environmental measurement can be performed by using monitoring devices of intelligent fixed equipment such as intelligent street lamps along the road, so that PM2.5 sampling data along the road can be obtained.

Specifically, referring to fig. 1 and fig. 2, fig. 1 is a schematic flow chart of an embodiment of a simulation method for road PM2.5 fine pollution distribution according to the present invention.

As shown in fig. 1, the simulation method for road PM2.5 refined pollution distribution according to the present invention specifically includes the following steps:

s100, acquiring PM2.5 sampling data in a target area, wherein the PM2.5 sampling data comprise position information;

s200, acquiring an environment data variable corresponding to PM2.5 sampling data;

s300, inputting PM2.5 sampling data and environment data variables into a multiple collinearity detection model, and eliminating common variables and non-significant variables in the environment data variables;

s400, establishing a relation model of the environmental data variable and the PM2.5 sampling data after the common variable and the non-significant variable are removed in S300, and checking the relation model to obtain an optimal relation model of the PM2.5 sampling data and the environmental data variable;

s500, setting points in a road network of a target area, extracting environment data variables of the set points, inputting the environment data variables into the optimal relation model of S400, and calculating PM2.5 simulation data of the set points.

And S600, interpolating the PM2.5 simulation data of each set point in the S500 in the target area.

Specifically, in the following description, the steps of the simulation method are specifically described as follows:

in this embodiment, the PM2.5 sampling data is a data set integrated by a plurality of PM2.5 collection points. Each PM2.5 acquisition point at least comprises PM2.5 data and a corresponding spatial position coordinate point so as to give corresponding position information to the PM2.5 sampling data.

Preferably, in this embodiment, the PM2.5 sampling data of S100 is acquired by acquiring the device moving along the road in the target area. See in particular the above description of the way in which the PM2.5 sampled data is acquired.

In another implementation, the PM2.5 sampled data may be an integration of stationary monitored PM2.5 data and mobile collected PM2.5 data along a roadway.

Preferably, in this embodiment, the S100 further includes preprocessing the homogeneity data of the PM2.5 sampling data, specifically including the following steps:

s110, carrying out homogeneity data division on PM2.5 sampling data by using a half-variation function to obtain a plurality of sub-sampling data sets;

s120, calculating the data median of each sub-sampling data set;

and S130, taking the data median of all the sub-sampling data sets of S12 as new PM2.5 sampling data.

Due to the fact that the PM2.5 time resolution of the acquisition mode of the road-following mobile acquisition or fixed monitoring acquisition is high, such as 1/second or 1/10 seconds, the homogeneity or the repeatability of the road-following sampled data is too high, the data volume of the whole model is extremely large, and the PM2.5 sampled data needs to be subjected to redundant value elimination. According to the method, after PM2.5 sampling data are obtained, the data volume can be reduced and the quality of the model can be ensured by extracting the median of the similarity data. The method is a half-variation function, the spatial distribution range of the similarity data is determined by the method, the median of the corresponding data is extracted, and finally PM2.5 sampling point distribution of the monitored road after the homogeneity data processing is obtained.

Compared with the prior art, most researchers subjectively determine the elimination space range of the repeated data according to the road length or the sampling data amount, for example, the average value of the data in a certain range is obtained, so that comparability of different researches is reduced, and transverse comparison research is also limited.

Therefore, the method adopts a half-variation function method to determine the similar space threshold, extracts the median of the data in the space range as the final PM2.5 sampling point, forms new PM2.5 sampling data, and is more objective and reasonable in extraction method and result. This is achievable by a person skilled in the art and is not overrepresented here.

And S200, acquiring an environment data variable corresponding to the PM2.5 sampling data.

In this example, the environment data variables mainly include the built environment data in various aspects such as road network centrality, land utilization, road traffic, and the like in the target area. Specifically, the present S200 includes the steps of:

and S210, generating a plurality of corresponding buffer areas according to the PM2.5 sampling data.

In this example, the PM2.5 sampling points of the PM2.5 sampling data contain corresponding spatial position coordinate points, and a corresponding plurality of buffers are generated according to the position information of each PM2.5 sampling point.

In one embodiment, the plurality of buffers ranges from 25, 50, 75, 150, 200, 250m, etc.

And S220, extracting the built environment data in the buffer area to generate environment data variables.

And extracting the constructed environment data in the range of the plurality of buffer areas of each PM2.5 sampling point to obtain an environment data variable. The constructed environment data of the embodiment mainly includes road network centrality (mediacy, arrival, gravity, linearity), land utilization (industry, residence, business, greenbelt, water area), road traffic (elevation, gradient, road area, road width, vehicle speed, road grade), and the like.

Specifically, a spatial processing analysis program or the like can be used to extract multi-aspect built environment data such as road network centrality, land utilization, road traffic and the like in a plurality of buffer areas. For example, the road network centrality can be calculated and obtained based on an urban network analysis tool (UNA), and the calculation radius is 1-5 km.

S300, inputting the PM2.5 sampling data and the environment data variable into the multiple collinearity detection model, and eliminating the collinearity variable and the non-significant variable in the environment data variable.

Due to the problem that environment data variables may have a certain degree of autocorrelation and multiple contributions, the constructed environment data is subjected to co-linear detection for this reason. In one embodiment, a technical means of a stepwise regression method of the SPSS is adopted, so that environment data variables with significant contribution are obtained through calculation, and redundant variables are removed from a calculation result by using a variance expansion factor (VIF) less than or equal to 7 as a standard, so that common variables and insignificant variables in the environment data variables are removed.

Based on the guidance of the present invention, one skilled in the art can realize the elimination of common variables and insignificant variables in environment data variables.

S400, establishing a relation model of the environmental data variable and the PM2.5 sampling data after the common variable and the non-significant variable are removed in S300, and checking the relation model to obtain an optimal relation model of the PM2.5 sampling data and the environmental data variable.

In this embodiment, the PM2.5 sampling data and the environment data variable are incorporated into a causal relationship model constructed by a machine learning algorithm (such as a multi-layer perceptron model), and after a simulation predicted value is compared with an actual measurement value, a relationship model containing an optimal independent variable combination is obtained.

As one implementation mode, PM2.5 sampling data is used as a dependent variable, an environment data variable without an insignificant variable is used as an independent variable, and a multi-layer perceptron model is used for constructing a causal link between the two. Wherein, 60% of the sample size is used as a training set, 20% is used as a verification set, and 20% is used as a test set. After the measured values are compared with the model results (the root mean square error RMSE, the mean absolute error MAE and the fitting degree R2 of the measured values and the model results are calculated and compared), the optimal relationship model is obtained.

Specifically, the learning rate of the optimal relationship model parameter is 0.3, the weight update momentum is set to 0.2, and the number of training times is 1000.

Compared with the prior art, the method has the advantages that in the aspect of model results, the model precision (adj R2) is improved by at least 40% by a multi-layer Perceptron (MLP) method compared with the conventional LUR multiple regression model.

As shown in fig. 3, in this embodiment, 1 simulation point (i.e., the set point) is randomly selected at every preset distance in the road network of the target area, and the generation of the simulation points for all roads in the road network is completed. The simulation point includes built environment data (i.e., environment data variable of the set point) corresponding to the road, and the PM2.5 simulation data corresponding to the road is calculated by substituting into the optimal relationship model of S400 and simulating the model.

In one example, the as-built environmental index is extracted by randomly taking 1 simulation point every 100m units of the road. In yet another example, the built-up environment index is extracted by randomly taking 1 simulation point every 50m or 150m unit of the road. Those skilled in the art can adjust the simulated dot spacing based on actual studies, and are not overly addressed herein.

And S600, interpolating the PM2.5 simulation data of each set point in the S500 in the target area to generate the PM2.5 simulation distribution of the road.

Preferably, in this embodiment, S600 further includes performing interpolation calculation on the PM2.5 simulation data of each set point in the target area according to the PM2.5 simulation data of each set point in S500, so as to obtain a fine simulation distribution of the road PM2.5 in the target area.

As shown in fig. 4, in the above method, in S600, 1 simulation point is randomly selected in a road network of the target area in units of a certain distance, for example, 1 simulation point is randomly selected every 100m units, so that there is still a portion where no point is present in the target area. Therefore, after the PM2.5 simulation data of each 100m simulation point of all the road networks are calculated by using the optimal relationship model, on the basis, the PM2.5 simulation data of the non-sampled part in the preset range around each random simulation point is calculated by using a kriging interpolation method, so that the continuous spatial distribution of the PM2.5 concentration data of all roads in the whole target area is obtained.

The main parameters of the kriging interpolation method are selected to be a spherical half-variation function model, the search radius (namely a preset range) is 100m, and the resolution is 10 m. This is achievable by the person skilled in the art on the basis of the description of the invention and is not overrepresented here.

Compared with the prior art, the simulation method for road PM2.5 refined pollution distribution provided by the invention also has the following technical effects:

1. in the aspect of specific application, because the method has higher spatial resolution (based on along-road mobile acquisition or fixed monitoring acquisition, the resolution ratio can reach below a meter level), compared with the traditional fixed site monitoring and remote sensing image method with kilometer resolution, the method can better analyze the environmental exposure problem and the exposure process in daily life of residents, and can provide more targeted and careful thinking countermeasures for healthy city construction, city environment treatment and improvement and the like.

2. The simulation method can be applied to similar regions or can be used for comparing models of a plurality of regions to obtain a relation model with higher universality, and finally a PM2.5 pollutant distribution map of all roads with a larger range and more refinement is obtained with the minimum monitoring cost, so that the daily pollution early warning effect of the corresponding region is better.

Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases.

Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (which may be a computer, a server, or a network device) to execute the simulation method according to the embodiments of the present invention.

The embodiment further provides an automatic processing system for fusing the GPS track and the activity log data, which is used for implementing the above embodiments and preferred embodiments, and the description of the system is omitted.

As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the system described in the embodiments below is preferably implemented in software, implementations in hardware, or a combination of software and hardware are also possible and contemplated.

Therefore, the invention also provides a simulation system for road PM2.5 refined pollution distribution, which comprises:

The environment variable module further comprises a homogeneity data preprocessing unit, and the homogeneity data preprocessing unit utilizes the half-variation function to perform homogeneity data cleaning and representative data extraction on the PM2.5 sampling data. After PM2.5 acquisition data are obtained, the data volume can be reduced and the quality of the model can be ensured by extracting median of the similarity data, and finally PM2.5 sampling point distribution of the monitored road after the homogeneity data processing is obtained.

The environment variable module extracts multi-aspect built environment data such as road network centrality, land utilization, road traffic and the like in a plurality of buffer areas by utilizing space processing analysis software based on PM2.5 sampling points to obtain environment data variables.

The simulation distribution module comprises an interpolation unit, and the interpolation unit is used for carrying out interpolation calculation on the PM2.5 simulation data of each set point in the target area according to the PM2.5 simulation data of each set point to obtain fine simulation distribution of the road PM2.5 of the target area.

The specific working principle of each module of the road PM2.5 refined pollution distribution simulation system can be seen in the specific steps of the simulation method.

The present invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by the processor, implements the steps S100 to S600 of the simulation method as described above.

The Processor may be, for example, a Central Processing Unit (CPU), a general purpose Processor, a Digital Signal Processor (DSP), an Application-Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, a transistor logic device, a hardware component, or any combination thereof. Which may implement or perform the various illustrative logical blocks, modules, and circuits described in connection with the disclosure. The processor may also be a combination of computing functions, e.g., comprising one or more microprocessors, DSPs, and microprocessors, among others.

Accordingly, the memory may be a memory unit inside the processor, an external memory unit independent of the processor, or a component including a memory unit inside the processor and an external memory unit independent of the processor.

The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way, so that any modification, equivalent change and modification made to the above embodiment according to the technical spirit of the present invention are within the scope of the technical solution of the present invention.

Claims

1. A simulation method for road PM2.5 refined pollution distribution is characterized by comprising the following steps:

2. The method for simulating road PM2.5 refined pollution distribution according to claim 1, wherein said S6 further includes:

3. The method for simulating road PM2.5 refined pollution distribution according to claim 1, wherein the method comprises the following steps:

and S5 points are arranged in the road network of the target area, and the distance between adjacent points is 50-150 m.

4. The method for simulating road PM2.5 refined pollution distribution according to claim 1, wherein the method comprises the following steps:

the S1 further includes preprocessing of homogeneity data for the PM2.5 sample data.

5. The method for simulating road PM2.5 refined pollution distribution according to claim 4, wherein the homogeneity data preprocessing specifically comprises the following steps:

s12, calculating the data median of each sub-sampling data set;

6. The method for simulating road PM2.5 refined pollution distribution according to claim 1, wherein said S2 specifically includes the following steps:

8. A road PM2.5 refines simulation system of pollution distribution, characterized by, includes:

9. The system for simulating road PM2.5 refined pollution distribution according to claim 8, wherein said simulated distribution module further comprises:

10. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, which computer program, when being executed by the processor, carries out the steps of the simulation method according to one of claims 1 to 7.