An atmospheric pollution hotspot identification method, device, equipment and medium

Abstract

This application relates to methods, devices, equipment, and media for identifying air pollution hotspots, belonging to the field of environmental monitoring technology. The method includes: acquiring and preprocessing air pollutant concentration data and meteorological data; calculating the backward trajectory of air masses at each observation time based on the preprocessed meteorological data; calculating the upstream background index based on the air pollutant concentrations corresponding to the positions of each trajectory point on the backward trajectory; selecting a training sample set based on the upstream background index; training a meteorological response model; determining baseline meteorological conditions; inputting the baseline meteorological conditions into the meteorological response model; outputting a baseline concentration field; inputting real-time meteorological conditions into the meteorological response model; outputting concentration response values; calculating a normalization factor based on the baseline concentration field and concentration response values; normalizing the air pollutant concentration data to obtain normalized concentration data; calculating a relative emission intensity index based on the normalized concentration data and the baseline concentration field; and identifying air pollution hotspots based on the relative emission intensity index.

Description

Technical Field

[0001] This application belongs to the field of environmental monitoring technology, and specifically relates to a method, device, equipment and medium for identifying air pollution hotspots. Background Technology

[0002] In recent years, as air pollution prevention and control work has gradually developed towards refinement and precision, accurately identifying regional air pollution hotspots has become an important foundation for pollution source supervision, pollution cause analysis, and pollution prevention and control decision-making. Satellite remote sensing technology has the advantages of wide coverage, strong continuous observation capability, and high temporal resolution, and has been widely used in regional-scale air pollution monitoring. By acquiring information on the concentrations of pollutants such as nitrogen dioxide, formaldehyde, and carbon monoxide, it provides an important data source for identifying pollution hotspots.

[0003] However, the concentration of atmospheric pollutants obtained by satellite remote sensing is not only affected by local emission sources, but also by meteorological conditions such as wind speed, wind direction, boundary layer height, temperature, and humidity, as well as the pollution transport process between regions. Existing technologies that directly use pollutant concentration data to identify pollution hotspots, or use methods such as background difference method and statistical threshold method for hotspot identification, cannot effectively eliminate the influence of meteorological conditions and external pollution transport on pollutant concentrations. The hotspot identification results obtained under different spatiotemporal conditions have poor comparability and are prone to misjudgment or omission. They cannot accurately reflect the local emission level of the target area and cannot meet the application needs of refined environmental supervision. Summary of the Invention

[0004] Based on the above analysis, the embodiments of the present invention aim to provide a method, apparatus, device and medium for identifying air pollution hotspots, so as to solve the technical problem that pollution hotspot identification in the prior art is prone to misjudgment or omission.

[0005] The objective of this invention is achieved as follows: A first aspect of the present invention provides a method for identifying air pollution hotspots, comprising: Acquire atmospheric pollutant concentration data and meteorological data for the target area, and preprocess them into a unified spatiotemporal grid; The backward trajectory of the air mass at each observation time is calculated based on the preprocessed meteorological data and the backward trajectory model. The upstream background index at each observation time is calculated based on the atmospheric pollutant concentration corresponding to the position of each trajectory point on the backward trajectory of the air mass. The atmospheric pollutant concentration data and meteorological data that meet the preset screening conditions are selected according to the upstream background index as a training sample set. The meteorological response model is trained based on the training sample set. Based on historical meteorological statistics of the target area, baseline meteorological conditions are determined. These baseline meteorological conditions are then input into the meteorological response model, which outputs a baseline concentration field. Real-time meteorological conditions are input into the meteorological response model, which outputs the real-time concentration response value corresponding to the real-time meteorological conditions. A normalization factor is calculated based on the baseline concentration field and the real-time concentration response value. The atmospheric pollutant concentration data is then normalized based on the normalization factor to obtain normalized concentration data. The relative emission intensity index is calculated based on the normalized concentration data and the baseline concentration field, and air pollution hotspots in the target area are identified based on the relative emission intensity index.

[0006] Furthermore, the atmospheric pollutant concentration data includes column concentration data or stratified concentration data of one or more pollutants among nitrogen dioxide, formaldehyde, and carbon monoxide acquired by satellite remote sensing; the meteorological data includes one or more of wind speed, wind direction, boundary layer height, relative humidity, temperature, and air pressure; the preprocessing to a unified spatiotemporal grid includes: performing cloud detection, quality control, and spatial interpolation on the atmospheric pollutant concentration data; and performing spatiotemporal interpolation and resampling on the meteorological data to the same spatial grid and time node as the preprocessed atmospheric pollutant concentration data.

[0007] Furthermore, the calculation of the air mass backward trajectory at each observation time based on the preprocessed meteorological data and the backward trajectory model includes: using the backward trajectory model to calculate the air mass backward trajectory at each observation time based on the preprocessed meteorological data, and performing the calculation according to a preset backtracking duration. H and time step Generate trajectory point sequence ,in, Represents trajectory points k coordinates Represents trajectory points k The observation time This represents the total number of trajectory points.

[0008] Furthermore, the step of calculating the upstream background index at each observation time based on the atmospheric pollutant concentrations corresponding to the positions of each trajectory point on the backward trajectory of the air mass includes: extracting the atmospheric pollutant concentrations corresponding to each trajectory point on the backward trajectory of the air mass. According to the trajectory points k With the current observation time Time difference between Calculate time decay weight ,in, The attenuation coefficient is calculated based on the time attenuation weight, and is expressed as: The upstream background index characterizes the overall pollution background level of the area that the air mass passes through before reaching the target area.

[0009] Further, training the meteorological response model based on the training sample set includes: dividing the target area into multiple meteorological homogeneous regions and constructing a mixed-effects model, represented as: ,in, express t Time Grid g The concentration of air pollutants, j Indicates a sub-region index. express t Time of the first p Each meteorological element value, P Indicates the total number of meteorological elements. Represents the global reference intercept. Indicates the first p The global baseline slope of each meteorological element Subregion j The random intercept, Represents a grid g The random intercept, Subregion j The p The random slope of each meteorological element Represents a grid g The p The random slope of each meteorological element The residual term is represented by the pollutant concentration in the training sample set as the response variable and the meteorological elements as the input variable. The parameters of the mixed-effects model are estimated by the maximum likelihood method to obtain the meteorological response model.

[0010] Furthermore, the normalization factor characterizes the concentration scaling relationship from current meteorological conditions to baseline meteorological conditions, and is expressed as follows: ,in, Represents the grid in the reference concentration field g concentration, Represents a grid g The real-time concentration response value; based on the normalization factor, the atmospheric pollutant concentration data is normalized to obtain normalized concentration data, expressed as: , express t Time Grid g The concentration of air pollutants, wherein the normalized concentration data represents the equivalent concentration value adjusted from the actual observed concentration to the baseline meteorological conditions.

[0011] Furthermore, the step of calculating the relative emission intensity index based on the normalized concentration data and the baseline concentration field, and identifying air pollution hotspots in the target area based on the relative emission intensity index, includes: a grid.g The relative emission intensity index is , This represents the average value of the baseline concentration field in the target area; high-frequency hotspot identification, source attribution hotspot identification, and risk warning hotspot identification are performed based on the relative emission intensity index; wherein, high-frequency hotspot identification includes: counting the number of times the relative emission intensity index of each grid exceeds a preset threshold within a statistical period, and identifying high-frequency hotspots based on the comparison result of the number of occurrences with the preset threshold; source attribution hotspot identification includes: calculating the pollution contribution potential of each grid based on the relative emission intensity index and the meteorological transport probability function, and identifying source attribution hotspots based on the pollution contribution potential, wherein the pollution contribution potential is expressed as: ,in, , representing the mean of the relative emission intensity index, Represents a grid g The total number of valid observation times within the statistical period. Represents the probability function of meteorological transmission. Indicates wind direction. Represents the wind direction frequency density function. Represents a grid g To receptor target R geometric distance, Indicates along the wind direction The characteristic transmission distance, where, Indicates along the wind direction ,from g arrive R The average wind speed along the path during the statistical period. Indicates the attenuation coefficient. The grid area is represented by the following: risk warning hotspot identification includes: performing trend analysis on the time series of the relative emission intensity index of each grid, calculating the regression slope and performing a significance test; calculating the ratio of the regression slope to the mean of the relative emission intensity index as the relative rate of change; when the regression slope is greater than zero, the significance level is lower than a preset threshold, and the relative rate of change is greater than a preset relative rate of change threshold, the corresponding grid is identified as a risk warning hotspot.

[0012] A second aspect of the present invention provides an air pollution hotspot identification device, comprising: The data acquisition and preprocessing module is used to acquire atmospheric pollutant concentration data and meteorological data of the target area and preprocess them to a unified spatiotemporal grid. The meteorological response model training module is used to calculate the backward trajectory of the air mass at each observation time based on the preprocessed meteorological data and the backward trajectory model, calculate the upstream background index at each observation time based on the atmospheric pollutant concentration corresponding to the position of each trajectory point on the backward trajectory of the air mass, and select atmospheric pollutant concentration data and meteorological data that meet the preset screening conditions according to the upstream background index as a training sample set, and train the meteorological response model based on the training sample set. The concentration normalization module is used to determine the baseline meteorological conditions based on historical meteorological statistics of the target area, input the baseline meteorological conditions into the meteorological response model, and output the baseline concentration field; input the real-time meteorological conditions into the meteorological response model and output the real-time concentration response value corresponding to the real-time meteorological conditions; calculate the normalization factor based on the baseline concentration field and the real-time concentration response value, and perform normalization processing on the atmospheric pollutant concentration data based on the normalization factor to obtain normalized concentration data; The pollution hotspot identification module is used to calculate the relative emission intensity index based on the normalized concentration data and the benchmark concentration field, and to identify air pollution hotspots in the target area based on the relative emission intensity index.

[0013] A third aspect of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the atmospheric pollution hotspot identification method described in any embodiment.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the atmospheric pollution hotspot identification method described in any embodiment.

[0015] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects: By calculating the upstream background index based on the backward trajectory of air masses, and training a meteorological response model by selecting training samples that meet preset screening conditions based on the upstream background index, a baseline concentration field under baseline meteorological conditions is established using the meteorological response model. The actual observed atmospheric pollutant concentrations are then subjected to meteorological normalization processing, and a relative emission intensity index is further constructed. This reduces the impact of different meteorological conditions and external pollution transport on pollutant concentrations, improves the comparability of pollutant concentration data between different times and regions, and can more accurately reflect the relative emission level of the target area. This improves the accuracy and reliability of atmospheric pollution hotspot identification and provides more reliable data support for refined pollution source supervision and pollution prevention and control decisions. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a flowchart of the atmospheric pollution hotspot identification method provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the air pollution hotspot identification device provided in Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of the electronic device architecture provided in Embodiment 3 of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. It should be noted that, unless otherwise specified, the implementation methods and features in the implementation methods in this disclosure can be combined, separated, interchanged, and / or rearranged. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] Example 1 A specific embodiment of the present invention, such as Figure 1 As shown, a method for identifying air pollution hotspots is disclosed, including the following steps: S1. Acquire atmospheric pollutant concentration data and meteorological data of the target area, and preprocess them into a unified spatiotemporal grid.

[0020] In this embodiment, the atmospheric pollutant concentration data includes column concentration data or stratified concentration data of one or more pollutants such as nitrogen dioxide, formaldehyde, and carbon monoxide obtained by satellite remote sensing, and the meteorological data includes one or more of wind speed, wind direction, boundary layer height, relative humidity, temperature, and air pressure. The preprocessing to a unified spatiotemporal grid includes: performing cloud detection, quality control, and spatial interpolation on the atmospheric pollutant concentration data; and performing spatiotemporal interpolation and resampling on the meteorological data to the same spatial grid and time node as the preprocessed atmospheric pollutant concentration data.

[0021] Specifically, after the above preprocessing, for each observation time t and each spatial grid gThe concentrations of air pollutants obtained were obtained after quality control and spatial gridding. The corresponding meteorological element values, along with these, constitute the basic data units for subsequent calculations in this embodiment.

[0022] S2. Calculate the air mass backward trajectory at each observation time based on the preprocessed meteorological data and backward trajectory model. Calculate the upstream background index at each observation time based on the atmospheric pollutant concentration corresponding to the position of each trajectory point on the air mass backward trajectory. Select atmospheric pollutant concentration data and meteorological data that meet the preset selection conditions according to the upstream background index as a training sample set. Train the meteorological response model based on the training sample set.

[0023] Specifically, step S2 includes S201-S203, which are as follows: S201. Based on the preprocessed meteorological data and the backward trajectory model, calculate the backward trajectory of the air mass at each observation time, including: The backward trajectory model is used to calculate the backward trajectory of air masses at each observation time based on the preprocessed meteorological data, according to a preset backtracking duration. H and time step Generate trajectory point sequence ,in, Represents trajectory points k coordinates Represents trajectory points k The observation time This represents the total number of trajectory points.

[0024] Specifically, each grid in the target area is used as the receiver target, and each observation time is used as the trajectory start time. The movement trajectory of the air mass is calculated based on the meteorological data at the corresponding time. The air mass is then tracked backward according to a preset backtracking duration and a preset time step to obtain the backward trajectory of the air mass corresponding to each observation time. Each backward trajectory of the air mass consists of multiple trajectory points arranged in chronological order. Each trajectory point includes at least the corresponding spatial coordinates and backtracking time information. The backward trajectory of the air mass can be calculated using the HYSPLIT model, and the backtracking duration can be set according to the scale of the study area, such as 24 hours, 48 ​​hours, or 72 hours. This invention does not limit this.

[0025] S202. Calculate the upstream background index for each observation time based on the atmospheric pollutant concentrations corresponding to the positions of each trajectory point on the backward trajectory of the air mass, including: Extract the atmospheric pollutant concentrations corresponding to each trajectory point on the backward trajectory of the air mass. According to the trajectory points k With the current observation time Time difference between Calculate time decay weight ,in, The attenuation coefficient is calculated based on the time attenuation weight, and is expressed as: The upstream background index characterizes the overall pollution background level of the area that the air mass passes through before reaching the target area.

[0026] The weight decay over time setting ensures that trajectory points closer to the target time contribute more to the index, reflecting the relative importance of the recent air mass path's potential impact on the target area. The higher the value, the greater the likelihood of a high pollution background along the upstream path of the air mass, and the greater the risk of the target area being affected by potential external transport.

[0027] Specifically, for each backward trajectory of an air mass, atmospheric pollutant concentration data at corresponding locations and times are extracted from each trajectory point. As background concentration, the corresponding time decay weight is determined based on the backtracking time of each trajectory point from the receptor target. Subsequently, the upstream background index corresponding to each observation time is calculated based on the atmospheric pollutant concentration at each trajectory point and the corresponding time decay weight. The larger the upstream background index, the higher the overall atmospheric pollutant concentration along the backward trajectory, and the higher the contribution of external pollution transport to the pollutant concentration of the receptor target. The smaller the upstream background index, the lower the contribution of external pollution transport, and the more significant the impact of local emissions on the pollutant concentration of the receptor target.

[0028] The preferred preset screening condition is to compare the upstream background index corresponding to each observation time with the preset background index threshold. If and only if the upstream background index is lower than the preset background index threshold determined by statistical analysis of historical data, the preprocessed atmospheric pollutant concentration data and the corresponding preprocessed meteorological data of the target area at the corresponding observation time are added as an observation sample to the training sample set.

[0029] S203. Training a meteorological response model based on the aforementioned training sample set, including: The target area is divided into multiple meteorological isoproton regions, and a mixed-effect model is constructed, represented as follows: ,in, express t Time Grid g The concentration of air pollutants, j Indicates a sub-region index. express t Time of the first p Each meteorological element value, P Indicates the total number of meteorological elements. Represents the global reference intercept. Indicates the first p The global baseline slope of each meteorological element Subregion j The random intercept, Represents a grid g The random intercept, Subregion j The p The random slope of each meteorological element Represents a grid g The p The random slope of each meteorological element Represents the residual term; Using the pollutant concentrations in the training sample set as response variables and meteorological elements as input variables, the parameters of the mixed-effects model are estimated using the maximum likelihood method to obtain the meteorological response model.

[0030] Specifically, the atmospheric pollutant concentrations corresponding to each observation sample in the training sample set are used as response variables, and the meteorological data of the corresponding observation samples are used as input variables to establish the response relationship between pollutant concentrations and meteorological conditions, thus obtaining a meteorological response model. The meteorological response model is used to describe the variation law of pollutant concentrations under different meteorological conditions, so as to characterize the influence of meteorological factors on the variation of pollutant concentrations and provide a model basis for subsequent meteorological normalization processing.

[0031] Preferably, the meteorological response model is constructed using a multi-layered nested mixed-effects model. The fixed effects are used to characterize the overall impact of meteorological elements such as wind speed, wind direction, temperature, boundary layer height, and relative humidity on pollutant concentrations, while the random effects are used to describe the random differences between different regions, seasons, or time periods, thereby improving the model's ability to fit the changes in pollutant concentrations under different regional and time conditions.

[0032] Furthermore, the model parameters can be solved using restricted maximum likelihood estimation (REML) to obtain the fixed effects parameters and random effects parameters. After the model training is completed, the model can be tested based on the residual distribution and goodness of fit. When the model meets the preset accuracy requirements, the trained model can be used as the meteorological response model for subsequent baseline concentration field calculation and normalization processing.

[0033] S3. Determine the baseline meteorological conditions based on historical meteorological statistics of the target area, input the baseline meteorological conditions into the meteorological response model, and output the baseline concentration field; input the real-time meteorological conditions into the meteorological response model, and output the real-time concentration response value corresponding to the real-time meteorological conditions; calculate the normalization factor based on the baseline concentration field and the real-time concentration response value, and normalize the atmospheric pollutant concentration data based on the normalization factor to obtain normalized concentration data.

[0034] Specifically, after obtaining the meteorological response model, in order to eliminate the influence of different meteorological conditions on pollutant concentrations, it is necessary to uniformly convert the atmospheric pollutant concentrations at different observation times to the same meteorological conditions for comparison. Therefore, in this embodiment, the baseline meteorological conditions are determined based on long-term meteorological statistics of the target area, and the baseline concentration field and the real-time concentration response value corresponding to the current meteorological conditions are calculated based on the meteorological response model, thereby completing the meteorological normalization processing of pollutant concentrations. The specific implementation steps of step S3 include: S301. Determine the baseline meteorological conditions. Specifically, statistical analysis is conducted on meteorological elements such as wind speed, wind direction, temperature, boundary layer height, relative humidity, and air pressure within a preset time period in the target area, and baseline values ​​for each meteorological element are determined. The baseline values ​​of each meteorological element together constitute the baseline meteorological conditions of the target area. The baseline meteorological conditions are used to uniformly characterize the diffusion conditions of the target area under standard meteorological conditions, so that pollutant concentrations obtained at different observation times can be converted to consistent meteorological conditions for comparison.

[0035] Preferably, the baseline value of each meteorological element can be determined by the median, mode, average or other statistical characteristic value of the historical observation data of the corresponding meteorological element, and the present invention does not limit this.

[0036] S302. Calculate the baseline concentration field and real-time concentration response value. Specifically, the baseline meteorological conditions are input into the meteorological response model to calculate the baseline concentration field corresponding to each grid, and the actual meteorological data corresponding to each observation time are input into the meteorological response model to calculate the real-time concentration response value under the current meteorological conditions corresponding to each observation time.

[0037] The baseline concentration field represents the pollutant concentration distribution of each grid in the target area under unified baseline meteorological conditions; the real-time concentration response value represents the response of the current meteorological conditions to the pollutant concentration. Since both the baseline concentration field and the real-time concentration response value are calculated by the same meteorological response model, they have a consistent model basis, providing a unified data source for subsequent calculation of the normalization factor.

[0038] S303, Meteorological Normalization Processing In this embodiment, the normalization factor characterizes the concentration scaling relationship from the current meteorological conditions to the baseline meteorological conditions, and is expressed as follows: ,in, Represents the grid in the reference concentration field g concentration, Represents a grid g The real-time concentration response value; based on the normalization factor, the atmospheric pollutant concentration data is normalized to obtain normalized concentration data, expressed as: , express t Time Grid g The concentration of air pollutants, wherein the normalized concentration data represents the equivalent concentration value adjusted from the actual observed concentration to the baseline meteorological conditions.

[0039] Specifically, after processing with S303, the atmospheric pollutant concentrations at each observation time correspond to uniform baseline meteorological conditions, effectively reducing the fluctuations in pollutant concentrations caused by meteorological factors such as wind speed, wind direction, boundary layer height, temperature, and humidity, and providing a unified data foundation for the subsequent construction of the relative emission intensity index.

[0040] S4. Calculate the relative emission intensity index based on the normalized concentration data and the benchmark concentration field, and identify air pollution hotspots in the target area based on the relative emission intensity index.

[0041] In this embodiment, step S4 specifically includes: Grid g The relative emission intensity index is , This represents the average value of the baseline concentration field in the target region; Based on the relative emission intensity index, perform high-frequency hotspot identification, source attribution hotspot identification, and risk warning hotspot identification; The high-frequency hotspot identification includes: counting the number of times the relative emission intensity index of each grid exceeds a preset threshold within a statistical period, and identifying high-frequency hotspots based on the comparison between the number of occurrences and the preset threshold; the source attribution hotspot identification includes: calculating the pollution contribution potential of each grid based on the relative emission intensity index and the meteorological transport probability function, and identifying source attribution hotspots based on the pollution contribution potential, wherein the pollution contribution potential is expressed as: ,in, , representing the mean of the relative emission intensity index, Represents a grid g The total number of valid observation times within the statistical period. Represents the probability function of meteorological transmission. Indicates wind direction. Represents the wind direction frequency density function. Represents a grid g To receptor target R geometric distance, Indicates along the wind direction The characteristic transmission distance, where Indicates along the wind direction ,from g arrive R The average wind speed along the path during the statistical period. Indicates the attenuation coefficient. The grid area is represented by the pollution contribution potential to identify source attribution hotspots; risk warning hotspot identification includes: performing trend analysis based on the time series of the relative emission intensity index of each grid, calculating the regression slope and performing a significance test; calculating the ratio of the regression slope to the mean of the relative emission intensity index as the relative rate of change; when the slope is significantly greater than zero, the significance level is lower than a preset threshold, and the relative rate of change is greater than a preset relative rate of change threshold, the corresponding grid is determined to be a risk warning hotspot.

[0042] Specifically, a larger relative emission intensity index indicates that the pollutant emission level of the corresponding grid is higher relative to the regional average emission level; a smaller relative emission intensity index indicates that the relative emission level of the corresponding grid is lower. Since the normalized concentration has eliminated or weakened the concentration changes caused by different meteorological conditions, the relative emission intensity index can more realistically reflect the relative differences in emission intensity of pollution sources between different regions, rather than the differences in observed concentrations caused by meteorological conditions. Furthermore, the relative emission intensity indices calculated at each observation time can be used to construct a relative emission intensity index time series, providing basic data for subsequent high-frequency hotspot identification, source attribution hotspot identification, and risk warning hotspot identification.

[0043] This embodiment can identify air pollution hotspots in the target area using at least one of the following methods, namely, high-frequency hotspot identification, source attribution hotspot identification, and risk warning hotspot identification, depending on different application requirements. The identification processes for each method are as follows: (1) High-frequency hotspot identification The number of times the relative emission intensity index of each grid exceeds the preset hotspot threshold within the statistical period is counted. Grids that exceed the preset threshold number of times are identified as high-frequency hotspots. High-frequency hotspots are used to reflect areas where emission anomalies exist for a long time.

[0044] (2) Source attribution hotspot identification The grids are sorted according to their pollution contribution potential. Grids with pollution contribution potential exceeding a preset contribution threshold are identified as source attribution hotspots. These hotspots are used to identify potential pollution source areas that significantly contribute to the pollution of the target area. The calculation of pollution contribution potential involves characteristic transmission distances. This represents the characteristic distance a pollutant travels to reach its initial concentration due to diffusion, deposition, and chemical transformation during its transport along a given wind direction. Calculate, or take values ​​empirically based on the climate and topography characteristics of the target area.

[0045] (3) Risk warning hotspot identification Trend analysis is performed on the time series of the relative emission intensity index of each grid, the regression slope is calculated and a significance test is conducted; the ratio of the regression slope to the mean of the relative emission intensity index is taken as the relative rate of change; when the regression slope is greater than zero, the significance level is lower than a preset threshold, and the relative rate of change exceeds a preset threshold, the corresponding grid is identified as a risk warning hotspot; risk warning hotspots are used to reflect areas that may continue to deteriorate or have abnormal emission trends in the future.

[0046] Compared with existing technologies, the atmospheric pollution hotspot identification method provided in this embodiment calculates the upstream background index based on the backward trajectory of air masses, and trains a meteorological response model by selecting training samples that meet preset screening conditions based on the upstream background index. Then, it uses the meteorological response model to establish a benchmark concentration field under benchmark meteorological conditions, performs meteorological normalization processing on the actual observed atmospheric pollutant concentrations, and further constructs a relative emission intensity index. This reduces the impact of different meteorological conditions and external pollution transport on pollutant concentrations, improves the comparability of pollutant concentration data between different times and different regions, and can more accurately reflect the relative emission level of the target area. This improves the accuracy and reliability of atmospheric pollution hotspot identification and provides more reliable data support for refined pollution source supervision and pollution prevention and control decisions.

[0047] Example 2 This embodiment provides an air pollution hotspot identification device, such as... Figure 2 As shown, it includes: The data acquisition and preprocessing module is used to acquire atmospheric pollutant concentration data and meteorological data of the target area and preprocess them to a unified spatiotemporal grid. The meteorological response model training module is used to calculate the backward trajectory of the air mass at each observation time based on the preprocessed meteorological data and the backward trajectory model, calculate the upstream background index at each observation time based on the atmospheric pollutant concentration corresponding to the position of each trajectory point on the backward trajectory of the air mass, and select atmospheric pollutant concentration data and meteorological data that meet the preset screening conditions according to the upstream background index as a training sample set, and train the meteorological response model based on the training sample set. The concentration normalization module is used to determine the baseline meteorological conditions based on historical meteorological statistics of the target area, input the baseline meteorological conditions into the meteorological response model, and output the baseline concentration field; input the real-time meteorological conditions into the meteorological response model and output the real-time concentration response value corresponding to the real-time meteorological conditions; calculate the normalization factor based on the baseline concentration field and the real-time concentration response value, and perform normalization processing on the atmospheric pollutant concentration data based on the normalization factor to obtain normalized concentration data; The pollution hotspot identification module is used to calculate the relative emission intensity index based on the normalized concentration data and the benchmark concentration field, and to identify air pollution hotspots in the target area based on the relative emission intensity index.

[0048] Example 3 This embodiment provides an electronic device, such as... Figure 3 As shown, it includes a memory and a processor. The memory stores a computer program, which, when executed by the processor, implements the atmospheric pollution hotspot identification method as described in any of the above embodiments.

[0049] Example 4 This embodiment provides a computer-readable storage medium storing a computer program thereon. When the program is executed by a processor, it implements the atmospheric pollution hotspot identification method as described in any of the above embodiments.

[0050] Computer-readable storage media include both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0051] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0052] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented in hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0053] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for identifying air pollution hotspots, characterized in that, include: Acquire atmospheric pollutant concentration data and meteorological data for the target area, and preprocess them into a unified spatiotemporal grid; The backward trajectory of the air mass at each observation time is calculated based on the preprocessed meteorological data and the backward trajectory model. The upstream background index at each observation time is calculated based on the atmospheric pollutant concentration corresponding to the position of each trajectory point on the backward trajectory of the air mass. The atmospheric pollutant concentration data and meteorological data that meet the preset screening conditions are selected according to the upstream background index as a training sample set. The meteorological response model is trained based on the training sample set. Based on historical meteorological statistics of the target area, baseline meteorological conditions are determined, and the baseline meteorological conditions are input into the meteorological response model to output a baseline concentration field. Input the real-time meteorological conditions into the meteorological response model and output the real-time concentration response value corresponding to the real-time meteorological conditions; A normalization factor is calculated based on the reference concentration field and the real-time concentration response value. The atmospheric pollutant concentration data is then normalized based on the normalization factor to obtain normalized concentration data. The relative emission intensity index is calculated based on the normalized concentration data and the baseline concentration field, and air pollution hotspots in the target area are identified based on the relative emission intensity index.

2. The method for identifying air pollution hotspots according to claim 1, characterized in that, The atmospheric pollutant concentration data includes column concentration data or stratified concentration data of one or more pollutants such as nitrogen dioxide, formaldehyde, and carbon monoxide obtained by satellite remote sensing, and the meteorological data includes one or more of wind speed, wind direction, boundary layer height, relative humidity, temperature, and air pressure. The preprocessing to a unified spatiotemporal grid includes: performing cloud detection, quality control, and spatial interpolation on the atmospheric pollutant concentration data; and performing spatiotemporal interpolation and resampling on the meteorological data to the same spatial grid and time node as the preprocessed atmospheric pollutant concentration data.

3. The method for identifying air pollution hotspots according to claim 1, characterized in that, The calculation of the air mass backward trajectory at each observation time based on preprocessed meteorological data and a backward trajectory model includes: The backward trajectory model is used to calculate the backward trajectory of air masses at each observation time based on the preprocessed meteorological data, according to a preset backtracking duration. H and time step Generate trajectory point sequence ,in, Represents trajectory points k coordinates Represents trajectory points k The observation time This represents the total number of trajectory points.

4. The method for identifying air pollution hotspots according to claim 3, characterized in that, The calculation of the upstream background index at each observation time based on the atmospheric pollutant concentrations corresponding to the positions of each trajectory point on the backward trajectory of the air mass includes: Extract the atmospheric pollutant concentrations corresponding to each trajectory point on the backward trajectory of the air mass. According to the trajectory points k With the current observation time Time difference between Calculate time decay weight ,in, The attenuation coefficient is calculated based on the time attenuation weight, and is expressed as: The upstream background index characterizes the overall pollution background level of the area that the air mass passes through before reaching the target area.

5. The method for identifying air pollution hotspots according to claim 1, characterized in that, The training of the meteorological response model based on the training sample set includes: The target area is divided into multiple meteorological isoproton regions, and a mixed-effect model is constructed, represented as follows: ,in, express t Time Grid g The concentration of air pollutants, j Indicates a sub-region index. express t Time of the first p Each meteorological element value, P Indicates the total number of meteorological elements. Represents the global reference intercept. Indicates the first p The global baseline slope of each meteorological element Subregion j The random intercept, Represents a grid g The random intercept, Subregion j The p The random slope of each meteorological element Represents a grid g The p The random slope of each meteorological element Represents the residual term; Using the pollutant concentrations in the training sample set as response variables and meteorological elements as input variables, the parameters of the mixed-effects model are estimated using the maximum likelihood method to obtain the meteorological response model.

6. The method for identifying air pollution hotspots according to claim 1, characterized in that, The normalization factor characterizes the concentration scaling relationship from current meteorological conditions to baseline meteorological conditions, and is expressed as follows: ,in, Represents the grid in the reference concentration field g concentration, Represents a grid g The real-time concentration response value; based on the normalization factor, the atmospheric pollutant concentration data is normalized to obtain normalized concentration data, expressed as: , express t Time Grid g The concentration of air pollutants, wherein the normalized concentration data represents the equivalent concentration value adjusted from the actual observed concentration to the baseline meteorological conditions.

7. The method for identifying air pollution hotspots according to claim 6, characterized in that, The step of calculating the relative emission intensity index based on the normalized concentration data and the baseline concentration field, and identifying air pollution hotspots in the target area based on the relative emission intensity index, includes: Grid g The relative emission intensity index is , This represents the average value of the baseline concentration field in the target region; Based on the relative emission intensity index, perform high-frequency hotspot identification, source attribution hotspot identification, and risk warning hotspot identification; The high-frequency hotspot identification includes: counting the number of times the relative emission intensity index of each grid exceeds a preset threshold during the statistical period, and identifying high-frequency hotspots based on the comparison between the number of occurrences and the preset threshold. Source attribution hotspot identification includes: calculating the pollution contribution potential of each grid based on the relative emission intensity index and the meteorological transport probability function; and identifying source attribution hotspots based on the pollution contribution potential, wherein the pollution contribution potential is expressed as: ,in, , representing the mean of the relative emission intensity index, Represents a grid g The total number of valid observation times within the statistical period. Represents the probability function of meteorological transmission. Indicates wind direction. Represents the wind direction frequency density function. Represents a grid g To receptor target R geometric distance, Indicates along the wind direction The characteristic transmission distance, where, Indicates along the wind direction ,from g arrive R The average wind speed along the path during the statistical period. Indicates the attenuation coefficient. Indicates the grid area; Risk warning hotspot identification includes: performing trend analysis on the time series of the relative emission intensity index of each grid, calculating the regression slope and performing a significance test; calculating the ratio of the regression slope to the mean of the relative emission intensity index as the relative rate of change; when the regression slope is greater than zero, the significance level is lower than a preset threshold, and the relative rate of change is greater than a preset relative rate of change threshold, the corresponding grid is identified as a risk warning hotspot.

8. An air pollution hotspot identification device, characterized in that, The device includes: The data acquisition and preprocessing module is used to acquire atmospheric pollutant concentration data and meteorological data of the target area and preprocess them to a unified spatiotemporal grid. The meteorological response model training module is used to calculate the backward trajectory of the air mass at each observation time based on the preprocessed meteorological data and the backward trajectory model, calculate the upstream background index at each observation time based on the atmospheric pollutant concentration corresponding to the position of each trajectory point on the backward trajectory of the air mass, and select atmospheric pollutant concentration data and meteorological data that meet the preset screening conditions according to the upstream background index as a training sample set, and train the meteorological response model based on the training sample set. The concentration normalization module is used to determine the baseline meteorological conditions based on historical meteorological statistics of the target area, input the baseline meteorological conditions into the meteorological response model, and output the baseline concentration field; input the real-time meteorological conditions into the meteorological response model and output the real-time concentration response value corresponding to the real-time meteorological conditions; calculate the normalization factor based on the baseline concentration field and the real-time concentration response value, and perform normalization processing on the atmospheric pollutant concentration data based on the normalization factor to obtain normalized concentration data; The pollution hotspot identification module is used to calculate the relative emission intensity index based on the normalized concentration data and the benchmark concentration field, and to identify air pollution hotspots in the target area based on the relative emission intensity index.

9. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, implements the atmospheric pollution hotspot identification method as described in any one of claims 1-7.

10. A storage medium, characterized in that, It stores a computer program, which, when executed by a processor, implements the atmospheric pollution hotspot identification method according to any one of claims 1-7.

Images