Construction method of regional air pollutant transmission network
By acquiring and pre-processing air pollutant monitoring data, defining single-source diffusion impact factors and screening paths based on causal mechanisms, a regional air pollutant transmission network is built, which solves the problems of data acquisition difficulties and influencing factors in the existing methods. Improves analysis efficiency and accuracy.
Patent Information
- Application Number
- CN202410931040.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-12
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-07-12
AI Technical Summary
The existing regional air pollutant transmission network construction method relies on pollution source lists and remote sensing data. It is difficult to obtain data, lagging up updates, high calculation costs, insufficient data accuracy and integrity, and failure to fully consider various influencing factors such as wind force and temperature.
By obtaining air pollutant monitoring data and geographical information data, performing data preprocessing, defining single-source diffusion impact factors, digging out the basic paths for air pollutant transmission, and screening the real paths based on causal mechanisms to build a regional air pollutant transmission network.
It improves the efficiency of air pollutant interaction relationship analysis and the accuracy and real-timeness of the transmission network, reduces the dependence on traditional methods, realizes a pollutant transmission path network that takes into account influencing factors more comprehensively, and enhances the reference value of pollution prevention and control.
Smart Images

Figure CN119025713A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of air quality data mining, and relates to a method for constructing a pollutant transmission network, in particular to a method for constructing a regional air pollutant transmission network. Background Art
[0002] Air pollution is a serious environmental problem faced by the world. The mutual transmission of pollutants between different regions leads to environmental degradation and difficulty in governance. Air is everywhere and has no specific form. The transmission process of air pollutants is highly volatile, uncontrolled, and involves the interaction of multiple factors.
[0003] The movement of air pollutants is affected by multi-dimensional factors, and their transmission trajectories are difficult to describe. Therefore, the process of constructing an air pollutant transmission path network is very complex.
[0004] Some methods for constructing air pollutant transmission path networks have been proposed in the prior art. For example, the VG method proposed by Lacasa et al., which converts time series into a visual graph of the network, can retain a large number of attributes in the original series and provide a simple and easy way of thinking for network modeling of air quality. In the VG method, time series are used as nodes. When the correlation between nodes is within the visible range, there is a path. Cabezas et al. used the VG method to convert the time series of tropospheric ozone into a transmission network, and verified the feasibility of this method in the field of air quality by the dynamic changes of pollutants in urban and rural areas.
[0005] After extracting various characteristics of air pollution, Song et al. established a complex network model, conducted a comprehensive analysis of the existence of paths between node pairs in the network, and mined the key nodes and paths of pollutant transmission based on the improved PageRank algorithm, and found that they were basically consistent with terrain characteristics and atmospheric flow changes.
[0006] However, the above existing method for constructing a regional air pollutant transmission network still has the following defects:
[0007] 1. The establishment of the air pollutant transmission path network relies on pollution source lists and remote sensing data. However, pollution source lists and remote sensing data are difficult to obtain, and the data labeling and processing processes are complicated, resulting in delayed updates of data for building the air pollutant transmission path network and high computational costs.
[0008] 2. Insufficient data accuracy and integrity. Air quality, meteorological and other data are monitored and recorded by two types of monitoring stations respectively, with differences in monitoring frequency and location. Existing methods cannot effectively integrate multi-source data, and the data composition structure is not comprehensive. Moreover, due to the imperfect monitoring records, there may be missing data in some areas or time periods, resulting in inaccurate results in the construction of the air pollutant transmission network.
[0009] 3. The model structure constructed by the existing method is simple and fails to fully consider the impact of various influencing factors such as wind force, temperature, and distance during the transmission process.
[0010] After searching, no public documents of the prior art identical or similar to the present invention were found. Summary of the invention
[0011] The purpose of the present invention is to overcome the shortcomings of the prior art and propose a method for constructing a regional air pollutant transmission network, which can effectively improve the analysis efficiency of the interaction relationship of regional air pollutants and the accuracy and real-time performance of the construction of an inter-regional air pollutant transmission network.
[0012] The present invention solves the practical problem by adopting the following technical solutions:
[0013] A method for constructing a regional air pollutant transmission network comprises the following steps:
[0014] Step 1: Obtain a large amount of air pollutant monitoring data and geographic information data in the region, perform data preprocessing, obtain the monitoring data of effective air quality monitoring stations, and then perform data filling, normalization, and data balancing;
[0015] Step 2: Based on the various factors that affect the spread of pollutants, define the single-source diffusion impact factor between a certain effective air quality monitoring station and other effective air quality monitoring stations, and solve the single-source diffusion impact factor between any two effective air quality monitoring stations at the corresponding time according to the preprocessed data in step 1, and mine the possible basic path of air pollutant transmission between any two effective air quality monitoring stations based on the single-source diffusion impact factor;
[0016] Step 3: Based on the possible basic paths of air pollutant transmission between any two effective air quality monitoring stations obtained in step 2, the real paths are screened with the causal mechanism as the guide, and then the regional air pollutant transmission network is constructed.
[0017] Moreover, the specific steps of step 1 include:
[0018] (1) Grid division of the area containing the air quality monitoring station: Based on the size of the area and the division accuracy, the area is divided into m*n squares separated by n horizontal lines and m vertical lines using distance or longitude and latitude as horizontal and vertical lines. Each square is numbered 1 to m*n starting from the upper left corner.
[0019] (2) Based on the regional grid division result of step (1), select effective air quality monitoring stations: count the number of air quality monitoring stations ng in each grid, set the number of monitoring stations in the effective grid nsta, if the total number of monitoring stations ngi in grid i is less than nsta, then grid i is regarded as an invalid grid; for effective grid j, if ngj=nsta, then all monitoring stations in the grid are selected as effective; if ngj>nsta, then nsta air quality monitoring stations are selected from it.
[0020] (3) For the effective air quality monitoring station selected in step (2), find the meteorological monitoring station closest to it, and connect the meteorological data and air quality monitoring data at the same time as the monitoring data of the effective air quality monitoring station.
[0021] (4) filling and normalizing the monitoring data of the effective air quality monitoring stations obtained in step (3);
[0022] (5) Finally, the data is balanced by randomly deleting the data items of valid air quality monitoring stations and timestamp indexes with large data volumes, so that the number of data items contained in each index is basically the same.
[0023] Moreover, the specific method of step (4) of step 1 is:
[0024] Compare the list of valid air quality monitoring stations, keep only the pollutant and meteorological information of the valid air quality monitoring stations in the list, clean up the redundant information, and calculate the mean value within 24 hours before and after the missing values and fill them in;
[0025] Normalize the monitoring data of effective air quality monitoring stations, and standardize each item in the monitoring data according to the following formula:
[0026]
[0027] In the above formula, Con std is the standardized value, x is the data value that needs to be normalized, min is the minimum value of all data in this data project, and max is the maximum value of all data in this data project;
[0028] Moreover, the specific steps of step 2 include:
[0029] (1) Comprehensively considering different influencing factors, under the influence of wind, distance and concentration, the single-source diffusion influence factor Inf of effective air quality monitoring station i on effective air quality monitoring station j at time t is calculated as ij (t) is defined as:
[0030]
[0031] In the above formula, ΔcenDij (t) is the concentration difference between sites i and j at time t; ΔF ij (t) is the wind force difference between sites i and j at time t; dis ij is the distance between sites i and j; Inf ij The larger the value of (t), the higher the possibility of the existence of the transmission path. Set the threshold a, when Inf ij When (t)>a, it is considered that there is a transmission basis between the two monitoring stations;
[0032] (2) Based on the distance between effective air quality monitoring stations and the value of the single-source diffusion impact factor, the basic path of air pollutant transmission that may exist between any two effective air quality monitoring stations can be obtained.
[0033] Moreover, the specific steps of step 3 include:
[0034] (1) Setting the time interval: In a given time interval, the pollutant concentration at monitoring station j must increase with that at monitoring station i; the time interval is defined as the interval (m, n) after the current time, where the start time and end time are defined as follows:
[0035]
[0036] In the above two formulas, m is the starting time point of the interval, and n is the ending time point of the interval. ij is the distance between monitoring stations i and j.
[0037] (2) The requirement of pollutant concentration change is: according to the time interval range of step (1), when the pollutant concentration at monitoring station j increases by more than 10% compared with the initial time within the specified time range, that is, when the following formula is established, the transmission path is considered to exist:
[0038] make
[0039] In the formula, m, n are time interval markers, con j (k) is the concentration of monitoring station j at time k, con j (start) is the pollutant concentration at the monitoring station j when calculating the single source impact factor, i.e. the initial moment;
[0040] (3) Accumulate the real paths at each moment and construct a regional air pollutant transmission network.
[0041] Advantages and beneficial effects of the present invention:
[0042] 1. The present invention proposes a method for constructing a regional air pollutant transmission network, taking the network construction of air pollutant transmission paths as the research object, starting from the pollutant concentration data of the monitoring station, comprehensively considering various factors, accumulating and screening the air pollutant transmission paths based on statistics, constructing the pollutant transmission paths between different regions in the region, generating an air quality network structure, and visualizing and analyzing the characteristics of the pollutant transmission process. The present invention effectively improves the analysis efficiency of the interaction between regional pollutants and provides an effective reference for pollution prevention and air governance.
[0043] 2. The present invention comprehensively considers various factors that affect the transmission of air pollutants, screens the transmission paths of air pollutants based on the causal mechanism, and constructs an air pollutant transmission network structure through the statistical accumulation method, thereby being able to visualize and analyze the characteristics of the air pollutant transmission process. In the implementation process of the present invention, the air pollutant transmission mechanism is fully analyzed, the influencing factors are comprehensively considered, and only the data of the ground monitoring station is used, which can be obtained in real time, thereby improving the accuracy and real-time nature of the construction of the air pollutant transmission relationship between regions.
[0044] 3. The present invention gets rid of the limitations of pollution source lists and remote sensing data in traditional methods. In the process of building the air pollutant transmission network, the data of ground monitoring stations are used as input. These data are published in real time on the official website of the Ministry of Ecology and Environment and are easy to obtain.
[0045] 4. The present invention comprehensively considers a variety of influencing factors, and the obtained air pollutant transmission path network is in line with reality and has high accuracy. The present invention proposes the concept of single-source diffusion influencing factors and constructs an air pollutant transmission path network based on a causal mechanism. Taking air quality monitoring stations as nodes and air pollutant transmission processes as edges, the air quality system is constructed as a graph structure model that is easy to analyze and calculate. The extraction method of spatiotemporal influencing factors that affect pollutant transmission is analyzed, and the concept of single-source diffusion influencing factors is proposed based on the kinetic diffusion mechanism of pollutants, and the pollutant transmission process is probabilistically characterized. The causal behavior of pollutant concentrations is extracted from historical data to screen reasonable transmission paths. The path screening process is reasonable, the factors are fully considered, the network structure is consistent with the actual situation, and the accuracy is high.
[0046] 5. The actual effect achieved by the present invention shows that the present invention is based on the historical time series to carry out in-depth mining, corresponding to the actual geographical distribution and spatiotemporal transmission state, and based on the air pollutant transmission network structure, it can analyze the transmission properties of different spatial scales from the monitoring station to the region, which effectively promotes the research on the dynamic behavior of pollutants. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 A basic framework schematic diagram of a method for constructing a regional air pollutant transmission network according to the present invention;
[0048] Figure 2 This is an example diagram of the causal mechanism path screening process of the present invention;
[0049] Figure 3 A sample diagram of the network construction result of the present invention;
[0050] Figure 4 Schematic diagram of transmission paths corresponding to different single-source diffusion influence factors of the present invention (taking February 15, 2022 as an example) The smaller the threshold of the single-source diffusion factor, the more paths there are in the network;
[0051] Figure 5 It is the pollutant circumferential transmission path diagram of the present invention. DETAILED DESCRIPTION
[0052] The embodiments of the present invention are further described in detail below with reference to the accompanying drawings:
[0053] The present invention is based on meteorological factors, geographical factors and historical data of pollutants in monitoring stations, and establishes a method for constructing a regional air pollutant transmission network based on a causal mechanism, such as Figure 1 As shown, the following steps are included:
[0054] Step 1: Obtain a large amount of air pollutant monitoring data and geographic information data in the region, perform data preprocessing, obtain the monitoring data of effective air quality monitoring stations, and then perform data filling, normalization, and data balancing;
[0055] In this embodiment, the data resources include massive meteorological monitoring data, pollutant concentration detection data, and site geographic information data obtained from various monitoring stations. It is necessary to screen effective monitoring stations through gridding and connect pollutant concentration monitoring information and meteorological information. If all monitoring information is directly used as basic data, it will cause data imbalance problems in location and time, and then cause imbalance in path distribution. Therefore, first select effective monitoring stations according to location distribution, then connect meteorological and air quality monitoring stations according to time, one by one, and finally balance and clean all data, normalize and delete redundant data.
[0056] The specific steps of step 1 include:
[0057] (1) Grid division of the area containing the air quality monitoring station: Based on the size of the area and the division accuracy, the area is divided into m*n squares separated by n horizontal lines and m vertical lines using distance or longitude and latitude as horizontal and vertical lines. Each square is numbered 1 to m*n starting from the upper left corner.
[0058] (2) Based on the regional grid division result of step (1), select effective air quality monitoring stations: count the number of air quality monitoring stations ng in each grid, set the number of monitoring stations in the effective grid nsta, if the total number of monitoring stations ngi in grid i is less than nsta, then grid i is regarded as an invalid grid; for effective grid j, if ngj=nsta, then all monitoring stations in the grid are selected as effective; if ngj>nsta, then nsta air quality monitoring stations are selected from it.
[0059] (3) For the effective air quality monitoring station selected in step (2), find the meteorological monitoring station closest to it, and connect the meteorological data and air quality monitoring data at the same time as the monitoring data of the effective air quality monitoring station.
[0060] For the effective air quality monitoring stations selected in step (2), such stations cannot monitor meteorological information. That is, the air quality monitoring stations and meteorological monitoring stations are built in different locations, and the pollutant information does not correspond to the meteorological information. For a certain air quality monitoring station, it is necessary to determine its temperature, wind speed and other related meteorological information based on its neighboring meteorological monitoring stations. The connection method is as follows: For a certain air quality monitoring station, find the meteorological monitoring station closest to it, connect the meteorological data and air quality monitoring data at the same time, and use them as the monitoring data of the air quality monitoring station.
[0061] (4) filling and normalizing the monitoring data of the effective air quality monitoring stations obtained in step (3);
[0062] The specific method of step (4) is:
[0063] Compare the list of valid air quality monitoring stations, keep only the pollutant and meteorological information of the valid air quality monitoring stations in the list, clean up the redundant information, and calculate the mean value within 24 hours before and after the missing values and fill them in;
[0064] Normalize the monitoring data of effective air quality monitoring stations, and standardize each item in the monitoring data according to the following formula:
[0065]
[0066] In the above formula, Con std is the standardized value, x is the data value that needs to be normalized, min is the minimum value of all data in this data project, and max is the maximum value of all data in this data project;
[0067] (5) Finally, the data is balanced by randomly deleting the data items of valid air quality monitoring stations and timestamp indexes with large data volumes, so that the number of data items contained in each index is basically the same, thereby reducing the impact caused by data imbalance.
[0068] Step 2: Based on the various factors that affect the spread of pollutants, define the single-source diffusion impact factor between a certain effective air quality monitoring station and other effective air quality monitoring stations, and solve the single-source diffusion impact factor between any two effective air quality monitoring stations at the corresponding time according to the preprocessed data in step 1. Based on the single-source diffusion impact factor, the basic path of air pollutant transmission that may exist between any two effective air quality monitoring stations is mined.
[0069] The specific steps of step 2 include:
[0070] (1) Comprehensively considering different influencing factors, under the influence of wind, distance and concentration, the single-source diffusion influence factor Inf of effective air quality monitoring station i on effective air quality monitoring station j at time t is calculated as ij (t) is defined as:
[0071]
[0072] In the above formula, ΔcenD ij (t) is the concentration difference between sites i and j at time t; ΔF ij (t) is the wind force difference between sites i and j at time t; dis ij is the distance between sites i and j; Inf ij The larger the value of (t), the higher the possibility of the existence of the transmission path. Set the threshold a, when Inf ij When (t)>a, it is considered that there is a transmission basis between the two monitoring stations;
[0073] (2) Based on the distance between effective air quality monitoring stations and the value of the single-source diffusion impact factor, the basic path of air pollutant transmission that may exist between any two effective air quality monitoring stations can be obtained.
[0074] In this embodiment, various factors affecting the transmission of air pollutants are considered separately. Different factors are comprehensively considered. Under the influence of wind, distance and concentration, the transmission paths of air pollutants are different:
[0075] 1) Distance factor. Distance affects the diffusion of pollutants between two places. For source monitoring station A, the farther the distance to destination monitoring station B is, the weaker its influence is. Since the earth is a sphere, the distance between two points is not the intuitive straight line distance on a two-dimensional plane, but the arc length on a spherical surface.
[0076] The location of two points is represented by the longitude and latitude in the geographic coordinate system. The specific method for determining the distance arc is as follows: Let the center of the earth be O, and the coordinates of station i be (Lon i ,lat i ), the coordinates of site j are (lon j ,latj ), the distance calculation formula between points i and j is as follows:
[0077] dis ij = R*arccos[sin(lat i )*sin(lat j )*cos(lon i -lon j )+cos(lat i )*cos(lat j )]
[0078] In the above formula, R is the radius of the earth.
[0079] 2) Wind factor. Wind can promote the spread of pollutants. The greater the wind difference between two places, the lower the spread cost. Pollutants will spread to other areas along the direction of the wind. The greater the wind difference between different areas, the faster the spread speed. When i is the source diffusion area and j is the destination, the wind difference ΔF in the direction of the line connecting the two places is ij (t) is:
[0080] ΔF ij (t)=|F i (T)|*COS(β-α)+|F j (t)|*cos(β-γ)
[0081] In the above formula, F i is the vector wind at site i, and its wind direction angle is α. j is the vector wind at site j, the wind direction angle is γ, and β is the angle between the line from site i to j and the north direction.
[0082] 3) Pollutant concentration difference factor. The gas diffusion process is actually the process of molecules moving to equilibrium between high-concentration and low-concentration areas. The influence of concentration difference is introduced in the process of air pollutant propagation. We define the new concentration difference as:
[0083]
[0084] Among them, cen i (t) is the pollutant concentration at site i at time t, ΔcenD ij (t) is the concentration difference between sites i and j at time t.
[0085] 4) Comprehensively consider different influencing factors, under the influence of wind, distance and concentration, the single-source diffusion influence factor Inf of effective air quality monitoring station i on effective air quality monitoring station j at time t is calculated as ij (t) is defined as:
[0086]
[0087] In the above formula, ΔcenD ij (t) is the concentration difference between sites i and j at time t. ΔF ij (t) is the wind force difference between sites i and j at time t. ij is the distance between sites i and j. Inf ij The larger the value of (t), the higher the possibility of the existence of the transmission path. Set the threshold a, when Inf ij When (t)>a, it is considered that there is a transmission basis between the two monitoring stations.
[0088] The present invention solves the transmission process of near-ground pollutants. Near-ground transmission will be hindered by surface objects. The farther the distance, the more pollutants will attenuate. The maximum distance in near-ground transmission is set to 200km. At the same time, when the distance is too close, a large number of closed loops are easily formed between monitoring stations. The minimum distance of the transmission path is set to 20km. The basic transmission path between monitoring stations can be obtained based on the distance and the value of the single-source diffusion impact factor.
[0089] Step 3: Based on the possible basic paths of air pollutant transmission between any two effective air quality monitoring stations obtained in step 2, the real paths are screened with the causal mechanism as the guide, and then the regional air pollutant transmission network is constructed.
[0090] At time t, assuming that the transmission path P ij If the path really exists, the spread of pollutants from place i to place j will inevitably lead to changes in pollutant concentrations. We believe that this path only exists if the pollutant concentration in place j increases in a certain period in the future.
[0091] The specific steps of step 3 include:
[0092] (1) Setting the time interval: In a given time interval, the pollutant concentration at monitoring station j must increase with that at monitoring station i; the time interval is defined as the interval (m, n) after the current time, where the start time and end time are defined as follows:
[0093]
[0094] In the above two formulas, m is the starting time point of the interval, and n is the ending time point of the interval. ij is the distance between monitoring stations i and j.
[0095] (2) The requirement of pollutant concentration change is: according to the time interval range of step (1), when the pollutant concentration at monitoring station j increases by more than 10% compared with the initial time within the specified time range, that is, when the following formula is established, the transmission path is considered to exist:
[0096] make
[0097] In the formula, m, n are time interval markers, con j (k) is the concentration of monitoring station j at time k, con j (start) is the pollutant concentration at the monitoring station j when calculating the single source impact factor, i.e. the initial moment;
[0098] (3) Accumulate the real paths at each moment and construct a regional air pollutant transmission network.
[0099] In this embodiment, the working principle of step 3 is:
[0100] 1) Time interval setting. Since the wind force changes all the time, the longer the distance, the longer the transmission time required, and the more uncertain factors there are in the transmission process, the specified distance interval is divided into 3 sections. As the distance increases, we gradually relax the distance interval and time interval. The time interval is determined as (m, n). That is, if monitoring station i transmits pollutants to monitoring station j, the pollutant concentration at monitoring station j will increase in the future time period (m, n). The start time and end time are specified as follows:
[0101]
[0102] In the above two formulas, m is the starting time point of the interval, and n is the ending time point of the interval. ij is the distance between monitoring stations i and j.
[0103] 2) Changes in concentration values. According to the above time interval range, when the concentration of pollutants at monitoring station j increases by more than 10% compared with the initial time within the specified time range, that is, when the following formula is established, the transmission path is considered to exist:
[0104] make
[0105] In the formula, m, n are time interval markers, con j (k) is the concentration of monitoring station j at time k, con j (start) is the pollutant concentration at the monitoring station j when calculating the single source impact factor, i.e. the initial moment;
[0106] Figure 2 The figure shows an example of path screening. For the four monitoring stations at time t, there may be three transmission paths. The time interval is solved for the distance between each monitoring station, and the change in pollutant concentration within the interval is calculated. The two edges (1→2, 3→4) whose pollutant concentration changes do not meet the requirements are deleted. After screening by the causal mechanism, the corresponding transmission path between the monitoring stations at this time (2→3) will be obtained.
[0107] 3) For steps 2) and 3), the solution is the propagation path at a single moment. The network can only describe the dynamics at a certain moment, and cannot characterize the regional laws and characteristics hidden behind large-scale data. For a pollutant transmission path network within a period of time or a certain cycle, it is necessary to map the propagation paths at each moment (t, t+1, t+2,…, t+n) to the entire valid site set, overlap the transmission paths corresponding to each timestamp in the time period, and accumulate them in sequence. The number of occurrences of the path is used as the weight of the corresponding edge. The nodes in the network are air quality monitoring stations, the edges are pollutant transmission paths, and the edges point to the direction of the wind difference. For effective storage in the computer, the paths are mapped one by one to the nodes and converted into a transmission matrix.
[0108] In this embodiment, combined with the single-source diffusion impact factor, based on the causal relationship of historical data on the monitoring station, the path is screened, the network structure at each moment is generated, and the network structure in different time periods can be obtained after statistical accumulation, and the transmission relationship is displayed in the form of a graph. Taking February 15, 2022 as an example, a transmission network between monitoring stations is constructed. The nodes in the network are air quality monitoring stations, and the edges represent the pollutant transmission relationship between nodes at that moment. The transmission path network for that day is as follows: Figure 3 As shown in the figure, the monitoring station number is marked on the node, and the node color indicates the degree of the node in the network. The deeper the color, the greater the degree. The size of the directed edge arrow represents the weight of the edge. The larger the weight, the more times the propagation path appears, indicating that the propagation of pollutants between the two places occurs more frequently.
[0109] The working principle of the present invention is:
[0110] The present invention relates to a method for constructing a regional air pollutant transmission network. The specific process of the method is: first, a large amount of monitoring data and geographic information data are processed, including grid division, selection of effective monitoring stations, data normalization, connection and filling. Secondly, the single-source diffusion influence factor is defined and the basic path is solved. According to the various factors affecting the propagation of pollutants, the influence factor of a certain monitoring station on other monitoring stations is defined. The single-source diffusion influence factor between any two monitoring stations at the corresponding time is solved to obtain the basic transmission path between the monitoring stations. Finally, the basic transmission path is selected and the network is constructed through the causal mechanism. Combined with the single-source diffusion influence factor, the path is selected based on the causal relationship measured by historical data on the historical data of the monitoring station, and the network structure at each time is generated. The network structure at all times in the specified time period is counted to generate a pollutant transmission path network in the area.
[0111] The invention is innovative in that:
[0112] Step A: Perform data processing on a large amount of air quality related data.
[0113] Step B: According to the various factors that affect the spread of pollutants, define the single-source diffusion impact factor of a certain effective monitoring station on other effective monitoring stations, and solve the basic transmission path between monitoring stations based on this factor.
[0114] Step C: Based on the possible transmission basic probability paths between monitoring stations, the real path screening is carried out guided by the causal mechanism.
[0115] Step D: Accumulate the paths at each moment to generate a pollutant transport network.
[0116] The data processing process of step A is as follows:
[0117] The area containing the air quality monitoring stations is divided into grids using longitude and latitude as separators. An equal number of monitoring stations are selected in each grid as valid monitoring stations.
[0118] When connecting meteorological data and pollutant concentration data, for a certain air quality monitoring station, find the meteorological monitoring station closest to it, and connect the meteorological data and air quality monitoring data at the same time as the monitoring data of the air quality monitoring station.
[0119] Normalize the monitoring data of effective air quality monitoring stations, and standardize each item in the monitoring data according to the following formula:
[0120]
[0121] In the above formula, Con std is the standardized value, x is the data value that needs to be normalized, min is the minimum value of all data in this data project, and max is the maximum value of all data in this data project;
[0122] The calculation and solution process of the single-source diffusion impact factor in step B is as follows:
[0123] Comprehensively considering different influencing factors, under the effects of wind, distance and concentration, the single-source diffusion impact factor Inf of effective air quality monitoring station i on effective air quality monitoring station j at time t is calculated as ij (t) is defined as:
[0124]
[0125] In the above formula, ΔcenD ij (t) is the concentration difference between sites i and j at time t. ΔF ij (t) is the wind force difference in the direction of the line connecting the two places, dis ij is the distance between sites i and j.
[0126] The path screening process in step C is as follows:
[0127] In a given time interval, the pollutant concentration at monitoring station j increases with i. The time interval is defined as the interval (m, n) after the current moment. The start time and end time are defined as follows:
[0128]
[0129] In the above two formulas, m is the starting time point of the interval, and n is the ending time point of the interval. ij is the distance between monitoring stations i and j.
[0130] The change requirement of pollutant concentration is: according to the above interval range, when the concentration of pollutants at monitoring station j increases by more than 10% compared with the initial time within the specified time range, that is, when the following formula is established, the transmission path is considered to exist:
[0131] make
[0132] In the formula, m, n are time interval markers, con j (k) is the concentration of monitoring station j at time k, con j (start) is the pollutant concentration at the monitoring station j when calculating the single source impact factor, i.e. the initial moment;
[0133] The present invention takes air quality monitoring stations as nodes and transmission paths as edges, and designs a method for constructing a regional air pollutant transmission network. First, multi-source data is processed to extract effective air pollutant transmission related information; then, the single-source diffusion influence factor is defined according to various factors affecting the transmission of pollutants, and possible paths between monitoring stations are found; secondly, the causal mechanism is used to screen the paths; finally, the paths at each moment are accumulated to generate a network structure. The different values of the single-source diffusion influence factor proposed in the present invention represent different probabilities of the existence of the path. The higher the single-source influence factor, the higher the probability of the existence of the path. Different single-source diffusion influence factors are selected to display the path. Low factors correspond to more paths, and high factors filter paths with low probabilities and only display fewer paths. Figure 4 Shown are the transmission paths corresponding to different single-source diffusion influence factors.
[0134] The pollutant propagation path is calculated with a cycle of one week. The pollutant concentration changes little at adjacent moments. In order to reduce the running time of the algorithm, data is collected every 3 hours. The path with high single-source impact factor is selected to generate the daily path. Figure 5 This is the path network for the four weeks in January. Since January has 31 days, the fourth time period contains a 10-day path collection. Figure 5The topographic structure of the region is also marked in the figure. The green part is the North China Plain, and the yellow part is the mountains and hills. It can be seen from the figure that the transport processes of different distances are widely distributed in the region. Especially in the eastern coastal areas, Cangzhou and Tianjin, which are located in the plains, when heavy pollution weather occurs, there are many transmission paths between the two and they occur frequently. The air quality in Shijiazhuang and Beijing is also poor, but the excavated interactive transport paths between the regions are not frequent. This is closely related to the geographical situation of the two places surrounded by mountains. Pollutants are difficult to transport outward in this semi-basin terrain. Comparing the 4 pictures, there are large differences in the number of transmission paths in different time periods. The reason is that the transmission process is directly affected by the concentration of pollutants and wind force. The uncertainty of wind force and direction and pollutant emissions causes drastic changes in the paths. High concentrations of pollutants and strong winds will produce more long-distance transmission paths.
[0135] It should be emphasized that the embodiments of the present invention are illustrative rather than restrictive. Therefore, the present invention includes but is not limited to the embodiments described in the specific implementation modes. Any other implementation modes derived by those skilled in the art based on the technical solutions of the present invention also fall within the scope of protection of the present invention.
Claims
1. A method for constructing a regional air pollutant transmission network, characterized in that: The following steps are involved: Step 1: Obtain a large amount of air pollutant monitoring data and geographic information data in the region, perform data preprocessing, obtain the monitoring data of effective air quality monitoring stations, and then perform data filling, normalization, and data balancing; Step 2: Based on the various factors that affect the spread of pollutants, define the single-source diffusion impact factor between a certain effective air quality monitoring station and other effective air quality monitoring stations, and solve the single-source diffusion impact factor between any two effective air quality monitoring stations at the corresponding time according to the preprocessed data in step 1, and mine the possible basic path of air pollutant transmission between any two effective air quality monitoring stations based on the single-source diffusion impact factor; Step 3: Based on the possible basic paths of air pollutant transmission between any two effective air quality monitoring stations obtained in step 2, the real paths are screened with the causal mechanism as the guide, and then the regional air pollutant transmission network is constructed.
2. The method for constructing a regional air pollutant transmission network according to claim 1, characterized in that: The specific steps of step 1 include: (1) Grid the area containing the air quality monitoring station: Based on the size and division accuracy of the area, use distance or longitude and latitude as horizontal and vertical lines to divide the entire area into m*n squares divided by n horizontal lines and m vertical lines. Number each square from 1 to m*n starting from the upper left corner; (2) Based on the regional grid division result of step (1), select effective air quality monitoring stations: count the number of air quality monitoring stations ng in each grid, set the number of monitoring stations in the effective grid nsta, if the total number of monitoring stations ngi in grid i is less than nsta, then grid i is regarded as an invalid grid; for effective grid j, if ngj=nsta, then all monitoring stations in the grid are selected as effective; if ngj>nsta, then select nsta air quality monitoring stations from it; (3) For the effective air quality monitoring station selected in step (2), find the meteorological monitoring station closest to it, and connect the meteorological data and air quality monitoring data at the same time as the monitoring data of the effective air quality monitoring station; (4) filling and normalizing the monitoring data of the effective air quality monitoring stations obtained in step (3); (5) Finally, the data is balanced by randomly deleting the data items of valid air quality monitoring stations and timestamp indexes with large data volumes, so that the number of data items contained in each index is basically the same.
3. The method for constructing a regional air pollutant transmission network according to claim 2, characterized in that: The specific method of step (4) of step 1 is: Compare the list of valid air quality monitoring stations, keep only the pollutant and meteorological information of the valid air quality monitoring stations in the list, clean up the redundant information, and calculate the mean value within 24 hours before and after the missing values and fill them in; Normalize the monitoring data of effective air quality monitoring stations, and standardize each item in the monitoring data according to the following formula: In the above formula, Con std is the standardized value, x is the data value that needs to be normalized, min is the minimum value of all the data in this data item, and max is the maximum value of all the data in this data item.
4. The method for constructing a regional air pollutant transmission network according to claim 1, characterized in that: The specific steps of step 2 include: (1) Comprehensively considering different influencing factors, under the influence of wind, distance and concentration, the single-source diffusion influence factor Inf of effective air quality monitoring station i on effective air quality monitoring station j at time t is calculated as ij (t) is defined as: In the above formula, ΔcenD ij (t) is the concentration difference between sites i and j at time t; ΔF ij (t) is the wind force difference between sites i and j at time t; dis ij is the distance between sites i and j; Inf ij The larger the value of (t), the higher the possibility of the existence of the transmission path. Set the threshold a, when Inf ij When (t)>a, it is considered that there is a transmission basis between the two monitoring stations; (2) Based on the distance between effective air quality monitoring stations and the value of the single-source diffusion impact factor, the basic path of air pollutant transmission that may exist between any two effective air quality monitoring stations can be obtained.
5. The method for constructing a regional air pollutant transmission network according to claim 1, characterized in that: The specific steps of step 3 include: (1) Setting the time interval: In a given time interval, the pollutant concentration at monitoring station j must increase with that at monitoring station i; the time interval is defined as the interval (m, n) after the current time, where the start time and end time are defined as follows: In the above two formulas, m is the starting time point of the interval, and n is the ending time point of the interval; dist ij is the distance between monitoring stations i and j; (2) The requirement of pollutant concentration change is: according to the time interval range of step (1), when the pollutant concentration at monitoring station j increases by more than 10% compared with the initial time within the specified time range, that is, when the following formula is established, the transmission path is considered to exist: In the formula, m, n are time interval markers, cin j (k) is the concentration of monitoring station j at time k, con j (start) is the pollutant concentration at the monitoring station j when calculating the single source impact factor, i.e. the initial moment; (3) Accumulate the real paths at each moment and construct a regional air pollutant transmission network.
Citation Information
Patent Citations
Atmospheric pollution monitoring and management method as well as system based on high-density deployment of sensors
CN105181898A
Atmospheric pollutant tracing method and system
CN113777223A
Atmospheric pollution area transmission analysis method and system
CN114154828A
Traceability monitoring method for atmospheric pollution
CN116340447A
Cited By
Atmospheric pollution transmission modeling and pollution driving factor identification method based on multiple networks
CN121543008A
Air-breathing smoke-sensing fire hazard grading detection method and system based on wireless networking
CN122336904A
