Power distribution network node topological structure construction method and device and readable storage medium
By clustering and differential analysis of the voltage time series data of distribution network nodes, node topology structure is constructed, which solves the problems of inefficiency, safety hazards and node omissions in the existing methods, and realizes efficient and accurate node topology structure construction.
Patent Information
- Application Number
- CN202411528502.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-30
- Publication Date
- 2025-03-14
AI Technical Summary
The existing distribution network node topology structure construction method wastes human resources, is inefficient, has safety risks, and is prone to node omission problems, affecting the accuracy of node topology structure.
By obtaining the voltage time series data of all nodes in the distribution network, clustering and dividing node clusters, establishing node connection relationships based on the difference of voltage time series data, building node topology structures, and avoiding field investigations.
The efficiency and accuracy of node topology structure construction are improved, safety hazards and node omission problems are avoided, and the topology structure is directly constructed using the electrical quantity characteristics.
Smart Images
Figure CN119627854A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of distribution network node division, and in particular to a method and device for constructing a distribution network node topology structure, and a computer-readable storage medium. Background Art
[0002] The node topology of the distribution network is a virtual node topology that reflects the distribution and connection relationship of each node in the distribution network. By analyzing the power data of each node in the node topology, the current flow and power distribution in the distribution network can be obtained, thereby reflecting the operating status of the distribution network. In addition, the node topology can also provide a basis for the flow calculation and network reconstruction of the distribution network. Therefore, how to construct an accurate node topology is crucial to the stable operation of the distribution network.
[0003] The existing method for constructing a node topology structure usually requires staff to conduct on-site inspections of the location and connection relationships of each node in the distribution network, and then draw a node topology structure. However, due to the large size of the distribution network system and the complex distribution and connection relationships of each node, a large number of staff are required to spend a lot of time on field inspections, and some remote or difficult-to-reach areas may not be fully covered, resulting in the omission of some nodes, making the node topology structure incomplete. In addition, some nodes may be located at high altitudes, underground or in dangerous areas, and staff may also face electrical safety risks during on-site inspections.
[0004] In summary, the existing node topology construction method not only wastes human resources, is inefficient, and has potential safety hazards, but is also prone to node omissions, affecting the accuracy of the node topology. Summary of the invention
[0005] To this end, the technical problem to be solved by the present invention is to overcome the problem that the node topology structure construction method in the prior art wastes human resources, is inefficient, has safety hazards, and is prone to node omissions, affecting the accuracy of the node topology structure.
[0006] In order to solve the above technical problems, the present invention provides a method for constructing a distribution network node topology structure, comprising: Obtain voltage time series data of all nodes in the distribution network; The voltage time series data of k nodes are selected from all nodes as cluster centers, and the voltage time series data of all nodes are clustered to obtain k initial branch line node clusters; k ≥ 1; The voltage time series data of each node in each initial branch line node cluster are arranged from large to small to obtain the arrangement order of each node; each node is arranged and connected according to the arrangement order to obtain the branch node topology structure corresponding to the initial branch line node cluster; For a head-end node in any branch node topology structure, a node set whose voltage time series data is greater than the voltage time series data of the head-end node is obtained in other branch node topologies; a node whose voltage time series data has the smallest difference with the voltage time series data of the head-end node is selected from the node set as a target node, and the head-end node is connected to the target node; The head end node in each branch node topology structure is connected to its corresponding target node to obtain the node topology structure of the distribution network.
[0007] Preferably, the process of acquiring the voltage time series data of each node in the distribution network includes: At each preset time interval, the voltage time series data of the node is collected once based on a preset sampling frequency; based on the voltage time series data obtained by each sampling, a voltage time series data set of the node is obtained; The voltage time series data with missing values in the voltage time series data set are used as the voltage time series data to be supplemented; and the Euclidean distances between the voltage time series data to be supplemented and other voltage time series data are calculated respectively; Select M voltage time series data with the smallest Euclidean distance as target voltage time series data, and calculate the weight of each target voltage time series data, as well as the product between the weight and the mean of each target time series data; Based on the product between the weight and the mean of the M target time series data, the filling value of the missing value in the voltage time series data to be supplemented is calculated, and based on the filled voltage time series data to be supplemented and other voltage time series data, the voltage time series data of the node is obtained.
[0008] Preferably, the calculation formula of the Euclidean distance between the voltage time series data to be supplemented and other voltage time series data is: , in, represents the Euclidean distance between the voltage time series data to be supplemented at the nth node and the ith other voltage time series data; represents the time series data of the voltage to be supplemented at the nth node; represents the i-th other voltage time series data of the n-th node; express The data value of the oth sampling point in ; express The data value of the oth sampling point in ; Indicates the number of sampling points of the voltage time series data to be supplemented or other voltage time series data; , Indicates the number of nodes in the distribution network; The calculation formula for the weight of the target voltage time series data is: , in, represents the weight of the mth target voltage time series data; represents the Euclidean distance between the time series data of the voltage to be supplemented of the nth node and the time series data of the target voltage of the mth node; Indicates the number of target voltage time series data; The calculation formula for filling the missing values in the voltage time series data to be supplemented is: , in, represents the filling value of the missing value in the voltage time series data to be supplemented at the nth node; Represents the mean of the mth target voltage time series data.
[0009] Preferably, after obtaining the voltage time series data of all nodes in the distribution network, the method further includes: Based on the data values corresponding to the same sampling point in the voltage time series data of all nodes, the average value and standard deviation of the node voltage measurement data of the sampling point are calculated; Based on the average value and standard deviation of the node voltage measurement data at the sampling point, the data value of the sampling point in the voltage time series data of each node is normalized to obtain the normalized voltage value of each node at the sampling point; Based on the normalized voltage value of each node at each sampling point, the normalized voltage time series data of each node is obtained.
[0010] Preferably, the calculation formula for the average value of the node voltage measurement data at the t-th sampling point is: , in, represents the average value of the node voltage measurement data at the t-th sampling point; Indicates j The data value of the node at the tth sampling point; Indicates the number of nodes in the distribution network; The calculation formula for the standard deviation of the node voltage measurement data at the tth sampling point is: , in, represents the standard deviation of the node voltage measurement data at the tth sampling point; The normalized voltage value of each node at the tth sampling point is expressed as: , in, Represents the normalized voltage value of the j-th node at the t-th sampling point.
[0011] Preferably, after obtaining k initial branch line node clusters, the method further includes: Based on the class centers of the k initial branch line node clusters, k new cluster centers are obtained; Re-clustering the voltage time series data of all nodes based on the k new cluster centers to obtain k new branch line node clusters, and determining whether the k new branch line node clusters are the same as the k initial branch line node clusters; If the k new branch line node clusters are different from the k initial branch line node clusters, the class centers of the k new branch line node clusters are used as k new clustering centers, and the voltage time series data of all nodes are clustered again until the k branch line node clusters obtained by clustering are equal to the k new branch line node clusters.
[0012] Preferably, after obtaining the node topology structure of the distribution network, the method further includes: Calculating the Pearson correlation coefficient between each two connected nodes in the node topology structure respectively, and determining the magnitude of the Pearson correlation coefficient and a preset threshold; If the Pearson correlation coefficient between connected nodes is less than the preset threshold, k=k+1 is updated, and the voltage time series data of k nodes are reselected from all nodes as clustering centers, and the voltage time series data of all nodes are clustered to update the node topology structure until the Pearson correlation coefficient between any connected nodes is greater than or equal to the preset threshold, and the current node topology structure is used as the target node topology structure of the distribution network.
[0013] Preferably, the calculation formula of the Pearson correlation coefficient between connected nodes is: , in, represents the Pearson correlation coefficient between the connected nodes j and k; represents the voltage time series data of node j; represents the voltage time series data of node k; express The data value of the tth sampling point in ; express The mean of express The data value of the tth sampling point in ; express The mean of Represents the total number of sampling points of the voltage time series data of each node; , , Indicates the number of nodes in the distribution network.
[0014] The present invention also provides a device for constructing a distribution network node topology structure, comprising: A data acquisition module is used to obtain voltage time series data of all nodes in the distribution network; The clustering module is used to select the voltage time series data of k nodes from all nodes as cluster centers, cluster the voltage time series data of all nodes, and obtain k initial branch line node clusters; k≥1; The first node connection module is used to arrange the voltage time series data of each node in each initial branch line node cluster from large to small to obtain the arrangement order of each node; arrange and connect each node according to the arrangement order to obtain the branch node topology structure corresponding to the initial branch line node cluster; A second node connection module is used for obtaining, for a head-end node in any branch node topology structure, a node set whose voltage time series data is greater than the voltage time series data of the head-end node in other branch node topologies; selecting a node whose voltage time series data has the smallest difference with the voltage time series data of the head-end node from the node set as a target node, and connecting the head-end node to the target node; The node topology structure acquisition module is used to connect the head-end node in each branch node topology structure with its corresponding target node to obtain the node topology structure of the distribution network.
[0015] The present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned method for constructing a distribution network node topology structure are implemented.
[0016] The method for constructing the node topology structure of the distribution network provided in the present application obtains the voltage data of each node in the distribution network. Since the nodes on the same branch line have similar characteristics and electrical quantity data, the nodes are first divided into different clusters by clustering the voltage data of each node, and the nodes in each cluster are the nodes on the same branch line; since on the same branch line, the voltage amplitude of the node will decrease as the electrical distance from the head-end node increases, therefore, the voltage time series data of each node in each initial branch line node cluster is arranged from large to small to obtain the arrangement order of each node in the cluster, and each node is arranged and connected according to the arrangement order to obtain the branch node corresponding to the initial branch line node cluster. point topology structure; finally, for the head node of each branch node topology structure, obtain a set of nodes whose voltage time series data is greater than the voltage time series data of the head node in other branch node topologies, and select the node with the smallest difference in voltage time series data with the head node from the node set as the target node, connect the target node with the head node, and then complete the connection of all branch node topologies to obtain the node topology structure of the distribution network; the present application directly uses the electrical quantity characteristics of each node to construct the node topology structure, without the need for staff to conduct on-site inspections of the nodes in the distribution network, avoiding safety hazards and node omissions during the inspection process, and at the same time improving the efficiency and accuracy of building the node topology structure. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein: Figure 1 A schematic diagram of a single-phase line of a distribution network provided for this application; Figure 2 A flow chart of the method for constructing the distribution network node topology structure provided in this application; Figure 3 A schematic diagram of voltage time series data of an IEEE33-node distribution network provided for this application; Figure 4 for Figure 3 The schematic diagram of the voltage amplitude of each node in the node distribution network at a certain sampling point is shown; Figure 5 for Figure 3 The node topology diagram of the node distribution network shown; Figure 6 for Figure 5 The heat map of correlation coefficients between nodes in the node topology shown; Figure 7 A schematic diagram of the structure of the distribution network node topology structure construction device provided in this application. DETAILED DESCRIPTION
[0018] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.
[0019] Since the nodes on the same branch line in the distribution network have similar characteristics and electrical quantity data, the electrical quantity data of all nodes in the distribution network can be clustered to divide the node clusters on different branch lines.
[0020] Furthermore, due to the differences in bus voltage and comprehensive load characteristics in different distribution networks, after clustering is completed, in order to improve the node topology of the distribution network, it is also necessary to analyze the upstream and downstream relationships of each node in the feeder according to the magnitude of the node voltage fluctuation amplitude; Specifically, Figure 1 The figure shows a schematic diagram of a single-phase line of a distribution network provided by the present application. They represent the bus voltage of the distribution network and the voltage of node m respectively; They represent the resistance and reactance of branch m respectively; Respectively represent the active power and reactive power of node m; They represent the transmission active power and transmission reactive power of branch m respectively; According to the voltage drop formula, the voltage change of the connected nodes in the distribution network at time t can be obtained as: , in, It represents the voltage amplitude change between node m and the connected node m+1 in the distribution network at time t; represents the voltage of node m+1 at time t; represents the voltage of node m at time t; represents the resistance of the m+1th line; represents the transmission active power of the m+1th line at time t; Based on the above formula, the spatial characteristics of the node voltage can be summarized as follows: the voltage amplitude of a node on the line is related to the bus voltage of that phase, the line load and the electrical distance from the head-end node. That is, at the same time of a single power line, the node voltage amplitude gradually decreases as the electrical distance from the head-end node increases.
[0021] Therefore, based on the spatial characteristics of the node voltage, the nodes on the same branch line can be sorted and connected to obtain the branch node topology of each branch line. Then, based on the spatial characteristics of the node voltage, the node topology of each branch line can be connected to obtain the node topology of the distribution network.
[0022] See also Figure 2 , Figure 2 The flowchart of the method for constructing the distribution network node topology structure provided by the present application based on the above principle is shown, and the method specifically includes: S10: Obtain voltage time series data of all nodes in the distribution network; S20: selecting the voltage time series data of k nodes from all nodes as cluster centers, clustering the voltage time series data of all nodes, and obtaining k initial branch line node clusters; k ≥ 1; In some embodiments of the present application, when selecting the voltage time series data of k nodes as the cluster center, the value of k may be a preset value or may be determined by the elbow method; Specifically, the voltage time series data of all nodes are scanned first, the k value is determined by the elbow method, the voltage time series data of k nodes are randomly selected as cluster centers, and the distance between the voltage time series data of each node and the cluster center is calculated using the Euclidean distance formula, so that the Euclidean distance between the voltage time series data of each node and the cluster center of its cluster is minimized, thereby obtaining k initial branch line node clusters; S30: Arrange the voltage time series data of each node in each initial branch line node cluster from large to small to obtain an arrangement order of each node; arrange and connect each node according to the arrangement order to obtain a branch node topology structure corresponding to the initial branch line node cluster; Specifically, when sorting and connecting the nodes in each initial branch line node cluster, the voltage time series data of each node in the cluster are first arranged from large to small, each voltage time series data corresponds to a node, and each node is arranged and connected according to its corresponding voltage time series data, so as to obtain the branch node topology structure corresponding to the initial branch line node cluster; Optionally, when sorting the voltage time series data, the sorting may be performed based on the mean value of each voltage time series data, or the data values of specific sampling points in two voltage time series data may be compared to obtain the arrangement order of the voltage time series data; S40: For a head-end node in any branch node topology structure, obtain a node set whose voltage time series data is greater than the voltage time series data of the head-end node in other branch node topologies; select a node whose voltage time series data has the smallest difference with the voltage time series data of the head-end node from the node set as a target node, and connect the head-end node to the target node; Specifically, the head node in the branch node topology structure refers to the node with the largest voltage time series data in the branch node topology structure; For example, the mean value of the voltage time series data of the head-end node in the first branch node topology is The mean of the voltage time series data of the second node in the second branch node topology is , the mean of the voltage time series data of the nth node in the third branch node topology is , and Both greater than ,at the same time , then the second node in the second branch topology structure is used as the target node, and the head end node in the first branch node topology structure is connected to the target node; S50: Connecting the head-end node in each branch node topology structure to its corresponding target node to obtain a node topology structure of the distribution network.
[0023] This application directly uses the electrical quantity characteristics of each node to construct the node topology structure, without the need for staff to conduct on-site inspections of the nodes in the distribution network, avoiding safety hazards and node omissions during the inspection process, while improving the efficiency and accuracy of building the node topology structure.
[0024] Specifically, human operational errors, environmental factors such as electromagnetic interference, and excessive temperature may affect the normal operation of the data acquisition equipment during the data collection and transmission process. At the same time, communication failures between the data acquisition equipment and the data center may also lead to data transmission interruptions, resulting in data loss. Therefore, after obtaining the voltage time series data of each node, the missing values need to be filled.
[0025] Specifically, in some embodiments of the present application, the process of acquiring voltage time series data of each node in the distribution network includes: At each preset time interval, the voltage time series data of the node is collected once based on a preset sampling frequency; based on the voltage time series data obtained by each sampling, a voltage time series data set of the node is obtained; For example, the voltage time series data of each node may be collected once every 12 hours, so as to construct a voltage time series data set of the node based on the voltage time series data of each node every day. In a specific example of the present application, the voltage time series data set of a certain node is represented in matrix form: , in, represents the voltage time series data set of node n; Indicates the number of days of collection; represents the number of sampling points per day; each row in the matrix represents a voltage time series data of node n; in one embodiment, the preset sampling frequency is 15 minutes, that is, there are 96 data values in a voltage time series data of node n; The voltage time series data with missing values in the voltage time series data set are used as the voltage time series data to be supplemented; and the Euclidean distances between the voltage time series data to be supplemented and other voltage time series data are calculated respectively; Taking the above matrix as an example, calculating the Euclidean distance between the voltage time series data to be supplemented and other voltage time series data means calculating the Euclidean distance between the row containing missing values and other rows in the matrix; Specifically, the calculation formula for the Euclidean distance between the voltage time series data to be supplemented and other voltage time series data is: , in, represents the Euclidean distance between the voltage time series data to be supplemented at the nth node and the ith other voltage time series data; represents the time series data of the voltage to be supplemented at the nth node; represents the i-th other voltage time series data of the n-th node; express The data value of the oth sampling point in ; express The data value of the oth sampling point in ; Indicates the number of sampling points of the voltage time series data to be supplemented or other voltage time series data; , Indicates the number of nodes in the distribution network; Select M voltage time series data with the smallest Euclidean distance as target voltage time series data, and calculate the weight of each target voltage time series data, as well as the product between the weight and the mean of each target time series data; Specifically, the calculation formula for the weight of the target voltage time series data is: , in, represents the weight of the mth target voltage time series data; represents the Euclidean distance between the time series data of the voltage to be supplemented of the nth node and the time series data of the target voltage of the mth node; Indicates the number of target voltage time series data; Based on the product between the weight and the mean of the M target time series data, the filling value of the missing value in the voltage time series data to be supplemented is calculated, and based on the filled voltage time series data to be supplemented and other voltage time series data, the voltage time series data of the node is obtained; Specifically, the calculation formula for filling the missing values in the voltage time series data to be supplemented is: , in, represents the filling value of the missing value in the voltage time series data to be supplemented at the nth node; Represents the mean of the mth target voltage time series data.
[0026] Compared with the common mean and standard deviation filling method, the data filling method used in the embodiment of the present application focuses on the correlation and similarity between data, can fill in according to the actual change trend of relevant data, retains the change law of data, and has better robustness and interpretability.
[0027] In an active power distribution system, the closer the electrical distance between loads, the smaller the line loss, and the smaller the voltage difference between adjacent nodes. If the node topology of the distribution network is directly established using unprocessed measurement data, the accuracy of the constructed node topology is low. Therefore, in some embodiments of the present application, after obtaining the voltage time series data of each node, the data is also normalized and scaled to a similar range, which is conducive to comparison and analysis between different features.
[0028] Specifically, after step S10, the method further includes: Based on the data values corresponding to the same sampling point in the voltage time series data of all nodes, the average value and standard deviation of the node voltage measurement data of the sampling point are calculated; For example, the voltage data of all nodes at the same sampling point are selected, and the Z-Score normalization method is used to calculate the average value of the node voltage measurement data and the standard deviation of the node voltage measurement data at the sampling point, so as to normalize the voltage data of all nodes at the sampling point; in a specific example of the present application, the voltage data of all nodes at the same sampling point are organized into a matrix form: , in, Indicates the number of nodes in the distribution network; Represents the data value of the Nth node at the tth sampling point; Specifically, the calculation formula for the average value of the node voltage measurement data at the t-th sampling point is: , in, represents the average value of the node voltage measurement data at the t-th sampling point; Indicates The data value of the node at the tth sampling point; Indicates the number of nodes in the distribution network; The calculation formula for the standard deviation of the node voltage measurement data at the tth sampling point is: , in, represents the standard deviation of the node voltage measurement data at the tth sampling point; Based on the average value and standard deviation of the node voltage measurement data at the sampling point, the data value of the sampling point in the voltage time series data of each node is normalized to obtain the normalized voltage value of each node at the sampling point; Specifically, the normalized voltage value of each node at the tth sampling point is expressed as: , in, represents the normalized voltage value of the jth node at the tth sampling point; Based on the normalized voltage value of each node at each sampling point, the normalized voltage time series data of each node is obtained.
[0029] Optionally, in order to make the clustering result more accurate, thereby ensuring that the nodes in each initial branch line node cluster have similar characteristics and electrical quantity data, after obtaining k initial branch line node clusters, the following steps are further included: Based on the class centers of the k initial branch line node clusters, k new cluster centers are obtained; Specifically, the cluster center of each initial branch line node cluster is the mean of the voltage time series data of all nodes in the cluster; Based on k new cluster centers, the voltage time series data of all nodes are re-clustered to obtain k new branch line node clusters, and it is determined whether the k new branch line node clusters are the same as the k initial branch line node clusters; If the k new branch line node clusters are different from the k initial branch line node clusters, the class centers of the k new branch line node clusters are used as k new clustering centers, and the voltage time series data of all nodes are clustered again until the k branch line node clusters obtained by clustering are equal to the k new branch line node clusters.
[0030] Furthermore, after obtaining the node topology of the distribution network, it is also necessary to verify the accuracy of the node topology. When performing the node voltage curve morphology analysis, the Pearson correlation coefficient is a statistical tool used to measure the strength and direction of the linear relationship between two variables. It provides a value between -1 and 1 to indicate the strength of the correlation between the variables. In this application, the Pearson correlation coefficient is used to identify node pairs with similar voltage curves in shape but different amplitudes, thereby verifying the above node topology.
[0031] Specifically, after obtaining the node topology structure of the distribution network, the following steps are also included: S60: respectively calculating the Pearson correlation coefficient between every two connected nodes in the node topology structure, and determining the magnitude of the Pearson correlation coefficient and a preset threshold; S70: If the Pearson correlation coefficient between any connected nodes is less than the preset threshold, update k=k+1 and return to execute step S20 to update the node topology structure until the Pearson correlation coefficient between any connected nodes is greater than or equal to the preset threshold, and use the current node topology structure as the target node topology structure of the distribution network.
[0032] The calculation formula of the Pearson correlation coefficient between connected nodes is: , in, represents the Pearson correlation coefficient between the connected nodes j and k; represents the voltage time series data of node j; represents the voltage time series data of node k; express The data value of the tth sampling point in ; express The mean of express The data value of the tth sampling point in ; express The mean of Represents the total number of sampling points of the voltage time series data of each node; , , Indicates the number of nodes in the distribution network.
[0033] Specifically, the larger the Pearson correlation coefficient between the connected nodes, the stronger the linear relationship between the samples, and the smaller the Pearson correlation coefficient between the connected nodes, the weaker the linear relationship between the surface samples. In the embodiment of the present application, a preset threshold is set to identify whether the connection relationship between the nodes is correct. If the difference in the direction of the voltage curve shape of the two connected nodes (i.e., the Pearson correlation coefficient) is greater than or equal to the preset threshold, it is considered that the two nodes are similar in voltage behavior and can be regarded as directly connected. If the Pearson correlation coefficient of the two connected nodes is less than the preset threshold, it is considered that the two nodes are not directly connected.
[0034] In order to verify the effectiveness of the method for constructing the node topology structure of the distribution network provided in this application, this embodiment also takes the IEEE 33-node distribution network as an example and uses the above method to construct its node topology structure: First, the smart meter data of each node in the IEEE 33-node distribution network is obtained, and the missing value filling and normalization processing are performed on it to obtain the voltage time series data of the 33 nodes after processing, such as Figure 3 shown.
[0035] The optimal number of clusters k is determined to be 6. The K-means clustering algorithm based on the similarity measurement between samples divides the IEEE33 nodes into 6 categories. Table 1 shows the specific branch nodes in each cluster after clustering: Table 1
[0036] like Figure 4 The voltage amplitude of each node at a certain sampling point is shown in Figure 2. The voltage amplitude is used to determine the upstream and downstream positions and connection relationships of each node, and the following is obtained: Figure 5 The node topology is shown.
[0037] Finally, the correlation between each two connected nodes is calculated according to the Pearson correlation coefficient, and a preset threshold P is set. Figure 5 The node topology shown is verified, as Figure 6 The figure shows a heat map of correlation coefficients obtained by calculating the Pearson correlation coefficients between connected nodes in this embodiment, wherein the size of the correlation coefficients of connected nodes is reflected by the height. The larger the correlation coefficient, the greater the possibility that the two nodes are connected. If the correlation coefficient of two directly connected nodes is greater than the set threshold, it means that their connection relationship is correct.
[0038] If the accuracy of the topological structure connection relationship is low, return to the clustering step to change the k value and cluster again.
[0039] The threshold required for each node is generally set by selecting the larger value among all the correlation coefficients of the node. Based on the method for constructing a distribution network node topology structure provided in the above embodiment, the present application embodiment also provides a device for constructing a distribution network node topology structure, such as Figure 7 As shown, the device specifically includes: The data acquisition module 10 is used to acquire the voltage time series data of all nodes in the distribution network; The clustering module 20 is used to select the voltage time series data of k nodes from all nodes as cluster centers, cluster the voltage time series data of all nodes, and obtain k initial branch line node clusters; k≥1; The first node connection module 30 is used to arrange the voltage time series data of each node in each initial branch line node cluster from large to small to obtain the arrangement order of each node; arrange and connect each node according to the arrangement order to obtain the branch node topology structure corresponding to the initial branch line node cluster; The second node connection module 40 is used for obtaining, for a head-end node in any branch node topology structure, a node set whose voltage time series data is greater than the voltage time series data of the head-end node in other branch node topologies; selecting a node whose voltage time series data has the smallest difference with the voltage time series data of the head-end node as a target node in the node set, and connecting the head-end node to the target node; The node topology structure acquisition module 50 is used to connect the head-end node in each branch node topology structure with its corresponding target node to obtain the node topology structure of the distribution network.
[0040] An embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for constructing a distribution network node topology structure are implemented.
[0041] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.
[0042] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0043] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0044] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0045] Obviously, the above embodiments are merely examples for the purpose of clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the present invention.
Claims
1. A method for constructing a distribution network node topology structure, characterized in that: include: Obtain voltage time series data of all nodes in the distribution network; The voltage time series data of k nodes are selected from all nodes as cluster centers, and the voltage time series data of all nodes are clustered to obtain k initial branch line node clusters; k ≥ 1; The voltage time series data of each node in each initial branch line node cluster are arranged from large to small to obtain the arrangement order of each node; each node is arranged and connected according to the arrangement order to obtain the branch node topology structure corresponding to the initial branch line node cluster; For a head-end node in any branch node topology structure, a node set whose voltage time series data is greater than the voltage time series data of the head-end node is obtained in other branch node topologies; a node whose voltage time series data has the smallest difference with the voltage time series data of the head-end node is selected from the node set as a target node, and the head-end node is connected to the target node; The head end node in each branch node topology structure is connected to its corresponding target node to obtain the node topology structure of the distribution network.
2. The method for constructing a distribution network node topology structure according to claim 1, characterized in that: The process of acquiring voltage time series data of each node in the distribution network includes: At each preset time interval, the voltage time series data of the node is collected once based on a preset sampling frequency; based on the voltage time series data obtained by each sampling, a voltage time series data set of the node is obtained; The voltage time series data with missing values in the voltage time series data set are used as the voltage time series data to be supplemented; and the Euclidean distances between the voltage time series data to be supplemented and other voltage time series data are calculated respectively; Select M voltage time series data with the smallest Euclidean distance as target voltage time series data, and calculate the weight of each target voltage time series data, as well as the product between the weight and the mean of each target time series data; Based on the product between the weight and the mean of the M target time series data, the filling value of the missing value in the voltage time series data to be supplemented is calculated, and based on the filled voltage time series data to be supplemented and other voltage time series data, the voltage time series data of the node is obtained.
3. The method for constructing a distribution network node topology structure according to claim 2, characterized in that: The calculation formula of the Euclidean distance between the voltage time series data to be supplemented and other voltage time series data is: ， in, represents the Euclidean distance between the voltage time series data to be supplemented at the nth node and the ith other voltage time series data; represents the time series data of the voltage to be supplemented at the nth node; represents the i-th other voltage time series data of the n-th node; express The data value of the oth sampling point in ; express The data value of the oth sampling point in ; Indicates the number of sampling points of the voltage time series data to be supplemented or other voltage time series data; , Indicates the number of nodes in the distribution network; The calculation formula for the weight of the target voltage time series data is: ， in, represents the weight of the mth target voltage time series data; represents the Euclidean distance between the time series data of the voltage to be supplemented of the nth node and the time series data of the target voltage of the mth node; Indicates the number of target voltage time series data; The calculation formula for filling the missing values in the voltage time series data to be supplemented is: ， in, represents the filling value of the missing value in the voltage time series data to be supplemented at the nth node; Represents the mean of the mth target voltage time series data.
4. The method for constructing a distribution network node topology structure according to claim 1, characterized in that: After obtaining the voltage time series data of all nodes in the distribution network, it also includes: Based on the data values corresponding to the same sampling point in the voltage time series data of all nodes, the average value and standard deviation of the node voltage measurement data of the sampling point are calculated; Based on the average value and standard deviation of the node voltage measurement data at the sampling point, the data value of the sampling point in the voltage time series data of each node is normalized to obtain the normalized voltage value of each node at the sampling point; Based on the normalized voltage value of each node at each sampling point, the normalized voltage time series data of each node is obtained.
5. The method for constructing a distribution network node topology structure according to claim 4, characterized in that: The calculation formula for the average value of the node voltage measurement data at the tth sampling point is: ， in, represents the average value of the node voltage measurement data at the t-th sampling point; Indicates j The data value of the node at the tth sampling point; Indicates the number of nodes in the distribution network; The calculation formula for the standard deviation of the node voltage measurement data at the tth sampling point is: ， in, represents the standard deviation of the node voltage measurement data at the tth sampling point; The normalized voltage value of each node at the tth sampling point is expressed as: ， in, Represents the normalized voltage value of the j-th node at the t-th sampling point.
6. The method for constructing a distribution network node topology structure according to claim 1, characterized in that: After obtaining k initial branch line node clusters, the following steps are also included: Based on the class centers of the k initial branch line node clusters, k new cluster centers are obtained; Re-clustering the voltage time series data of all nodes based on the k new cluster centers to obtain k new branch line node clusters, and determining whether the k new branch line node clusters are the same as the k initial branch line node clusters; If the k new branch line node clusters are different from the k initial branch line node clusters, the class centers of the k new branch line node clusters are used as k new clustering centers, and the voltage time series data of all nodes are clustered again until the k branch line node clusters obtained by clustering are equal to the k new branch line node clusters.
7. The method for constructing a distribution network node topology structure according to claim 1, characterized in that: After obtaining the node topology of the distribution network, it also includes: Calculating the Pearson correlation coefficient between each two connected nodes in the node topology structure respectively, and determining the magnitude of the Pearson correlation coefficient and a preset threshold; If the Pearson correlation coefficient between connected nodes is less than the preset threshold, k=k+1 is updated, and the voltage time series data of k nodes are reselected from all nodes as clustering centers, and the voltage time series data of all nodes are clustered to update the node topology structure until the Pearson correlation coefficient between any connected nodes is greater than or equal to the preset threshold, and the current node topology structure is used as the target node topology structure of the distribution network.
8. The method for constructing a distribution network node topology structure according to claim 7, characterized in that: The calculation formula of the Pearson correlation coefficient between connected nodes is: ， in, represents the Pearson correlation coefficient between the connected nodes j and k; represents the voltage time series data of node j; represents the voltage time series data of node k; express The data value of the tth sampling point in ; express The mean of express The data value of the tth sampling point in ; express The mean of Represents the total number of sampling points of the voltage time series data of each node; , , Indicates the number of nodes in the distribution network.
9. A device for constructing a distribution network node topology structure, characterized in that: include: A data acquisition module is used to obtain voltage time series data of all nodes in the distribution network; The clustering module is used to select the voltage time series data of k nodes from all nodes as cluster centers, cluster the voltage time series data of all nodes, and obtain k initial branch line node clusters; k≥1; The first node connection module is used to arrange the voltage time series data of each node in each initial branch line node cluster from large to small to obtain the arrangement order of each node; Arrange and connect each node in the order of arrangement to obtain a branch node topology structure corresponding to the initial branch line node cluster; A second node connection module is used for obtaining, for a head-end node in any branch node topology structure, a node set whose voltage time series data is greater than the voltage time series data of the head-end node in other branch node topologies; selecting a node whose voltage time series data has the smallest difference with the voltage time series data of the head-end node from the node set as a target node, and connecting the head-end node to the target node; The node topology structure acquisition module is used to connect the head-end node in each branch node topology structure with its corresponding target node to obtain the node topology structure of the distribution network.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for constructing a distribution network node topology structure according to any one of claims 1 to 8 are implemented.