Method and device for determining atmospheric pollution hotspots, computer equipment and medium

Abstract

Embodiments of the present application provide a method and device for determining atmospheric pollution hotspots, computer equipment and a medium, relating to the technical field of atmospheric pollution monitoring, which comprises: dividing atmospheric pollution concentration data into grid data of a preset resolution; in all grids in the grid data, determining a grid with an atmospheric pollution concentration value higher than a first dynamic concentration threshold as a first potential hotspot grid; determining a second potential hotspot grid in the first potential hotspot grid based on a time-dimension anomaly identification algorithm; determining a third potential hotspot grid in the first potential hotspot grid based on the dynamic change of a time-space dynamic anomaly index in a time-space coupling dimension; determining the intersection of the second potential hotspot grid and the third potential hotspot grid as a candidate hotspot grid, verifying the environmental rationality of each candidate hotspot grid using multi-source data, and determining a trusted hotspot grid. The scheme can improve the accuracy of hotspot identification, has strong dynamic adaptability, and reduces false positives of hotspots.

Description

Technical Field

[0001] This invention relates to the field of air pollution monitoring technology, and in particular to a method, apparatus, computer equipment, and medium for determining air pollutant hotspots. Background Technology

[0002] Current hotspot identification technologies for air pollutants aim to capture sudden increases in pollutant concentrations, providing data support for environmental management, pollution source investigation, and pollution control. In the temporal dimension, statistical thresholding methods (such as sliding window dynamic calculation and quantile threshold setting) are primarily used to quickly identify sudden increases in concentration at single sites. In the spatial dimension, comparisons of spatial correlations between sites and their surrounding environment are relied upon, or machine learning clustering models are applied to analyze spatial anomaly patterns. Some studies attempt to combine temporal and spatial anomaly analysis results to improve the reliability of hotspot identification through collaborative verification. However, existing technologies still face significant challenges: limited ability to capture small-scale anomalous hotspots; insufficient adaptability of spatiotemporal combined anomaly analysis algorithms to dynamically changing pollution scenarios; and the existence of false hotspot reports due to meteorological interference.

[0003] (1) Limited ability to capture small-scale anomalies: Current technology is mainly based on spatial correlation analysis of discrete monitoring stations, and its identification accuracy is highly dependent on the density of station distribution. In sparsely populated areas, current technology is more suitable for identifying anomalies in large-scale areas, but it is difficult to accurately locate sudden small-scale pollution anomaly points and pollution sources.

[0004] (2) Insufficient adaptability to dynamically changing pollution scenarios: Although existing technologies attempt to combine temporal and spatial anomaly analysis, they have not achieved deep integration at the algorithm level. The typical logic is "when the time scale of a certain point exceeds the threshold and the spatial scale is higher than the surrounding concentration, an alarm is triggered", but this logic does not integrate the historical dynamic changes of the surrounding area concentration.

[0005] (3) Hotspot false alarm problem: Since the existing technology mainly relies on the pollutant concentration to determine the anomaly, it lacks a collaborative verification mechanism for the distribution of pollution sources and meteorological conditions. Therefore, it is easy to cause hotspot false alarms due to regional pollution, pollutant accumulation caused by inversion layer, and insufficient ability to identify regular periodic fluctuations. Summary of the Invention

[0006] In view of this, embodiments of the present invention provide a method for determining atmospheric pollutant hotspots to solve the technical problems of low accuracy, poor dynamic adaptability, and false alarms in hotspot identification in the prior art. The method includes:

[0007] Acquire atmospheric pollutant concentration data and multi-source data related to atmospheric pollution from multiple scale stations within the target area, divide the atmospheric pollutant concentration data into grid data of a preset resolution, and perform spatial consistency processing on the grid data and the multi-source data;

[0008] Among all grids in the grid data, the grids with atmospheric pollutant concentration values ​​higher than the first dynamic concentration threshold are identified as the first potential hotspot grids;

[0009] A second potential hotspot grid is determined in the first potential hotspot grid based on a time-dimensional anomaly identification algorithm;

[0010] Calculate the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids. Based on the dynamic changes of the spatiotemporal dynamic anomaly index in the spatiotemporal coupling dimension, determine the third potential hotspot grid in the first potential hotspot grids. The spatiotemporal dynamic anomaly index represents the degree of deviation of the current atmospheric pollutant concentration value of the target grid from the historical dynamic baseline of atmospheric pollutant concentration in the surrounding area.

[0011] The intersection of the second potential hotspot grid and the third potential hotspot grid is determined as a candidate hotspot grid, and the environmental rationality of each candidate hotspot grid is verified using the multi-source data. The candidate hotspot grid that passes the verification is determined as a reliable hotspot grid.

[0012] This invention also provides a device for determining atmospheric pollutant hotspots, addressing the technical problems of low accuracy, poor dynamic adaptability, and false alarms in existing hotspot identification technologies. The device includes:

[0013] The data processing module is used to acquire atmospheric pollutant concentration data and multi-source data related to atmospheric pollution from multi-scale stations within the target area, divide the atmospheric pollutant concentration data into grid data of a preset resolution, and perform spatial consistency processing on the grid data and the multi-source data.

[0014] The first hotspot determination module is used to identify grids with atmospheric pollutant concentration values ​​higher than a first dynamic concentration threshold as first potential hotspot grids among all grids in the grid data.

[0015] The second hotspot determination module is used to determine the second potential hotspot grid in the first potential hotspot grid based on the time-dimensional anomaly identification algorithm.

[0016] The third hotspot determination module is used to calculate the spatiotemporal dynamic anomaly index of each of the first potential hotspot grids. Based on the dynamic changes of the spatiotemporal dynamic anomaly index in the spatiotemporal coupling dimension, the third potential hotspot grid is determined in the first potential hotspot grid. The spatiotemporal dynamic anomaly index represents the degree of deviation of the current atmospheric pollutant concentration value of the target grid from the historical dynamic baseline of atmospheric pollutant concentration in the surrounding area.

[0017] The fourth hotspot determination module is used to determine the intersection of the second potential hotspot grid and the third potential hotspot grid as a candidate hotspot grid, and to verify the environmental rationality of each candidate hotspot grid using the multi-source data, and to determine the candidate hotspot grid that passes the verification as a reliable hotspot grid.

[0018] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method for determining any of the above-mentioned atmospheric pollutant hotspots, thereby solving the technical problems of low accuracy, poor dynamic adaptability, and false alarms in hotspot identification in the prior art.

[0019] This invention also provides a computer-readable storage medium storing a computer program that executes any of the above-described methods for determining atmospheric pollutant hotspots, in order to solve the technical problems of low accuracy, poor dynamic adaptability, and false alarms in hotspot identification in the prior art.

[0020] Compared with the prior art, the beneficial effects that the above-mentioned at least one technical solution adopted in the embodiments of this specification can achieve include at least the following: This application proposes to divide atmospheric pollutant concentration data into grid data of a preset resolution, and to perform spatial consistency processing on the grid data and multi-source data, realizing the transformation of data from the station scale to a fine grid scale, constructing a hotspot identification architecture based on high-resolution pollutant grid data, which can support hotspot identification at a small scale and effectively improve the accuracy of hotspot location; at the same time, by eliminating the spatial constraints of the traditional discrete station distribution, the ability to trace pollution sources can be significantly enhanced, the spatial accuracy of hotspot identification can be improved, and more precise spatial targets can be provided for environmental supervision; in the process of identifying hotspot grids, this application, in addition to adopting a time-dimensional anomaly identification algorithm, also proposes a basic Based on the dynamic changes of the spatiotemporal dynamic anomaly index in the spatiotemporal coupling dimension, a third potential hotspot grid is determined in the first potential hotspot grid, realizing a dynamic anomaly identification mechanism based on spatiotemporal coupling. This achieves effective separation of background pollution fluctuations and sudden anomalies, and enables targeted responses to pollution characteristics in different regions and time periods. It is beneficial to enhance the sensitivity of capturing sudden point source pollution and mobile pollution sources, thereby enhancing the adaptability of dynamic pollution scenarios. In addition, this application also proposes to use the multi-source data to verify the environmental rationality of each candidate hotspot grid, realizing a collaborative verification mechanism for multi-source data. This effectively reduces hotspot misjudgments caused by abnormal meteorological conditions, land use type conflicts, and spatial mismatch of pollution sources, enhances the system's ability to distinguish between real emission sources and background pollution accumulation, and reduces environmental hotspot identification misjudgments. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of a method for determining atmospheric pollutant hotspots provided in an embodiment of the present invention;

[0023] Figure 2 This is an example flowchart of a method for determining atmospheric pollutant hotspots provided in an embodiment of the present invention;

[0024] Figure 3 This is a structural block diagram of a computer device provided in an embodiment of the present invention;

[0025] Figure 4 This is a structural block diagram of an air pollutant hotspot determination device provided in an embodiment of the present invention. Detailed Implementation

[0026] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0027] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. This application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0028] In this embodiment of the invention, a method for determining atmospheric pollutant hotspots is provided, such as... Figure 1 As shown, the method includes:

[0029] Step S101: Obtain atmospheric pollutant concentration data and multi-source data related to atmospheric pollution from multi-scale stations within the target area, divide the atmospheric pollutant concentration data into grid data of a preset resolution, and perform spatial consistency processing on the grid data and the multi-source data.

[0030] Step S102: Among all grids in the grid data, the grids with atmospheric pollutant concentration values ​​higher than the first dynamic concentration threshold are identified as the first potential hotspot grids;

[0031] Step S103: Determine the second potential hotspot grid in the first potential hotspot grid based on the time-dimensional anomaly identification algorithm;

[0032] Step S104: Calculate the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids. Based on the dynamic changes of the spatiotemporal dynamic anomaly index in the spatiotemporal coupling dimension, determine the third potential hotspot grid in the first potential hotspot grids. The spatiotemporal dynamic anomaly index represents the degree of deviation of the current atmospheric pollutant concentration value of the target grid from the historical dynamic baseline of atmospheric pollutant concentration in the surrounding area.

[0033] Step S105: The intersection of the second potential hotspot grid and the third potential hotspot grid is determined as a candidate hotspot grid, and the environmental rationality of each candidate hotspot grid is verified using the multi-source data. The candidate hotspot grid that passes the verification is determined as a reliable hotspot grid.

[0034] Depend on Figure 1 As shown in the flowchart, in this embodiment of the invention, this application proposes to divide atmospheric pollutant concentration data into grid data of a preset resolution and perform spatial consistency processing on the grid data and multi-source data, realizing the transformation of data from the station scale to a fine grid scale, constructing a hotspot identification architecture based on high-resolution pollutant grid data, which can support hotspot identification at a small scale and effectively improve the accuracy of hotspot location; at the same time, by eliminating the spatial constraints of traditional discrete station distribution, the ability to trace pollution sources can be significantly enhanced, the spatial accuracy of hotspot identification can be improved, and more precise spatial targets can be provided for environmental supervision; in the process of identifying hotspot grids, in addition to using a time-dimensional anomaly identification algorithm, this application also proposes a method based on the spatiotemporal dynamic anomaly index... The dynamic changes in the spatiotemporal coupling dimension are used to determine the third potential hotspot grid in the first potential hotspot grid, realizing a dynamic anomaly identification mechanism based on spatiotemporal coupling. This achieves effective separation of background pollution fluctuations and sudden anomalies, and enables targeted responses to pollution characteristics in different regions and time periods. It is beneficial to enhance the sensitivity of capturing sudden point source pollution and mobile pollution sources, thereby enhancing the adaptability to dynamic pollution scenarios. In addition, this application also proposes to use the multi-source data to verify the environmental rationality of each candidate hotspot grid, realizing a collaborative verification mechanism for multi-source data. This effectively reduces hotspot misjudgments caused by abnormal meteorological conditions, land use type conflicts, and spatial mismatch of pollution sources, and enhances the system's ability to distinguish between real emission sources and background pollution accumulation, thereby reducing environmental hotspot identification misjudgments.

[0035] In practice, during the acquisition of atmospheric pollutant concentration data and multi-source data related to air pollution from multiple scale stations within the target area, multi-scale station observation data—including real-time monitoring records of atmospheric pollutant concentrations from national, provincial, and high-density monitoring stations—can be acquired as the aforementioned atmospheric pollutant concentration data. Simultaneously, meteorological elements that integrate the Global Forecast System (GFS) and ERA5 reanalysis datasets, covering key atmospheric diffusion parameters such as rainfall, wind speed, and wind direction, as well as pollution source data and high-precision land use type raster data that integrate spatial distribution information of industrial, mobile, catering, and dust sources, can be used as the aforementioned multi-source data related to air pollution.

[0036] In specific implementation, during the process of dividing the atmospheric pollutant concentration data into grid data of a preset resolution and performing spatial consistency processing on the grid data and the multi-source data, algorithms such as physicochemical transport models or machine learning can be used to train the model. Then, the collected multi-scale station real-time pollutant monitoring data (i.e., atmospheric pollutant concentration data), meteorological data, and high-precision land use data are used as inputs to the model to establish a grid dataset of pollutant concentration data at a preset resolution (e.g., a 1 km high resolution). The coefficient of determination R is used as the input.2 Verify the accuracy of the model.

[0037] In practical implementation, after constructing grid data from atmospheric pollutant concentration data, the pollutant grid can be used as a spatial reference to perform resolution adaptation processing on meteorological and land use data. Simultaneously, pollution source data is converted into grid-type data formats, and grids without pollution sources are labeled "non-emission zones," achieving spatial consistency in the representation of various data types. Furthermore, a unified coding system covering the entire region can be constructed, assigning a unique identifier to each grid and simultaneously recording the coordinates of the grid's center point, the latitude and longitude of the grid's four sides, and its administrative region information to support subsequent personnel scheduling and management.

[0038] In practice, grids with air pollutant concentrations exceeding a first dynamic concentration threshold are designated as first potential hotspot grids, thus enabling the selection of these grids based on real-time pollutant concentration distribution characteristics. For example, ... Figure 2 As shown, after dividing the atmospheric pollutant concentration data into grid data with a preset resolution (such as 1×1 km), the entire grid can be scanned based on the real-time pollutant concentration distribution characteristics, using the 80th percentile of the concentration values ​​of all grids in the study area (i.e. the target area mentioned above) as a dynamic discrimination threshold. Grids with concentration values ​​higher than this threshold are marked as the first potential hotspot grid PG1, which constitutes the object set for subsequent spatiotemporal coupling analysis.

[0039] In specific implementation, during the process of determining the second potential hotspot grid in the first potential hotspot grid based on the time-dimensional anomaly identification algorithm, such as... Figure 2 As shown, the second potential hotspot grid may include potential hotspot grids identified using one or more methods based on temporal anomalies, for example,

[0040] The sliding quartile method can be used:

[0041] In the time-dimensional anomaly identification stage, the sliding quartile method is used as the basic discrimination tool to capture sudden increases in atmospheric pollutant concentrations in a single grid over time. For the target grid point, the sliding quartile method uses the current monitoring time as the end of the sliding window and selects a continuous data sequence of length H backwards to form the current analysis window (for short-term anomaly identification, H can be set to 5). The interquartile range (IQR) of the pollutant concentration sequence within the window is calculated as a key statistic to measure the degree of data dispersion. Simultaneously, a dynamically adjusted anomaly determination coefficient α1 is set, adaptively determined based on the real-time Air Quality Index (AQI): when the AQI value is less than 100 (indicating good to lightly polluted air quality), α1 is 3.0; when the AQI value is greater than or equal to 100 (indicating moderate or more severe air pollution), α1 increases to 4.5. This design aims to moderately relax the anomaly determination criteria when the background pollution concentration is high, avoiding misjudging normal fluctuations as anomalies due to an overall increase in pollution levels. Finally, the current atmospheric pollutant concentration observation value C at the target grid point is... T Compare with the α1×IQR threshold calculated within this window. If C T Strictly greater than this threshold, i.e., satisfying C T If the value is greater than α1×IQR, then the grid point is determined to have a significant high value anomaly in the time dimension at the current time, and it is identified as a second potential hotspot grid PG2.

[0042] Time series constraints can also be used:

[0043] In the time-dimensional anomaly identification stage, to improve the accuracy of capturing real sudden pollution events and reduce false positives, a time-series constraint is introduced as one of the discrimination conditions. The core purpose of this constraint is to effectively eliminate false anomaly signals caused by continuous fluctuations in normal background concentration. Specifically, this method sets the atmospheric pollutant concentration C at the target grid point at the current time T as follows: T It needs to meet specific recent evolution settings, namely: C T-2 <C T-1 <C T And C T-3 and C T-4 All less than C T C T-1 C T-2 C T-3 C T-4 These represent the atmospheric pollutant concentration values ​​at times 1, 2, 3, and 4 before time T. Grids that meet the above criteria are identified as another type of potential hotspot grid, PG3.

[0044] In specific implementation, based on the dynamic changes of the spatiotemporal dynamic anomaly index in the spatiotemporal coupling dimension, in the process of determining the third potential hotspot grid in the first potential hotspot grid, to overcome the limitations of traditional static spatial comparison methods and improve adaptability to dynamic changes in regional background pollution, this application introduces the Spatiotemporal Dynamic Anomaly Index (STDAI) as a core discrimination indicator. This method aims to quantify the relative deviation of the current atmospheric pollutant concentration at the target grid point from the historical dynamic baseline of its surrounding area. The calculation of the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids includes:

[0045] Calculate the first Z-score normalized result of the atmospheric pollutant concentration value sequence for each of the first potential hotspot grids within a time window of a preset time length;

[0046] Calculate the average pollutant concentration of the atmospheric pollutant concentration value sequence of all grids within the surrounding spatial buffer of each first potential hot spot grid within a time window of a preset time length, and calculate the second Z-score normalized result of the average pollutant concentration of all grids.

[0047] Based on the first Z-score normalization result and the second Z-score normalization result, the spatiotemporal dynamic anomaly index of each of the first potential hotspot grids is calculated.

[0048] For example, the spatiotemporal dynamic anomaly index of each of the first potential hotspot grids is calculated using the following formula:

[0049]

[0050] in, The spatiotemporal dynamic anomaly index for each of the first potential hotspot grids; The first Z-score normalized result (e.g., the first Z-score normalized result of the atmospheric pollutant concentration value sequence within a dynamic time window of length 5 [t-4, t] for target grid point g). The second Z-score normalized result is the average Z-score normalized result of the atmospheric pollutant concentration sequence of all grid points within the same time window [t-4, t] in the spatial buffer (e.g., the radius of the spatial buffer is 3km) around the target grid point g, i.e., the regional dynamic baseline.

[0051] In specific implementation, the surrounding spatial buffer of each of the first potential hotspot grids can be a spatial buffer that includes a preset range of the first potential hotspot grid. For example, a square spatial buffer with a radius of r (e.g., r=3km) can be established with the first potential hotspot grid as the center, and the grid points within this spatial buffer are used to calculate the regional dynamic baseline.

[0052] In specific implementation, to further quantify and accurately determine the dynamic changes of the spatiotemporal dynamic anomaly index in the spatiotemporal coupling dimension, so as to improve the screening accuracy of potential hotspot grids, a third potential hotspot grid is determined in the first potential hotspot grid based on the dynamic changes of the spatiotemporal dynamic anomaly index in the spatiotemporal coupling dimension, including:

[0053] For each of the first potential hotspot grids, the dynamic spatiotemporal threshold of each first potential hotspot grid is calculated based on the mean and standard deviation of the spatiotemporal dynamic anomaly index of the historical period of each first potential hotspot grid.

[0054] Based on the relationship between the spatiotemporal dynamic anomaly index and the dynamic spatiotemporal threshold of each of the first potential hotspot grids, a third potential hotspot grid is determined in the first potential hotspot grids.

[0055] For example, based on the mean and standard deviation of the spatiotemporal dynamic anomaly index of each of the first potential hotspot grids during the same historical period, the dynamic spatiotemporal threshold of each first potential hotspot grid is calculated, including:

[0056] The dynamic spatiotemporal threshold for each of the first potential hotspot grids is calculated using the following formula:

[0057]

[0058] in, For dynamic spatiotemporal thresholds; The mean of the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids during the same historical period (e.g., the same period within the past month); The standard deviation of the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids during the historical period; This is the sensitivity coefficient (for example, it can be set to 2).

[0059] In specific implementation, in order to accurately identify hotspot grids where the increase in air pollutant concentration is not only sudden in time but also exceeds the dynamic average concentration level of the surrounding buffer zone during the same period in history, a third potential hotspot grid is determined in the first potential hotspot grid based on the relationship between the spatiotemporal dynamic anomaly index and the dynamic spatiotemporal threshold of each first potential hotspot grid, including:

[0060] The first potential hotspot grid whose spatiotemporal dynamic anomaly index is greater than the dynamic spatiotemporal threshold is identified as the third potential hotspot grid PG4. For example... Figure 2 As shown, that is, if > If the first potential hotspot grid is found to have a significant abnormal deviation in the spatiotemporal coupling dimension, it indicates that the increase in pollutant concentration in the grid is not only sudden in time, but also higher than the dynamic average concentration level of its surrounding area in the same period in history, thus constituting a potential pollution hotspot core, namely the third potential hotspot grid PG4.

[0061] In specific implementation, the intersection of the second potential hotspot grid and the third potential hotspot grid PG4 is determined as the candidate hotspot grid, such as... Figure 2 As shown, the identification of pollutant hotspot grids in the spatiotemporal dimension can simultaneously satisfy the anomaly feature determination of the above methods. That is, the second potential hotspot grid can include the two potential hotspot grids PG2 and PG3 mentioned above. The potential hotspot grids PG2 and PG3 identified by conventional time dimension anomaly identification and the potential hotspot grid PG4 identified by spatiotemporal dynamic deviation analysis method are integrated, and the intersection of PG2, PG3 and PG4, PG2∩PG3∩PG4, is taken as the candidate hotspot grid PG51.

[0062] In practice, to further improve the accuracy of hotspot identification, optimize the practicality of hotspot reporting, and adapt to the needs of different application scenarios, a further screening can be performed on the candidate hotspot grid PG51. For example, if there are multiple potential hotspot grids in a square area with a radius of R (e.g., R=3km) centered on the target candidate hotspot grid PG51, only the single grid with the largest Spatiotemporal Dynamic Anomaly Index (STDAI) value is retained as the final candidate hotspot grid PG52 for identification in that area.

[0063] In practical implementation, to further improve the accuracy of hotspot identification and reduce false alarms, a process is proposed to verify the rationality of each candidate hotspot grid environment using the multi-source data, thereby achieving multi-source collaborative environment credibility verification and eliminating false hotspots caused by environmental interference, such as:

[0064] If each candidate hotspot grid (which can be PG51 or PG52) does not fall under any of the following four conditions, then the candidate hotspot grid is determined to have passed the environmental rationality verification and is designated as a trustworthy hotspot grid PGF:

[0065] At the current moment, the ground wind speed in the area where the candidate hotspot grid is located is greater than level 4 (used to eliminate strong wind interference).

[0066] At the current moment, the area where the candidate hotspot grid is located is experiencing rain or snow (to eliminate interference from precipitation processes).

[0067] The land use type of the candidate hotspot grid does not conform to the typical emission or accumulation characteristics of the target pollutant (e.g., hotspot grids should not exist in environments such as water bodies, forests, and wetlands) (used to eliminate land use type conflicts).

[0068] Simultaneously satisfying the following two sub-conditions: there are no pollution sources within the preset range (e.g., radius > 10km) where the candidate hotspot grid is located, and there are no pollution sources upwind of the candidate hotspot grid (e.g., a fan-shaped area ±45° of the prevailing wind direction) (used to eliminate spatial mismatch of pollution sources).

[0069] In specific implementation, such as Figure 2 As shown, based on the finally identified reliable hotspot grid PGF, information integration operations are performed to associate the corresponding attributes of the hotspot grid (unique grid code, geographic coordinates of the grid center point, latitude and longitude range around the grid, and information on the administrative region to which it belongs) and generate a structured hotspot grid management record output.

[0070] In this embodiment, a computer device is provided, such as... Figure 3 As shown, it includes a memory 301, a processor 302, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method for determining any of the above-mentioned atmospheric pollutant hotspots.

[0071] Specifically, the computer device can be a computer terminal, a server, or a similar computing device.

[0072] In this embodiment, a computer-readable storage medium is provided, which stores a computer program that performs any of the above-described methods for determining atmospheric pollutant hotspots.

[0073] Specifically, computer-readable storage media, including both permanent and non-permanent, removable and non-removable media, can store information using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable 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 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 storage media does not include transient media, such as modulated data signals and carrier waves.

[0074] Based on the same inventive concept, this invention also provides an apparatus for determining atmospheric pollutant hotspots, as described in the following embodiments. Since the principle of the apparatus for determining atmospheric pollutant hotspots is similar to that of the method for determining atmospheric pollutant hotspots, the implementation of the apparatus can refer to the implementation of the method for determining atmospheric pollutant hotspots, and repeated details will not be elaborated further. As used below, the terms "unit" or "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0075] Figure 4 This is a structural block diagram of an air pollutant hotspot determination device according to an embodiment of the present invention, such as... Figure 4 As shown, it includes:

[0076] Data processing module 401 is used to acquire atmospheric pollutant concentration data and atmospheric pollution-related multi-source data from multi-scale stations within the target area, divide the atmospheric pollutant concentration data into grid data of a preset resolution, and perform spatial consistency processing on the grid data and the multi-source data.

[0077] The first hotspot determination module 402 is used to determine the grids with atmospheric pollutant concentration values ​​higher than the first dynamic concentration threshold as the first potential hotspot grids among all grids in the grid data.

[0078] The second hotspot determination module 403 is used to determine the second potential hotspot grid in the first potential hotspot grid based on the time-dimensional anomaly identification algorithm.

[0079] The third hotspot determination module 404 is used to calculate the spatiotemporal dynamic anomaly index of each of the first potential hotspot grids, and determine the third potential hotspot grid in the first potential hotspot grid based on the dynamic changes of the spatiotemporal dynamic anomaly index in the spatiotemporal coupling dimension. The spatiotemporal dynamic anomaly index represents the degree of deviation of the current atmospheric pollutant concentration value of the target grid from the historical dynamic baseline of atmospheric pollutant concentration in the surrounding area.

[0080] The fourth hotspot determination module 405 is used to determine the intersection of the second potential hotspot grid and the third potential hotspot grid as a candidate hotspot grid, and to verify the environmental rationality of each candidate hotspot grid using the multi-source data, and to determine the candidate hotspot grid that passes the verification as a reliable hotspot grid.

[0081] In one embodiment, the third hotspot determination module is configured to: calculate a first Z-score normalized result of the atmospheric pollutant concentration value sequence of each of the first potential hotspot grids within a preset time window; calculate the average pollutant concentration of the atmospheric pollutant concentration value sequence of all grids within the surrounding spatial buffer of each of the first potential hotspot grids within a preset time window; and calculate a second Z-score normalized result of the average pollutant concentration of all grids; and calculate the spatiotemporal dynamic anomaly index of each of the first potential hotspot grids based on the first Z-score normalized result and the second Z-score normalized result.

[0082] In one embodiment, the third hotspot determination module is used to calculate the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids using the following formula:

[0083]

[0084] in, The spatiotemporal dynamic anomaly index for each of the first potential hotspot grids; This is the standardized result of the first Z-score; This is the standardized result of the second Z-score.

[0085] In one embodiment, the third hotspot determination module is configured to, for each of the first potential hotspot grids, calculate a dynamic spatiotemporal threshold for each first potential hotspot grid based on the mean and standard deviation of the spatiotemporal dynamic anomaly index of each first potential hotspot grid during the same historical period; and determine a third potential hotspot grid among the first potential hotspot grids according to the relationship between the spatiotemporal dynamic anomaly index and the dynamic spatiotemporal threshold of each first potential hotspot grid.

[0086] In one embodiment, the third hotspot determination module is used to calculate the dynamic spatiotemporal threshold of each of the first potential hotspot grids using the following formula:

[0087]

[0088] in, For dynamic spatiotemporal thresholds; The mean of the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids during the same historical period; The standard deviation of the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids during the historical period; This is the sensitivity coefficient.

[0089] In one embodiment, the third hotspot determination module is used to determine the first potential hotspot grids whose spatiotemporal dynamic anomaly index is greater than the dynamic spatiotemporal threshold as the third potential hotspot grids.

[0090] In one embodiment, the fourth hotspot determination module is used to determine that a candidate hotspot grid passes the environmental rationality verification if each candidate hotspot grid does not fall under the following four conditions: the ground wind speed in the area where the candidate hotspot grid is located is greater than level 4 at the current time; rain or snow occurs in the area where the candidate hotspot grid is located at the current time; the land use type of the candidate hotspot grid does not conform to the typical emission or accumulation characteristics of the target pollutant; and the following two sub-conditions are met simultaneously: there is no pollution source within the preset range where the candidate hotspot grid is located and there is no pollution source upwind of the candidate hotspot grid.

[0091] The embodiments of this invention achieve the following technical effects: This application proposes dividing atmospheric pollutant concentration data into grid data of a preset resolution and performing spatial consistency processing on the grid data and multi-source data, realizing the transformation of data from the station scale to a fine grid scale, constructing a hotspot identification architecture based on high-resolution pollutant grid data, which can support hotspot identification at a small scale and effectively improve the accuracy of hotspot location; at the same time, by eliminating the spatial constraints of traditional discrete station distribution, it can significantly enhance the ability to trace pollution sources, improve the spatial accuracy of hotspot identification, and provide more precise spatial targets for environmental supervision; in the process of identifying hotspot grids, in addition to using a time-dimensional anomaly identification algorithm, this application also proposes a method based on the spatiotemporal dynamic anomaly index in time... The dynamic changes in the spatial coupling dimension are used to determine the third potential hotspot grid in the first potential hotspot grid, realizing a dynamic anomaly identification mechanism based on spatiotemporal coupling. This achieves effective separation of background pollution fluctuations and sudden anomalies, and enables targeted responses to pollution characteristics in different regions and time periods. This enhances the sensitivity of capturing sudden point source pollution and mobile pollution sources, thereby improving the adaptability to dynamic pollution scenarios. In addition, this application also proposes to use the multi-source data to verify the environmental rationality of each candidate hotspot grid, realizing a collaborative verification mechanism for multi-source data. This effectively reduces hotspot misjudgments caused by abnormal meteorological conditions, land use type conflicts, and spatial mismatch of pollution sources, and enhances the system's ability to distinguish between real emission sources and background pollution accumulation, thereby reducing environmental hotspot identification misjudgments.

[0092] Obviously, those skilled in the art should understand that the modules or steps of the above-described embodiments of the present invention can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the embodiments of the present invention are not limited to any particular hardware and software combination.

[0093] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations can be made to the embodiments 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 protection scope of the present invention.

Claims

1. A method for determining atmospheric pollutant hotspots, characterized in that, include: Acquire atmospheric pollutant concentration data and multi-source data related to atmospheric pollution from multiple scale stations within the target area, divide the atmospheric pollutant concentration data into grid data of a preset resolution, and perform spatial consistency processing on the grid data and the multi-source data; Among all grids in the grid data, the grids with atmospheric pollutant concentration values ​​higher than the first dynamic concentration threshold are identified as the first potential hotspot grids; A second potential hotspot grid is determined in the first potential hotspot grid based on a time-dimensional anomaly identification algorithm; Calculate the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids. Based on the dynamic changes of the spatiotemporal dynamic anomaly index in the spatiotemporal coupling dimension, determine the third potential hotspot grid in the first potential hotspot grids. The spatiotemporal dynamic anomaly index represents the degree of deviation of the current atmospheric pollutant concentration value of the target grid from the historical dynamic baseline of atmospheric pollutant concentration in the surrounding area. The intersection of the second potential hotspot grid and the third potential hotspot grid is determined as a candidate hotspot grid, and the environmental rationality of each candidate hotspot grid is verified using the multi-source data. The candidate hotspot grids that pass the verification are determined as credible hotspot grids. Calculate the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids, including: Calculate the first Z-score normalized result of the atmospheric pollutant concentration value sequence for each of the first potential hotspot grids within a time window of a preset time length; Calculate the average pollutant concentration of the atmospheric pollutant concentration value sequence of all grids within the surrounding spatial buffer of each first potential hot spot grid within a time window of a preset time length, and calculate the second Z-score normalized result of the average pollutant concentration of all grids. Based on the first Z-score normalization result and the second Z-score normalization result, calculate the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids; Based on the first Z-score normalization result and the second Z-score normalization result, the spatiotemporal dynamic anomaly index of each of the first potential hotspot grids is calculated, including: The spatiotemporal dynamic anomaly index for each of the first potential hotspot grids is calculated using the following formula: in, The spatiotemporal dynamic anomaly index for each of the first potential hotspot grids; This is the standardized result of the first Z-score; The second Z-score normalization result is given, where g is the target grid point and t is the time of the preset time window.

2. The method for determining air pollutant hotspots as described in claim 1, characterized in that, Based on the dynamic changes of the spatiotemporal dynamic anomaly index in the spatiotemporal coupling dimension, a third potential hotspot grid is determined in the first potential hotspot grid, including: For each of the first potential hotspot grids, the dynamic spatiotemporal threshold of each first potential hotspot grid is calculated based on the mean and standard deviation of the spatiotemporal dynamic anomaly index of the historical period of each first potential hotspot grid. Based on the relationship between the spatiotemporal dynamic anomaly index and the dynamic spatiotemporal threshold of each of the first potential hotspot grids, a third potential hotspot grid is determined in the first potential hotspot grids.

3. The method for determining air pollutant hotspots as described in claim 2, characterized in that, Based on the mean and standard deviation of the spatiotemporal dynamic anomaly index of each of the first potential hotspot grids during the same historical period, the dynamic spatiotemporal threshold of each of the first potential hotspot grids is calculated, including: The dynamic spatiotemporal threshold for each of the first potential hotspot grids is calculated using the following formula: in, For dynamic spatiotemporal thresholds; The mean of the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids during the same historical period; The standard deviation of the spatiotemporal dynamic anomaly index for each of the first potential hotspot grids during the historical period; This is the sensitivity coefficient.

4. The method for determining air pollutant hotspots as described in claim 2, characterized in that, Based on the relationship between the spatiotemporal dynamic anomaly index and the dynamic spatiotemporal threshold of each of the first potential hotspot grids, a third potential hotspot grid is determined from the first potential hotspot grids, including: The first potential hotspot grid whose spatiotemporal dynamic anomaly index is greater than the dynamic spatiotemporal threshold is identified as the third potential hotspot grid.

5. The method for determining atmospheric pollutant hotspots as described in any one of claims 1 to 4, characterized in that, Verifying the rationality of each candidate hotspot grid environment using the multi-source data includes: If each candidate hotspot grid does not fall under any of the following four conditions, then the candidate hotspot grid is determined to have passed the environmental rationality verification: At the current moment, the ground wind speed in the area where the candidate hotspot grid is located is greater than level 4; At the current moment, the area where the candidate hotspot grid is located is experiencing rain or snow. The land use types of the candidate hotspot grids do not conform to the typical emission or accumulation characteristics of the target pollutants; The following two sub-conditions must be met simultaneously: there are no pollution sources within the preset range where the candidate hotspot grid is located, and there are no pollution sources upwind of the candidate hotspot grid.

6. A device for determining atmospheric pollutant hotspots, characterized in that, include: The data processing module is used to acquire atmospheric pollutant concentration data and multi-source data related to atmospheric pollution from multi-scale stations within the target area, divide the atmospheric pollutant concentration data into grid data of a preset resolution, and perform spatial consistency processing on the grid data and the multi-source data. The first hotspot determination module is used to identify grids with atmospheric pollutant concentration values ​​higher than a first dynamic concentration threshold as first potential hotspot grids among all grids in the grid data. The second hotspot determination module is used to determine the second potential hotspot grid in the first potential hotspot grid based on the time-dimensional anomaly identification algorithm. The third hotspot determination module is used to calculate the spatiotemporal dynamic anomaly index of each of the first potential hotspot grids. Based on the dynamic changes of the spatiotemporal dynamic anomaly index in the spatiotemporal coupling dimension, the third potential hotspot grid is determined in the first potential hotspot grid. The spatiotemporal dynamic anomaly index represents the degree of deviation of the current atmospheric pollutant concentration value of the target grid from the historical dynamic baseline of atmospheric pollutant concentration in the surrounding area. The fourth hotspot determination module is used to determine the intersection of the second potential hotspot grid and the third potential hotspot grid as a candidate hotspot grid, and to verify the environmental rationality of each candidate hotspot grid using the multi-source data, and to determine the candidate hotspot grid that passes the verification as a reliable hotspot grid. The third hotspot determination module is used to calculate the first Z-score normalized result of the atmospheric pollutant concentration value sequence of each first potential hotspot grid within a preset time window; calculate the average pollutant concentration of the atmospheric pollutant concentration value sequence of all grids within the surrounding spatial buffer of each first potential hotspot grid within a preset time window, and calculate the second Z-score normalized result of the average pollutant concentration of all grids; calculate the spatiotemporal dynamic anomaly index of each first potential hotspot grid based on the first Z-score normalized result and the second Z-score normalized result; and calculate the spatiotemporal dynamic anomaly index of each first potential hotspot grid based on the first Z-score normalized result and the second Z-score normalized result using the following formula: in, The spatiotemporal dynamic anomaly index for each of the first potential hotspot grids; This is the standardized result of the first Z-score; The second Z-score normalization result is given, where g is the target grid point and t is the time of the preset time window.

7. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for determining atmospheric pollutant hotspots as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that performs the method for determining atmospheric pollutant hotspots according to any one of claims 1 to 5.

Images