A clustering method of air quality monitoring sites and related devices

Abstract

The application relates to the technical field of air quality monitoring, and provides a clustering method for air quality monitoring stations and related equipment, which comprises the following steps: obtaining pollutant monitoring data and spatial geographic information of a plurality of air quality monitoring stations in a target area; calculating a space-time consistency value between each two air quality monitoring stations based on all the pollutant monitoring data, and calculating a spatial correlation value between each two air quality monitoring stations based on all the spatial geographic information; setting a plurality of clustering partitions, and constructing a membership matrix; constructing a partition target function based on all the space-time consistency values and all the spatial correlation values, and iteratively updating the membership matrix by using the partition target function to obtain a final membership matrix; and clustering all the air quality monitoring stations by using the final membership matrix to obtain a station clustering result of the target area. The method can improve the accuracy of clustering of air quality monitoring stations.

Description

Technical Field

[0001] This application relates to the field of air quality monitoring technology, and in particular to a clustering method and related equipment for air quality monitoring stations. Background Technology

[0002] With the increasing demand for air pollution prevention and control and refined environmental management, various regions have established air quality monitoring networks covering urban areas, suburbs, and regional backgrounds. These monitoring stations continuously generate high-frequency time-series data on pollutants such as PM2.5 and ozone. Reasonable zoning and clustering of these monitoring stations helps identify the scope of pollution impact and joint prevention and control units, supporting pollution source tracing analysis, optimal allocation of monitoring resources, and evaluation of regional governance effectiveness, thus possessing significant engineering application value.

[0003] Existing site clustering methods are mostly spatial in nature, typically constructing similarity scores based on latitude and longitude distances, spatial adjacency, or static geographic attributes, and then using algorithms such as K-means, spectral clustering, and hierarchical clustering to obtain spatial partitions. Even when a temporal dimension is introduced, common practices often involve summarizing features such as mean value, quantile statistics, or periodic averages of time series data, or simply weighting and merging temporal and spatial distances / adding smoothing terms. This approach easily weakens the phased nature of pollution processes, short-term anomalies, and potential delayed propagation relationships between sites, resulting in insufficient distinguishing ability in cases where sites are "spatially close but have different temporal sequences" or "spatially distant but affected by the same pollution process and have consistent temporal sequences." Furthermore, these methods are sensitive to noise and anomalies, leading to poor partitioning stability.

[0004] On the other hand, the spatial distribution of monitoring stations is usually uneven and there are differences in groups such as administrative divisions, station types, and exposure levels. If the goal is only to optimize the overall similarity, some zones are easily dominated by densely populated areas, some groups are not representative enough within the zone, and the size of the zones is extremely unbalanced or even "cluster collapse", which leads to low accuracy in the clustering of air quality monitoring stations. Summary of the Invention

[0005] This application provides a clustering method and related equipment for air quality monitoring stations, which can solve the problem of low accuracy in clustering air quality monitoring stations.

[0006] In a first aspect, embodiments of this application provide a clustering method for air quality monitoring stations, the clustering method comprising: Acquire pollutant monitoring data and spatial geographic information from multiple air quality monitoring stations in the target area; The spatiotemporal consistency value between every two air quality monitoring stations is calculated based on all pollutant monitoring data, and the spatial correlation value between every two air quality monitoring stations is calculated based on all spatial geographic information. Multiple clustering partitions are defined, and a membership matrix is ​​constructed; the elements in the membership matrix represent the membership degree between air quality monitoring stations and clustering partitions. A partitioning objective function is constructed based on all spatiotemporal consistency values ​​and all spatial correlation values, and the membership matrix is ​​iteratively updated using the partitioning objective function to obtain the final membership matrix. The final membership matrix was used to cluster all air quality monitoring stations to obtain the station clustering results for the target area.

[0007] Optionally, the spatiotemporal consistency value between every two air quality monitoring stations is calculated based on all pollutant monitoring data, including: Through the formula: ; Calculate the first The air quality monitoring station and the first Spatiotemporal consistency values ​​among air quality monitoring stations ; in, Indicates the first The air quality monitoring station and the first The optimal time delay that maximizes the correlation coefficient between air quality monitoring stations within a given time lag range. Indicates the first Pollutant monitoring data from several air quality monitoring stations Indicates the first Pollutant monitoring data from several air quality monitoring stations The time delay is represented by an integer value. Indicates the first The pollutant time series from the first air quality monitoring station and the second Pollutant time series from several air quality monitoring stations, in terms of time lag The correlation coefficient between the shifted sequences Indicates the first The pollutant time series from the first air quality monitoring station and the second Pollutant time series from several air quality monitoring stations, in terms of time lag The correlation coefficient between the shifted sequences; Indicates the first The autocorrelation coefficient between the pollutant time series and its own time series at each air quality monitoring station. For time series displacement operators, Indicates the first The set of neighboring stations of an air quality monitoring station No. The set of neighboring stations of an air quality monitoring station This represents the maximum time delay.

[0008] Optionally, the spatial geographic information includes the geographic location data of air quality monitoring stations; The spatial correlation value between every two air quality monitoring stations is calculated based on all spatial geographic information, including: Based on the distance between the geographical locations of every two air quality monitoring stations, a spatial correlation value between every two air quality monitoring stations is calculated.

[0009] Optionally, the spatial correlation value between every two air quality monitoring stations can be calculated based on distance, including: Through the formula: ; Calculate the first The air quality monitoring station and the first Spatial correlation values ​​between air quality monitoring stations ; in, Represents the scale distance function. Indicates the first The air quality monitoring station and the first The distance between air quality monitoring stations.

[0010] Optionally, the partitioning objective function is: ; in, This represents the value of the objective function. Indicates the first The air quality monitoring station and the first Spatiotemporal consistency values ​​among air quality monitoring stations Indicates the first The air quality monitoring station and the first Spatial correlation values ​​between air quality monitoring stations Represents the membership matrix of the th element. Membership vector of each air quality monitoring station Indicates the first Membership vector of each air quality monitoring station Represents spatial transport consistency weights. Indicates environmental equity weighting. Indicates coverage of balanced weights. This indicates the number of air quality monitoring stations. Indicates the number of cluster partitions. Indicates the number of site groups. Indicates the first In the cluster partition, the th The percentage of each site group Indicates the first The baseline percentage of each site group in the entire domain. Indicates the first The coverage ratio of each cluster partition, Indicates the first The air quality monitoring station and the first Membership degree between cluster partitions.

[0011] Optionally, the membership matrix is ​​iteratively updated using the partitioning objective function to obtain the final membership matrix, including: The membership matrix is ​​differentiated using the partitioning objective function; The membership matrix is ​​updated using gradient descent based on the derivative result; Determine whether the updated membership matrix satisfies the termination condition; If so, the updated membership matrix will be used as the final membership matrix; Otherwise, use the updated membership matrix as the membership matrix and return the steps for constructing the partition objective function based on all spatiotemporal consistency values ​​and all spatial correlation values.

[0012] Optionally, the membership matrix can be updated using gradient descent based on the derivative result, including: Through the formula: ; Calculate the first The iteration of the ... The intermediate membership vector of each air quality monitoring station ; in, Indicates the first The iteration of the ... Membership vector of each air quality monitoring station , The maximum number of iterations. Indicates the first The iteration step size of the next iteration Represents the membership matrix of the th element. Membership vector of each air quality monitoring station Represents the membership matrix. Indicates the first The partition membership matrix obtained in the next iteration; Projecting and integrating all intermediate membership vectors yields the updated membership matrix.

[0013] Optionally, the termination condition is: ; in, Indicates the first The value of the objective function in the next iteration Indicates the first The value of the objective function in the next iteration Indicates the convergence threshold. Indicates the fairness constraint threshold. This indicates the coverage equalization threshold.

[0014] Secondly, embodiments of this application provide a clustering device for air quality monitoring stations, comprising: The acquisition module is used to acquire pollutant monitoring data and spatial geographic information from multiple air quality monitoring stations in the target area; The calculation module is used to calculate the spatiotemporal consistency value between every two air quality monitoring stations based on all pollutant monitoring data, and to calculate the spatial correlation value between every two air quality monitoring stations based on all spatial geographic information; The module is used to define multiple clustering partitions and construct a membership matrix; the elements in the membership matrix represent the membership degrees between air quality monitoring stations and clustering partitions. The update module is used to construct a partitioning objective function based on all spatiotemporal consistency values ​​and all spatial correlation values, and to iteratively update the membership matrix using the partitioning objective function to obtain the final membership matrix. The clustering module is used to cluster all air quality monitoring stations using the final membership matrix to obtain the station clustering results for the target area.

[0015] Thirdly, embodiments of this application provide a terminal 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 aforementioned clustering method for air quality monitoring stations.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned clustering method for air quality monitoring stations.

[0017] The above-mentioned solution in this application has the following beneficial effects: In the embodiments of this application, pollutant monitoring data and spatial geographic information from multiple air quality monitoring stations in the target area are acquired. Then, based on all pollutant monitoring data, the spatiotemporal consistency value between every two air quality monitoring stations is calculated, and based on all spatial geographic information, the spatial correlation value between every two air quality monitoring stations is calculated. Multiple clustering partitions are then set, and a membership matrix is ​​constructed. A partitioning objective function is then constructed based on all spatiotemporal consistency values ​​and all spatial correlation values, and the membership matrix is ​​iteratively updated using the partitioning objective function to obtain the final membership matrix. Finally, the final membership matrix is ​​used to cluster all air quality monitoring stations to obtain the station clustering results for the target area. Specifically, analyzing the spatiotemporal consistency and spatial correlation values ​​between air quality monitoring stations based on pollutant monitoring data and spatial geographic information considers the dynamic correlation and spatial influence relationships between monitoring stations, improving the information richness of station clustering. Constructing a partitioning objective function and using iteratively updating the membership matrix fully explores the connection between spatial quality monitoring stations and clustering partitions, improving the accuracy of air quality monitoring station clustering.

[0018] Other beneficial effects of this application will be described in detail in the following detailed description section. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, 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 of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating a clustering method for air quality monitoring stations provided in an embodiment of this application; Figure 2 A schematic diagram of the structure of a clustering device for an air quality monitoring station provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. Detailed Implementation

[0021] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0022] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0023] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0024] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0025] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0026] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0027] To address the low accuracy of existing air quality monitoring station clustering methods, this application provides a clustering method for air quality monitoring stations. This method acquires pollutant monitoring data and spatial geographic information from multiple air quality monitoring stations in a target area. Then, it calculates the spatiotemporal consistency value between every two air quality monitoring stations based on all pollutant monitoring data, and calculates the spatial correlation value between every two air quality monitoring stations based on all spatial geographic information. Multiple clustering partitions are then defined, and a membership matrix is ​​constructed. A partition objective function is then constructed based on all spatiotemporal consistency values ​​and all spatial correlation values. The membership matrix is ​​iteratively updated using the partition objective function to obtain the final membership matrix. Finally, the final membership matrix is ​​used to cluster all air quality monitoring stations to obtain the station clustering results for the target area. Among them, the spatiotemporal consistency and spatial correlation values ​​of air quality monitoring stations were analyzed based on pollutant monitoring data and spatial geographic information. The dynamic correlation and spatial influence between monitoring stations were considered to improve the information richness of station clustering. A partition objective function was constructed and used to iteratively update the membership matrix. The connection between spatial quality monitoring stations and cluster partitions was fully explored to improve the accuracy of air quality monitoring station clustering.

[0028] The following is an illustrative example of the clustering method for air quality monitoring stations provided in this application.

[0029] like Figure 1 As shown, the clustering method for air quality monitoring stations provided in this application includes the following steps: Step 11: Obtain pollutant monitoring data and spatial geographic information from multiple air quality monitoring stations in the target area.

[0030] The aforementioned target areas are those requiring air quality monitoring. The pollutant monitoring data includes the monitoring values ​​of pollutant indicators from air quality monitoring stations at multiple historical time points, such as PM2.5 concentration and ozone concentration. The aforementioned spatial geographic information includes the geographical location data of the air quality monitoring stations, administrative divisions, functional types, or population exposure levels (referring to the comprehensive degree of exposure of the population to risk or harmful factors within the area where the air quality monitoring station is located).

[0031] In some embodiments of this application, pollutant monitoring data of air quality monitoring stations can be obtained by accessing the control system of the air quality monitoring station, and spatial geographic information of the air quality monitoring station can be obtained by geospatial systems, positioning systems, etc.

[0032] Step 12: Calculate the spatiotemporal consistency value between every two air quality monitoring stations based on all pollutant monitoring data, and calculate the spatial correlation value between every two air quality monitoring stations based on all spatial geographic information.

[0033] The spatiotemporal consistency value described above describes the temporal and spatial consistency of air quality in the areas where two air quality monitoring stations are located. The higher the spatiotemporal consistency value, the greater the temporal and spatial consistency of air quality between the two areas. The spatial correlation value described above describes the degree of geographical correlation between the areas where two air quality monitoring stations are located. The higher the spatial correlation value, the greater the geographical correlation between the two areas.

[0034] In some embodiments of this application, the steps of calculating the spatiotemporal consistency value between every two air quality monitoring stations based on all pollutant monitoring data and calculating the spatial correlation value between every two air quality monitoring stations based on all spatial geographic information include: The first step is to calculate the spatiotemporal consistency value between every two air quality monitoring stations based on all pollutant monitoring data.

[0035] Specifically, through the formula: ; Calculate the first The air quality monitoring station and the first Spatiotemporal consistency values ​​among air quality monitoring stations .

[0036] in, Indicates the first The air quality monitoring station and the first The optimal time delay that maximizes the correlation coefficient between air quality monitoring stations within a given time lag range. Indicates the first Pollutant monitoring data from several air quality monitoring stations Indicates the first Pollutant monitoring data from several air quality monitoring stations The time delay is represented by an integer value. Indicates the first The pollutant time series from the first air quality monitoring station and the second Pollutant time series from several air quality monitoring stations, in terms of time lag The correlation coefficient between the shifted sequences Indicates the first The pollutant time series from the first air quality monitoring station and the second Pollutant time series from several air quality monitoring stations, in terms of time lag The correlation coefficient between the shifted sequences Indicates the first The autocorrelation coefficient between the pollutant time series and its own time series at each air quality monitoring station. For time series displacement operators, Indicates the first The set of neighboring stations of an air quality monitoring station No. The set of neighboring stations of an air quality monitoring station This represents the maximum time delay.

[0037] The second step involves calculating the spatial correlation value between each pair of air quality monitoring stations based on the distance between their geographical locations.

[0038] Specifically, through the formula: ; Calculate the first The air quality monitoring station and the first Spatial correlation values ​​between air quality monitoring stations .

[0039] in, Represents the scale distance function. Indicates the first The air quality monitoring station and the first The distance between air quality monitoring stations.

[0040] For example, the distance between the geographical locations of two air quality monitoring stations can be calculated using the Euclidean distance formula, etc.

[0041] Step 13: Set multiple clustering partitions and construct a membership matrix.

[0042] The elements in the membership matrix above represent the membership degree between air quality monitoring stations and cluster partitions.

[0043] In some embodiments of this application, multiple clustering partitions can be preset, the number of clustering partitions is a preset value, and an initialized membership matrix can be constructed.

[0044] Step 14: Construct a partitioning objective function based on all spatiotemporal consistency values ​​and all spatial correlation values, and use the partitioning objective function to iteratively update the membership matrix to obtain the final membership matrix.

[0045] The objective function described above is used to describe the accuracy of the membership matrix. The elements in the final membership matrix represent the final membership degrees between air quality monitoring stations and cluster partitions.

[0046] In some embodiments of this application, the steps of constructing a partitioning objective function based on all spatiotemporal consistency values ​​and all spatial correlation values, and iteratively updating the membership matrix using the partitioning objective function to obtain the final membership matrix include: The first step is to construct a partitioning objective function based on all spatiotemporal consistency values ​​and all spatial correlation values.

[0047] Specifically, the partitioning objective function is: ; in, This represents the value of the objective function. Indicates the first The air quality monitoring station and the first Spatiotemporal consistency values ​​among air quality monitoring stations Indicates the first The air quality monitoring station and the first Spatial correlation values ​​between air quality monitoring stations Represents the membership matrix of the th element. Membership vector of each air quality monitoring station Indicates the first Membership vector of each air quality monitoring station Represents spatial transport consistency weights. Indicates environmental equity weighting. Indicates coverage of balanced weights. This indicates the number of air quality monitoring stations. Indicates the number of cluster partitions. Indicates the number of site groups. Indicates the first In the cluster partition, the th The percentage of each site group Indicates the first The baseline percentage of each site group in the entire domain. Indicates the first The coverage ratio of each cluster partition, Indicates the first The air quality monitoring station and the first Membership degree between cluster partitions.

[0048] The membership vector of the aforementioned air quality monitoring stations is a vector composed of all the membership degrees corresponding to each air quality monitoring station. The station groups are based on attributes such as administrative divisions, functional types, or population exposure levels. and The expression is: ; The second step is to use the partition objective function to differentiate the membership matrix.

[0049] Specifically, the derivative result is: ; in, Indicates the first The air quality monitoring station and the first Membership degree between cluster partitions Indicates the first The soft size of the cluster partition, i.e., all air quality monitoring stations for the first cluster partition. The sum of the membership degrees of each cluster partition. Indicates the first The site group number to which each air quality monitoring station belongs.

[0050] For example, the first two terms of the partitioning objective function can be combined and denoted as: ; in, Indicates the first The air quality monitoring station and the first The air quality monitoring stations are consistent with each other in terms of time and space. Spatial correlation value The overall correlation strength obtained by linear combination.

[0051] In addition, to handle the proportional form in the fairness term, the following is introduced: ; in, Indicates the first The air quality monitoring station and the first Membership degree between cluster partitions Indicates the first In the nth cluster partition belonging to the th Soft count of each site group, Indicates that the first The air quality monitoring station belongs to the group number [number missing]. The summation condition corresponding to all site indices of the group.

[0052] get: ; in This is a site index used to iterate through all sites during the summation, distinguishing it from a fixed site index. .

[0053] First of all Find the partial derivative: Utilization of spatiotemporal consensus term and spatial transmission smoothing term as well as We can obtain: ; Fairness item have to: ; Apply the law of commerce to businesses: ; in For indicator functions, when If the value is 1, then the value is 0; otherwise, the value is 0.

[0054] Again Using the chain rule: ; Coverage balance terms are have to Therefore: ; Therefore, the above derivative result is obtained.

[0055] The third step is to update the membership matrix using gradient descent based on the derivative results.

[0056] Specifically, through the formula: ; Calculate the first The iteration of the ... The intermediate membership vector of each air quality monitoring station .

[0057] in, Indicates the first The iteration of the ... Membership vector of each air quality monitoring station , The maximum number of iterations. Indicates the first The iteration step size of the next iteration Represents the membership matrix of the th element. Membership vector of each air quality monitoring station Represents the membership matrix. Indicates the first The membership matrix obtained in the next iteration; Projecting and integrating all intermediate membership vectors yields the updated membership matrix.

[0058] Specifically, will Projection onto the set of probability constraints The updated membership vector is obtained. Then, all the updated membership vectors are integrated into a single matrix to obtain the updated membership matrix.

[0059] The fourth step is to determine whether the updated membership matrix satisfies the termination condition.

[0060] Specifically, the termination conditions are: ; in, Indicates the first The value of the objective function in the next iteration Indicates the first The value of the objective function in the next iteration This represents the convergence threshold, used to limit the upper limit of the change in the objective function value between two consecutive iterations. This represents the fairness constraint threshold, used to limit the upper limit of the maximum deviation between the proportion of each site group within each cluster partition and the overall baseline proportion. This represents the coverage balance threshold, used to limit the coverage ratio of each cluster partition from the uniform coverage ratio. The upper limit of the maximum deviation between them.

[0061] If so, the updated membership matrix will be used as the final membership matrix.

[0062] Otherwise, use the updated membership matrix as the membership matrix and return the steps for constructing the partition objective function based on all spatiotemporal consistency values ​​and all spatial correlation values.

[0063] For example, the third method can also be used. Next iteration step size Adaptive determination using backtracking search: Let Candidate solutions are obtained by following the above "gradient step-projection" method. If the candidate solution does not satisfy the descent condition of the objective function. Then let And recalculate until satisfied .

[0064] Step 15: Use the final membership matrix to cluster all air quality monitoring stations to obtain the station clustering results for the target area.

[0065] In one embodiment of this application, taking 66 air quality monitoring stations within the target area as an example, after clustering using the final membership matrix, the air quality monitoring stations can be divided into three clustering partitions, with 9, 18, and 39 stations in each partition, respectively. For example, if the partition label of an air quality monitoring station is 1, it indicates that the air quality monitoring station belongs to the first clustering partition. Specifically, for the clustering analysis of a certain province, the first clustering partition contains 9 air quality monitoring stations, mainly composed of stations in cities A and B, and is located in the southern part of the target area; the second clustering partition contains 18 air quality monitoring stations, mainly composed of stations in cities C, D, E, F, and G, and is located in the central-western transitional area; the third clustering partition contains 39 air quality monitoring stations, mainly composed of stations in cities F, I, J, K, L, M, and N, covering the northern and northeastern regions of the province. Therefore, the method of this application can classify stations with similar environmental characteristics and regional connections into the same cluster partition based on the temporal similarity of pollutants and spatial proximity of air quality monitoring stations, thereby reflecting the attribute differences between different cluster partitions.

[0066] For example, through the formula: ; Determine the zone labels for air quality monitoring stations The group label is used to describe the corresponding cluster partition. For example, if the group label is 1, it means that the air quality monitoring station belongs to the first cluster partition.

[0067] in, Indicates the first The air quality monitoring station and the first The final membership degree between cluster partitions.

[0068] It should be noted that after obtaining the site clustering results, the air quality of the target area can be analyzed based on the site clustering results, such as analyzing the changes in air quality monitoring data in different clustering zones.

[0069] For example, data from air quality monitoring stations in a certain province will be used as an illustration. The data used comes from a daily-scale station observation and meteorological element summary data table, which includes daily records from 66 monitoring stations (identified by site code) in a certain province from March 1, 2023 to February 29, 2024, totaling 23,973 "site-date" samples. The number of days recorded by each station during the observation period varies (approximately 239–366 days per station), which can be considered as including some missing data. Each sample includes pollutant indicators (such as PM2.5, ozone, etc.) and multiple meteorological elements (such as 2m air temperature, boundary layer height, relative humidity, sea level pressure, shortwave radiation, total cloud cover, 10m wind field, 500hPa wind field, precipitation, etc.), and includes basic information such as station latitude, longitude, and elevation. The geographical location (latitude and longitude) of the site can be provided by the site location table (China_airlocation.xls) and corresponds one-to-one with the site codes mentioned above; the site covers 14 cities in a certain province and can be used to define group indexes by city and other attributes to carry out fairness constraint related analysis.

[0070] Based on the above data, the site clustering results of the province are obtained according to the method of this application. The site clustering results can be evaluated by using the silhouette coefficient, average fairness deviation and cluster size equilibrium deviation.

[0071] It is worth mentioning that the analysis of the spatiotemporal consistency and spatial correlation values ​​among air quality monitoring stations based on pollutant monitoring data and spatial geographic information takes into account the dynamic correlation and spatial influence between monitoring stations, improves the information richness of station clustering, constructs a partition objective function and uses it to iteratively update the membership matrix, fully explores the connection between spatial quality monitoring stations and cluster partitions, and improves the accuracy of air quality monitoring station clustering.

[0072] Furthermore, the method of this application has the following advantages: By characterizing the temporal consistency between sites through spatiotemporal consensus, the clustering results can reflect the dynamic correlation of the pollution process, avoiding the problem of "degrading to spatial clustering after time averaging"; By characterizing the spatial influence relationships between sites through spatial transmission priors, the zoning results are made more consistent with the spatial continuity of pollution diffusion and transmission. Fairness is characterized by comparing the proportion of groups within a partition with the baseline proportion across the entire region, thereby improving the representativeness of the partitioning results and reducing the phenomenon of over- or under-representation of groups. By characterizing the partition size status through partition coverage ratio, a basis is provided for suppressing extreme imbalances in partition size and improving the stability of results.

[0073] The following is an exemplary description of the clustering device for air quality monitoring stations provided in this application.

[0074] like Figure 2 As shown, this application embodiment provides a clustering device for air quality monitoring stations. The clustering device 200 for air quality monitoring stations includes: The acquisition module 201 is used to acquire pollutant monitoring data and spatial geographic information from multiple air quality monitoring stations in the target area; The calculation module 202 is used to calculate the spatiotemporal consistency value between every two air quality monitoring stations based on all pollutant monitoring data, and to calculate the spatial correlation value between every two air quality monitoring stations based on all spatial geographic information; Module 203 is used to set multiple clustering partitions and construct a membership matrix; the elements in the membership matrix are the membership degrees between air quality monitoring stations and clustering partitions. The update module 204 is used to construct a partitioning objective function based on all spatiotemporal consistency values ​​and all spatial correlation values, and to iteratively update the membership matrix using the partitioning objective function to obtain the final membership matrix. Clustering module 205 is used to cluster all air quality monitoring stations using the final membership matrix to obtain the station clustering results for the target area.

[0075] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0076] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0077] like Figure 3 As shown, an embodiment of this application provides a terminal device, wherein the terminal device D10 of this embodiment includes: at least one processor D100 ( Figure 3The diagram shows only one processor, a memory D101, and a computer program D102 stored in the memory D101 and executable on the at least one processor D100, wherein the processor D100 executes the computer program D102 to implement the steps in any of the above method embodiments.

[0078] Specifically, when the processor D100 executes the computer program D102, it acquires pollutant monitoring data and spatial geographic information from multiple air quality monitoring stations in the target area. Then, based on all pollutant monitoring data, it calculates the spatiotemporal consistency value between every two air quality monitoring stations and the spatial correlation value between every two air quality monitoring stations based on all spatial geographic information. Multiple clustering partitions are then set, and a membership matrix is ​​constructed. Next, a partitioning objective function is constructed based on all spatiotemporal consistency values ​​and all spatial correlation values. The membership matrix is ​​iteratively updated using the partitioning objective function to obtain the final membership matrix. Finally, the final membership matrix is ​​used to cluster all air quality monitoring stations to obtain the station clustering results for the target area. In this process, the analysis of spatiotemporal consistency and spatial correlation values ​​between air quality monitoring stations based on pollutant monitoring data and spatial geographic information considers the dynamic correlation and spatial influence relationships between monitoring stations, improving the information richness of station clustering. The construction of the partitioning objective function and its iterative updating of the membership matrix fully explores the connection between spatial quality monitoring stations and clustering partitions, improving the accuracy of air quality monitoring station clustering.

[0079] The processor D100 can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0080] In some embodiments, the memory D101 may be an internal storage unit of the terminal device D10, such as a hard disk or memory of the terminal device D10. In other embodiments, the memory D101 may be an external storage device of the terminal device D10, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the terminal device D10. Furthermore, the memory D101 may include both internal and external storage units of the terminal device D10. The memory D101 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory D101 can also be used to temporarily store data that has been output or will be output.

[0081] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0082] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0083] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above-described embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a clustering method device / terminal device at an air quality monitoring station, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, such as a USB flash drive, a portable hard drive, a magnetic disk, or an optical disk.

[0084] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0085] Those skilled in the art will 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 or a combination of computer software and electronic hardware. 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 implementation should not be considered beyond the scope of this application.

[0086] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention.

Claims

1. A clustering method for air quality monitoring stations, characterized in that, include: Acquire pollutant monitoring data and spatial geographic information from multiple air quality monitoring stations in the target area; The spatiotemporal consistency value between every two air quality monitoring stations is calculated based on all pollutant monitoring data, and the spatial correlation value between every two air quality monitoring stations is calculated based on all spatial geographic information. Define multiple clustering partitions and construct a membership matrix; The elements in the membership matrix represent the membership degree between air quality monitoring stations and clustering partitions. A partitioning objective function is constructed based on all spatiotemporal consistency values ​​and all spatial correlation values, and the membership matrix is ​​iteratively updated using the partitioning objective function to obtain the final membership matrix. The final membership matrix is ​​used to cluster all air quality monitoring stations to obtain the station clustering results for the target area.

2. The clustering method according to claim 1, characterized in that, The calculation of the spatiotemporal consistency value between every two air quality monitoring stations based on all pollutant monitoring data includes: Through the formula: Calculate the first The air quality monitoring station and the first Spatiotemporal consistency values ​​among air quality monitoring stations ; in, Indicates the first The air quality monitoring station and the first The optimal time delay that maximizes the correlation coefficient between air quality monitoring stations within a given time lag range. Indicates the first Pollutant monitoring data from several air quality monitoring stations Indicates the first Pollutant monitoring data from several air quality monitoring stations The time delay is represented by an integer value. Indicates the first The pollutant time series from the first air quality monitoring station and the second Pollutant time series from several air quality monitoring stations, in terms of time lag The correlation coefficient between the shifted sequences Indicates the first The pollutant time series from the first air quality monitoring station and the second Pollutant time series from several air quality monitoring stations, in terms of time lag The correlation coefficient between the shifted sequences Indicates the first The autocorrelation coefficient between the pollutant time series and its own time series at each air quality monitoring station. For time series displacement operators, Indicates the first The set of neighboring stations of an air quality monitoring station No. The set of neighboring stations of an air quality monitoring station This represents the maximum time delay.

3. The clustering method according to claim 1, characterized in that, The spatial geographic information includes the geographical location data of air quality monitoring stations; The calculation of the spatial correlation value between every two air quality monitoring stations based on all spatial geographic information includes: Based on the distance between the geographical locations of every two air quality monitoring stations, a spatial correlation value between every two air quality monitoring stations is calculated according to the distance.

4. The clustering method according to claim 3, characterized in that, The calculation of the spatial correlation value between every two air quality monitoring stations based on the distance includes: Through the formula: Calculate the first The air quality monitoring station and the first Spatial correlation values ​​between air quality monitoring stations ; in, Represents the scale distance function. Indicates the first The air quality monitoring station and the first The distance between air quality monitoring stations.

5. The clustering method according to claim 1, characterized in that, The objective function for partitioning is: in, This represents the value of the objective function. Indicates the first The air quality monitoring station and the first Spatiotemporal consistency values ​​among air quality monitoring stations Indicates the first The air quality monitoring station and the first Spatial correlation values ​​between air quality monitoring stations Represents the membership matrix of the th element. Membership vector of each air quality monitoring station Indicates the first Membership vector of each air quality monitoring station Represents spatial transport consistency weights. Indicates environmental equity weighting. Indicates coverage of balanced weights. This indicates the number of air quality monitoring stations. Indicates the number of cluster partitions. Indicates the number of site groups. Indicates the first In the cluster partition, the th The percentage of each site group Indicates the first The baseline percentage of each site group in the entire domain. Indicates the first The coverage ratio of each cluster partition, Indicates the first The air quality monitoring station and the first Membership degree between cluster partitions.

6. The clustering method according to claim 5, characterized in that, The step of iteratively updating the membership matrix using the partitioning objective function to obtain the final membership matrix includes: The membership matrix is ​​differentiated using the partitioning objective function; The membership matrix is ​​updated using gradient descent based on the derivative result. Determine whether the updated membership matrix satisfies the termination condition; If so, the updated membership matrix shall be used as the final membership matrix; Otherwise, the updated membership matrix is ​​used as the membership matrix, and the step of constructing the partition objective function based on all spatiotemporal consistency values ​​and all spatial correlation values ​​is returned.

7. The clustering method according to claim 6, characterized in that, The step of updating the membership matrix using gradient descent based on the derivative result includes: Through the formula: Calculate the first The iteration of the ... The intermediate membership vector of each air quality monitoring station ; in, Indicates the first The iteration of the ... Membership vector of each air quality monitoring station , The maximum number of iterations. Indicates the first The iteration step size of the next iteration Represents the membership matrix of the th element. Membership vector of each air quality monitoring station Represents the membership matrix. Indicates the first The membership matrix obtained from the next iteration; Projecting and integrating all intermediate membership vectors yields the updated membership matrix.

8. The clustering method according to claim 7, characterized in that, The termination condition is: in, Indicates the first The value of the objective function in the next iteration Indicates the first The value of the objective function in the next iteration Indicates the convergence threshold. Indicates the fairness constraint threshold. This indicates the coverage equalization threshold.

9. A terminal 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 clustering method for air quality monitoring stations as described in any one of claims 1 to 8.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the clustering method for air quality monitoring stations as described in any one of claims 1 to 8.

Images