Intelligent statistical analysis method and analysis system for power grid operation data

Abstract

The invention relates to the technical field of power grid data analysis, in particular to a power grid operation data intelligent statistical analysis method and system, and the method comprises the steps: obtaining a power grid operation data set corresponding to each sub-geographic region in a preset time period, and obtaining a power grid operation data set of each sub-geographic region based on a preset index weight corresponding to each power grid operation index; clustering vectors corresponding to the plurality of power grid operation data sets to obtain a plurality of initial vector clusters, screening part of the power grid operation vectors from the initial vector clusters according to the adjacent condition of the sub-geographic areas, dividing the initial vector clusters into a plurality of intermediate vector clusters, adding the screened power grid operation vectors into corresponding target vector clusters, and obtaining a plurality of sub-geographic areas; determining a plurality of sub-geographic regions corresponding to the same target vector cluster as geographic regions of the same type and combining the sub-geographic regions into a key geographic region; according to the method, intelligent statistical analysis is performed on the operation data of the power grid, so that areas with similar power consumption conditions can be divided, and unified power consumption management and data analysis of geographic areas of the same type are facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid data analysis, and particularly relates to an intelligent statistical analysis method and analysis system for power grid operation data. Background Art

[0002] In recent years, with the advancement of smart grids and large operation systems, real-time monitoring of the power grid operation status and environmental conditions is essential. Rich power grid operation data resources have gradually been formed in each level of dispatching centers. These data record the long-term operation conditions of the power systems in each region. With the development of the economy, the electricity demand for people's production and living increases year by year. Therefore, it is necessary to strengthen the operation management of the power grid, which is of great significance for ensuring the safe operation of the power grid and the distribution of electricity. However, the current electricity consumption management is usually hierarchical management from large regions to small regions, without fully considering the actual electricity consumption situations between different small regions, resulting in inaccurate division of electricity consumption areas and being unfavorable for the unified management and data analysis of the power grid operation data in areas with the same electricity consumption level. Summary of the Invention

[0003] In view of the above technical problems, the present invention provides an intelligent statistical analysis method and analysis system for power grid operation data. By intelligently statistically analyzing the power grid operation data, it is possible to divide areas with similar electricity consumption situations, which is beneficial for unified electricity consumption management and data analysis of the same type of geographical regions.

[0004] According to a first aspect of the present invention, there is provided an intelligent statistical analysis method for power grid operation data, including the following steps:

[0005] S100, for any preset time period, obtain the power grid operation data set corresponding to each sub-geographical region within the target geographical region during the preset time period; the power grid operation data set includes data corresponding to a number of power grid operation indicators.

[0006] S200, convert each power grid operation data set into a power grid operation vector, and based on the preset index weights corresponding to each power grid operation indicator, cluster a number of power grid operation vectors to obtain an initial clustering result; the initial clustering result includes a number of initial vector clusters.

[0007] S300, for any initial vector cluster, according to the adjacent situations of the sub-geographical regions corresponding to the power grid operation vectors in the initial vector cluster, screen out the power grid operation vectors corresponding to the non-adjacent sub-geographical regions, and divide the remaining a number of power grid operation vectors in the initial vector cluster into a number of intermediate vector clusters.

[0008] S400. For any screened power grid operation vector, determine the target vector cluster corresponding to the screened power grid operation vector according to the target similarity between the screened power grid operation vector and the centroids of all intermediate vector clusters and the adjacency with several power grid operation vectors in each intermediate vector cluster, and add the screened power grid operation vector to the corresponding target vector cluster.

[0009] S500. Determine the sub - geographical regions corresponding to several power grid operation vectors in the same target vector cluster as the same - type geographical regions and merge them into a key geographical region.

[0010] According to the second aspect of the present invention, there is provided a system for intelligent statistical analysis of power grid operation data, and the system includes:

[0011] A first acquisition module, configured to obtain, for any preset time period, the power grid operation data set corresponding to each sub - geographical region within the target geographical region in the preset time period; the power grid operation data set includes data corresponding to several power grid operation indicators.

[0012] A clustering module, configured to convert each power grid operation data set into a power grid operation vector, and perform clustering on several power grid operation vectors based on the preset index weights corresponding to each power grid operation indicator to obtain an initial clustering result; the initial clustering result includes several initial vector clusters.

[0013] A first partitioning module, configured to, for any initial vector cluster, screen out the power grid operation vectors corresponding to the sub - geographical regions that are not adjacent according to the adjacency of the sub - geographical regions corresponding to the power grid operation vectors in the initial vector cluster, and divide the remaining several power grid operation vectors in the initial vector cluster into several intermediate vector clusters.

[0014] An adding module, configured to, for any screened power grid operation vector, determine the target vector cluster corresponding to the screened power grid operation vector according to the target similarity between the screened power grid operation vector and the centroids of all intermediate vector clusters and the adjacency with several power grid operation vectors in each intermediate vector cluster, and add the screened power grid operation vector to the corresponding target vector cluster.

[0015] A merging module, configured to determine the sub - geographical regions corresponding to several power grid operation vectors in the same target vector cluster as the same - type geographical regions and merge them into a key geographical region.

[0016] The present invention has at least the following beneficial effects:

[0017] The present invention provides an intelligent statistical analysis method for power grid operation data. First, power grid operation data sets corresponding to each sub-geographical area within a target geographical area in a preset time period are obtained and converted into power grid operation vectors. Based on the preset index weights corresponding to each power grid operation index, several power grid operation vectors are clustered to obtain several initial vector clusters. By introducing the weights of each power grid operation index, the clustering result is more reasonable and reliable, which is beneficial to the accurate division of regions. Then, according to the adjacency of sub-geographical areas, some power grid operation vectors are screened out from each initial vector cluster and divided into several intermediate vector clusters. Then, the target vector cluster corresponding to each screened power grid operation vector is obtained and added to the corresponding target vector cluster. Several sub-geographical areas corresponding to the same target vector cluster are determined as the same type of geographical area and merged into key geographical areas. Through the intelligent statistical analysis of power grid operation data, areas with similar power consumption situations can be divided, which is beneficial to the unified power consumption management and data analysis of the same type of geographical areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is a flowchart of the intelligent statistical analysis method for power grid operation data provided by the embodiment of the present invention;

[0020] Figure 2 It is a schematic diagram of the intelligent statistical analysis system for power grid operation data provided by the embodiment of the present invention;

[0021] Figure 3 It is a schematic diagram of the structure of the weight acquisition module provided by the embodiment of the present invention;

[0022] Figure 4 It is a schematic diagram of the structure of the first division module provided by the embodiment of the present invention;

[0023] Figure 5 It is a schematic diagram of the structure of the addition module provided by the embodiment of the present invention;

[0024] Figure 6 It is a schematic diagram of the structure of the third determination module provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0026] This embodiment provides an intelligent statistical analysis method for power grid operation data. As Figure 1 shown, the method includes the following steps:

[0027] S100, for any preset time period, obtain the power grid operation data sets corresponding to each sub-geographical area within the target geographical area during the preset time period; the power grid operation data sets include data corresponding to several power grid operation indicators. Among them, the sub-geographical area is a pre-divided geographical area, such as a residential area, an industrial area, a shopping mall area, etc.

[0028] Specifically, the preset time period can be any time period within a day; it can be understood that there are several preset time periods set within a day. For example, 9:00 - 10:00 am, 14:00 - 15:00 pm, 20:00 - 21:00 pm.

[0029] Furthermore, the power grid operation indicator is any one of rated voltage, rated current, rated electric power, actual voltage, actual current, and actual electric power; it can be understood that the power grid operation indicator data is collected at any time point within the preset time period, or the power grid operation indicator data is collected at several time periods within the preset time period and the average value is taken.

[0030] S200, convert each power grid operation data set into a power grid operation vector, and cluster several power grid operation vectors based on the preset index weights corresponding to each power grid operation indicator to obtain an initial clustering result; the initial clustering result includes several initial vector clusters.

[0031] Specifically, the k-means clustering model is used to cluster the power grid operation vectors; those skilled in the art know the specific implementation manner of the k-means clustering model, which will not be elaborated here.

[0032] Furthermore, the preset index weights corresponding to each power grid operation indicator are determined through the following steps:

[0033] S001, obtain n historical power grid operation data set samples from several sub-geographical areas, and perform standardization processing on the data in each historical power grid operation data set sample to obtain n target power grid operation data set samples; it can be understood that performing standardization processing on the data means eliminating the influence of different dimensions.

[0034] S002. If the data corresponding to all power grid operation indicators in n target power grid operation dataset samples are all within the corresponding preset numerical intervals, calculate the entropy value of each power grid operation indicator, and calculate the corresponding preset indicator weight for each power grid operation indicator based on the entropy value of each power grid operation indicator. It can be understood that each power grid operation indicator corresponds to a preset numerical interval, and the preset numerical intervals corresponding to the rated voltage, rated current, and rated electric power can be set as several fixed values according to actual needs.

[0035] Among them, the preset indicator weight corresponding to the power grid operation indicator meets the following conditions:

[0036] W j =(1 - H j ) / (∑ m j＝1 (1 - H j ))), where W j represents the preset indicator weight corresponding to the j-th power grid operation indicator, H j is the entropy value of the j-th power grid operation indicator, and m is the number of power grid operation indicators. Those skilled in the art know the specific calculation method of the entropy value, which will not be elaborated here.

[0037] S003. If there is data in the data corresponding to all power grid operation indicators in n target power grid operation dataset samples that is not within the corresponding preset numerical interval, calculate the average data value and the index dispersion degree corresponding to each power grid operation indicator. It can be understood that the index dispersion degree refers to the standard deviation of several data corresponding to the power grid operation indicator.

[0038] S004. Respectively determine the ratio of the average data value and the index dispersion degree corresponding to each power grid operation indicator as the index priority of the power grid operation indicator itself, and normalize each index priority to obtain the preset indicator weight corresponding to each power grid operation indicator.

[0039] As described above, introducing the preset indicator weight in k-means clustering makes the obtained similarity more accurate, and the clustering result is more reasonable and reliable. Moreover, when obtaining the preset indicator weight, different weight calculation methods are adopted according to whether there are outliers in the data, which overcomes the disadvantage that calculating the weight using the entropy value will affect the weight accuracy when there are outliers, makes the obtained indicator weight more accurate, and further makes the obtained similarity more accurate and reliable, which is beneficial to the accurate division of clustering.

[0040] S300. For any initial vector cluster, according to the adjacency of the sub - geographic regions corresponding to the power grid operation vectors in the initial vector cluster, screen out the power grid operation vectors corresponding to the non - adjacent sub - geographic regions, and divide several remaining power grid operation vectors in the initial vector cluster into several intermediate vector clusters. It can be understood that the adjacency of the sub - geographic regions is pre - configured, and those skilled in the art know the specific implementation methods for collecting whether any two sub - geographic regions are adjacent. For example, it can be collected on a map app, which will not be elaborated here.

[0041] Further, divide several remaining power grid operation vectors in the initial vector cluster into several intermediate vector clusters through the following steps:

[0042] S301. Obtain the sub - geographic regions corresponding to each power grid operation vector in the remaining power grid operation vectors, and obtain k sub - geographic regions. It can be understood that what is obtained is the sub - geographic region identifier.

[0043] S302. According to the adjacency relationship between the k sub - geographic regions, perform a merging and zoning process on the k sub - geographic regions to obtain several geographic region zones. It can be understood that merge all adjacent sub - geographic regions in sequence, and different merging regions generate different geographic region zones. That is, there are no adjacent sub - geographic regions between different two geographic region zones.

[0044] S303. Divide the power grid operation vectors corresponding to several sub - geographic regions within each geographic region zone into one intermediate vector cluster to obtain several intermediate vector clusters.

[0045] As mentioned above, since several power grid operation vectors within the initial vector cluster are similar, they can be divided into the same power consumption level for power grid operation data statistical analysis. Screen out independent sub - geographic regions, and perform a merging and zoning process on regions with adjacent sub - geographic regions. The power consumption situations of sub - geographic regions within each geographic region zone are similar and can be used as a whole region for subsequent unified data analysis, which is beneficial to unified power consumption management of the zone.

[0046] S400. For any screened - out power grid operation vector, determine the target vector cluster corresponding to the screened - out power grid operation vector according to the target similarity between the screened - out power grid operation vector and the centroids of all intermediate vector clusters and the adjacency with several power grid operation vectors in each intermediate vector cluster, and add the screened - out power grid operation vector to the corresponding target vector cluster. It can be understood that all intermediate vector clusters refer to the sum of several intermediate vector clusters divided from each initial vector cluster.

[0047] Specifically, determine the target vector cluster corresponding to the screened - out power grid operation vector through the following steps:

[0048] S401. For any screened power grid operation vector, calculate the target similarity between the screened power grid operation vector and the centroid of all intermediate vector clusters respectively, and determine the intermediate vector cluster corresponding to the maximum target similarity as the key vector cluster. It can be understood that the target similarity is calculated through the Euclidean distance and the preset index weight corresponding to each power grid operation index.

[0049] S402. When the maximum target similarity is greater than the preset similarity threshold, obtain the sub - geographical regions corresponding to all power grid operation vectors in the key vector cluster. Those skilled in the art set the preset similarity threshold according to actual needs, which will not be elaborated here.

[0050] S403. When there is a geographical region adjacent to the sub - geographical region corresponding to the screened power grid operation vector among the sub - geographical regions corresponding to all power grid operation vectors in the key vector cluster, take the key vector cluster as the target vector cluster corresponding to the screened power grid operation vector.

[0051] As mentioned above, since deleting some power grid operation vectors will cause the centroid of the intermediate vector cluster to change, it is necessary to recalculate the target similarity and add the qualified power grid operation vectors to the corresponding target vector cluster according to the adjacent situation, so as to realize the expansion of the power grid operation vectors within the cluster, and thus realize the expansion of the geographical area, which is beneficial to the unified data analysis of more sub - geographical regions and reduces the redundancy of data analysis.

[0052] S500. Determine the sub - geographical regions corresponding to several power grid operation vectors in the same target vector cluster as the same - type geographical regions and merge them into the key geographical region.

[0053] As mentioned above, since several power grid operation vectors in the same target vector cluster are similar, it indicates that the electricity consumption levels of several sub - geographical regions corresponding to the same target vector cluster are similar and can be considered as the same - type geographical regions. By merging the same - type and adjacent sub - geographical regions, it is beneficial to the unified electricity consumption management and electricity consumption data analysis of the same - type geographical regions.

[0054] Furthermore, in another specific embodiment, the method also determines the same - type geographical regions through the following steps:

[0055] S10. Obtain several target vector clusters corresponding to each preset time period;

[0056] S20. For any two power grid operation vectors, when the two power grid operation vectors exist in the same target vector cluster in each preset time period, determine the sub - geographical regions corresponding to the two power grid operation vectors as the same - type geographical regions.

[0057] As described above, by introducing multiple preset time periods, that is, when the power consumption situations of two sub-geographical regions are similar in each preset time period, the two sub-geographical regions can be determined as the same type of geographical region, making the confirmation result of the same type of region more accurate and reliable, and further making the power consumption management of the same type of geographical region more reasonable.

[0058] An embodiment of the present invention further provides an intelligent statistical analysis system for power grid operation data, as Figure 2 shown. The system includes:

[0059] A first acquisition module 100, configured to acquire, for any preset time period, a power grid operation data set corresponding to each sub-geographical region within a target geographical region in the preset time period; the power grid operation data set includes data corresponding to a plurality of power grid operation indicators. Among them, the sub-geographical region is a pre-divided geographical region, such as a residential area, an industrial area, a shopping mall area, etc.

[0060] Specifically, the preset time period can be any time period within a day; it can be understood that several preset time periods are set within a day, for example, 9:00 - 10:00 am, 14:00 - 15:00 pm, 20:00 - 21:00 pm.

[0061] Furthermore, the power grid operation indicator is any one of rated voltage, rated current, rated electric power, actual voltage, actual current, and actual electric power; it can be understood that the power grid operation indicator data is collected at any time point within the preset time period, or the power grid operation indicator data is collected in several time periods within the preset time period and the average value is taken.

[0062] A clustering module 200, configured to convert each power grid operation data set into a power grid operation vector, and cluster a plurality of power grid operation vectors based on a preset index weight corresponding to each power grid operation indicator to obtain an initial clustering result; the initial clustering result includes several initial vector clusters.

[0063] Specifically, the k-means clustering model is used to cluster the power grid operation vectors; those skilled in the art know the specific implementation manners of the k-means clustering model, which will not be elaborated here.

[0064] Furthermore, the system further includes a weight acquisition module, as Figure 3 shown. The weight acquisition module includes:

[0065] A second acquisition module 001, configured to acquire n historical power grid operation data set samples from several sub-geographical regions, and perform standardization processing on the data in each historical power grid operation data set sample to obtain n target power grid operation data set samples; it can be understood that performing standardization processing on the data means eliminating the influence of different dimensions.

[0066] The first calculation module 002 is configured to calculate the entropy value of each grid operation index if the data corresponding to all grid operation indexes in the n target grid operation data set samples are within the corresponding preset value ranges, and calculate the preset index weight corresponding to each grid operation index according to the entropy value of each grid operation index; it can be understood that each grid operation index corresponds to a preset value range respectively, and the preset value ranges corresponding to the rated voltage, rated current, and rated electric power can be set as several fixed values according to actual requirements.

[0067] Among them, the preset index weight corresponding to the grid operation index meets the following conditions:

[0068] W j =(1 - H j ) / (∑ m j＝1 (1 - H j ))), where W j represents the preset index weight corresponding to the jth grid operation index, H j is the entropy value of the jth grid operation index, and m is the number of grid operation indexes; those skilled in the art know the specific calculation method of the entropy value, which will not be elaborated here.

[0069] The second calculation module 003 is configured to calculate the average data value and the index dispersion degree corresponding to each grid operation index if there is data in the n target grid operation data set samples that is not within the corresponding preset value range; it can be understood that the index dispersion degree refers to the standard deviation of several data corresponding to the grid operation index.

[0070] The processing module 004 is configured to respectively determine the ratio of the average data value and the index dispersion degree corresponding to each grid operation index as the index priority of the grid operation index itself, and normalize each index priority to obtain the preset index weight corresponding to each grid operation index.

[0071] As described above, introducing the preset index weight in k-means clustering makes the obtained similarity more accurate, and the clustering result is more reasonable and reliable. Moreover, when obtaining the preset index weight, different weight calculation methods are adopted according to whether there are outliers in the data, which overcomes the disadvantage that using the entropy value to calculate the weight will affect the weight accuracy when there are outliers, makes the obtained index weight more accurate, and further makes the obtained similarity more accurate and reliable, which is beneficial to the accurate division of clustering.

[0072] The first partitioning module 300 is configured to, for any initial vector cluster, screen out the grid operation vectors corresponding to the sub-geographical regions that do not have adjacent sub-geographical regions according to the adjacency of the sub-geographical regions corresponding to the grid operation vectors in the initial vector cluster, and divide a number of remaining grid operation vectors in the initial vector cluster into a number of intermediate vector clusters; it can be understood that the adjacency of the sub-geographical regions is pre-configured, and those skilled in the art know the specific implementation manners of collecting whether any two sub-geographical regions are adjacent, such as collecting on a map app, which will not be elaborated here.

[0073] Further, as Figure 4 shown, the first partitioning module 300 includes:

[0074] The third acquisition module 301 is configured to acquire the sub-geographical region corresponding to each grid operation vector in the remaining grid operation vectors, obtaining k sub-geographical regions; it can be understood that what is acquired is the sub-geographical region identifier.

[0075] The slicing module 302 is configured to perform a merging and slicing process on the k sub-geographical regions according to the adjacency relationship between the k sub-geographical regions, obtaining a number of geographical region slices; it can be understood that all adjacent sub-geographical regions are merged in sequence, and different merged regions generate different geographical region slices, that is, there are no adjacent sub-geographical regions between different two geographical region slices.

[0076] The second partitioning module 303 is configured to divide the grid operation vectors corresponding to the sub-geographical regions within each geographical region slice into one intermediate vector cluster, so as to obtain a number of intermediate vector clusters.

[0077] As described above, since a number of grid operation vectors within the initial vector cluster are all similar and can be divided into the same power consumption level for grid operation data statistical analysis, the independent sub-geographical regions are screened out, and the regions with adjacent sub-geographical regions are merged and sliced. The power consumption situations of the sub-geographical regions within each geographical region slice are similar and can be used as a whole region for subsequent unified data analysis, which is beneficial to the unified power consumption management of the slices.

[0078] The adding module 400 is configured to, for any screened-out grid operation vector, determine the target vector cluster corresponding to the screened-out grid operation vector according to the target similarity between the screened-out grid operation vector and the centroids of all the intermediate vector clusters and the adjacency with a number of grid operation vectors in each intermediate vector cluster, and add the screened-out grid operation vector to the corresponding target vector cluster; it can be understood that all the intermediate vector clusters refer to the sum of a number of intermediate vector clusters divided from each initial vector cluster.

[0079] Specifically, as Figure 5 shown, the adding module 400 includes:

[0080] The first determination module 401 is configured to calculate the target similarity between the screened power grid operation vectors and the centroids of all intermediate vector clusters respectively for any screened power grid operation vector, and determine the intermediate vector cluster corresponding to the maximum target similarity as the key vector cluster; it can be understood that the target similarity is calculated through the Euclidean distance and the preset index weights corresponding to each power grid operation index.

[0081] The fourth acquisition module 402 is configured to, when the maximum target similarity is greater than the preset similarity threshold, acquire the sub-geographical regions corresponding to all the power grid operation vectors in the key vector cluster; those skilled in the art set the preset similarity threshold according to actual needs, which will not be elaborated here.

[0082] The second determination module 403 is configured to, when there is a geographical region adjacent to the sub-geographical region corresponding to the screened power grid operation vector among the sub-geographical regions corresponding to all the power grid operation vectors in the key vector cluster, determine the key vector cluster as the target vector cluster corresponding to the screened power grid operation vector.

[0083] As mentioned above, since deleting some power grid operation vectors will cause the centroid of the intermediate vector cluster to change, it is necessary to recalculate the target similarity and add the power grid operation vectors that meet the requirements to the corresponding target vector cluster according to the adjacency situation, so as to realize the expansion of the power grid operation vectors within the cluster, and thus realize the expansion of the geographical region area, which is beneficial to unified data analysis of a larger number of sub-geographical regions and reduces the redundancy of data analysis.

[0084] The merging module 500 is configured to determine the sub-geographical regions corresponding to several power grid operation vectors in the same target vector cluster as the same type of geographical regions and merge them into the key geographical region.

[0085] As mentioned above, since several power grid operation vectors in the same target vector cluster are similar, it indicates that the electricity consumption levels of several sub-geographical regions corresponding to the same target vector cluster are similar, and they can be considered as the same type of geographical regions. By merging the same type of adjacent sub-geographical regions, it is beneficial to unified electricity consumption management and electricity consumption data analysis of the same type of geographical regions.

[0086] Furthermore, the system further includes a third determination module, as Figure 6 shown, the third determination module includes:

[0087] The fifth acquisition module 10 acquires several target vector clusters corresponding to each preset time period.

[0088] The fourth determination module 20 determines, for any two grid operation vectors, that the sub-geographical regions corresponding to the two grid operation vectors are the same type of geographical region when the two grid operation vectors exist in the same target vector cluster in each preset time period.

[0089] As described above, by introducing multiple preset time periods, that is, when the electricity consumption situations of two sub-geographical regions are similar in each preset time period, can it be determined that the two sub-geographical regions are the same type of geographical region, making the confirmation result of the same type of region more accurate and reliable, and thus making the electricity consumption management of the same type of geographical region more reasonable.

[0090] Although some specific embodiments of the present invention have been described in detail by way of examples, those skilled in the art should understand that the above examples are only for illustration and not for limiting the scope of the present invention. Those skilled in the art should also understand that various modifications can be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. An intelligent statistical analysis method for power grid operation data, characterized in that, The method includes the following steps: S100. For any preset time period, obtain the power grid operation data sets corresponding to each sub-geographical area within the target geographical area during the preset time period; the power grid operation data sets include data corresponding to a number of power grid operation indicators; S200. Convert each power grid operation data set into a power grid operation vector, and based on the preset index weights corresponding to each power grid operation indicator, cluster a number of power grid operation vectors to obtain an initial clustering result; the initial clustering result includes a number of initial vector clusters; S300. For any initial vector cluster, according to the adjacency of the sub-geographical areas corresponding to the power grid operation vectors in the initial vector cluster, filter out the power grid operation vectors corresponding to the sub-geographical areas that do not have adjacency, and divide the remaining power grid operation vectors in the initial vector cluster into a number of intermediate vector clusters; S400. For any filtered power grid operation vector, determine the target vector cluster corresponding to the filtered power grid operation vector according to the target similarity between the filtered power grid operation vector and the centroids of all intermediate vector clusters and the adjacency with a number of power grid operation vectors in each intermediate vector cluster, and add the filtered power grid operation vector to the corresponding target vector cluster; S500. Determine the sub-geographical areas corresponding to the power grid operation vectors in the same target vector cluster as the same type of geographical areas and merge them into key geographical areas.

2. The intelligent statistical analysis method for power grid operation data according to claim 1, characterized in that The power grid operation indicator is any one of rated voltage, rated current, rated electric power, actual voltage, actual current, and actual electric power.

3. The intelligent statistical analysis method for power grid operation data according to claim 1, wherein The preset index weights corresponding to each power grid operation indicator are obtained through the following steps: S001. Obtain n historical power grid operation data set samples from a number of sub-geographical areas, and perform standardization processing on the data in each historical power grid operation data set sample to obtain n target power grid operation data set samples; S002. If the data corresponding to all power grid operation indicators in the n target power grid operation data set samples are within the corresponding preset numerical intervals, calculate the entropy value of each power grid operation indicator, and calculate the preset index weight corresponding to each power grid operation indicator according to the entropy value of each power grid operation indicator; Among them, the preset index weights corresponding to the power grid operation indicators meet the following conditions: W j = (1 - H j ) / (∑ m j＝1 (1 - H j ))), where W j represents the preset index weight corresponding to the jth power grid operation index, H j is the entropy value of the jth power grid operation index, and m is the number of power grid operation indices; S003. If there is data in the n target power grid operation data set samples that is not within the corresponding preset numerical intervals for the data corresponding to all power grid operation indicators, calculate the average data value and the index dispersion degree corresponding to each power grid operation indicator; S004. Determine the ratio of the average data value and the index dispersion degree corresponding to each power grid operation indicator as the index priority of the power grid operation indicator itself, and normalize each index priority to obtain the preset index weight corresponding to each power grid operation indicator.

4. The intelligent statistical analysis method for power grid operation data according to claim 1, wherein The remaining power grid operation vectors in the initial vector cluster are divided into a number of intermediate vector clusters through the following steps: S301. Obtain the sub-geographical areas corresponding to each power grid operation vector in the remaining power grid operation vectors to obtain k sub-geographical areas; S302. According to the adjacency relationship between k sub-geographical regions, perform a merging and slicing process on the k sub-geographical regions to obtain several geographical region slices. S303. Divide the power grid operation vectors corresponding to several sub-geographical regions within each geographical region slice into an intermediate vector cluster to obtain several intermediate vector clusters.

5. The intelligent statistical analysis method for power grid operation data according to claim 1, characterized in that The target vector cluster corresponding to the filtered power grid operation vector is determined through the following steps: S401. For any filtered power grid operation vector, calculate the target similarity between the filtered power grid operation vector and the centroids of all intermediate vector clusters, and determine the intermediate vector cluster corresponding to the maximum target similarity as the key vector cluster. S402. When the maximum target similarity is greater than the preset similarity threshold, obtain the sub-geographical regions corresponding to all the power grid operation vectors in the key vector cluster. S403. When there is a geographical region adjacent to the sub-geographical region corresponding to the filtered power grid operation vector among the sub-geographical regions corresponding to all the power grid operation vectors in the key vector cluster, use the key vector cluster as the target vector cluster corresponding to the filtered power grid operation vector.

6. The intelligent statistical analysis method for power grid operation data according to claim 1, wherein The method further includes the following steps: S10. Obtain several target vector clusters corresponding to each preset time period. S20. For any two power grid operation vectors, when the two power grid operation vectors are in the same target vector cluster in each preset time period, determine the sub-geographical regions corresponding to the two power grid operation vectors as the same type of geographical regions.

7. An intelligent statistical analysis system for power grid operation data, characterized in that, The system includes: A first acquisition module, configured to, for any preset time period, acquire the power grid operation data set corresponding to each sub-geographical region within the target geographical region during the preset time period; the power grid operation data set includes data corresponding to several power grid operation indicators. A clustering module, configured to convert each power grid operation data set into a power grid operation vector, and perform clustering on several power grid operation vectors based on the preset index weights corresponding to each power grid operation indicator to obtain an initial clustering result; the initial clustering result includes several initial vector clusters. A first division module, configured to, for any initial vector cluster, filter out the power grid operation vectors corresponding to the non-adjacent sub-geographical regions according to the adjacency situation of the sub-geographical regions corresponding to the power grid operation vectors in the initial vector cluster, and divide several remaining power grid operation vectors in the initial vector cluster into several intermediate vector clusters. An addition module, configured to, for any filtered power grid operation vector, determine the target vector cluster corresponding to the filtered power grid operation vector according to the target similarity between the filtered power grid operation vector and the centroids of all intermediate vector clusters and the adjacency situation with several power grid operation vectors in each intermediate vector cluster, and add the filtered power grid operation vector to the corresponding target vector cluster. A merging module, configured to determine the sub-geographical regions corresponding to several power grid operation vectors in the same target vector cluster as the same type of geographical regions and merge them into a key geographical region.

8. The intelligent statistical analysis system for power grid operation data according to claim 7, wherein The power grid operation indicator is any one of rated voltage, rated current, rated electric power, actual voltage, actual current, and actual electric power.

9. The intelligent statistical analysis system for power grid operation data according to claim 7, wherein The system further includes a weight acquisition module, and the weight acquisition module includes: A second acquisition module, configured to acquire n historical power grid operation dataset samples from a number of sub-geographical regions, and perform normalization processing on the data in each historical power grid operation dataset sample to obtain n target power grid operation dataset samples; A first calculation module, configured to calculate the entropy value of each power grid operation index and calculate the corresponding preset index weight of each power grid operation index according to the entropy value of each power grid operation index if the data corresponding to all power grid operation indexes in the n target power grid operation dataset samples are within the corresponding preset numerical intervals; Wherein, the preset index weight corresponding to the power grid operation index meets the following conditions: W j = (1 - H j ) / (∑ m j＝1 (1 - H j ))), where W j represents the preset index weight corresponding to the j-th power grid operation index, H j is the entropy value of the j-th power grid operation index, and m is the number of power grid operation indices; A second calculation module, configured to calculate the average data value and the index dispersion degree corresponding to each power grid operation index if there is data in the data corresponding to all power grid operation indexes in the n target power grid operation dataset samples that is not within the corresponding preset numerical intervals; A processing module, configured to respectively determine the ratio of the average data value and the index dispersion degree corresponding to each power grid operation index as the index priority of the power grid operation index itself, and normalize each index priority to obtain the preset index weight corresponding to each power grid operation index.

10. The intelligent statistical analysis system for power grid operation data according to claim 7, wherein The first partitioning module includes: A third acquisition module, configured to acquire the sub-geographical region corresponding to each power grid operation vector in the remaining power grid operation vectors to obtain k sub-geographical regions; A slicing module, configured to perform a merging and slicing process on the k sub-geographical regions according to the adjacent relationship between the k sub-geographical regions to obtain a number of geographical region slices; A second partitioning module, configured to partition the power grid operation vectors corresponding to a number of sub-geographical regions within each geographical region slice into an intermediate vector cluster to obtain a number of intermediate vector clusters.

Images