Power distribution network station area user portrait generation method, device, equipment and medium

Abstract

The method is mainly applied to the technical field of artificial intelligence. The invention discloses a power distribution network district user portrait generation method, device and equipment and a medium, and the method comprises the steps: obtaining power data in a target district, and generating a power utilization behavior feature set based on the power data; clustering feature data of the power consumption behavior feature set to obtain a plurality of first data sets; combining the plurality of first data sets into a second data set, wherein the second data set comprises a plurality of data layer nodes; dividing the second data set into a plurality of data subsets based on a target data layer node in the plurality of data layer nodes; and determining the user characteristics corresponding to each data subset and displaying the user characteristics to obtain a power distribution network area user portrait including each user characteristic. The method can generate the accurate power distribution network area user portrait, facilitates the optimization of the distribution of power resources in a power system according to the power distribution network area user portrait, and improves the operation efficiency of a power distribution network.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method, device, equipment and medium for generating user portraits in a distribution network substation area. Background Art

[0002] With the continuous growth of electricity demand and the acceleration of urbanization, the planning and optimization of distribution networks have become increasingly important. As an important part of the power system, the distribution network is mainly responsible for receiving electricity from the transmission network or regional power plants, and distributing it locally or step by step according to voltage to various users through distribution facilities. The substation is the last level of the distribution network and the part that is in direct contact with users. It usually refers to the area in the distribution network that is powered by a single transformer, and all users in this area are connected to the distribution network through the same transformer. By analyzing the electricity consumption data of users in the substation, power companies can understand the electricity consumption characteristics and behavior patterns of different users in detail. However, traditional electricity consumption data analysis methods rely on experience and simple statistics, lack in-depth analysis of user characteristics, and are difficult to identify different types of user groups, resulting in low accuracy and insufficient effectiveness of distribution network planning. Summary of the invention

[0003] The present invention provides a method, device, equipment and medium for generating user portraits of distribution network substations, which can generate accurate user portraits of distribution network substations, and is conducive to optimizing the allocation of power resources in the power system according to the user portraits of the distribution network substations, thereby improving the operating efficiency of the distribution network.

[0004] The present invention provides a method for generating a user portrait of a distribution network area, the method comprising: Acquire power data in the target area, and generate a power consumption behavior feature set based on the power data; Clustering the feature data of the power usage behavior feature set to obtain a plurality of first data sets; Merging the plurality of first data sets into a second data set, the second data set comprising a plurality of data layer nodes; Based on a target data layer node among the multiple data layer nodes, dividing the second data set into multiple data subsets; The user characteristics corresponding to each of the data subsets are determined and displayed to obtain a distribution network substation user profile including each of the user characteristics.

[0005] Furthermore, the acquiring of power data in the target area and generating a power usage behavior feature set based on the power data includes: Dividing the power data into a plurality of characteristic parameters; Based on the degree of correlation between each of the characteristic parameters, the multiple characteristic parameters are screened and the power consumption behavior feature set is generated according to the screened characteristic parameters.

[0006] Further, each of the characteristic parameters includes a plurality of parameter values; The method for generating user portraits in distribution network substations also includes the step of determining the degree of correlation; The determining of the degree of correlation comprises: Obtaining a sample value from a plurality of parameter values ​​in the first characteristic parameter, and determining an average value of the plurality of parameter values ​​in the first characteristic parameter; Obtaining a sample value from a plurality of parameter values ​​in the second characteristic parameter, and determining an average value of the plurality of parameter values ​​in the second characteristic parameter; Based on the correlation between the sample value and average value corresponding to the first characteristic parameter and the sample value and average value corresponding to the second characteristic parameter, the correlation value between the first characteristic parameter and the second characteristic parameter is determined to determine the degree of correlation between the first characteristic parameter and the second characteristic parameter.

[0007] Furthermore, the method for generating user portraits of distribution network substations further includes the step of determining the number of clusters when clustering the feature data of the power consumption behavior feature set; The step of determining the number of clusters when clustering the feature data of the power usage behavior feature set includes: When clustering the feature data of the power usage behavior feature set, the power usage behavior feature set is divided into a plurality of data clusters; Determine a sum of squared errors in each of the data clusters, wherein the sum of squared errors is the sum of squared distances between each sample data point in the data cluster and a center point of the data cluster; Based on the number of the data clusters and each of the square error sum values, a relationship curve graph is generated, wherein the relationship curve graph includes a horizontal axis and a vertical axis, wherein the horizontal axis represents the number of the data clusters, and the vertical axis represents the square error sum value; The inflection point of the target curve is determined in the relationship curve diagram, and the abscissa value corresponding to the inflection point of the target curve is used as the optimal number of clusters.

[0008] Further, the plurality of first data sets are merged into a second data set, wherein the second data set includes a plurality of data layer nodes, including: Based on the distance between each data point in the first data set, a merging operation is performed on each data point until each data point of the first data set is merged into the same cluster to obtain the second data set, which includes multiple data layer nodes.

[0009] Furthermore, the specific method of performing the merging operation on each of the data points is: Based on each link distance, a distance matrix is ​​generated, wherein the distance matrix includes a plurality of clusters, each cluster includes at least one data point, and the link distance is the distance between two clusters; Merge the two clusters with the smallest link distance in the distance matrix into a new cluster; Based on the distances between the new cluster and other clusters, the distance matrix is ​​updated; According to the updated distance matrix, the step of merging the two clusters with the smallest link distance in the distance matrix into a new cluster is returned until each of the data points is merged into the same cluster, thereby obtaining a hierarchical clustering tree structure, which is used to divide the data layer nodes into multiple clusters.

[0010] Furthermore, the method for generating a user portrait of a distribution network area further includes: identifying abnormal data including abnormal values ​​in the power data; The abnormal data is deleted according to the abnormal value, and the power data after the abnormal data is deleted is used as the pre-processed power data.

[0011] The present invention also provides a device for generating a user portrait of a distribution network area, the device comprising: An acquisition module, used to acquire power data in a target area and generate a power consumption behavior feature set based on the power data; A first clustering module, used for clustering the feature data of the power consumption behavior feature set to obtain a plurality of first data sets; A second clustering module, configured to merge the plurality of first data sets into a second data set, wherein the second data set includes a plurality of data layer nodes; A hierarchical partitioning module, configured to partition the second data set into a plurality of data subsets based on a target data layer node among the plurality of data layer nodes; A display module is used to determine the user characteristics corresponding to each of the data subsets and display them to obtain a distribution network substation user portrait including each of the user characteristics.

[0012] The present invention also provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method for generating a user portrait of a distribution network substation as described in any one of the above items is implemented.

[0013] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for generating a user portrait of a distribution network substation as described in any one of the above items is implemented.

[0014] The present invention has at least the following beneficial effects: The technical solution of this application can generate accurate user portraits of distribution network substations. First, by collecting power data in the target substation, a feature set of power consumption behavior is automatically generated, avoiding the subjectivity of artificial feature selection; then, a clustering algorithm is used to group the feature data. This method is not affected by personal experience and can objectively identify the natural grouping of power consumption behavior; then, by merging and dividing data sets, the user groups are further refined. This method is more in-depth than simple statistical analysis and can identify different types of user groups; finally, the user characteristics corresponding to each data subset are displayed to construct an accurate user portrait of the distribution network substation. The technical solution of this application reduces the reliance on expert experience, improves the accuracy and effectiveness of user portraits, and realizes in-depth analysis of user characteristics, which is conducive to optimizing the allocation of power resources in the power system according to the user portraits of the distribution network substations, thereby improving the operating efficiency of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the technical solution of the present invention and do not constitute a limitation on the technical solution of the present invention.

[0016] Figure 1 It is a flow chart of the steps of the method for generating user portraits in distribution network substations; Figure 2 It is a flow chart of the steps for determining the degree of relevance in the method for generating user profiles in distribution network substations; Figure 3 It is a flow chart of the steps for determining the number of clusters in the method for generating user profiles in distribution network substations; Figure 4 It is a flow chart of the steps of merging each data point in the method for generating user portraits in distribution network substations; Figure 5 It is a flowchart of the steps when the distribution network substation user portrait generation method is implemented in the application scenario; Figure 6 It is a structural schematic diagram of a user portrait generating device in a distribution network station area; Figure 7 It is a structural diagram of an electronic device. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0018] In the relevant technical field, user profiling in the substation area is a very important part of distribution network management and optimization, and its accuracy will directly affect the economy, stability and safety of the power system. In user profiling, traditional cluster analysis is a commonly used method, which identifies different types of user groups by analyzing historical electricity consumption data. However, traditional cluster analysis has some shortcomings. For example, many existing methods rely on expert experience for feature selection. This method has low reliability and is easily affected by subjective factors, resulting in insufficient accuracy and stability of the profiling results.

[0019] Please refer to Figure 1 , Figure 1 It is a flow chart of the steps of the method for generating user portraits in distribution network substations.

[0020] This embodiment provides a method for generating a user profile of a distribution network area, including: S101: Obtain power data in a target substation, and generate a power consumption behavior feature set based on the power data.

[0021] S102: Clustering feature data of a power usage behavior feature set to obtain a plurality of first data sets.

[0022] S103: Merge multiple first data sets into a second data set, where the second data set includes multiple data layer nodes.

[0023] S104. Divide the second data set into multiple data subsets based on the target data layer node among the multiple data layer nodes.

[0024] S105. Determine the user characteristics corresponding to each data subset and display them to obtain a distribution network substation user portrait including each user characteristic.

[0025] In step S101 of some embodiments, the user's electricity consumption data, including electricity consumption, electricity consumption time, and electricity consumption frequency, is collected in real time through a smart meter, and historical electricity consumption records are used to analyze electricity consumption behavior patterns in combination with seasonal factors and weather changes, or in combination with user feedback and survey data to supplement the electricity consumption behavior feature set to increase dimensions such as user satisfaction and electricity usage habits.

[0026] In step S102 of some embodiments, a K-means clustering algorithm may be used to divide users into different groups according to the similarity of electricity usage characteristics. In addition to the K-means algorithm, there are many other clustering algorithms to choose from, such as DBSCAN (Density-Based Spatial Clustering of Applications with Noise), Gaussian Mixture Models (GMM), Spectral Clustering, etc.

[0027] In step S103 of some embodiments, a dimensionality reduction technique such as principal component analysis (PCA) is used to merge multiple first data sets and reduce the data dimension to form a second data set.

[0028] Optionally, multiple first data sets can be merged into the second data set by a multi-level clustering algorithm. Multi-level clustering is a technique that combines multiple clustering algorithms to perform cluster analysis on data at different levels in order to more finely identify and classify patterns and features in the data. It can more comprehensively capture the hierarchical structure and complexity of the data by gradually refining the clustering results of the data. Hierarchical clustering can be divided into two types: agglomerative hierarchical clustering (bottom-up) and divisive hierarchical clustering (top-down). Agglomerative hierarchical clustering starts from each data point, merges the two nearest clusters, and gradually builds larger clusters until all data points are merged into one cluster.

[0029] It can be understood that the second data set is not a single large cluster, but a hierarchy of clusters. This structure includes multiple data hierarchy nodes, and each data hierarchy node can be cut at different levels to obtain clusters of different numbers and sizes. The multiple clusters obtained by cutting, each cluster represents a group of similar users. For each cluster, by analyzing the common characteristics of its members, these characteristics can be used to build a user portrait. For example, we can calculate the average age, average consumption amount, most common purchase categories, etc. of each cluster. Based on the characteristic analysis of the clusters, a user portrait can be built for each cluster. User portraits can include demographic information, behavior patterns, preferences, etc.

[0030] Optionally, in addition to traditional clustering algorithms, deep learning models can also be used to create user portraits. For example, a convolutional neural network (CNN) can be used to extract spatial features from user electricity usage data, or a recurrent neural network (RNN) can be used to capture the time series features of user electricity usage behavior. These deep learning models can more comprehensively capture the complex patterns of user electricity usage behavior and improve the accuracy of user portraits.

[0031] Optionally, the power data refers to data such as the name of the substation, rated capacity of the transformer, annual maximum and minimum load rates of the substation, etc. In addition, by incorporating the annual cumulative number of power outages in the substation, the three-phase imbalance rate, etc. into the analysis dimensions, the aim is to generate user portraits through cluster analysis of more accurate substation characteristics, so as to better reflect user needs and carry out distribution network planning and optimization.

[0032] In some embodiments, obtaining power data in a target area and generating a power usage behavior feature set based on the power data include: The power data is divided into a plurality of characteristic parameters; based on the correlation between each characteristic parameter, the plurality of characteristic parameters are screened and a power consumption behavior feature set is generated according to the screened characteristic parameters.

[0033] In some embodiments, each characteristic parameter includes multiple parameter values.

[0034] Please refer to Figure 2 , Figure 2 It is a flow chart of the steps for determining the degree of relevance in the method for generating user profiles in distribution network substations.

[0035] The method for generating user portraits in distribution network areas also includes a step of determining the degree of relevance, and the specific implementation of this step includes: S201. Obtain a sample value from a plurality of parameter values ​​in a first characteristic parameter, and determine an average value of the plurality of parameter values ​​in the first characteristic parameter.

[0036] S202: Obtain a sample value from a plurality of parameter values ​​in the second characteristic parameter, and determine an average value of the plurality of parameter values ​​in the second characteristic parameter.

[0037] S203. Determine the correlation value between the first characteristic parameter and the second characteristic parameter based on the correlation relationship between the sample value and the average value corresponding to the first characteristic parameter and the sample value and the average value corresponding to the second characteristic parameter to determine the degree of correlation between the first characteristic parameter and the second characteristic parameter.

[0038] Optionally, in addition to using correlation analysis, other feature selection and dimensionality reduction methods can be combined to optimize the selection of user power consumption features, such as principal component analysis (PCA), linear discriminant analysis (LDA), mutual information method, etc. These methods can more effectively extract and select features in specific situations and improve the accuracy and reliability of user portraits.

[0039] It can be understood that this embodiment introduces a correlation analysis method to reduce the dimension and optimize the selection of user electricity consumption characteristics, and comprehensively considers the correlation between features to select the most closely related and concise feature set to the user category, thereby improving the accuracy and reliability of user portraits, thereby effectively reducing redundant features and improving data processing efficiency and the accuracy of analysis results.

[0040] Please refer to Figure 3 , Figure 3 It is a flow chart of the steps for determining the number of clusters in the method for generating user profiles in distribution network substations.

[0041] In some embodiments, the method for generating a user profile of a distribution network area further includes a step of determining the number of clusters when clustering the feature data of the power consumption behavior feature set. The specific implementation of this step includes: S301 . When clustering feature data of a power usage behavior feature set, the power usage behavior feature set is divided into a plurality of data clusters.

[0042] S302 , determining a sum of squared errors in each data cluster, where the sum of squared errors is the sum of squared distances between each sample data point in the data cluster and the center point of the data cluster.

[0043] S303 , generating a relationship curve graph based on the number of data clusters and each error sum of squares value, wherein the relationship curve graph includes a horizontal axis and a vertical axis, wherein the horizontal axis represents the number of data clusters, and the vertical axis represents the error sum of squares value.

[0044] S304: Determine the inflection point of the target curve in the relationship curve diagram, and use the abscissa value corresponding to the inflection point of the target curve as the optimal number of clusters.

[0045] It should be noted that when traditional clustering algorithms process large-scale data, it is usually difficult to determine the optimal number of clusters, and the clustering results are easily disturbed by noise data, affecting their accuracy. For example, although the K-means clustering algorithm is widely used, it is sensitive to the selection of the initial cluster center and is prone to fall into the local optimal solution. In addition, the presence of noise data and outliers will significantly affect the accuracy and stability of the clustering results.

[0046] In this embodiment, first, by dividing the power consumption behavior feature set into multiple data clusters, a preliminary clustering process is performed on large-scale data; then, the error sum of squares of each data cluster is calculated. This step quantifies the quality of clustering and makes the evaluation of clustering results more objective; then, a relationship curve is generated to visualize the relationship between the number of data clusters and the error sum of squares, which helps to identify the changing trend between the number of clusters and the quality of clustering; finally, the inflection point of the target curve is determined in the relationship curve. This inflection point represents the optimal solution for the number of clusters, that is, at this point, increasing the number of clusters will no longer significantly improve the quality of clustering. This method reduces the reliance on expert experience, reduces the impact of noise data, and improves the accuracy and reliability of clustering results.

[0047] In some embodiments, a plurality of first data sets are merged into a second data set, the second data set comprising a plurality of data layer nodes, including: Based on the distance between each data point in the first data set, a merging operation is performed on each data point until each data point of the first data set is merged into the same cluster to obtain a second data set, which includes multiple data layer nodes.

[0048] In some embodiments, a K-means clustering method is used when generating the first data set, and a hierarchical clustering method is used when generating the second data set. K-means clustering and hierarchical clustering are combined to perform multi-level clustering analysis on user data. Different types of user groups are identified through preliminary clustering and further refined clustering. This method can more comprehensively capture the multi-level characteristics of user electricity consumption behavior and improve the accuracy and stability of clustering results.

[0049] Please refer to Figure 4 , Figure 4 It is a flow chart of the steps of merging each data point in the method for generating user portraits in distribution network substations.

[0050] In some embodiments, the specific manner of performing the merging operation on each data point is: S401. Generate a distance matrix based on each link distance, wherein the distance matrix includes multiple clusters, each cluster includes at least one data point, and the link distance is the distance between two clusters.

[0051] S402: Merge the two clusters with the smallest link distance in the distance matrix into a new cluster.

[0052] S403: Update the distance matrix based on the distances between the new cluster and other clusters.

[0053] S404, returning to step S402 according to the updated distance matrix, until each data point is merged into the same cluster, and a hierarchical clustering tree structure is obtained, wherein the hierarchical clustering tree structure is used to divide the data layer nodes into multiple clusters.

[0054] It should be noted that existing user profiling methods usually only focus on a single-level cluster analysis, which makes it difficult to fully capture the multi-level characteristics of user electricity consumption behavior. Single-level cluster analysis cannot fully reflect the complexity and diversity of user electricity consumption behavior, resulting in insufficient detail and accuracy of the profiling results.

[0055] This embodiment solves the limitations of existing user portrait methods when processing multi-dimensional data through a hierarchical clustering method. First, by generating a distance matrix, the similarity between clusters can be quantified, which helps to fully capture the multi-level characteristics of user electricity consumption behavior; then, the two clusters with the smallest link distance are merged. This step can gradually integrate information and reduce feature redundancy while retaining the correlation between features. After updating the distance matrix, the merging process is repeated until all data points are merged into the same cluster to form a hierarchical clustering tree structure. This tree structure can not only reveal the hierarchy of user electricity consumption behavior, but also divide it on different data layer nodes to obtain multiple clusters, each cluster representing a group of users with similar electricity consumption behavior. This method effectively processes multi-dimensional data, improves the effectiveness and simplicity of feature selection, and makes user portraits more comprehensive and accurate.

[0056] In some embodiments, the method for generating a user profile of a distribution network area further includes: Abnormal data including abnormal values ​​in the power data are identified; the abnormal data are deleted according to the abnormal values, and the power data after the abnormal data is deleted is used as the pre-processed power data.

[0057] It can be understood that this embodiment preprocesses the original data set with the purpose of removing noise data and outliers to ensure the quality and consistency of the data. Among them, outliers are used to identify and process abnormal points in the data to reduce their impact on the clustering results, which is conducive to ensuring the accuracy of multi-level clustering and providing a reliable data basis for the subsequent generation of user portraits.

[0058] In some embodiments, when displaying user characteristics and user portraits of distribution network stations, visualization tools such as radar charts and bar charts can be used to intuitively display user power consumption characteristics and behavior patterns, helping power companies to better understand and manage user needs. Visualized user portraits not only improve the intuitiveness of data analysis, but also provide a scientific basis for the optimization and management of power systems.

[0059] It is understandable that through in-depth analysis of user portraits in distribution network areas, user groups with potential for regulation can be identified, supporting power companies to effectively regulate during peak periods of power demand, balance power supply and demand, and ensure stable operation of the power grid. At the same time, personalized services and solutions are provided for different types of users to improve user satisfaction and loyalty.

[0060] Please refer to Figure 5 , Figure 5 It is a flowchart of the steps when the distribution network substation user portrait generation method is implemented in the application scenario.

[0061] like Figure 5As shown, this embodiment provides a method for generating user portraits in a distribution network substation area in an application scenario.

[0062] First, extract the power data in the target substation for analysis, including the following characteristic parameters: substation name, transformer GISID, transformer rated capacity (kVA), number of low-voltage users in the substation, rated current (A), 10 limit-crossing days, load evaluation, time of annual maximum load rate in the substation, annual maximum load rate in the substation (%), three-phase imbalance rate when the load rate is maximum, time of annual minimum load rate in the substation, annual minimum load rate in the substation, three-phase imbalance rate when the load rate is minimum, annual cumulative power outage duration (hours), annual cumulative number of power outages, transformer operating years, topology change sign, monthly line loss in the substation (%), power supply radius, and 10 distribution 2.0 substations.

[0063] Secondly, the above power data is preprocessed, including data cleaning and outlier detection.

[0064] Next, the correlation analysis method is used to identify and quantify the relationship between each feature, thereby reducing the dimensionality of the original data. In the correlation analysis, the Pearson correlation coefficient is used to measure the linear correlation between two variables. The formula of the Pearson correlation coefficient is:

[0065] in, and The variables are and The sample value of and They are and The value range of the Pearson correlation coefficient is [-1, 1], where 1 indicates a complete positive correlation, -1 indicates a complete negative correlation, and 0 indicates no correlation. The Pearson correlation coefficient between each feature and the user category is calculated, and features are selected based on the size of the correlation coefficient. The selected feature parameters are added to the electricity consumption behavior feature set.

[0066] After the selection is completed, a multi-level clustering analysis is performed on the feature parameters by combining K-means clustering and hierarchical clustering methods. First, the K-means clustering algorithm is used to perform preliminary clustering on the data and divide the data into several preliminary clustering groups. In order to determine the optimal number of clusters for K-means clustering, the present invention adopts the Elbow Method. The Elbow Method calculates the total sum of squared errors (SSE) under different numbers of clusters, and plots the relationship between the number of clusters and SSE. The inflection point where the SSE drops significantly is selected as the inflection point of the target curve, and the horizontal coordinate value corresponding to the inflection point of the target curve is used as the optimal number of clusters. The goal of the K-means clustering algorithm is to minimize the distance from the data point to the cluster center in each cluster, and its objective function is:

[0067] in, is the number of clusters, It is clusters, It is The centers of the clusters, is a data point.

[0068] After the preliminary clustering is completed, the hierarchical clustering algorithm is further used to refine the results of the preliminary clustering. Agglomerative hierarchical clustering starts from each data point, merges the two closest clusters, and gradually builds larger clusters until all data points are merged into one cluster. Its main steps and formulas are as follows: 1. Calculate the distance matrix. Calculate the distance between all data points to form a distance matrix. Commonly used distance metrics include Euclidean distance, Manhattan distance, etc. The formula for Euclidean distance is:

[0069] in, and are two data points, is the dimension of the feature.

[0070] 2. Merge the two closest clusters. Find the two clusters with the smallest distance in the distance matrix and merge them into a new cluster. Update the distance matrix and calculate the distance between the new cluster and other clusters. Common distance update methods include minimum distance (single link), maximum distance (full link) and average distance (average link). The formula for single link is:

[0071] in, and There are two clusters, and are the data points in the cluster.

[0072] 3. Repeat step 2 until all data points are merged into one cluster.

[0073] The goal of hierarchical clustering is to identify the hierarchical structure and complexity in the data by gradually refining the clustering results of the data. Through multi-level clustering methods, different types of user groups in the distribution area can be identified, such as high-load users, low-load users, users with stable electricity consumption behavior, and users with large fluctuations in electricity consumption behavior.

[0074] Finally, by using visualization tools such as radar charts and bar charts, user characteristics are displayed to obtain a distribution network user profile that includes the characteristics of each user.

[0075] It can be understood that any of the above embodiments can generate an accurate user portrait of the distribution network substation. First, by collecting power data in the target substation, a power consumption behavior feature set is automatically generated, avoiding the subjectivity of artificial feature selection; then, the feature data is grouped using a clustering algorithm. This method is not affected by personal experience and can objectively identify the natural grouping of power consumption behavior; then, by merging and dividing data sets, the user groups are further refined. This method is more in-depth than simple statistical analysis and can identify different types of user groups; finally, the user characteristics corresponding to each data subset are displayed to construct an accurate user portrait of the distribution network substation. The technical solution of this application reduces the reliance on expert experience, improves the accuracy and effectiveness of user portraits, and realizes in-depth analysis of user characteristics, which is conducive to optimizing the allocation of power resources in the power system according to the user portraits of the distribution network substation, thereby improving the operating efficiency of the distribution network.

[0076] Please refer to Figure 6 , Figure 6 It is a structural schematic diagram of a user portrait generating device in a distribution network substation area.

[0077] This embodiment provides a distribution network area user profile generation device including: The acquisition module 601 is used to acquire power data in the target substation and generate a power consumption behavior feature set based on the power data.

[0078] The first clustering module 602 is used to cluster the feature data of the electricity usage behavior feature set to obtain a plurality of first data sets.

[0079] The second clustering module 603 is used to merge the multiple first data sets into a second data set, where the second data set includes multiple data layer nodes.

[0080] The hierarchical division module 604 is used to divide the second data set into multiple data subsets based on the target data layer node among the multiple data layer nodes.

[0081] The display module 605 is used to determine the user characteristics corresponding to each data subset and display them to obtain a user portrait of the distribution network area including each user characteristic.

[0082] It will be appreciated by those skilled in the art that all or some of the steps and devices in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or transient medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. As is well known to those skilled in the art, communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0083] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0084] An embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, it implements any of the above methods for generating user portraits of distribution network substations.

[0085] refer to Figure 7 , Figure 7 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes: The processor 701 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application; The memory 702 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 702 can store operating devices and other applications. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 702, and the processor 701 calls and executes the distribution network substation user portrait generation method of the embodiment of this application; Input / output interface 703, used to implement information input and output; Communication interface 704, used to realize communication interaction between the device and other devices, which can be realized through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.); A bus 705 that transmits information between various components of the device (e.g., the processor 701, the memory 702, the input / output interface 703, and the communication interface 704); The processor 701 , the memory 702 , the input / output interface 703 and the communication interface 704 are connected to each other in communication within the device via a bus 705 .

[0086] It can be understood that the contents of the above method embodiments are all applicable to the electronic device embodiment, the functions specifically implemented by the electronic device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0087] An embodiment of the present application also provides a computer-readable storage medium, which stores a program executable by a processor. When the program executable by the processor is executed by the processor, it is used to implement the method for generating user portraits of distribution network substations as described in any one of the above-mentioned specific embodiments.

[0088] An embodiment of the present application also discloses a computer program product, including a computer program or computer instructions, wherein the computer program or computer instructions are stored in a computer-readable storage medium, and a processor of a computer device reads the computer program or computer instructions from the computer-readable storage medium, and the processor executes the computer program or computer instructions, so that the computer device executes the method for generating user portraits of distribution network substations as described in any of the previous embodiments.

[0089] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0090] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein, for example. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, device, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. It should be understood that in the present application, "at least one (item)" refers to one or more, and "a plurality" refers to two or more.

[0091] In the several embodiments provided in the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0092] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0093] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0094] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store program codes.

[0095] Although the description of the present application has been quite detailed and specifically describes several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, but should be regarded as providing a broad possible interpretation of these claims by reference to the attached claims, taking into account the prior art, so as to effectively cover the intended scope of the present application. In addition, the above description of the present application is based on the embodiments foreseeable by the inventor, and its purpose is to provide a useful description, and those non-substantial changes to the present application that have not yet been foreseen may still represent equivalent changes to the present application.

Claims

1. A method for generating user portraits in a distribution network area, characterized in that: The method comprises: Acquire power data in the target area, and generate a power consumption behavior feature set based on the power data; Clustering the feature data of the power usage behavior feature set to obtain a plurality of first data sets; Merging the plurality of first data sets into a second data set, the second data set comprising a plurality of data layer nodes; Based on a target data layer node among the multiple data layer nodes, dividing the second data set into multiple data subsets; The user characteristics corresponding to each of the data subsets are determined and displayed to obtain a distribution network substation user profile including each of the user characteristics.

2. A method for generating user portraits in a distribution network area according to claim 1, characterized in that: The acquiring of power data in the target area and generating a power usage behavior feature set based on the power data includes: Dividing the power data into a plurality of characteristic parameters; Based on the degree of correlation between each of the characteristic parameters, the multiple characteristic parameters are screened and the power consumption behavior feature set is generated according to the screened characteristic parameters.

3. A method for generating user portraits of distribution network areas according to claim 2, characterized in that: Each of the characteristic parameters includes a plurality of parameter values; The method further comprises the steps of determining the degree of correlation; The determining of the degree of correlation comprises: Obtaining a sample value from a plurality of parameter values ​​in the first characteristic parameter, and determining an average value of the plurality of parameter values ​​in the first characteristic parameter; Obtaining a sample value from a plurality of parameter values ​​in the second characteristic parameter, and determining an average value of the plurality of parameter values ​​in the second characteristic parameter; Based on the correlation between the sample value and average value corresponding to the first characteristic parameter and the sample value and average value corresponding to the second characteristic parameter, the correlation value between the first characteristic parameter and the second characteristic parameter is determined to determine the degree of correlation between the first characteristic parameter and the second characteristic parameter.

4. A method for generating user portraits in a distribution network area according to claim 1, characterized in that: The method further comprises the step of determining the number of clusters when clustering the feature data of the power usage behavior feature set; The step of determining the number of clusters when clustering the feature data of the power usage behavior feature set includes: When clustering the feature data of the power usage behavior feature set, the power usage behavior feature set is divided into a plurality of data clusters; Determine a sum of squared errors in each of the data clusters, wherein the sum of squared errors is the sum of squared distances between each sample data point in the data cluster and a center point of the data cluster; Based on the number of the data clusters and each of the square error sum values, a relationship curve graph is generated, wherein the relationship curve graph includes a horizontal axis and a vertical axis, wherein the horizontal axis represents the number of the data clusters, and the vertical axis represents the square error sum value; The inflection point of the target curve is determined in the relationship curve diagram, and the abscissa value corresponding to the inflection point of the target curve is used as the optimal number of clusters.

5. A method for generating user portraits in a distribution network area according to claim 1, characterized in that: The step of merging the plurality of first data sets into a second data set, wherein the second data set includes a plurality of data layer nodes, including: Based on the distance between each data point in the first data set, a merging operation is performed on each data point until each data point of the first data set is merged into the same cluster to obtain the second data set, which includes multiple data layer nodes.

6. A method for generating user portraits in a distribution network area according to claim 5, characterized in that: The specific method of performing the merging operation on each of the data points is: Based on each link distance, a distance matrix is ​​generated, wherein the distance matrix includes a plurality of clusters, each cluster includes at least one data point, and the link distance is the distance between two clusters; Merge the two clusters with the smallest link distance in the distance matrix into a new cluster; Based on the distances between the new cluster and other clusters, the distance matrix is ​​updated; According to the updated distance matrix, the step of merging the two clusters with the smallest link distance in the distance matrix into a new cluster is returned until each of the data points is merged into the same cluster, thereby obtaining a hierarchical clustering tree structure, which is used to divide the data layer nodes into multiple clusters.

7. A method for generating user portraits in a distribution network area according to claim 1, characterized in that: The method further comprises: identifying abnormal data including abnormal values ​​in the power data; The abnormal data is deleted according to the abnormal value, and the power data after the abnormal data is deleted is used as the pre-processed power data.

8. A distribution network area user portrait generation device, characterized in that: The device comprises: An acquisition module, used to acquire power data in a target area and generate a power consumption behavior feature set based on the power data; A first clustering module, used for clustering the feature data of the power consumption behavior feature set to obtain a plurality of first data sets; A second clustering module, configured to merge the plurality of first data sets into a second data set, wherein the second data set includes a plurality of data layer nodes; A hierarchical partitioning module, configured to partition the second data set into a plurality of data subsets based on a target data layer node among the plurality of data layer nodes; A display module is used to determine the user characteristics corresponding to each of the data subsets and display them to obtain a distribution network substation user portrait including each of the user characteristics.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the method for generating user portraits of distribution network substations as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for generating user portraits of distribution network substations according to any one of claims 1 to 7 is implemented.

Images