Power distribution network medium and low voltage topology model generation method based on spectral clustering and related device

Abstract

The invention belongs to the technical field of electric power information, and discloses a spectral clustering-based power distribution network medium-low voltage topology model generation method and related device, and the method comprises the steps: taking a transformer and a low-voltage meter box as nodes, and taking switch equipment as edges to construct a low-voltage topological graph; determining a connection weight coefficient of edges between the nodes according to the type and number of switches between the nodes, constructing an adjacent weight coefficient matrix of the low-voltage topological graph according to the connection weight coefficient, and constructing a degree matrix of the low-voltage topological graph according to the number of the edges connected with the nodes; and according to the adjacency weight coefficient matrix and the degree matrix of the low-voltage topological graph, clustering each node of the low-voltage topological graph by adopting a spectral clustering method to obtain a clustering result, and based on the low-voltage topological graph, dividing the nodes with the same category identifier into the same sub-graph to obtain a plurality of low-voltage topological sub-graphs and taking the low-voltage topological sub-graphs as low-voltage topological models. Not only is effective segmentation and simplification of the low-voltage topology in the power distribution network realized, but also powerful technical support is provided for intelligent management, rapid fault positioning and resource optimization configuration of the power grid.

Description

Technical Field

[0001] The present invention belongs to the field of electric power information technology, and relates to a method for generating a low-voltage topology model in a distribution network based on spectral clustering and a related device. Background Art

[0002] The low-voltage topology in the distribution network refers to the topological relationship between the low-voltage side outgoing line of the distribution transformer, the low-voltage distribution cabinet, and then the various user access points. This topological relationship includes the electrical connection relationship of various low-voltage equipment, the geographical location information of the equipment, and the geographical path information of the line. The low-voltage topology is of great significance to the stable operation and troubleshooting of the power system. Its design needs to consider multiple factors such as electrical safety, power quality, and power loss, and be as simple, reliable, economical, and safe as possible.

[0003] The low-voltage topology in the distribution network contains a large number of types and devices, which leads to a complex and large-scale low-voltage topology structure, making it difficult to carry out full-volume topology analysis and calculation. According to research, the low-voltage topology model of the distribution network composed of millions of devices has a result return time of hours in scenarios such as version switching and path search. This performance cannot meet the needs of topology query and analysis. In addition, due to the diverse connection methods between distribution network devices, the frequency of real-time changes in connection methods is fast, and the density of connections and interlacing between lines is extremely high, resulting in a huge performance cost in describing and storing the full topology. Therefore, how to reduce the complexity of topological connections, and thus reduce the complexity of the entire topological model, has become an urgent problem to be solved. Summary of the invention

[0004] The purpose of the present invention is to overcome the above-mentioned shortcomings of the prior art and to provide a method for generating a low voltage topology model in a distribution network based on spectral clustering and a related device.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present invention provides a method for generating a low-voltage topology model in a distribution network based on spectral clustering, comprising: taking transformers and low-voltage meter boxes as nodes, taking switch-type devices as edges, and determining the correspondence between nodes and edges in combination with device terminal information, and constructing a low-voltage topology graph according to the correspondence; determining the connection weight coefficients of the edges between nodes according to the types and numbers of switches between nodes, and constructing an adjacency weight coefficient matrix of the low-voltage topology graph according to the connection weight coefficients of the edges between nodes; and constructing a degree matrix of the low-voltage topology graph according to the number of edges connected to the nodes; clustering each node of the low-voltage topology graph using a spectral clustering method according to the adjacency weight coefficient matrix and the degree matrix of the low-voltage topology graph to obtain a clustering result, and determining a category identifier of each node according to the clustering result; based on the low-voltage topology graph, dividing nodes with the same category identifier into the same subgraph to obtain a plurality of low-voltage topology subgraphs as a low-voltage topology model.

[0007] Optionally, the transformer and the low-voltage meter box are used as nodes, the switch-type equipment is used as edges, and the correspondence between the nodes and the edges is determined in combination with the equipment terminal information. When constructing a low-voltage topology diagram according to the correspondence, several edges between the two nodes are merged, and attributes describing the basic information of multiple edges are added to the merged edges.

[0008] Optionally, determining the connection weight coefficient of the edge between nodes according to the switch type and number between the nodes includes: obtaining the longest path in the low-voltage topology graph as the target path; and obtaining a first preset weight coefficient p and a second preset weight coefficient q according to the following formula:

[0009]

[0010] Wherein, a is the number of non-connecting switches on the target path, and b is the number of connecting switches on the target path.

[0011] The connection weight coefficient between nodes is obtained by the following formula:

[0012] V i,j =A*p+B*q

[0013] Among them, V i,j is the connection weight coefficient of the edge between node i and node j, A is the number of non-connecting switches on the edge between node i and node j, and B is the number of connecting switches on the edge between node i and node j.

[0014] Optionally, obtaining the longest path in the low-voltage topology map includes: traversing the low-voltage topology map using a depth traversal algorithm to obtain the longest path in the low-voltage topology map.

[0015] Optionally, the adjacency weight coefficient matrix V of the low voltage topology graph is:

[0016]

[0017] Among them, V i,j is the connection weight coefficient of the edge between node i and node j, and n is the total number of nodes.

[0018] The degree matrix D of the low voltage topology graph is:

[0019]

[0020] Among them, D i is the number of edges connected to node i.

[0021] Optionally, the method of clustering the nodes of the low-voltage topology graph using a spectral clustering method according to the adjacency weight coefficient matrix and degree matrix of the low-voltage topology graph to obtain a clustering result includes: obtaining a Laplace matrix L=DV according to the adjacency weight coefficient matrix V and the degree matrix D of the low-voltage topology graph, and selecting the first k eigenvectors of the Laplace matrix L in ascending order of eigenvalues ​​to obtain an eigenvector matrix Y; clustering the nodes of the low-voltage topology graph according to the eigenvector matrix Y and using a spectral clustering method to obtain a clustering result; wherein, when clustering the nodes of the low-voltage topology graph using the spectral clustering method, a K-means algorithm is used to cluster the eigenvectors.

[0022] In a second aspect, the present invention provides a system for generating a low-voltage topology model in a distribution network based on spectral clustering, comprising: a topology map construction module, which is used to use transformers and low-voltage meter boxes as nodes, switch-type devices as edges, and determine the correspondence between nodes and edges in combination with device terminal information, and construct a low-voltage topology map according to the correspondence; a matrix construction module, which is used to determine the connection weight coefficients of the edges between nodes according to the types and numbers of switches between nodes, and construct an adjacency weight coefficient matrix of the low-voltage topology map according to the connection weight coefficients of the edges between nodes; and construct a degree matrix of the low-voltage topology map according to the number of edges connected to the nodes; a spectral clustering module, which is used to cluster each node of the low-voltage topology map using a spectral clustering method according to the adjacency weight coefficient matrix and degree matrix of the low-voltage topology map to obtain a clustering result, and determine the category identifier of each node according to the clustering result; a model generation module, which is used to divide nodes with the same category identifier into the same subgraph based on the low-voltage topology map, obtain several low-voltage topology subgraphs and use them as a low-voltage topology model.

[0023] Optionally, the transformer and the low-voltage meter box are used as nodes, the switch-type equipment is used as edges, and the correspondence between the nodes and the edges is determined in combination with the equipment terminal information. When constructing a low-voltage topology diagram according to the correspondence, several edges between the two nodes are merged, and attributes describing the basic information of multiple edges are added to the merged edges.

[0024] Optionally, determining the connection weight coefficient of the edge between nodes according to the switch type and number between the nodes includes: obtaining the longest path in the low-voltage topology graph as the target path; and obtaining a first preset weight coefficient p and a second preset weight coefficient q according to the following formula:

[0025]

[0026] Wherein, a is the number of non-connecting switches on the target path, and b is the number of connecting switches on the target path.

[0027] The connection weight coefficient between nodes is obtained by the following formula:

[0028] V i,j =A*p+B*q

[0029] Among them, V i,j is the connection weight coefficient of the edge between node i and node j, A is the number of non-connecting switches on the edge between node i and node j, and B is the number of connecting switches on the edge between node i and node j.

[0030] Optionally, the transformer and the low-voltage meter box are used as nodes, the switch-type equipment is used as edges, and the correspondence between the nodes and the edges is determined in combination with the equipment terminal information. When constructing a low-voltage topology diagram according to the correspondence, several edges between the two nodes are merged, and attributes describing the basic information of multiple edges are added to the merged edges.

[0031] Optionally, obtaining the longest path in the low-voltage topology map includes: traversing the low-voltage topology map using a depth traversal algorithm to obtain the longest path in the low-voltage topology map.

[0032] Optionally, the adjacency weight coefficient matrix V of the low voltage topology graph is:

[0033]

[0034] Among them, V i,j is the connection weight coefficient of the edge between node i and node j, and n is the total number of nodes.

[0035] The degree matrix D of the low voltage topology graph is:

[0036]

[0037] Among them, D i is the number of edges connected to node i.

[0038] Optionally, the method of clustering the nodes of the low-voltage topology graph using a spectral clustering method according to the adjacency weight coefficient matrix and degree matrix of the low-voltage topology graph to obtain a clustering result includes: obtaining a Laplace matrix L=DV according to the adjacency weight coefficient matrix V and the degree matrix D of the low-voltage topology graph, and selecting the first k eigenvectors of the Laplace matrix L in ascending order of eigenvalues ​​to obtain an eigenvector matrix Y; clustering the nodes of the low-voltage topology graph according to the eigenvector matrix Y and using a spectral clustering method to obtain a clustering result; wherein, when clustering the nodes of the low-voltage topology graph using the spectral clustering method, a K-means algorithm is used to cluster the eigenvectors.

[0039] In a third aspect of the present invention, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for generating a low-voltage topology model in a distribution network based on spectral clustering are implemented.

[0040] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, wherein 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 generating a low-voltage topology model in a distribution network based on spectral clustering are implemented.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The present invention is based on the spectrum clustering distribution network medium and low voltage topology model generation method, with transformers and low voltage meter boxes as nodes, switch type equipment as edges, and combined with the equipment terminal information to determine the corresponding relationship between nodes and edges, and construct a low voltage topology graph according to the corresponding relationship, which can effectively avoid the problem that the model is too large and difficult to analyze and calculate due to the complex low voltage topology of the distribution network and the existing modeling method directly using the equipment as a node to construct a graph structure model. At the same time, the complex low voltage topology of the distribution network is effectively divided into multiple easy-to-manage subgraphs, thereby further reducing the complexity of the topological structure, providing a solid support for subsequent power grid analysis, operation and optimization. In addition, by introducing the type and number of switches as the basis for calculating the connection weight coefficient between nodes, an innovative node similarity evaluation method that is closer to the actual electrical connection characteristics is constructed. This method not only takes into account the physical connection between devices, but also more deeply reflects the relevance and closeness of the devices at the electrical level, thereby overcoming the limitation of the traditional graph partitioning method that only relies on node distance or simple connectivity to evaluate node closeness. At the same time, the nodes are clustered in combination with the spectral clustering algorithm, and the nodes are divided into different categories according to the clustering results, thereby generating multiple low-voltage topology subgraphs. This subgraph segmentation method based on electrical connection characteristics not only makes the connections between nodes within each subgraph closer, but also facilitates the regional and hierarchical management and analysis of the distribution network. Compared with the traditional spectral clustering algorithm based on node distance, the method of the present invention is more in line with the actual connection between medium and low voltage equipment, and is more in line with the actual needs of electrical business. It can be seen that the method for generating a medium and low voltage topology model of a distribution network based on spectral clustering not only realizes the effective segmentation and simplification of the medium and low voltage topology of the distribution network, but also provides strong technical support for the intelligent management of the power grid, rapid fault location and optimal resource allocation, and has significant practicality and broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a flow chart of a method for generating a low-voltage topology model in a distribution network based on spectral clustering according to an embodiment of the present invention.

[0044] Figure 2 This is a topological diagram of the connection between devices in an embodiment of the present invention.

[0045] Figure 3 This is a multiple edge merge graph where multiple edges exist between two nodes in an embodiment of the present invention.

[0046] Figure 4 It is a longest path search graph between any two points in the embodiment of the present invention.

[0047] Figure 5 The present invention provides a block diagram of a system structure for generating a low voltage topology model in a distribution network based on spectral clustering according to an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0049] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention 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 interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0050] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0051] See also Figure 1 In one embodiment of the present invention, a method for generating a low-voltage topology model in a distribution network based on spectral clustering is provided, which can effectively simplify the low-voltage topology model in the distribution network, reduce the difficulty of generating and storing the low-voltage topology model, and thus provide a basic data base support for distribution network analysis applications based on the low-voltage topology model.

[0052] Specifically, the method for generating a low voltage topology model in a distribution network based on spectral clustering of the present invention comprises the following steps:

[0053] S1: Take transformers and low-voltage meter boxes as nodes, switch devices as edges, and determine the correspondence between nodes and edges in combination with device terminal information, and construct a low-voltage topology diagram based on the correspondence.

[0054] S2: Determine the connection weight coefficient of the edge between nodes according to the switch type and number between the nodes, and construct the adjacency weight coefficient matrix of the low-voltage topology graph according to the connection weight coefficient of the edge between the nodes; and construct the degree matrix of the low-voltage topology graph according to the number of edges connected to the nodes.

[0055] S3: According to the adjacency weight coefficient matrix and degree matrix of the low-voltage topology graph, a spectral clustering method is used to cluster the nodes of the low-voltage topology graph to obtain clustering results, and the category identification of each node is determined according to the clustering results.

[0056] S4: Based on the low-voltage topology graph, nodes with the same category identifier are divided into the same subgraph to obtain several low-voltage topology subgraphs as the low-voltage topology model.

[0057] The invention discloses a method for generating a low-voltage topology model in a distribution network based on spectral clustering. The method uses transformers and low-voltage meter boxes as nodes, switch-type devices as edges, and combines device terminal information to determine the correspondence between nodes and edges. A low-voltage topology graph is constructed based on the correspondence. This can effectively avoid the problem that the model is too large and difficult to analyze and calculate because the low-voltage topology in the distribution network is relatively complex, and the existing modeling method directly uses devices as nodes to construct a graph structure model. At the same time, the complex low-voltage topology in the distribution network is effectively divided into multiple easy-to-manage subgraphs, thereby further reducing the complexity of the topological structure and providing solid support for subsequent power grid analysis, operation and maintenance, and optimization.

[0058] In addition, by introducing the type and number of switches as the basis for calculating the weight coefficient of the connection between nodes, an innovative node similarity evaluation method that is closer to the actual electrical connection characteristics is constructed. This method not only takes into account the physical connection between devices, but also more deeply reflects the relevance and closeness of the devices at the electrical level, thereby overcoming the limitations of the traditional graph partitioning method that only relies on node distance or simple connectivity to evaluate node closeness. At the same time, the nodes are clustered in combination with the spectral clustering algorithm, and the nodes are divided into different categories based on the clustering results, thereby generating multiple low-voltage topology subgraphs. This subgraph segmentation method based on electrical connection characteristics not only makes the connection between nodes within each subgraph closer, but also facilitates the regional and hierarchical management and analysis of the distribution network. Compared with the traditional spectral clustering algorithm based on node distance, the method of the present invention is more in line with the actual connection between medium and low voltage equipment, and is more in line with the actual needs of electrical business.

[0059] It can be seen that the method for generating low-voltage topology models in distribution networks based on spectral clustering in the present invention not only realizes the effective segmentation and simplification of low-voltage topologies in distribution networks, but also provides strong technical support for intelligent management of power grids, rapid fault location and optimal resource allocation. It has significant practicality and broad application prospects.

[0060] In a possible implementation, when constructing a low-voltage topology graph, attribute information of nodes and attribute information of edges in the low-voltage topology graph are both determined based on low-voltage topology data of the distribution network.

[0061] Among them, the low-voltage topology data of the distribution network includes two types: basic equipment file information and measurement data. For example, the basic equipment file information may include equipment number, equipment name, equipment type, equipment subtype, voltage level, equipment terminal information, and longitude and latitude coordinates. The measurement data varies according to different equipment types. For example, the measurement data of transformer equipment includes the current and voltage collection information of the transformer; the measurement data of switch equipment includes the operating status collection information of the switch; the measurement data of low-voltage meter box equipment includes the current, voltage and power collection information of the low-voltage meter box.

[0062] When constructing a low-voltage topology graph, initialize the graph model G = [V, E, p], where V is a node in the graph, E is an edge in the graph, and p is a node / edge attribute. The low-voltage topology in the distribution network covers many types of equipment. In order to carry out the mapping of the topology-graph model, the mapping rules of equipment types and nodes and edges are defined, that is, transformers and low-voltage meter boxes are mapped as nodes in the graph, and switch-type equipment is mapped as edges in the graph. Exemplarily, switch-type equipment includes span sections, distribution circuit breakers, disconnectors, load switches, fuses, pole-mounted circuit breakers, disconnectors, load switches, and pole-mounted fuses.

[0063] In a possible implementation, the transformer and the low-voltage meter box are used as nodes, the switch-type equipment is used as edges, and the correspondence between the nodes and the edges is determined in combination with the equipment terminal information. When a low-voltage topology diagram is constructed according to the correspondence, several edges between the two nodes are merged, and attributes describing the basic information of the multiple edges are added to the merged edges.

[0064] Specifically, the correspondence between nodes and edges is determined based on the device terminal information. Figure 2 , device 1 is a transformer, and its device terminal information is [1929873001, 7747002345]. Device 2 is also a transformer, and its device terminal information is [4255632620, 7747002345]. Device 1 and device 2 have the same terminal number 7747002345, so it is considered that the two devices are physically connected. Therefore, in the low-voltage topology diagram, the node corresponding to device 1 and the edge corresponding to device 2 should be directly connected.

[0065] See also Figure 3 In the actual line topology, there may be more than one switch device between a single transformer or low-voltage meter box. According to the above definition, each switch device is mapped to an edge, so there will be multiple edges between two nodes, which is redundant. In this case, these redundant edges are merged, and attributes describing the basic information of multiple edges are added to the merged edges.

[0066] Exemplarily, a new attribute length is created for each edge, and the value of length is the number of switch-type devices in the edge. If the edge contains multiple switch-type devices, then length>1, otherwise it is 1.

[0067] In a possible implementation, determining the connection weight coefficient of the edge between nodes according to the switch type and number between the nodes includes: obtaining the longest path in the low voltage topology map as the target path; and obtaining a first preset weight coefficient p and a second preset weight coefficient q according to the following formula:

[0068]

[0069] Wherein, a is the number of non-connecting switches on the target path, and b is the number of connecting switches on the target path.

[0070] The connection weight coefficient between nodes is obtained by the following formula:

[0071] V i,j =A*p+B*q

[0072] Among them, V i,j is the connection weight coefficient of the edge between node i and node j, A is the number of non-connecting switches on the edge between node i and node j, and B is the number of connecting switches on the edge between node i and node j.

[0073] Explanatory, in this embodiment, a first preset weight coefficient p and a second preset weight coefficient q are defined to determine the connection weight coefficient of the edge. Specifically, if the edge between two nodes contains a non-connecting switch, its connection weight coefficient increases by p, and if it contains a connecting switch, its connection weight coefficient increases by q, that is, 2p.

[0074] As for the calculation of specific values ​​of the first preset weight coefficient p and the second preset weight coefficient q, in this implementation, based on the longest path in the low-voltage topology diagram, the following determination method is given.

[0075] Obtain the longest path in the low voltage topology map as the target path, for example, see Figure 4 , where the path between transformer 1 and transformer 7 is the longest path between any two points in the low-voltage topology diagram. Assuming that the path between node i and node j is the longest, and the longest path contains a non-tie switches and b tie switches, then the following constraints are satisfied:

[0076]

[0077] Based on the above constraint conditions, specific values ​​of the first preset weight coefficient p and the second preset weight coefficient q are obtained, wherein the specific values ​​are retained to the minimum number of decimal places under the premise of satisfying the conditions.

[0078] In a possible implementation manner, obtaining the longest path in the low-voltage topology map includes: traversing the low-voltage topology map using a depth traversal algorithm to obtain the longest path in the low-voltage topology map.

[0079] Explanatory, the application of the deep traversal algorithm starts from any node of the low-voltage topology graph and recursively traverses along the edges of the graph, recording the length of the current path each time it traverses. During the traversal process, whenever a new node is encountered, the algorithm will continue to explore other nodes connected to the node in depth until it can no longer move forward. Then, the algorithm will backtrack to the previous node and try other unexplored edges. Through continuous comparison and updating, the algorithm can eventually find and record the longest path in a single path. This process ensures that every possible path in the low-voltage topology graph is considered, so as to accurately find the longest path.

[0080] In a possible implementation, the adjacency weight coefficient matrix V of the low voltage topology graph is:

[0081]

[0082] Among them, V i,j is the connection weight coefficient of the edge between node i and node j, where when i=j, V i,j =1; n is the total number of nodes.

[0083] The degree matrix D of the low voltage topology graph is:

[0084]

[0085] Among them, D i is the number of edges connected to node i.

[0086] In a possible implementation, the clustering results are obtained by clustering the nodes of the low-voltage topology graph using a spectral clustering method according to the adjacency weight coefficient matrix and degree matrix of the low-voltage topology graph, including: obtaining a Laplace matrix L=DV according to the adjacency weight coefficient matrix V and the degree matrix D of the low-voltage topology graph, and selecting the first k eigenvectors of the Laplace matrix L in order of eigenvalue from small to large to obtain an eigenvector matrix Y; clustering the nodes of the low-voltage topology graph according to the eigenvector matrix Y and using a spectral clustering method to obtain a clustering result; wherein, when clustering the nodes of the low-voltage topology graph using the spectral clustering method, a K-means algorithm is used to cluster the eigenvectors.

[0087] Specifically, solve the eigenvectors and eigenvalues ​​of the Laplace matrix L, sort the eigenvectors from small to large according to the eigenvalues, select the first k eigenvectors and organize them into an eigenvector matrix Y by row:

[0088]

[0089] For example, a specific process based on Python is given for clustering the nodes of the low-voltage topology graph according to the eigenvector matrix Y and the spectral clustering method: the SpectralClustering class in the scikit-learn library is introduced in Python, the eigenvector matrix Y is used as the input of the function SpectralClustering, and the number of clusters c is set. k Then clustering is achieved. In the process of implementing the spectral clustering method, the K-means algorithm is used to cluster the feature vectors. Finally, the nodes can be further processed or analyzed according to the obtained clustering labels. For example, nodes with the same clustering label can be represented by the same color or mark in the low-voltage topology map to intuitively observe the clustering results.

[0090] In view of the complex low-voltage topology of the distribution network, traditional graph modeling directly uses devices as nodes to construct a graph structure model, resulting in the problem that the graph structure is too large and difficult to analyze and calculate. The method for generating a low-voltage topology model in a distribution network based on spectral clustering proposed in the present invention is specifically a low-voltage topology graph segmentation method based on graph clustering, which can divide the low-voltage topology graph into multiple subgraphs. The subgraph segmentation result reduces the complexity of topology graph traversal and analysis. The present invention uses switches between nodes and defines switch coefficients to derive a correlation coefficient matrix as the input of the spectral clustering algorithm. Compared with the traditional spectral clustering algorithm that uses node distance to calculate the similarity between nodes, it is more in line with the connection between medium and low voltage equipment, more relevant to electrical business, and has higher practicality.

[0091] The following are device embodiments of the present invention, which can be used to implement the method embodiments of the present invention. For details not disclosed in the device embodiments, please refer to the method embodiments of the present invention.

[0092] See also Figure 5 In another embodiment of the present invention, a system for generating a medium- and low-voltage topology model in a distribution network based on spectral clustering is provided, which can be used to implement the above-mentioned method for generating a medium- and low-voltage topology model in a distribution network based on spectral clustering. Specifically, the system for generating a medium- and low-voltage topology model in a distribution network based on spectral clustering includes a topology graph construction module, a matrix construction module, a spectral clustering module, and a model generation module.

[0093] Among them, the topology map construction module is used to use transformers and low-voltage meter boxes as nodes, switch-type devices as edges, and determine the correspondence between nodes and edges in combination with device terminal information, and construct a low-voltage topology map based on the correspondence; the matrix construction module is used to determine the connection weight coefficients of the edges between nodes according to the types and numbers of switches between nodes, and construct the adjacency weight coefficient matrix of the low-voltage topology map based on the connection weight coefficients of the edges between nodes; and construct the degree matrix of the low-voltage topology map based on the number of edges connected to the nodes; the spectral clustering module is used to cluster the nodes of the low-voltage topology map using the spectral clustering method based on the adjacency weight coefficient matrix and degree matrix of the low-voltage topology map to obtain clustering results, and determine the category identification of each node based on the clustering results; the model generation module is used to divide nodes with the same category identification into the same subgraph based on the low-voltage topology map, obtain several low-voltage topology subgraphs and use them as low-voltage topology models.

[0094] In a possible implementation, the transformer and the low-voltage meter box are used as nodes, the switch-type equipment is used as edges, and the correspondence between the nodes and the edges is determined in combination with the equipment terminal information. When a low-voltage topology diagram is constructed according to the correspondence, several edges between the two nodes are merged, and attributes describing the basic information of the multiple edges are added to the merged edges.

[0095] In a possible implementation, determining the connection weight coefficient of the edge between nodes according to the switch type and number between the nodes includes: obtaining the longest path in the low voltage topology map as the target path; and obtaining a first preset weight coefficient p and a second preset weight coefficient q according to the following formula:

[0096]

[0097] Wherein, a is the number of non-connecting switches on the target path, and b is the number of connecting switches on the target path.

[0098] The connection weight coefficient between nodes is obtained by the following formula:

[0099] V i,j =A*p+B*q

[0100] Among them, V i,j is the connection weight coefficient of the edge between node i and node j, A is the number of non-connecting switches on the edge between node i and node j, and B is the number of connecting switches on the edge between node i and node j.

[0101] In a possible implementation manner, obtaining the longest path in the low-voltage topology map includes: traversing the low-voltage topology map using a depth traversal algorithm to obtain the longest path in the low-voltage topology map.

[0102] In a possible implementation, the adjacency weight coefficient matrix V of the low voltage topology graph is:

[0103]

[0104] Among them, V i,j is the connection weight coefficient of the edge between node i and node j, and n is the total number of nodes.

[0105] The degree matrix D of the low voltage topology graph is:

[0106]

[0107] Among them, D i is the number of edges connected to node i.

[0108] In a possible implementation, the clustering results are obtained by clustering the nodes of the low-voltage topology graph using a spectral clustering method according to the adjacency weight coefficient matrix and degree matrix of the low-voltage topology graph, including: obtaining a Laplace matrix L=DV according to the adjacency weight coefficient matrix V and the degree matrix D of the low-voltage topology graph, and selecting the first k eigenvectors of the Laplace matrix L in order of eigenvalue from small to large to obtain an eigenvector matrix Y; clustering the nodes of the low-voltage topology graph according to the eigenvector matrix Y and using a spectral clustering method to obtain a clustering result; wherein, when clustering the nodes of the low-voltage topology graph using the spectral clustering method, a K-means algorithm is used to cluster the eigenvectors.

[0109] All relevant contents of each step involved in the embodiment of the aforementioned method for generating a low-voltage topology model in a distribution network based on spectral clustering can be referred to the functional description of the functional modules corresponding to the system for generating a low-voltage topology model in a distribution network based on spectral clustering in the embodiment of the present invention, and will not be repeated here.

[0110] The division of modules in the embodiments of the present invention is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present invention may be integrated into one processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0111] In another embodiment of the present invention, a computer device is provided, the computer device including a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the low-voltage topology model generation method in the distribution network based on spectral clustering.

[0112] In another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It can be understood that the computer-readable storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the method for generating a low-voltage topology model in a distribution network based on spectral clustering in the above embodiment.

[0113] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take 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 code.

[0114] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks 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 produce 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.

[0115] 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.

[0116] 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.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for generating a low voltage topology model in a distribution network based on spectral clustering, characterized in that: include: Taking transformers and low-voltage meter boxes as nodes and switch devices as edges, the corresponding relationship between nodes and edges is determined in combination with device terminal information, and a low-voltage topology diagram is constructed based on the corresponding relationship; Determine the connection weight coefficient of the edge between nodes according to the switch type and number between the nodes, and construct the adjacency weight coefficient matrix of the low voltage topology graph according to the connection weight coefficient of the edge between the nodes; and constructing the degree matrix of the low voltage topology graph based on the number of edges connecting the nodes; According to the adjacency weight coefficient matrix and degree matrix of the low-voltage topology graph, the spectral clustering method is used to cluster the nodes of the low-voltage topology graph to obtain the clustering results, and the category identification of each node is determined according to the clustering results; Based on the low-voltage topology graph, nodes with the same category identifier are divided into the same subgraph, and several low-voltage topology subgraphs are obtained as low-voltage topology models.

2. The method for generating a low voltage topology model in a distribution network based on spectral clustering according to claim 1, characterized in that: The transformer and the low-voltage meter box are used as nodes, the switch-type equipment is used as the edge, and the correspondence between the nodes and the edges is determined in combination with the equipment terminal information. When constructing a low-voltage topology diagram according to the correspondence, several edges between the two nodes are merged, and attributes describing the basic information of multiple edges are added to the merged edges.

3. The method for generating a low voltage topology model in a distribution network based on spectral clustering according to claim 1, characterized in that: Determining the connection weight coefficient of the edge between nodes according to the switch type and number between the nodes includes: Obtain the longest path in the low voltage topology graph as the target path; The first preset weight coefficient p and the second preset weight coefficient q are obtained according to the following formula: Where a is the number of non-tie switches on the target path, and b is the number of tie switches on the target path; The connection weight coefficient between nodes is obtained by the following formula: V i,j =A*p+B*q Among them, V i,j is the connection weight coefficient of the edge between node i and node j, A is the number of non-connecting switches on the edge between node i and node j, and B is the number of connecting switches on the edge between node i and node j.

4. The method for generating a low voltage topology model in a distribution network based on spectral clustering according to claim 3, characterized in that: The obtaining of the longest path in the low-voltage topology map includes: using a depth traversal algorithm to traverse the low-voltage topology map to obtain the longest path in the low-voltage topology map.

5. The method for generating a low voltage topology model in a distribution network based on spectral clustering according to claim 1, characterized in that: The adjacency weight coefficient matrix V of the low voltage topology graph is: Among them, V i,j is the connection weight coefficient of the edge between node i and node j, and n is the total number of nodes; The degree matrix D of the low voltage topology graph is: Among them, D i is the number of edges connected to node i.

6. The method for generating a low voltage topology model in a distribution network based on spectral clustering according to claim 1, characterized in that: The clustering results obtained by clustering the nodes of the low-voltage topology graph using a spectral clustering method according to the adjacency weight coefficient matrix and the degree matrix of the low-voltage topology graph include: According to the adjacency weight coefficient matrix V and the degree matrix D of the low-voltage topology graph, the Laplace matrix L=DV is obtained, and the first k eigenvectors of the Laplace matrix L are selected in the order of eigenvalue from small to large to obtain the eigenvector matrix Y; According to the eigenvector matrix Y, each node of the low-voltage topology graph is clustered by the spectral clustering method to obtain a clustering result; wherein, when the nodes of the low-voltage topology graph are clustered by the spectral clustering method, the K-means algorithm is used to cluster the eigenvectors.

7. A system for generating low voltage topology models in distribution networks based on spectral clustering, characterized in that: include: A topology construction module is used to use transformers and low-voltage meter boxes as nodes, switch-type devices as edges, and determine the correspondence between nodes and edges in combination with device terminal information, and construct a low-voltage topology map based on the correspondence; A matrix construction module, used to determine the connection weight coefficients of the edges between nodes according to the switch types and numbers between the nodes, and to construct an adjacency weight coefficient matrix of the low voltage topology graph according to the connection weight coefficients of the edges between the nodes; and constructing the degree matrix of the low voltage topology graph based on the number of edges connecting the nodes; The spectral clustering module is used to cluster the nodes of the low-voltage topology map using the spectral clustering method according to the adjacency weight coefficient matrix and degree matrix of the low-voltage topology map to obtain clustering results, and determine the category identification of each node according to the clustering results; The model generation module is used to divide nodes with the same category identifier into the same subgraph based on the low-voltage topology graph, and obtain several low-voltage topology subgraphs as the low-voltage topology model.

8. The system for generating a low voltage topology model in a distribution network based on spectral clustering according to claim 7, characterized in that: The transformer and the low-voltage meter box are used as nodes, the switch-type equipment is used as the edge, and the correspondence between the nodes and the edges is determined in combination with the equipment terminal information. When constructing a low-voltage topology diagram according to the correspondence, several edges between the two nodes are merged, and attributes describing the basic information of multiple edges are added to the merged edges.

9. The system for generating a low voltage topology model in a distribution network based on spectral clustering according to claim 7, characterized in that: Determining the connection weight coefficient of the edge between nodes according to the switch type and number between the nodes includes: Obtain the longest path in the low voltage topology graph as the target path; The first preset weight coefficient p and the second preset weight coefficient q are obtained according to the following formula: Where a is the number of non-tie switches on the target path, and b is the number of tie switches on the target path; The connection weight coefficient between nodes is obtained by the following formula: V i,j =A*p+B*q Among them, V i,j is the connection weight coefficient of the edge between node i and node j, A is the number of non-connecting switches on the edge between node i and node j, and B is the number of connecting switches on the edge between node i and node j.

10. The system for generating a low voltage topology model in a distribution network based on spectral clustering according to claim 9, characterized in that: The obtaining of the longest path in the low-voltage topology map includes: using a depth traversal algorithm to traverse the low-voltage topology map to obtain the longest path in the low-voltage topology map.

11. The system for generating a low voltage topology model in a distribution network based on spectral clustering according to claim 7, characterized in that: The adjacency weight coefficient matrix V of the low voltage topology graph is: Among them, V i,j is the connection weight coefficient of the edge between node i and node j, and n is the total number of nodes; The degree matrix D of the low voltage topology graph is: Among them, D i is the number of edges connected to node i.

12. The system for generating a low voltage topology model in a distribution network based on spectral clustering according to claim 7, characterized in that: The clustering results obtained by clustering the nodes of the low-voltage topology graph using a spectral clustering method according to the adjacency weight coefficient matrix and the degree matrix of the low-voltage topology graph include: According to the adjacency weight coefficient matrix V and the degree matrix D of the low-voltage topology graph, the Laplace matrix L=DV is obtained, and the first k eigenvectors of the Laplace matrix L are selected in the order of eigenvalue from small to large to obtain the eigenvector matrix Y; According to the eigenvector matrix Y, each node of the low-voltage topology graph is clustered by the spectral clustering method to obtain a clustering result; wherein, when the nodes of the low-voltage topology graph are clustered by the spectral clustering method, the K-means algorithm is used to cluster the eigenvectors.

13. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for generating a low-voltage topology model in a distribution network based on spectral clustering are implemented as described in any one of claims 1 to 6.

14. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for generating a low-voltage topology model in a distribution network based on spectral clustering are implemented as claimed in any one of claims 1 to 6.

Images