A method, system, device and medium for identifying carbon emission concentration areas

Abstract

The present invention discloses a method, system, device and medium for identifying carbon emission agglomeration areas. First, the carbon emission area is divided into several spatial units, various carbon emission attributes are obtained, and a spatiotemporal carbon emission matrix is ​​constructed, thereby determining the emission level of each spatial unit under various carbon emission attributes; secondly, the carbon emission spatial clustering hyperparameter is determined by a density clustering algorithm, and then the density clustering algorithm is used to set the carbon emission spatial clustering hyperparameter to perform cluster analysis on the spatial units, and several spatial clustering clusters are obtained; then, according to the emission level and several spatial clustering clusters, the agglomeration emission level of each spatial unit under various carbon emission attributes is determined, and a spatiotemporal agglomeration emission level matrix is ​​constructed; finally, the K-means clustering algorithm is used to cluster the spatial units to identify the category of carbon emission agglomeration areas. The present invention can realize the hierarchical identification of carbon emission agglomeration areas, and improve the accuracy and stability of the carbon emission agglomeration area identification results.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, system, device and medium for identifying carbon emission concentration areas. Background Art

[0002] Existing technologies usually use methods such as spatial statistical analysis and cluster analysis to identify carbon emission concentration areas. Common spatial statistical methods include Moran's I and spatial autocorrelation analysis, which use indices to quantify the degree of emission concentration. Cluster analysis often uses unsupervised clustering algorithms such as K-means clustering and density clustering, which divide areas with similar carbon emission characteristics into the same concentration area based on artificially given hyperparameters.

[0003] However, the existing carbon emission cluster identification methods can only identify high-concentration areas of carbon emissions, but are unable to classify the clusters, resulting in the identification results being unable to fully and accurately reflect the actual situation of the carbon emission clusters; and the selection of hyperparameters has an important impact on the carbon emission cluster identification results. The existing methods require humans to continuously adjust the hyperparameters based on experience and experiments, which not only increases the workload but also affects the stability of the identification results. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method, system, equipment and medium for identifying carbon emission concentration areas, so as to improve the accuracy and stability of the results of carbon emission concentration area identification.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a method for identifying carbon emission concentration areas, comprising:

[0006] Divide the carbon emission target area into several spatial units, obtain the carbon emission attributes of each spatial unit, and construct a spatiotemporal carbon emission matrix;

[0007] Determining the emission level of each spatial unit under each carbon emission attribute according to the spatiotemporal carbon emission matrix;

[0008] According to the spatiotemporal carbon emission matrix, a preliminary clustering analysis is performed on the spatial units by a density clustering algorithm to determine the carbon emission spatial clustering hyperparameters; a density clustering algorithm is set according to the carbon emission spatial clustering hyperparameters, and a clustering analysis is performed on the spatial units again by the set density clustering algorithm to obtain a number of spatial clustering clusters;

[0009] Determining the aggregate emission level of each spatial unit under each carbon emission attribute according to the emission level and the plurality of spatial clustering clusters;

[0010] According to the agglomeration emission level, constructing a spatiotemporal agglomeration emission level matrix;

[0011] Based on the spatiotemporal agglomeration emission level matrix, taking the carbon emission attributes in the spatiotemporal carbon emission matrix as features, clustering analysis is performed on spatial units by using a K-means clustering algorithm that determines hyperparameter values ​​based on the elbow rule to obtain a number of category clusters;

[0012] According to the plurality of category clusters, each spatial unit is identified as a different carbon emission concentration area category.

[0013] Optionally, the carbon emission attributes of each spatial unit include the daily carbon emission of each spatial unit and the carbon emission in each characteristic time period.

[0014] Optionally, determining the emission level of each spatial unit under each carbon emission attribute according to the spatiotemporal carbon emission matrix includes:

[0015] Selecting each carbon emission attribute in the spatiotemporal carbon emission matrix in turn, calculating the corresponding emission coverage ratios at different quantiles under the selected carbon emission attributes, and obtaining a plurality of emission coverage ratios;

[0016] Determining emission level characteristic quantiles of various carbon emission attributes according to the plurality of emission coverage ratios;

[0017] The emission level of each spatial unit under each carbon emission attribute is determined according to the emission level characteristic quantiles of the various carbon emission attributes.

[0018] Optionally, performing a preliminary clustering analysis on spatial units by a density clustering algorithm according to the spatiotemporal carbon emission matrix to determine carbon emission spatial clustering hyperparameters includes:

[0019] Selecting each carbon emission attribute in the spatiotemporal carbon emission matrix in turn, and selecting a plurality of emission level characteristic quantiles under the selected carbon emission attributes;

[0020] For each selected emission level characteristic quantile, select the spatial unit whose carbon emission is higher than the selected emission level characteristic quantile;

[0021] Performing a preliminary clustering analysis on the selected spatial units by using a density clustering algorithm to obtain a preliminary spatial clustering result;

[0022] According to the preliminary spatial clustering results, the grid search method is used to determine the hyperparameters of carbon emission spatial clustering.

[0023] Optionally, the determining carbon emission spatial clustering hyperparameters by using a grid search method according to the preliminary spatial clustering results includes:

[0024] Perform a grid search in the determined hyperparameter space, and record the number of spatial clustering clusters and the proportion of clustered carbon emissions obtained by clustering the preliminary spatial clustering results using a density clustering algorithm under different hyperparameter combination values;

[0025] Convert the indicator distribution of the number of spatial clusters and the proportion of cluster carbon emissions into indicator quantiles respectively;

[0026] Selecting a hyperparameter combination value that makes the index quantile of the number of spatial clusters and the index quantile of the proportion of cluster carbon emissions closest as the carbon emission spatial clustering hyperparameter;

[0027] The cluster carbon emission ratio is the ratio of the sum of carbon emissions of spatial units classified into any spatial cluster under any carbon emission attribute to the sum of carbon emissions of all spatial units.

[0028] Optionally, determining the aggregate emission level of each spatial unit under each carbon emission attribute according to the emission level and the plurality of spatial clustering clusters includes:

[0029] When a spatial unit is classified into any of the plurality of spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is respectively the emission level of the spatial unit under each carbon emission attribute;

[0030] When a spatial unit is not classified into any of the plurality of spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is 0;

[0031] For the same carbon emission attribute, if a spatial unit is simultaneously classified into different spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is the highest emission level of the spatial unit under the carbon emission attributes corresponding to different spatial clusters.

[0032] Optionally, the identifying each spatial unit as a different carbon emission concentration area category according to the plurality of category clusters includes:

[0033] According to the agglomeration emission levels of the plurality of clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include primary agglomeration area, secondary agglomeration area and tertiary agglomeration area;

[0034] According to the carbon emissions in each characteristic time period of the several category clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include stable agglomeration area, daytime peak agglomeration area, nighttime peak agglomeration area and commuting peak agglomeration area;

[0035] According to the spatial forms of the several category clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include regional agglomeration areas and channel agglomeration areas.

[0036] The embodiment of the present invention further provides a carbon emission concentration area identification system, which is used to implement the carbon emission concentration area identification method as described in any one of the above items, including:

[0037] The spatiotemporal carbon emission matrix construction module is used to divide the carbon emission target area into several spatial units, obtain the carbon emission attributes of each spatial unit, and construct the spatiotemporal carbon emission matrix;

[0038] An emission level determination module, used to determine the emission level of each spatial unit under each carbon emission attribute according to the spatiotemporal carbon emission matrix;

[0039] A carbon emission spatial clustering module is used to perform preliminary clustering analysis on spatial units through a density clustering algorithm according to the spatiotemporal carbon emission matrix, and determine carbon emission spatial clustering hyperparameters; set a density clustering algorithm according to the carbon emission spatial clustering hyperparameters, and perform clustering analysis on spatial units again through the set density clustering algorithm to obtain a number of spatial clustering clusters;

[0040] An aggregate emission level determination module, used to determine the aggregate emission level of each spatial unit under various carbon emission attributes according to the emission level and the plurality of spatial clusters;

[0041] A spatiotemporal agglomeration emission level matrix construction module, used to construct a spatiotemporal agglomeration emission level matrix according to the agglomeration emission level;

[0042] A carbon emission category clustering module is used to perform clustering analysis on spatial units based on the spatiotemporal agglomeration emission level matrix and each carbon emission attribute in the spatiotemporal carbon emission matrix as a feature by using a K-means clustering algorithm that determines hyperparameter values ​​based on the elbow rule to obtain a number of category clustering clusters;

[0043] The carbon emission concentration area identification module is used to identify each spatial unit as a different carbon emission concentration area category according to the plurality of category clusters.

[0044] An embodiment of the present invention provides a carbon emission concentration area identification device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the carbon emission concentration area identification method described in any one of the above.

[0045] An embodiment of the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the carbon emission concentration area identification methods described above.

[0046] Compared with the prior art, the present invention provides a method, system, device and medium for identifying carbon emission agglomeration areas. First, the carbon emission area is divided into several spatial units, various carbon emission attributes are obtained, and a spatiotemporal carbon emission matrix is ​​constructed, so as to determine the emission level of each spatial unit under various carbon emission attributes; secondly, the carbon emission spatial clustering hyperparameters are determined by a density clustering algorithm, and then the density clustering algorithm is used to set the carbon emission spatial clustering hyperparameters to perform clustering analysis on the spatial units, and several spatial clustering clusters are obtained; then, according to the emission level and several spatial clustering clusters, the agglomeration emission level of each spatial unit under various carbon emission attributes is determined, and a spatiotemporal agglomeration emission level matrix is ​​constructed; finally, the K-means clustering algorithm is used to cluster the spatial units to identify the carbon emission agglomeration area category. By introducing the emission grade method, the present invention can better reflect the carbon emission characteristics of spatial units, realize the hierarchical identification of carbon emission agglomeration areas, and improve the analysis and management capabilities of carbon emission data; then, the carbon emission spatial clustering hyperparameters are determined through the DBSCAN algorithm, which can automatically adjust and optimize the hyperparameters of the carbon emission spatial clustering model, which not only reduces the workload of parameter tuning, but also improves the accuracy and stability of the identification results. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flow chart of a preferred embodiment of a method for identifying carbon emission concentration areas provided by the present invention;

[0048] Figure 2 It is a structural block diagram of a preferred embodiment of a carbon emission concentration area identification system provided by the present invention;

[0049] Figure 3 It is a structural block diagram of a preferred embodiment of a carbon emission concentration area identification device provided by the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this technical field without creative work are within the scope of protection of the present invention.

[0051] See also Figure 1 , Figure 1It is a flowchart of a preferred embodiment of a method for identifying carbon emission concentration areas provided by the present invention, and the method comprises steps S1 to S7:

[0052] S1. Divide the carbon emission target area into several spatial units, obtain the carbon emission attributes of each spatial unit, and construct a spatiotemporal carbon emission matrix;

[0053] As an optional embodiment, the carbon emission attributes of each spatial unit include the daily carbon emission of each spatial unit and the carbon emission in each characteristic time period.

[0054] Specifically, the carbon emission target area is divided into several spatial units according to the rules to obtain the spatial unit serial number; the total daily carbon emissions in each spatial unit and the carbon emissions in each characteristic time period are used as the attributes of each carbon emission to obtain the carbon emission attribute serial number; the spatial unit serial number is used as the row and the carbon emission attribute serial number is used as the column to construct a spatiotemporal carbon emission matrix;

[0055] For example, a given carbon emission concentration area is identified as a target area, and the area is divided into spatial units according to a regular (square / honeycomb) grid (the number of spatial units is set to k). The size of the spatial unit can be adjusted according to the granularity of the data source and the needs of the application scenario. A 1km grid is recommended for the regional scale, a 500m grid is recommended for the urban scale, and a 100m grid is recommended for the microscopic urban area.

[0056] Then, based on the historical carbon emission data in the carbon emission geographic database, the total daily carbon emission and 24-hour carbon emission in each spatial unit are calculated. If the database contains carbon emission data for multiple days in a time period, the average daily carbon emission and 24-hour carbon emission in the time period can be used to reflect the carbon emission intensity of each spatial unit. Select a group of m characteristic time periods {T i} m , use the carbon emissions in these time periods to reflect the changes in carbon emissions in different time periods within a day. The characteristic time period can be selected according to actual needs. Common characteristic time periods include morning (8:00-12:00), noon (12:00-14:00), afternoon (14:00-18:00), evening (18:00-22:00), and night (22:00-6:00 the next day). Calculate the carbon emissions of each spatial unit in the selected characteristic time period respectively. Each spatial unit contains a total daily carbon emissions and carbon emissions in each characteristic time period, totaling m+1 carbon emission attributes. For example, referring to Table 1, the spatial unit number is used as the row and the carbon emission attribute number is used as the column to construct the spatiotemporal carbon emissions matrix E∈R k×(m+1) .

[0057] Table 1 Spatiotemporal carbon emissions matrix

[0058]

[0059] It is worth noting that the carbon emission target area is refined into multiple spatial units, and the spatiotemporal carbon emission matrix is ​​constructed by combining the daily carbon emission data in each spatial unit and the carbon emissions in the characteristic period. In this way, more detailed and comprehensive recording and management of carbon emission data can be achieved. Carbon emission is a dynamic process, and its aggregation situation may change with time. The present invention is no longer limited to the traditional carbon emission monitoring method that only focuses on spatial distribution, but comprehensively considers the spatiotemporal characteristics of carbon emissions; the carbon emissions in the characteristic period are introduced into the spatial clustering and agglomeration area identification process, which can capture the dynamic changes of carbon emissions in time and space, and distinguish agglomeration areas with similar overall emission intensity but different emission periods. Increasing the consideration of the time dimension can make the identification results more comprehensively reflect the actual situation of carbon emission agglomeration areas.

[0060] S2. Determine the emission level of each spatial unit under each carbon emission attribute according to the spatiotemporal carbon emission matrix;

[0061] As an optional embodiment of the above solution, determining the emission level of each spatial unit under each carbon emission attribute according to the spatiotemporal carbon emission matrix includes:

[0062] Selecting each carbon emission attribute in the spatiotemporal carbon emission matrix in turn, calculating the corresponding emission coverage ratios at different quantiles under the selected carbon emission attributes, and obtaining a plurality of emission coverage ratios;

[0063] Determining emission level characteristic quantiles of various carbon emission attributes according to the plurality of emission coverage ratios;

[0064] The emission level of each spatial unit under each carbon emission attribute is determined according to the emission level characteristic quantiles of the various carbon emission attributes.

[0065] Specifically, the calculation of the emission coverage ratios corresponding to different quantiles under the selected carbon emission attribute to obtain a plurality of emission coverage ratios, and the determination of the emission level characteristic quantiles of each carbon emission attribute according to the emission coverage ratios include:

[0066] Under the selected carbon emission attribute, calculate the carbon emission corresponding to each spatial unit divided into several quantiles at equal intervals;

[0067] Select any quantile in turn, and select each spatial unit whose carbon emissions are higher than the selected quantile, calculate the ratio of the sum of carbon emissions of each spatial unit whose carbon emissions are higher than the selected quantile to the sum of carbon emissions of all spatial units, and obtain several emission coverage ratios under the selected carbon emission attribute;

[0068] A number of emission coverage ratios are selected to obtain a group of quantiles corresponding to the selected emission coverage ratios, respectively, as the emission level characteristic quantiles under the selected carbon emission attribute.

[0069] Exemplarily, any carbon emission attribute in the spatiotemporal carbon emission matrix is ​​selected, and the carbon emissions under it are divided into 100 quantiles at 1% intervals (wherein the X% quantile is equal to the value ranked X% after the sample values ​​are arranged from small to large, and other values ​​can also be selected according to actual conditions. The smaller the interval division, the greater the calculation amount, and the more accurate the final result). Then, the carbon emission values ​​in the selected carbon emission attribute are sorted in order from small to large, and then divided into 100 equally spaced quantiles, each quantile indicating how many spatial units in the sample have carbon emissions below the quantile; then the proportion of the sum of carbon emissions above each quantile to the total carbon emissions is calculated, and for each quantile, the corresponding sum of carbon emissions is calculated, which refers to the sum of carbon emissions of all spatial units with carbon emissions higher than the quantile; then the sum is divided by the sum of all carbon emissions to obtain a ratio, recorded as the emission coverage ratio p, indicating the proportion of carbon emissions of the spatial units above the ratio to the total carbon emissions; then a group of n emission coverage ratios worthy of attention {p i} n , we get a set of corresponding carbon emission quantiles, called emission level characteristic quantiles {q i} n , and then record the emission level characteristic quantiles corresponding to each carbon emission attribute; finally, define each characteristic quantile to correspond to an emission level, and the emission level value range is the interval [0,n]. The larger the characteristic quantile, the higher the emission level value.

[0070] It should be noted that the emission coverage ratio is not an attribute of each spatial unit, but a global value corresponding to the quantile when the target area, spatial unit division, and carbon emission attribute are selected. For example, given the above conditions, the 20% quantile corresponds to a global emission coverage ratio, which is equal to the ratio of the sum of emissions of spatial units with emissions higher than this quantile to the total carbon emissions. Therefore, the emission coverage ratio of the 20% quantile should be around 80%, but it is not strictly equal to 100%-20%.

[0071] As an optional embodiment of the above scheme, determining the emission level of each spatial unit under each carbon emission attribute according to the emission level characteristic quantile of each carbon emission attribute includes:

[0072] Arrange all emission level characteristic quantiles of each carbon emission attribute from small to large;

[0073] If the carbon emissions under the current carbon emissions attribute of the current spatial unit is greater than or equal to the i-th emission level characteristic quantile value under the current carbon emissions attribute and less than the i+1-th emission level characteristic quantile value, then the emission level of the current spatial unit under the current carbon emissions attribute is value i;

[0074] If the carbon emissions under the current carbon emissions attribute of the current spatial unit are less than the minimum value of the emission level characteristic quantile under the current carbon emissions attribute, the emission level under the current carbon emissions attribute of the current spatial unit is 0;

[0075] If the carbon emissions under the current carbon emissions attribute of the current spatial unit are greater than the maximum value of the emission level characteristic quantile under the current carbon emissions attribute, then the emission level of the current spatial unit under the current carbon emissions attribute is n; where n is the number of all emission level characteristic quantiles under the current carbon emissions attribute.

[0076] More specifically, the emission level characteristic quantiles {q i} n Arrange from small to large; if the carbon emissions under the current carbon emissions attribute of the current spatial unit belongs to the interval [q i ,q i+1 )(i=1,2,…,n-1), the emission level of the current carbon emission attribute of the current spatial unit is value i; if the carbon emission under the current carbon emission attribute of the current spatial unit is less than q1, the emission level of the current carbon emission attribute of the current spatial unit is 0; if the carbon emission under the current carbon emission attribute of the current spatial unit is greater than q n , the current carbon emission attribute emission level of the current spatial unit is n.

[0077] It is worth noting that by introducing the method of emission level characteristic quantiles, the graded identification of carbon emission concentration areas is realized. The graded identification method of carbon emission concentration areas can be generally applied to various application scenarios, can avoid the influence of outliers and extreme values, and the emission level classification standards can be given specific meanings according to needs.

[0078] S3. According to the spatiotemporal carbon emission matrix, a density clustering algorithm is used to perform a preliminary clustering analysis on the spatial units to determine the carbon emission spatial clustering hyperparameters; a density clustering algorithm is set according to the carbon emission spatial clustering hyperparameters, and the spatial units are clustered again by the set density clustering algorithm to obtain a number of spatial clustering clusters;

[0079] As an optional embodiment of the above scheme, performing a preliminary clustering analysis on the spatial units by a density clustering algorithm according to the spatiotemporal carbon emission matrix to determine the carbon emission spatial clustering hyperparameters includes:

[0080] Selecting each carbon emission attribute in the spatiotemporal carbon emission matrix in turn, and selecting a plurality of emission level characteristic quantiles under the selected carbon emission attributes;

[0081] For each selected emission level characteristic quantile, select the spatial unit whose carbon emission is higher than the selected emission level characteristic quantile;

[0082] Performing a preliminary clustering analysis on the selected spatial units by using a density clustering algorithm to obtain a preliminary spatial clustering result;

[0083] According to the preliminary spatial clustering results, the grid search method is used to determine the hyperparameters of carbon emission spatial clustering.

[0084] Furthermore, the grid search method is used to determine the carbon emission spatial clustering hyperparameters according to the preliminary spatial clustering results, including:

[0085] Perform a grid search in the determined hyperparameter space, and record the number of spatial clustering clusters and the proportion of clustered carbon emissions obtained by clustering the preliminary spatial clustering results using a density clustering algorithm under different hyperparameter combination values;

[0086] Convert the indicator distribution of the number of spatial clusters and the proportion of cluster carbon emissions into indicator quantiles respectively;

[0087] Selecting a hyperparameter combination value that makes the index quantile of the number of spatial clusters and the index quantile of the proportion of cluster carbon emissions closest as the carbon emission spatial clustering hyperparameter;

[0088] The cluster carbon emission ratio is the ratio of the sum of carbon emissions of spatial units classified into any spatial cluster under any carbon emission attribute to the sum of carbon emissions of all spatial units.

[0089] Specifically, each carbon emission attribute in the spatiotemporal carbon emission matrix is ​​selected in turn, and a plurality of emission level characteristic quantiles are selected under the selected carbon emission attributes; for each selected emission level characteristic quantile, a spatial unit whose carbon emission is higher than the current emission level characteristic quantile is selected; the spatial units whose selected carbon emission is higher than the current emission level characteristic quantile are preliminarily spatially clustered by the DBSCAN algorithm (density clustering algorithm), and the carbon emission spatial clustering hyperparameters of the current emission level characteristic quantile under the selected carbon emission attribute are determined by the grid search method;

[0090] Conduct grid search within the determined hyperparameter grid search value range, and record the number of spatial clusters and the proportion of clustered carbon emissions obtained by the DBSCAN algorithm under different hyperparameter value combinations;

[0091] The index distribution of the number of spatial clustering clusters and the proportion of clustered carbon emissions are converted into index quantiles respectively, and the hyperparameter value combination that makes the index quantiles of the number of spatial clustering clusters and the proportion of clustered carbon emissions closest is selected as the value of the carbon emission spatial clustering hyperparameter of the current emission level characteristic quantile under the selected carbon emission attribute;

[0092] It should be noted that the selection of model hyperparameters has an important impact on the identification results of carbon emission agglomeration areas. However, existing methods often lack a widely accepted and highly adaptable method to determine the optimal hyperparameter combination. This leads to the need for researchers to continuously adjust hyperparameters based on experience and experiments in practical applications, which not only increases the workload, but also may affect the stability of the identification results. Therefore, an embodiment of the present invention provides a highly suitable clustering model hyperparameter confirmation path. By introducing automated parameter adjustment technology, cross-validation and other strategies, it is possible to automatically adjust and optimize the hyperparameters of the carbon emission spatial clustering model to reduce the workload of parameter tuning, improve the stability and accuracy of the identification results, and make the technical solution of the present invention more flexible and efficient in practical applications.

[0093] It should be noted that the density clustering algorithm takes the DBSCAN algorithm as an example. The DBSCAN algorithm mainly involves two hyperparameters: the neighborhood distance threshold and the sample number threshold. The present invention defines the number of clusters and the proportion of cluster carbon emissions, and selects the hyperparameters of the model based on the principle of balance between these two indicators. Among them, the number of clusters refers to the number of clusters obtained by the DBSCAN algorithm under a given hyperparameter combination, and the proportion of cluster carbon emissions refers to the ratio of the sum of the selected carbon emissions of the unit grids included in the cluster to the total amount of selected carbon emissions. Generally, under certain other conditions, the larger the neighborhood distance threshold is set, the more clusters the number and the greater the proportion of cluster carbon emissions; the larger the sample number threshold is set, the smaller the number of clusters and the smaller the proportion of cluster carbon emissions, and the effect is opposite to the neighborhood distance threshold. Utilizing the above-mentioned adversarial characteristics, the value of the hyperparameter that best meets the balance conditions of the two indicators is automatically selected through grid search.

[0094] In the specific implementation, in actual operation, we can also plot the changes of the number of clusters and the proportion of cluster carbon emissions as the two hyperparameters change, and select the optimal hyperparameter value under visualization conditions.

[0095] S4. Determine the aggregate emission level of each spatial unit under each carbon emission attribute according to the emission level and the plurality of spatial clustering clusters;

[0096] As an optional embodiment of the above solution, determining the aggregate emission level of each spatial unit under each carbon emission attribute according to the emission level and the plurality of spatial clustering clusters includes:

[0097] When a spatial unit is classified into any of the plurality of spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is respectively the emission level of the spatial unit under each carbon emission attribute;

[0098] When a spatial unit is not classified into any of the plurality of spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is 0;

[0099] For the same carbon emission attribute, if a spatial unit is simultaneously classified into different spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is the highest emission level of the spatial unit under the carbon emission attributes corresponding to different spatial clusters.

[0100] Specifically, any emission level characteristic quantile under any carbon emission attribute is selected;

[0101] When a spatial unit is classified into any spatial cluster, if the carbon emissions of the current spatial unit under the current carbon emissions attribute are lower than the characteristic quantile of the current emission level under the current carbon emissions attribute, the agglomeration emission level of the current spatial unit under the current carbon emissions attribute is 0; otherwise, the agglomeration emission level of the current spatial unit under the current carbon emissions attribute is the emission level of the current carbon emissions attribute of the current spatial unit;

[0102] When a spatial unit is not classified into any spatial cluster, the agglomeration emission level of the current spatial unit under the current carbon emission attribute is 0;

[0103] For the same carbon emission attribute, if the current spatial unit belongs to different spatial clusters at the same time, the aggregate emission level of the current spatial unit under the current carbon emission attribute takes the highest value among them.

[0104] S5. Constructing a spatiotemporal agglomeration emission level matrix according to the agglomeration emission level;

[0105] Specifically, the spatial unit serial number is used as a row, and the carbon emission attribute serial number is used as a column to construct a spatiotemporal agglomeration emission level matrix; wherein the matrix elements in the spatiotemporal agglomeration emission level matrix are used to record the agglomeration emission level of each spatial unit under each carbon emission attribute;

[0106] For example, with the spatial unit numbers as rows and the carbon emission attribute numbers as columns, a spatiotemporal clustered emission level matrix H∈R is created as shown in Table 3: k×(m+1) .

[0107] Table 3 Spatiotemporal cluster emission level matrix

[0108]

[0109] It should be noted that although the dimension of this matrix is ​​the same as the spatiotemporal carbon emissions matrix, the difference is that the matrix elements are used to record the aggregate emission level of each carbon emission in each grid unit. The "aggregate emission level" of a carbon emission in a spatial unit refers to the highest emission level corresponding to the carbon emission agglomeration area to which the carbon emission in the spatial unit belongs.

[0110] S6. Based on the spatiotemporal agglomeration emission level matrix, taking the carbon emission attributes in the spatiotemporal carbon emission matrix as features, clustering analysis is performed on the spatial units by using a K-means clustering algorithm that determines the hyperparameter values ​​based on the elbow rule to obtain a plurality of category clusters;

[0111] Specifically, based on the spatiotemporal agglomeration emission level matrix, taking the various carbon emission attributes in the spatiotemporal carbon emission matrix as features, a K-means algorithm (K-means clustering algorithm) is used to perform cluster analysis on each spatial unit, and the value of the hyperparameter of the K-means algorithm is determined based on the elbow rule, and the category clustering cluster to which each spatial unit belongs is calculated, and each spatial unit is divided into different carbon emission agglomeration area categories according to the category clustering cluster;

[0112] Exemplarily, based on the spatiotemporal clustered emission level matrix H, with various carbon emission attributes as features, the K-means algorithm is used to perform cluster analysis on spatial units, and the value of the hyperparameter (number of clusters) of the K-means algorithm is confirmed based on the elbow rule, and the cluster number of the carbon emission graded spatiotemporal cluster area to which each spatial unit belongs is calculated.

[0113] It is worth noting that the mean of each agglomeration emission level of each cluster reflects the overall level of carbon emissions in this type of agglomeration area, and the changes in the agglomeration emission level in different characteristic time periods reflect the temporal distribution characteristics of the carbon emission agglomeration level of the agglomeration area. According to the characteristics of the emission level, emission time distribution, and agglomeration area spatial morphology of each cluster, each cluster is defined as an intuitive and explainable carbon emission spatiotemporal agglomeration area category. Finally, the embodiment of the present invention divides the carbon emission agglomeration area into different categories by combining the three characteristics of the agglomeration area emission level, emission time distribution, and spatial morphology, and matches them with the administrative divisions in space, so as to more accurately grasp the changing laws of carbon emissions and make the identification results more accurate and reliable.

[0114] S7. Identify each spatial unit as a different carbon emission concentration area category based on the plurality of category clusters.

[0115] It should be noted that, according to the different categories of carbon emission concentration areas, the carbon emissions in the carbon emission target areas can be monitored, so as to control the carbon emission equipment in the carbon emission target areas.

[0116] As an optional embodiment of the above solution, identifying each spatial unit as a different carbon emission concentration area category according to the plurality of category clusters includes:

[0117] According to the agglomeration emission levels of the plurality of clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include primary agglomeration area, secondary agglomeration area and tertiary agglomeration area;

[0118] According to the carbon emissions in each characteristic time period of the several category clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include stable agglomeration area, daytime peak agglomeration area, nighttime peak agglomeration area and commuting peak agglomeration area;

[0119] According to the spatial forms of the several category clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include regional agglomeration areas and channel agglomeration areas.

[0120] It is worth noting that, in general, according to the ranking of the average daily carbon emission level of the agglomeration area from high to low, it can be divided into first-level agglomeration area, second-level agglomeration area and third-level agglomeration area, etc. The emission level of the agglomeration area reflects the carbon emission intensity in the area, which can be used to determine the management and control strength of the corresponding carbon emission agglomeration area. In addition, by comparing and analyzing the areas with different emission levels, the main sources and key factors of carbon emissions can be found, providing data basis for formulating more effective emission reduction measures. According to the time distribution characteristics of carbon emissions, the agglomeration area can be divided into stable type, daytime peak type, night peak type, commuting peak type, etc. Differentiating the agglomeration area according to the time distribution of carbon emissions in the agglomeration area can help analyze the carbon emission law in the time dimension and improve the efficiency of carbon emission measures in a targeted manner. According to the differences in the spatial morphology of the agglomeration area, the agglomeration area can be divided into regional agglomeration area and channel agglomeration area. In specific implementation, the regional agglomeration area obtained by the present invention can also be spatially connected with administrative areas such as town boundaries and village boundaries, and the channel agglomeration area can be spatially connected with the road network to ensure that each agglomeration area matches the administrative divisions at all levels. The spatial form of the agglomeration area determines the logic and focus of carbon emission management measures. Regional agglomerations often cover a variety of industrial types, with diverse sources of carbon emissions, making management more difficult. Corresponding measures should focus on total volume control, industrial structure adjustment, and energy efficiency improvement. Channel-type agglomerations are mainly along major transportation routes, with mobile emissions as the main source, and have obvious cross-regional characteristics, requiring cross-regional collaboration for joint management. Corresponding measures are mainly focused on optimizing traffic organization and improving vehicle emission standards. The final category of the agglomeration area can also be a combination of the above three categories.

[0121] It is worth noting that the embodiment of the present invention divides carbon emission agglomeration areas into different categories by combining the three characteristics of emission level, emission time distribution and spatial morphology of the agglomeration areas, and matches them spatially with administrative divisions, thereby improving the accuracy and reliability of carbon emission identification results, thereby helping carbon emission measure formulators to grasp the changing patterns of carbon emissions more accurately.

[0122] like Figure 2 As shown, an embodiment of the present invention further provides a carbon emission concentration area identification system, which is used to implement any of the carbon emission concentration area identification methods described above, including:

[0123] The spatiotemporal carbon emission matrix construction module 21 is used to divide the carbon emission target area into a number of spatial units, obtain various carbon emission attributes of each spatial unit, and construct the spatiotemporal carbon emission matrix;

[0124] An emission level determination module 22, for determining the emission level of each spatial unit under each carbon emission attribute according to the spatiotemporal carbon emission matrix;

[0125] The carbon emission spatial clustering module 23 is used to perform preliminary clustering analysis on the spatial units through the density clustering algorithm according to the spatiotemporal carbon emission matrix, and determine the carbon emission spatial clustering hyperparameters; set the density clustering algorithm according to the carbon emission spatial clustering hyperparameters, and perform clustering analysis on the spatial units again through the set density clustering algorithm to obtain a number of spatial clustering clusters;

[0126] An aggregate emission level determination module 24 is used to determine the aggregate emission level of each spatial unit under various carbon emission attributes according to the emission level and the plurality of spatial clusters;

[0127] A spatiotemporal agglomeration emission level matrix construction module 25 is used to construct a spatiotemporal agglomeration emission level matrix according to the agglomeration emission level;

[0128] The carbon emission category clustering module 26 is used to perform cluster analysis on spatial units based on the spatiotemporal agglomeration emission level matrix and each carbon emission attribute in the spatiotemporal carbon emission matrix as a feature by using a K-means clustering algorithm that determines hyperparameter values ​​based on the elbow rule to obtain a plurality of category clusters;

[0129] The carbon emission concentration area identification module 27 is used to identify each spatial unit as a different carbon emission concentration area category according to the plurality of category clusters.

[0130] As an optional embodiment, the emission level determination module 22 is specifically used to:

[0131] Selecting each carbon emission attribute in the spatiotemporal carbon emission matrix in turn, calculating the corresponding emission coverage ratios at different quantiles under the selected carbon emission attributes, and obtaining a plurality of emission coverage ratios;

[0132] Determining emission level characteristic quantiles of various carbon emission attributes according to the plurality of emission coverage ratios;

[0133] The emission level of each spatial unit under each carbon emission attribute is determined according to the emission level characteristic quantiles of the various carbon emission attributes.

[0134] As an optional embodiment, the emission level determination module 22 is further specifically configured to:

[0135] Arrange all emission level characteristic quantiles of each carbon emission attribute from small to large;

[0136] If the carbon emissions under the current carbon emissions attribute of the current spatial unit is greater than or equal to the i-th emission level characteristic quantile value under the current carbon emissions attribute and less than the i+1-th emission level characteristic quantile value, then the emission level of the current spatial unit under the current carbon emissions attribute is value i;

[0137] If the carbon emissions under the current carbon emissions attribute of the current spatial unit are less than the minimum value of the emission level characteristic quantile under the current carbon emissions attribute, the emission level under the current carbon emissions attribute of the current spatial unit is 0;

[0138] If the carbon emissions under the current carbon emissions attribute of the current spatial unit are greater than the maximum value of the emission level characteristic quantile under the current carbon emissions attribute, then the emission level of the current spatial unit under the current carbon emissions attribute is n; where n is the number of all emission level characteristic quantiles under the current carbon emissions attribute.

[0139] As an optional embodiment, the carbon emission spatial clustering module 23 further includes a carbon emission spatial clustering hyperparameter selection unit, which is specifically used to:

[0140] Selecting each carbon emission attribute in the spatiotemporal carbon emission matrix in turn, and selecting a plurality of emission level characteristic quantiles under the selected carbon emission attributes;

[0141] For each selected emission level characteristic quantile, select the spatial unit whose carbon emission is higher than the selected emission level characteristic quantile;

[0142] Performing a preliminary clustering analysis on the selected spatial units by using a density clustering algorithm to obtain a preliminary spatial clustering result;

[0143] According to the preliminary spatial clustering results, the grid search method is used to determine the hyperparameters of carbon emission spatial clustering.

[0144] The carbon emission spatial clustering hyperparameter selection unit is further specifically used for:

[0145] Perform a grid search in the determined hyperparameter space, and record the number of spatial clustering clusters and the proportion of clustered carbon emissions obtained by clustering the preliminary spatial clustering results using a density clustering algorithm under different hyperparameter combination values;

[0146] Convert the indicator distribution of the number of spatial clusters and the proportion of cluster carbon emissions into indicator quantiles respectively;

[0147] Selecting a hyperparameter combination value that makes the index quantile of the number of spatial clusters and the index quantile of the proportion of cluster carbon emissions closest as the carbon emission spatial clustering hyperparameter;

[0148] The cluster carbon emission ratio is the ratio of the sum of carbon emissions of spatial units classified into any spatial cluster under any carbon emission attribute to the sum of carbon emissions of all spatial units.

[0149] As an optional embodiment, the aggregate emission level determination module 24 further includes an aggregate emission level determination unit, which is specifically configured to:

[0150] When a spatial unit is classified into any of the plurality of spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is respectively the emission level of the spatial unit under each carbon emission attribute;

[0151] When a spatial unit is not classified into any of the plurality of spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is 0;

[0152] For the same carbon emission attribute, if a spatial unit is simultaneously classified into different spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is the highest emission level of the spatial unit under the carbon emission attributes corresponding to different spatial clusters.

[0153] As an optional embodiment, the carbon emission category clustering module 26 further includes a carbon emission concentration area category identification unit, which is specifically used to:

[0154] According to the agglomeration emission levels of the plurality of clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include primary agglomeration area, secondary agglomeration area and tertiary agglomeration area;

[0155] According to the carbon emissions in each characteristic time period of the several category clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include stable agglomeration area, daytime peak agglomeration area, nighttime peak agglomeration area and commuting peak agglomeration area;

[0156] According to the spatial forms of the several category clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include regional agglomeration areas and channel agglomeration areas.

[0157] It should be noted that a carbon emission concentration area identification system provided in an embodiment of the present invention is used to execute all process steps of a carbon emission concentration area identification method in the above embodiment. The working principles and beneficial effects of the two correspond one to one, and thus will not be elaborated upon.

[0158] The embodiment of the present invention also provides a device for identifying carbon emission concentration areas, such as Figure 3 The figure is a structural block diagram of a preferred embodiment of a carbon emission concentration area identification device provided by the present invention. The carbon emission concentration area identification device comprises a processor 31, a memory 32, and a computer program stored in the memory 32 and configured to be executed by the processor 31. When the processor 31 executes the computer program, the carbon emission concentration area identification method described in any of the above embodiments is implemented.

[0159] In addition, an embodiment of the present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the carbon emission concentration area identification method described in any of the above embodiments.

[0160] When the processor 31 executes the computer program, the steps in the above-mentioned carbon emission concentration area identification method embodiment are implemented, for example: Figure 1 Alternatively, when the processor 31 executes the computer program, the functions of each module in the above-mentioned carbon emission concentration area identification system embodiment are realized, for example Figure 2 The functions of each unit of the carbon emission concentration area identification system are shown.

[0161] Preferably, the computer program can be divided into one or more modules / units, which are stored in the memory 32 and executed by the processor 31 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the carbon emission concentration area identification device.

[0162] The processor 31 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor 31 can be any conventional processor. The processor 31 is the control center of the carbon emission concentration area identification device, and various parts of the carbon emission concentration area identification device are connected by various interfaces and lines.

[0163] The memory 32 mainly includes a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory 32 can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, and a flash card (Flash Card), etc., or the memory 32 can also be other volatile solid-state storage devices.

[0164] It should be noted that the above-mentioned carbon emission concentration area identification device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that Figure 3 The structural block diagram shown is merely an example of the structure of the above-mentioned carbon emission concentration area identification device, and does not constitute a structural limitation of the above-mentioned carbon emission concentration area identification device. The above-mentioned carbon emission concentration area identification device may include more or fewer components than shown in the figure, or a combination of certain components, or different components.

[0165] In summary, the embodiments of the present invention provide a method, system, device and medium for identifying carbon emission agglomeration areas. First, the carbon emission area is divided into several spatial units, various carbon emission attributes are obtained, and a spatiotemporal carbon emission matrix is ​​constructed to determine the emission level of each spatial unit under various carbon emission attributes; secondly, the carbon emission spatial clustering hyperparameters are determined by a density clustering algorithm, and then the density clustering algorithm is used to set the carbon emission spatial clustering hyperparameters to perform clustering analysis on the spatial units to obtain several spatial clustering clusters; then, according to the emission level and several spatial clustering clusters, the agglomeration emission level of each spatial unit under various carbon emission attributes is determined, and a spatiotemporal agglomeration emission level matrix is ​​constructed; finally, the K-means clustering algorithm is used to cluster the spatial units to identify the carbon emission agglomeration area category. By introducing the emission grade method, the present invention can better reflect the carbon emission characteristics of spatial units, realize the hierarchical identification of carbon emission agglomeration areas, and improve the analysis and management capabilities of carbon emission data; then, the carbon emission spatial clustering hyperparameters are determined through the DBSCAN algorithm, which can automatically adjust and optimize the hyperparameters of the carbon emission spatial clustering model, which not only reduces the workload of parameter tuning, but also improves the accuracy and stability of the identification results.

[0166] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for identifying carbon emission concentration areas, characterized in that: include: Divide the carbon emission target area into several spatial units, obtain the carbon emission attributes of each spatial unit, and construct a spatiotemporal carbon emission matrix; Determining the emission level of each spatial unit under each carbon emission attribute according to the spatiotemporal carbon emission matrix; According to the spatiotemporal carbon emission matrix, a preliminary clustering analysis is performed on the spatial units by a density clustering algorithm to determine the carbon emission spatial clustering hyperparameters; a density clustering algorithm is set according to the carbon emission spatial clustering hyperparameters, and a clustering analysis is performed on the spatial units again by the set density clustering algorithm to obtain a number of spatial clustering clusters; Determining the aggregate emission level of each spatial unit under each carbon emission attribute according to the emission level and the plurality of spatial clustering clusters; According to the agglomeration emission level, constructing a spatiotemporal agglomeration emission level matrix; Based on the spatiotemporal agglomeration emission level matrix, taking the carbon emission attributes in the spatiotemporal carbon emission matrix as features, clustering analysis is performed on spatial units by using a K-means clustering algorithm that determines hyperparameter values ​​based on the elbow rule to obtain a number of category clusters; According to the plurality of category clusters, each spatial unit is identified as a different carbon emission concentration area category.

2. The method for identifying carbon emission concentration areas according to claim 1, characterized in that: The carbon emission attributes of each spatial unit include the daily carbon emission of each spatial unit and the carbon emission in each characteristic time period.

3. The method for identifying carbon emission concentration areas according to claim 2, characterized in that: Determining the emission level of each spatial unit under each carbon emission attribute according to the spatiotemporal carbon emission matrix includes: Selecting each carbon emission attribute in the spatiotemporal carbon emission matrix in turn, calculating the corresponding emission coverage ratios at different quantiles under the selected carbon emission attributes, and obtaining a plurality of emission coverage ratios; Determining emission level characteristic quantiles of various carbon emission attributes according to the plurality of emission coverage ratios; The emission level of each spatial unit under each carbon emission attribute is determined according to the emission level characteristic quantiles of the various carbon emission attributes.

4. The method for identifying carbon emission concentration areas according to claim 3, characterized in that: The method of performing a preliminary clustering analysis on the spatial units by using a density clustering algorithm according to the spatiotemporal carbon emission matrix to determine the carbon emission spatial clustering hyperparameters includes: Selecting each carbon emission attribute in the spatiotemporal carbon emission matrix in turn, and selecting a plurality of emission level characteristic quantiles under the selected carbon emission attributes; For each selected emission level characteristic quantile, select the spatial unit whose carbon emission is higher than the selected emission level characteristic quantile; Performing a preliminary clustering analysis on the selected spatial units by using a density clustering algorithm to obtain a preliminary spatial clustering result; According to the preliminary spatial clustering results, the grid search method is used to determine the hyperparameters of carbon emission spatial clustering.

5. The method for identifying carbon emission concentration areas according to claim 4, characterized in that: The method of determining carbon emission spatial clustering hyperparameters by using a grid search method based on the preliminary spatial clustering results includes: Perform a grid search in the determined hyperparameter space, and record the number of spatial clustering clusters and the proportion of clustered carbon emissions obtained by clustering the preliminary spatial clustering results using a density clustering algorithm under different hyperparameter combination values; Convert the indicator distribution of the number of spatial clusters and the proportion of cluster carbon emissions into indicator quantiles respectively; Selecting a hyperparameter combination value that makes the index quantile of the number of spatial clustering clusters and the index quantile of the proportion of clustered carbon emissions closest as the carbon emission spatial clustering hyperparameter; The cluster carbon emission ratio is the ratio of the sum of carbon emissions of spatial units classified into any spatial cluster under any carbon emission attribute to the sum of carbon emissions of all spatial units.

6. The method for identifying carbon emission concentration areas according to claim 5, characterized in that: The step of determining the aggregate emission level of each spatial unit under each carbon emission attribute according to the emission level and the plurality of spatial clustering clusters includes: When a spatial unit is classified into any of the plurality of spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is respectively the emission level of the spatial unit under each carbon emission attribute; When a spatial unit is not classified into any of the plurality of spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is 0; For the same carbon emission attribute, if a spatial unit is simultaneously classified into different spatial clusters, the aggregate emission level of the spatial unit under each carbon emission attribute is the highest emission level of the spatial unit under the carbon emission attributes corresponding to different spatial clusters.

7. The method for identifying carbon emission concentration areas according to claim 2 or 6, characterized in that: The identifying each spatial unit as a different carbon emission concentration area category according to the plurality of category clusters includes: According to the agglomeration emission levels of the plurality of clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include primary agglomeration area, secondary agglomeration area and tertiary agglomeration area; According to the carbon emissions in each characteristic time period of the several category clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include stable agglomeration area, daytime peak agglomeration area, nighttime peak agglomeration area and commuting peak agglomeration area; According to the spatial forms of the several category clusters, each spatial unit is identified, and the obtained carbon emission agglomeration area categories include regional agglomeration areas and channel agglomeration areas.

8. A carbon emission concentration area identification system, characterized in that: A method for identifying carbon emission concentration areas according to any one of claims 1 to 7, comprising: The spatiotemporal carbon emission matrix construction module is used to divide the carbon emission target area into several spatial units, obtain the carbon emission attributes of each spatial unit, and construct the spatiotemporal carbon emission matrix; An emission level determination module, used to determine the emission level of each spatial unit under each carbon emission attribute according to the spatiotemporal carbon emission matrix; A carbon emission spatial clustering module is used to perform preliminary clustering analysis on spatial units through a density clustering algorithm according to the spatiotemporal carbon emission matrix, and determine carbon emission spatial clustering hyperparameters; set a density clustering algorithm according to the carbon emission spatial clustering hyperparameters, and perform clustering analysis on spatial units again through the set density clustering algorithm to obtain a number of spatial clustering clusters; An aggregate emission level determination module, used to determine the aggregate emission level of each spatial unit under various carbon emission attributes according to the emission level and the plurality of spatial clusters; A spatiotemporal agglomeration emission level matrix construction module, used to construct a spatiotemporal agglomeration emission level matrix according to the agglomeration emission level; A carbon emission category clustering module is used to perform clustering analysis on spatial units based on the spatiotemporal agglomeration emission level matrix and each carbon emission attribute in the spatiotemporal carbon emission matrix as a feature by using a K-means clustering algorithm that determines hyperparameter values ​​based on the elbow rule to obtain a number of category clustering clusters; The carbon emission concentration area identification module is used to identify each spatial unit as a different carbon emission concentration area category according to the plurality of category clusters.

9. A device for identifying carbon emission concentration areas, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the method for identifying carbon emission concentration areas as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the carbon emission concentration area identification method according to any one of claims 1 to 7.

Images