Power distribution network operation situation prediction method and device and computer equipment

Abstract

This application relates to a method, apparatus, and computer equipment for predicting the operational status of a distribution network, belonging to the field of computer technology. The method includes: obtaining a current power grid hierarchical tree and a current power grid knowledge graph based on current operational data of the distribution network; extracting abnormal paths from the current power grid knowledge graph through the matching relationship between tree nodes in the current power grid hierarchical tree and graph nodes in the current power grid knowledge graph; determining the current operational status of the distribution network based on the distribution of matching tree nodes of abnormal graph nodes in the abnormal paths within the current power grid hierarchical tree; the current operational status is used to characterize the operational state of entities in the distribution network; and predicting the operational status of the distribution network based on the current operational status, the current power grid hierarchical tree, and the current power grid knowledge graph, obtaining the predicted operational status result of the distribution network. This method can improve the accuracy of the predicted operational status.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for predicting the operational status of a power distribution network. Background Technology

[0002] With the rapid development of smart grid technology, the effective prediction of the operation status of the distribution network is of great significance for ensuring the stability and reliability of power supply.

[0003] In related technologies, when predicting the operating status of a distribution network, it is difficult to dynamically adapt and adjust according to the real-time operating status of the distribution network, resulting in a mismatch between the predicted operating status and the actual operating status. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, device, computer equipment, computer-readable storage medium, and computer program product for predicting the operating status of a distribution network that can improve the accuracy of the predicted operating status, in order to address the technical problem of the mismatch between the predicted operating status and the actual operating status.

[0005] Firstly, this application provides a method for predicting the operational status of a power distribution network, including:

[0006] Based on the current operating data of the distribution network, the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network are reconstructed to obtain the current power grid hierarchy tree and current power grid knowledge graph of the distribution network. The power grid hierarchy tree of the distribution network is used to represent the hierarchical relationship between topological nodes in the physical topology of the distribution network. The power grid hierarchy tree includes multiple levels, each level includes at least one tree node, and each tree node represents a topological node. The power grid knowledge graph of the distribution network is used to represent the association relationship between entities in the distribution network. The entities in the distribution network include the topological nodes and corresponding fault information and operation and maintenance information. Each graph node in the power grid knowledge graph represents an entity.

[0007] By matching the tree nodes in the current power grid hierarchical tree with the graph nodes in the current power grid knowledge graph, abnormal paths are extracted from the current power grid knowledge graph; the graph nodes included in the abnormal paths are abnormal graph nodes.

[0008] Based on the distribution of matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree, the current operating status of the distribution network is determined; the matching tree nodes of the abnormal graph nodes are the tree nodes in the current power grid hierarchical tree that match the abnormal graph nodes; the current operating status is used to characterize the operating status of entities in the distribution network.

[0009] Based on the current operating status, the current power grid hierarchy tree, and the current power grid knowledge graph, the operating status of the distribution network is predicted, and the predicted operating status result of the distribution network is obtained.

[0010] In one embodiment, based on the current operating data of the distribution network, the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network are reconstructed to obtain the current power grid hierarchy tree and current power grid knowledge graph of the distribution network, including:

[0011] Based on the operational data, the hierarchical relationship between tree nodes in the historical power grid hierarchy tree is updated to obtain the current power grid hierarchy tree;

[0012] Based on the matching relationship between the tree nodes in the current power grid hierarchical tree and the graph nodes in the historical power grid knowledge graph, the association relationship between the graph nodes in the historical power grid knowledge graph is updated to obtain the current power grid knowledge graph.

[0013] In one embodiment, extracting abnormal paths from the current power grid knowledge graph through the matching relationship between tree nodes in the current power grid hierarchical tree and graph nodes in the current power grid knowledge graph includes:

[0014] From the current power grid knowledge graph, the association relationships between matching graph nodes of tree nodes in the current power grid hierarchical tree are extracted to obtain the current association relationship set; the matching graph nodes of the tree nodes are the graph nodes in the current power grid knowledge graph that match the tree nodes;

[0015] In the current set of associations, target associations that differ from the historical set of associations are identified; the historical set of associations is obtained based on the associations between tree nodes in the historical power grid hierarchical tree and matching graph nodes in the historical power grid knowledge graph.

[0016] The abnormal path is obtained based on the target association relationship.

[0017] In one embodiment, the number of abnormal paths is multiple;

[0018] The determination of the current operating status of the distribution network based on the distribution of matching tree nodes of the abnormal graph nodes in the abnormal path within the current power grid hierarchical tree includes:

[0019] For each abnormal path, based on the distribution of the matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree, the path penetration information, lateral impact information and vertical impact information of the abnormal path at each level are determined.

[0020] For each level, the horizontal and vertical impact information of the abnormal path at that level are fused to determine the path impact information of the abnormal path at that level;

[0021] By using the path influence information of each abnormal path at each level, each abnormal path is clustered to obtain at least one abnormal path cluster.

[0022] For each abnormal path cluster, the vertical diffusion speed information of the abnormal path cluster between each level and the horizontal influence information under each level are fused to obtain the path cluster influence information of the abnormal path cluster; the vertical diffusion speed information is determined based on the abnormal time of the abnormal path cluster under each level, the abnormal time of the abnormal path cluster under each level is the time when the target graph node of the abnormal path cluster under the level becomes abnormal, and the target graph node is the abnormal graph node that matches the matching tree node in the level;

[0023] By integrating the path cluster impact information of each abnormal path cluster, as well as the path penetration information and lateral impact information of each abnormal path at each level, the current operating status of the power distribution network is obtained.

[0024] In one embodiment, determining the path penetration information, lateral impact information, and vertical impact information of the abnormal path at each level based on the distribution of matching tree nodes of the abnormal graph nodes in the abnormal path within the current power grid hierarchical tree includes:

[0025] For each level, based on the number of abnormal graph nodes of the abnormal path, the number of each level, and the number of matching tree nodes of the abnormal path under the level, the path penetration information of the abnormal path under the level is obtained.

[0026] Based on the number of matching tree nodes of the abnormal path at the level and the number of tree nodes in the level, the lateral impact information of the abnormal path at the level is obtained.

[0027] Based on the lateral influence information of the abnormal path at the level and the level above the level, as well as the structural similarity information between the level above the level and the level, the hierarchical transmission information of the abnormal path between the level above the level and the level is obtained.

[0028] Based on the stated level and the levels preceding it, a target level set is obtained. Based on the hierarchical transfer information between the levels in the target level set, the vertical impact information of the abnormal path under the stated level is obtained.

[0029] In one embodiment, before fusing the vertical diffusion velocity information of the abnormal path cluster between each of the said levels and the lateral influence information of each abnormal path in the abnormal path cluster at each of the said levels to obtain the path cluster influence information of the abnormal path cluster, the method further includes:

[0030] Based on the time difference between the abnormal times corresponding to each adjacent level and the level difference between each adjacent level, the diffusion speed between each adjacent level is obtained. Based on the maximum diffusion speed among the diffusion speeds between each adjacent level, the longitudinal diffusion speed information of the abnormal path cluster between each level is obtained.

[0031] The path cluster influence information of the abnormal path cluster is obtained by fusing the vertical diffusion velocity information of the abnormal path cluster between each level and the lateral influence information of each abnormal path in the abnormal path cluster at each level, including:

[0032] The path cluster influence information of the abnormal path cluster is obtained by fusing the longitudinal diffusion speed information of the abnormal path cluster between each level and the average value of the lateral influence information of each abnormal path in the abnormal path cluster at each level.

[0033] In one embodiment, the step of predicting the operating status of the distribution network based on the current operating status, the current power grid hierarchy tree, and the current power grid knowledge graph to obtain the operating status prediction result of the distribution network includes:

[0034] Extract the target entity evolution strategy associated with the abnormal path from the current power grid knowledge graph;

[0035] Based on the hierarchical relationship of the current power grid hierarchy tree and the target entity evolution strategy, taking the current operating status as the initial operating state of the entities in the distribution network, the operating state of the entities in the distribution network is deduced to obtain the predicted operating state of the entities in the distribution network.

[0036] Based on the predicted operating status of the entities in the power distribution network, the predicted operating status of the power distribution network is obtained.

[0037] Secondly, this application also provides a power distribution network operation status prediction device, comprising:

[0038] The power grid information acquisition module is used to reconstruct the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network based on the current operating data of the distribution network, so as to obtain the current power grid hierarchy tree and current power grid knowledge graph of the distribution network. The power grid hierarchy tree of the distribution network is used to represent the hierarchical relationship between topological nodes in the physical topology structure of the distribution network. The power grid hierarchy tree includes multiple levels, each level includes at least one tree node, and each tree node represents a topological node. The power grid knowledge graph of the distribution network is used to represent the association relationship between entities in the distribution network. The entities in the distribution network include the topological nodes and corresponding fault information and operation and maintenance information. Each graph node in the power grid knowledge graph represents an entity.

[0039] An abnormal path extraction module is used to extract abnormal paths from the current power grid knowledge graph by matching the tree nodes in the current power grid hierarchy tree with the graph nodes in the current power grid knowledge graph; the abnormal path includes abnormal graph nodes.

[0040] The operational status determination module is used to determine the current operational status of the distribution network based on the distribution of matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree; the matching tree nodes of the abnormal graph nodes are the tree nodes in the current power grid hierarchical tree that match the abnormal graph nodes; the current operational status is used to characterize the operational status of entities in the distribution network;

[0041] The operation status prediction module is used to predict the operation status of the distribution network based on the current operation status, the current power grid hierarchy tree, and the current power grid knowledge graph, and obtain the operation status prediction result of the distribution network.

[0042] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0043] Based on the current operating data of the distribution network, the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network are reconstructed to obtain the current power grid hierarchy tree and current power grid knowledge graph of the distribution network. The power grid hierarchy tree of the distribution network is used to represent the hierarchical relationship between topological nodes in the physical topology of the distribution network. The power grid hierarchy tree includes multiple levels, each level includes at least one tree node, and each tree node represents a topological node. The power grid knowledge graph of the distribution network is used to represent the association relationship between entities in the distribution network. The entities in the distribution network include the topological nodes and corresponding fault information and operation and maintenance information. Each graph node in the power grid knowledge graph represents an entity.

[0044] By matching the tree nodes in the current power grid hierarchical tree with the graph nodes in the current power grid knowledge graph, abnormal paths are extracted from the current power grid knowledge graph; the graph nodes included in the abnormal paths are abnormal graph nodes.

[0045] Based on the distribution of matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree, the current operating status of the distribution network is determined; the matching tree nodes of the abnormal graph nodes are the tree nodes in the current power grid hierarchical tree that match the abnormal graph nodes; the current operating status is used to characterize the operating status of entities in the distribution network.

[0046] Based on the current operating status, the current power grid hierarchy tree, and the current power grid knowledge graph, the operating status of the distribution network is predicted, and the predicted operating status result of the distribution network is obtained.

[0047] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0048] Based on the current operating data of the distribution network, the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network are reconstructed to obtain the current power grid hierarchy tree and current power grid knowledge graph of the distribution network. The power grid hierarchy tree of the distribution network is used to represent the hierarchical relationship between topological nodes in the physical topology of the distribution network. The power grid hierarchy tree includes multiple levels, each level includes at least one tree node, and each tree node represents a topological node. The power grid knowledge graph of the distribution network is used to represent the association relationship between entities in the distribution network. The entities in the distribution network include the topological nodes and corresponding fault information and operation and maintenance information. Each graph node in the power grid knowledge graph represents an entity.

[0049] By matching the tree nodes in the current power grid hierarchical tree with the graph nodes in the current power grid knowledge graph, abnormal paths are extracted from the current power grid knowledge graph; the graph nodes included in the abnormal paths are abnormal graph nodes.

[0050] Based on the distribution of matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree, the current operating status of the distribution network is determined; the matching tree nodes of the abnormal graph nodes are the tree nodes in the current power grid hierarchical tree that match the abnormal graph nodes; the current operating status is used to characterize the operating status of entities in the distribution network.

[0051] Based on the current operating status, the current power grid hierarchy tree, and the current power grid knowledge graph, the operating status of the distribution network is predicted, and the predicted operating status result of the distribution network is obtained.

[0052] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0053] Based on the current operating data of the distribution network, the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network are reconstructed to obtain the current power grid hierarchy tree and current power grid knowledge graph of the distribution network. The power grid hierarchy tree of the distribution network is used to represent the hierarchical relationship between topological nodes in the physical topology of the distribution network. The power grid hierarchy tree includes multiple levels, each level includes at least one tree node, and each tree node represents a topological node. The power grid knowledge graph of the distribution network is used to represent the association relationship between entities in the distribution network. The entities in the distribution network include the topological nodes and corresponding fault information and operation and maintenance information. Each graph node in the power grid knowledge graph represents an entity.

[0054] By matching the tree nodes in the current power grid hierarchical tree with the graph nodes in the current power grid knowledge graph, abnormal paths are extracted from the current power grid knowledge graph; the graph nodes included in the abnormal paths are abnormal graph nodes.

[0055] Based on the distribution of matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree, the current operating status of the distribution network is determined; the matching tree nodes of the abnormal graph nodes are the tree nodes in the current power grid hierarchical tree that match the abnormal graph nodes; the current operating status is used to characterize the operating status of entities in the distribution network.

[0056] Based on the current operating status, the current power grid hierarchy tree, and the current power grid knowledge graph, the operating status of the distribution network is predicted, and the predicted operating status result of the distribution network is obtained.

[0057] The aforementioned methods, devices, computer equipment, computer-readable storage media, and computer program products for predicting the operational status of distribution networks dynamically reconstruct the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network based on its current operational status. This generates a current power grid hierarchy tree and current power grid knowledge graph adapted to the current actual operation of the distribution network. The power grid hierarchy tree accurately reflects the physical topology of the distribution network, while the power grid knowledge graph deeply integrates topological nodes from the physical topology with corresponding fault information and maintenance information, achieving a structured and systematic expression of the distribution network structure and operation. Through the matching relationship between tree nodes and graph nodes, abnormal paths can be extracted from the current power grid knowledge graph. By matching the distribution of abnormal graph nodes included in abnormal paths within the current power grid hierarchical tree, the current operating status of the distribution network can be objectively and accurately determined. Through the current power grid hierarchical tree and the current power grid knowledge graph, the internal relationships and anomaly propagation patterns of the distribution network can be fully explored, thereby enabling accurate prediction of the distribution network's operating status based on its current operating status. This prediction of the distribution network's operating status, through the reconstruction of the power grid hierarchical tree and power grid knowledge graph, avoids information lag, thus enabling accurate extraction of abnormal paths and assessment of the current operating status, ultimately leading to accurate prediction of the operating status, thereby improving the accuracy of the predicted operating status. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only one embodiment of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is an application environment diagram of the power distribution network operation status prediction method in one embodiment;

[0060] Figure 2 This is a flowchart illustrating a method for predicting the operational status of a power distribution network in one embodiment.

[0061] Figure 3 This is a flowchart illustrating the steps for determining the current operating status of a distribution network based on the distribution of matching tree nodes of abnormal graph nodes in abnormal paths within the current power grid hierarchy tree, as shown in one embodiment.

[0062] Figure 4 This is a flowchart illustrating the steps of predicting the operating status of a distribution network based on the current operating status, the current power grid hierarchy tree, and the current power grid knowledge graph in one embodiment, to obtain the predicted operating status result of the distribution network.

[0063] Figure 5 This is a structural block diagram of a power distribution network operation status prediction device in one embodiment;

[0064] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0065] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0066] It is understood that terms such as "first" and "second" in this application are used only to distinguish similar objects and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. The term "connection" in the embodiments of this application refers to various connection methods, such as direct or indirect connections, to achieve communication between devices; this application does not impose any limitations on this.

[0067] It is understandable that "at least one" refers to one or more, while "multiple" refers to two or more.

[0068] When used herein, the singular forms of “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “comprising / including” or “having,” etc., specify the presence of the stated features, wholes, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, wholes, steps, operations, components, parts, or combinations thereof. Meanwhile, the term “and / or” as used in this specification includes any and all combinations of the associated listed items.

[0069] The distribution network operation status prediction method provided in this application embodiment can be applied to, for example, Figure 1In the application environment shown, server 102 communicates with terminal 104 via a network. A data storage system can store the data that server 102 needs to process. This data storage system can be integrated onto server 102 or located on a cloud or other network server. Server 102 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. Terminal 104 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, projection devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. Head-mounted devices can be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. Furthermore, server 102 and terminal 104 can be directly or indirectly connected via wired or wireless communication, which is not limited herein.

[0070] It should be noted that both server 102 and terminal 104 can be used individually to execute the power distribution network operation status prediction method provided in this application embodiment, or they can be used together to execute the power distribution network operation status prediction method provided in this application embodiment.

[0071] For example, firstly, server 102 reconstructs the historical power grid hierarchy tree and historical power grid knowledge graph of the power distribution network based on the current operating data of the power distribution network, obtaining the current power grid hierarchy tree and current power grid knowledge graph of the power distribution network. The power grid hierarchy tree of the power distribution network is used to represent the hierarchical relationship between topological nodes in the physical topology of the power distribution network. The power grid hierarchy tree includes multiple levels, each level includes at least one tree node, and each tree node represents a topological node. The power grid knowledge graph of the power distribution network is used to represent the association relationship between entities in the power distribution network. Entities in the power distribution network include topological nodes and corresponding fault information and operation and maintenance information. Each graph node in the power grid knowledge graph represents an entity. Then, server 102 connects the tree nodes in the current power grid hierarchy tree with the current power grid knowledge graph. The matching relationships between graph nodes in the knowledge graph are used to extract abnormal paths from the current power grid knowledge graph. The graph nodes included in the abnormal paths are called abnormal graph nodes. Then, server 102 determines the current operating status of the distribution network based on the distribution of the matching tree nodes of the abnormal graph nodes in the abnormal paths in the current power grid hierarchical tree. The matching tree nodes of the abnormal graph nodes are the tree nodes in the current power grid hierarchical tree that match the abnormal graph nodes. The current operating status is used to characterize the operating state of entities in the distribution network. Then, server 102 predicts the operating status of the distribution network based on the current operating status, the current power grid hierarchical tree, and the current power grid knowledge graph, and obtains the operating status prediction result of the distribution network. Finally, server 102 pushes the operating status prediction result of the distribution network to terminal 104.

[0072] In one embodiment, such as Figure 2 As shown, a method for predicting the operational status of a power distribution network is provided, and this method is applied to... Figure 1 Taking a server as an example, it can be understood that this method can also be applied to a terminal, and can also be applied to a system that includes both a server and a terminal, and is implemented through the interaction between the server and the terminal. The method includes the following steps:

[0073] Step S202: Based on the current operating data of the distribution network, reconstruct the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network to obtain the current power grid hierarchy tree and current power grid knowledge graph of the distribution network.

[0074] The power grid hierarchy tree of the distribution network is used to represent the hierarchical relationship between topological nodes in the physical topology of the distribution network. The power grid hierarchy tree includes multiple levels, each level includes at least one tree node, and each tree node represents a topological node. In practical applications, the physical topology of the distribution network includes topological nodes such as those from substations to feeders, then to branch lines and user terminals.

[0075] The power grid knowledge graph of the distribution network is used to represent the relationship between entities in the distribution network. The entities in the distribution network include topology nodes and the corresponding fault information and operation and maintenance information of the topology nodes. Each graph node in the power grid knowledge graph represents an entity, and the connecting edges between graph nodes represent the relationship between the entities corresponding to the graph nodes. In practical applications, the entities in the distribution network include topology nodes in the physical topology of the distribution network and the fault records and operation and maintenance records of the topology nodes.

[0076] In this step, firstly, the server acquires the current operating data of the distribution network, including electrical quantity data such as voltage, current, and power, as well as non-electrical quantity data such as equipment temperature and ambient humidity. Next, the server preprocesses this operating data to remove noise and redundant information. Then, based on the preprocessed operating data, the server reconstructs the historical power grid hierarchy tree of the distribution network to obtain the current power grid hierarchy tree of the distribution network. Furthermore, based on the current power grid hierarchy tree of the distribution network, the server reconstructs the historical power grid knowledge graph of the distribution network to obtain the current power grid knowledge graph of the distribution network.

[0077] Step S204: Extract abnormal paths from the current power grid knowledge graph by matching the tree nodes in the current power grid hierarchical tree with the graph nodes in the current power grid knowledge graph; the graph nodes included in the abnormal paths are abnormal graph nodes.

[0078] In this system, each tree node in the power grid hierarchical tree represents a topology node in the distribution network topology, and each graph node in the power grid knowledge graph represents an entity, including the topology node and its corresponding fault information and operation and maintenance information. Therefore, tree nodes and graph nodes can be matched based on the topology nodes represented by tree nodes and graph nodes. For a matched tree node and graph node, the tree node is the matching tree node of the graph node, and the graph node is the matching graph node of the tree node.

[0079] In this step, firstly, the server locates the matching graph nodes of the tree nodes in the current power grid hierarchical tree within the current power grid knowledge graph, and extracts the associations between the matching graph nodes to obtain the current set of associations. Then, the server compares the current set of associations with the historical set of associations extracted based on the historical power grid hierarchical tree and the historical power grid knowledge graph. Based on the comparison results, the server determines the target associations that differ from the current set of associations. Next, the server obtains the abnormal paths based on the target associations and the graph nodes connected to them. The graph nodes included in the abnormal paths, that is, the graph nodes connected to the target associations, are the abnormal graph nodes.

[0080] Step S206: Based on the distribution of the matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree, determine the current operating status of the distribution network.

[0081] Among them, the matching tree node of the anomaly graph node is the tree node in the current power grid hierarchy tree that matches the anomaly graph node.

[0082] The current operational status is used to characterize the operational status of entities in the distribution network.

[0083] In this step, the server assesses the impact of abnormal paths on the distribution network based on the distribution of matching tree nodes of abnormal graph nodes in the current power grid hierarchy tree, and then determines the current operating status of the distribution network.

[0084] Step S208: Based on the current operating status, the current power grid hierarchy tree, and the current power grid knowledge graph, predict the operating status of the distribution network to obtain the predicted operating status result of the distribution network.

[0085] In this step, the server dynamically extrapolates the operating status of entities in the distribution network based on the current power grid hierarchy tree and the current power grid knowledge graph, taking the current operating status as the initial operating state of the entities in the distribution network, to obtain the predicted operating status of the entities in the distribution network, and thus obtain the operating status prediction result of the distribution network.

[0086] In the aforementioned method for predicting the operational status of distribution networks, the historical network hierarchy tree and historical network knowledge graph of the distribution network are dynamically reconstructed based on the current operational status of the distribution network. This generates a current network hierarchy tree and current network knowledge graph adapted to the current actual operation of the distribution network. The network hierarchy tree accurately reflects the physical topology of the distribution network, while the network knowledge graph deeply integrates topological nodes in the physical topology with corresponding fault information and maintenance information, achieving a structured and systematic expression of the distribution network structure and operation. Through the matching relationship between tree nodes and graph nodes, abnormal paths can be extracted from the current network knowledge graph; and the abnormal paths include anomalies... The distribution of matching tree nodes in the current power grid hierarchy tree can objectively and accurately determine the current operating status of the distribution network. Through the current power grid hierarchy tree and the current power grid knowledge graph, the internal relationships and anomaly propagation patterns of the distribution network can be fully explored, thereby enabling accurate prediction of the distribution network's operating status based on its current operating status. The above-mentioned prediction of the distribution network's operating status avoids information lag by reconstructing the power grid hierarchy tree and the power grid knowledge graph, thus enabling accurate extraction of abnormal paths and assessment of the current operating status, and consequently accurate prediction of the operating status, thereby improving the accuracy of the predicted operating status.

[0087] In one embodiment, step S202 above, which reconstructs the historical power grid hierarchy tree and historical power grid knowledge graph of the power grid based on the current operating data of the power grid, to obtain the current power grid hierarchy tree and current power grid knowledge graph of the power grid, includes the following steps: based on the operating data, updating the hierarchical relationship between tree nodes in the historical power grid hierarchy tree to obtain the current power grid hierarchy tree; based on the matching relationship between tree nodes in the current power grid hierarchy tree and graph nodes in the historical power grid knowledge graph, updating the association relationship between graph nodes in the historical power grid knowledge graph to obtain the current power grid knowledge graph.

[0088] In this embodiment, the process of updating the hierarchical relationships between tree nodes in the historical power grid hierarchy tree based on operational data to obtain the current power grid hierarchy tree is as follows: The server determines whether the hierarchical relationships between tree nodes in the historical power grid hierarchy tree have changed based on the operational data. If they have changed, the server updates the hierarchical relationships between tree nodes in the historical power grid hierarchy tree to obtain the current power grid hierarchy tree. For example, in the historical power grid hierarchy tree, tree node T1 is connected to tree nodes T2 and T3, and tree node T2 is connected to tree nodes T4 and T5. When the server identifies, based on the operational data of tree node T4, that tree node T4 no longer conforms to the original hierarchical relationship, it adjusts the hierarchical relationship between tree node T4 and the other tree nodes. For example, it adjusts tree node T4 to be directly connected to tree node T1.

[0089] In the embodiments provided in this application, multiple node attribute dimensions are preset for each topology node. These multiple node attribute dimensions include at least an inherent attribute dimension, which is used to characterize the attributes of the topology node that do not change with real-time state. Each tree node has corresponding tree node attribute information under each node attribute dimension, and each graph node has corresponding graph node attribute information under each node attribute dimension. When updating the hierarchical relationship between tree nodes in the historical power grid hierarchical tree, the server can synchronously update the tree node attribute information of the tree nodes. Based on the tree node attribute information of the tree nodes in the current power grid hierarchical tree and the graph node attribute information of the graph nodes in the historical power grid knowledge graph, the server can determine the node matching degree between the tree nodes in the current power grid hierarchical tree and the graph nodes in the historical power grid knowledge graph. The node matching degree characterizes the matching relationship between the tree nodes in the current power grid hierarchical tree and the graph nodes in the historical power grid knowledge graph. For example, the server can first calculate the dimensional matching degree between the tree node attribute information and the graph node attribute information under the same node attribute dimension for each node attribute dimension. Then, based on the dimensional weights corresponding to each node attribute dimension, the server can weightedly fuse the dimensional matching degrees corresponding to each node attribute dimension to obtain the node matching degree between the tree node and the graph node.

[0090] In this embodiment, the server updates the association relationships between graph nodes in the historical power grid knowledge graph based on the matching relationship between tree nodes in the current power grid hierarchy tree and graph nodes in the historical power grid knowledge graph, and obtains the current power grid knowledge graph as follows: First, based on the node matching degree between tree nodes in the current power grid hierarchy tree and graph nodes in the historical power grid knowledge graph, the server filters out tree nodes and graph nodes whose node matching degree is greater than or equal to a preset first node matching degree threshold from the tree nodes in the current power grid hierarchy tree and graph nodes in the historical power grid knowledge graph, and obtains a first node pair (each first node pair includes one tree node and one graph node). For example, the tree node attribute information of tree node T1 includes (topology node identifier "FXL1-2023-001", voltage level "10kV", load capacity "500kVA"), and the graph node attribute information of graph node G1 includes (topology node identifier "line number FXL1", voltage level "10kV", load capacity "450kVA"). Then, the server determines that tree node T1 and graph node G1 are a first node pair.

[0091] Then, for each first node pair, the server compares the tree node attribute information of the tree node in the first node pair under the inherent attribute dimension with the graph node attribute information of the graph node in the first node pair under the inherent attribute dimension. If the two do not conflict, the first node pair is determined as the second node pair. For example, taking the first node pair (tree node T1, graph node G1) as an example, assuming that the tree node attribute information of tree node T1 under the inherent attribute dimension includes "laying method underground" and "insulation level A", and the graph node attribute information of graph node G1 under the inherent attribute dimension includes "laying method underground" and "insulation level ≥ A", then the server determines the first node pair (tree node T1, graph node G1) as the second node pair. For example, taking the first node pair (tree node T2, graph node G2) as an example, assuming that the tree node attribute information of tree node T2 in the inherent attribute dimension includes "rated current 800A", and the graph node attribute information of graph node G2 in the inherent attribute dimension includes "rated current ≥ 1000A", then the server will not determine the first node pair (tree node T2, graph node G2) as the second node pair.

[0092] Next, for each second node pair, the server extracts the tree nodes in each second node pair and determines the hierarchical relationship between the tree nodes in each second node pair based on the current power grid hierarchical tree. The server also extracts the graph nodes in each second node pair and determines the hierarchical relationship between the graph nodes in each second node pair based on the historical power grid knowledge graph. The server compares the hierarchical relationship between the tree nodes in each second node pair and the hierarchical relationship between the graph nodes in each second node pair, and retains the second node pairs with consistent hierarchical relationships to obtain the third node pair. For example, the second node pair includes (tree node T1, graph node G1), (tree node T2, graph node G2), and (tree node T3, graph node G3). Assuming the hierarchical relationship between tree nodes in each second node pair in the current power grid hierarchy tree is tree node T1 → tree node T2 → tree node T3, and the hierarchical relationship between graph nodes in each second node pair in the historical power grid knowledge graph is graph node G1 → graph node G2 → graph node G3, then the server will store the second node pairs (tree node T1, graph node G1), (tree node T2, graph node G2), and (tree node T3) in the graph knowledge graph. Node T3 and graph node G3 are both determined as the third node pair; for example, suppose the second node pair also includes (tree node T4 and graph node G4), and the hierarchical relationship between the tree nodes in each second node pair in the current power grid hierarchy tree is tree node T1→tree node T2→tree node T3 and tree node T1→tree node T4, but the hierarchical relationship between the graph nodes in each second node pair in the historical power grid knowledge graph is graph node G3→graph node G4, then the server will not determine the second node pair (tree node T4 and graph node G4) as the third node pair.

[0093] Then, for each third node pair, the server takes the graph node in the third node pair as the starting point, searches for its adjacent graph node in the historical power grid knowledge graph, and if the node matching degree between the adjacent graph node and the remaining tree nodes in the current power grid hierarchy tree other than the tree nodes in each third node pair is greater than or equal to the preset second node matching degree threshold, the adjacent graph node and the remaining tree nodes are determined as a fourth node pair.

[0094] Next, the server combines each third node pair and each fourth node pair to obtain target node pairs. For each target node pair, the server extracts the tree nodes in each target node pair and determines the association relationships (such as connection relationships, power transmission relationships, and control relationships) between the tree nodes in each target node pair based on the current power grid hierarchical tree. It also determines the current association relationships (connection relationships, power transmission relationships, and control relationships) between the graph nodes in each target node pair and integrates the current topological relationships between the graph nodes in each target node pair into the historical power grid knowledge graph to update the association relationships between the graph nodes in the historical power grid knowledge graph, thus obtaining the current power grid knowledge graph. In practical applications, for graph nodes in a pair of target nodes, if the current association between these two graph nodes is the same as the historical association between them in the historical power grid knowledge graph, the server retains the updated or more accurate association (retaining the historical association or updating the historical association to the current association); if the current association between these two graph nodes is a new association compared to the historical association between them in the historical power grid knowledge graph, the server adds the current association to the historical power grid knowledge graph; if the current association between these two graph nodes conflicts with the association between them in the historical power grid knowledge graph, the server introduces a conflict resolution strategy to retain the historical association or update the historical association to the current association. The conflict resolution strategy is based on the timestamp of the association (including the historical association and the current association) and the credibility of the data source.

[0095] In this embodiment, the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network are dynamically reconstructed based on the current operating status of the distribution network. This enables the generation of a current power grid hierarchy tree and current power grid knowledge graph that are adapted to the current actual operation of the distribution network. The power grid hierarchy tree accurately reflects the physical topology of the distribution network, and the power grid knowledge graph deeply integrates the topological nodes in the physical topology with the corresponding fault information and operation and maintenance information, thereby achieving a structured and systematic expression of the distribution network structure and operation.

[0096] In one embodiment, step S204 above, which extracts abnormal paths from the current power grid knowledge graph by matching tree nodes in the current power grid hierarchy tree with graph nodes in the current power grid knowledge graph, includes the following steps: extracting the association relationships between matching graph nodes of tree nodes in the current power grid hierarchy tree from the current power grid knowledge graph to obtain a current association relationship set; the matching graph nodes of tree nodes are the graph nodes in the current power grid knowledge graph that match the tree nodes; in the current association relationship set, determining the target association relationships that differ from the historical association relationship set; the historical association relationship set is obtained based on the association relationships between matching graph nodes of tree nodes in the historical power grid hierarchy tree and matching graph nodes in the historical power grid knowledge graph; and obtaining the abnormal paths based on the target association relationships.

[0097] Among them, the set of historical relationships is obtained by matching the relationships between tree nodes in the historical power grid hierarchy tree of the distribution network and nodes in the historical power grid knowledge graph of the distribution network.

[0098] In various embodiments of this application, the matched tree node and graph node refer to the tree node and graph node in the same target node pair.

[0099] In this embodiment, the server locates matching graph nodes of tree nodes in the current power grid hierarchical tree within the current power grid knowledge graph, and extracts the associations between the matching graph nodes to obtain a current set of associations. Then, the server compares the current set of associations with a historical set of associations extracted based on the historical power grid hierarchical tree and the historical power grid knowledge graph. Based on the comparison results, the server identifies target associations that differ from the current set of associations. Next, based on the target associations and the graph nodes connected to them, an abnormal path is obtained. For example, in the current set of associations, there exists an association between graph node G1 and graph node G2, but in the historical set of associations, there is no association between graph node G1 and graph node G2. Therefore, the electronic device determines that the path from graph node G1 to graph node G2 is an abnormal path.

[0100] In this embodiment, by matching the tree nodes with the graph nodes, it is possible to extract the abnormal paths that have changed in the current power grid knowledge graph.

[0101] In one embodiment, there are multiple abnormal paths.

[0102] like Figure 3 As shown, step S206 above, based on the distribution of the matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree, determines the current operating status of the distribution network, including the following steps:

[0103] Step S302: For each abnormal path, based on the distribution of the matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree, determine the path penetration information, lateral impact information and vertical impact information of the abnormal path at each level.

[0104] Step S304: For each level, integrate the horizontal and vertical impact information of the abnormal path at the level to determine the path impact information of the abnormal path at the level.

[0105] Step S306: Cluster the abnormal paths by using the path influence information of each abnormal path at each level to obtain at least one abnormal path cluster.

[0106] Step S308: For each abnormal path cluster, the vertical diffusion speed information of the abnormal path cluster between each level and the horizontal influence information under each level are fused to obtain the path cluster influence information of the abnormal path cluster.

[0107] Step S310: Integrate the path cluster impact information of each abnormal path cluster, as well as the path penetration information and lateral impact information of each abnormal path at each level, to obtain the current operating status of the distribution network.

[0108] Among them, the vertical diffusion speed information is determined based on the abnormal time of the abnormal path cluster at each level. The abnormal time of the abnormal path cluster at each level is the time when the target graph node of the abnormal path cluster at the level becomes abnormal. The target graph node is the abnormal graph node that matches the matching tree node in the level. Furthermore, the abnormal time of the abnormal path cluster at each level refers to the time when the earliest abnormal target graph node among the target graph nodes becomes abnormal.

[0109] Among them, the path penetration information of the abnormal path under the level is used to characterize the level depth of the level and the number of matching tree nodes of the abnormal graph node in the abnormal path under the level; the matching tree node of the abnormal graph node in the abnormal path under the level refers to the matching tree node of the abnormal graph node in the abnormal path that is located under the level.

[0110] Among them, the horizontal impact information of the abnormal path at the level is used to characterize the horizontal impact range of the abnormal path at the level. For example, the proportion of the matching tree nodes of the abnormal graph nodes in the abnormal path at the level and the tree nodes at the level.

[0111] Among them, the vertical impact information of the abnormal path under the level is used to characterize the vertical impact depth of the abnormal path from the first level to the current level. For example, the transmission of the horizontal impact information of each level that the abnormal path passes through in the process of reaching the current level.

[0112] In this embodiment, for each abnormal path, firstly, the server determines the path penetration information, lateral impact information, and vertical impact information of the abnormal path at each level based on the distribution of the matching tree nodes of the abnormal graph nodes in the current power grid hierarchical tree; then, for each level, the server fuses the lateral impact information and vertical impact information of the abnormal path at that level to obtain the path impact information of the abnormal path at that level; finally, the server uses the path impact information of the abnormal path at each level as the path feature of the abnormal path.

[0113] Then, the server clusters the abnormal paths based on their path characteristics to obtain at least one abnormal path cluster, and each abnormal path cluster includes at least one abnormal path.

[0114] Next, for each abnormal path cluster, firstly, for each level, the server determines the target graph node that matches the matching tree node in the abnormal graph node of the abnormal path cluster, and identifies the earliest abnormal target graph node among the target graph nodes. Then, the time when the earliest abnormal target graph node becomes abnormal is determined as the abnormal time of the abnormal path cluster at that level. Then, based on the time difference between the abnormal times of adjacent levels, the server determines the vertical diffusion speed information of the abnormal path cluster between each level. Then, the server weightedly fuses the vertical diffusion speed information of the abnormal path cluster between each level and the horizontal influence information at each level to obtain the path cluster influence information of the abnormal path cluster.

[0115] Finally, the server integrates the path cluster impact information of each abnormal path cluster, as well as the path penetration information and lateral impact information of each abnormal path at each level, from multiple dimensions to obtain the current operating status of the distribution network.

[0116] In this embodiment, by matching the distribution of the abnormal graph nodes included in the abnormal path in the current power grid hierarchical tree, the current operating status of the distribution network can be objectively and accurately determined.

[0117] In one embodiment, step S302 above, based on the distribution of matching tree nodes of abnormal graph nodes in the abnormal path in the current power grid hierarchical tree, determines the path penetration information, lateral impact information, and vertical impact information of the abnormal path at each level, including the following steps: For each level, based on the number of abnormal graph nodes of the abnormal path, the number of each level, and the number of matching tree nodes of the abnormal path at the level, the path penetration information of the abnormal path at the level is obtained; based on the number of matching tree nodes of the abnormal path at the level and the number of tree nodes in the level, the lateral impact information of the abnormal path at the level is obtained; based on the lateral impact information of the abnormal path at the level and at the level above the level, as well as the structural similarity information between the levels above the level, the hierarchical transmission information of the abnormal path between the levels above the level is obtained; based on the level and the level preceding the level, the target level is obtained; based on the hierarchical transmission information between the target levels, the vertical impact information of the abnormal path at the level is obtained.

[0118] In this embodiment, for each abnormal path and each level, firstly, the server obtains the path penetration information of the abnormal path at that level based on the number of abnormal graph nodes of the abnormal path, the number of levels, and the number of matching tree nodes of the abnormal path at that level; the server calculates the path penetration information of the abnormal path at the level based on the following formula 1:

[0119] (Formula 1)

[0120] in, Indicates the number of each abnormal path. Indicates the first An abnormal path, ; Indicates the quantity of each level. Indicates the first Each level ; Indicates the first The abnormal path in the first... The path penetration information at each level is also called the path penetration contribution value; Indicates the first The number of abnormal graph nodes in each abnormal path; Indicates the first The abnormal path in the first... The number of matching tree nodes at each level.

[0121] Next, the server obtains the lateral impact information of the abnormal path at that level based on the number of matching tree nodes and the number of tree nodes in that level; the server calculates the lateral impact information of the abnormal path at that level based on the following formula 2:

[0122] (Formula 2)

[0123] in, Indicates the first The abnormal path in the first... The horizontal influence information at each level is also called the level coverage. Indicates the first The number of tree nodes in each level.

[0124] Then, based on the lateral impact information of the abnormal path at this level, the lateral impact information of the abnormal path at the level above this level, and the structural similarity information between the level above and this level, the server obtains the hierarchical propagation information of the abnormal path between the level above and this level; the server calculates the hierarchical propagation information of the abnormal path between two adjacent levels based on the following formula 3:

[0125] (Formula 3)

[0126] in, Indicates the first The abnormal path in the first... Each level and the The hierarchical transfer of information between levels is also called hierarchical value transfer. Indicates the first Each level and the The structural similarity information between levels, also known as structural similarity, can be obtained based on the tightness of the connection and functional relevance between the topological nodes corresponding to the tree nodes in the level.

[0127] Finally, the server combines this level and all levels preceding it into a target level set, and calculates the sum of the hierarchical information passed between levels in the target level set to obtain the vertical impact information of the abnormal path at this level; the server calculates the vertical impact information of the abnormal path at the level based on the following formula 4:

[0128] (Formula 4)

[0129] in, Indicates the first The abnormal path in the first... Vertical impact information at each level.

[0130] In this embodiment, in step S304 above, for each level, the horizontal and vertical impact information of the abnormal path under the level is fused to determine the path impact information of the abnormal path under the level. The server calculates the path impact information of the abnormal path under the level based on the following formula 5:

[0131] (Formula 5)

[0132] in, Indicates the first The abnormal path in the first... Information on the impact of paths at each level; Indicates the weight of horizontal influence. This indicates the weight of the vertical influence.

[0133] In this embodiment, the impact of abnormal paths on the levels in the current power grid hierarchy tree is evaluated from multiple dimensions (horizontal and vertical) by the distribution of matching tree nodes of abnormal graph nodes included in the abnormal path in the current power grid hierarchy tree.

[0134] In one embodiment, before fusing the longitudinal diffusion speed information of the abnormal path cluster between each level and the lateral influence information of each abnormal path in the abnormal path cluster at each level in step S308 above to obtain the path cluster influence information of the abnormal path cluster, the following steps are also included: obtaining the diffusion speed between each adjacent level based on the time difference between the abnormal times corresponding to each adjacent level and the level difference between each adjacent level, and obtaining the longitudinal diffusion speed information of the abnormal path cluster between each level based on the maximum diffusion speed among the diffusion speeds between each adjacent level.

[0135] Step S308 above, which integrates the longitudinal diffusion speed information of the abnormal path cluster between each level and the lateral influence information of each abnormal path in the abnormal path cluster at each level, to obtain the path cluster influence information of the abnormal path cluster, includes the following steps: integrating the average value of the longitudinal diffusion speed information of the abnormal path cluster between each level and the lateral influence information of each abnormal path in the abnormal path cluster at each level to obtain the path cluster influence information of the abnormal path cluster.

[0136] In this implementation, for each abnormal path cluster, firstly, for each group of adjacent levels, the server calculates the time difference between the abnormal times corresponding to the adjacent levels in that group and the level difference between the adjacent levels in that group (the level difference between each group of adjacent levels is 1). Then, the ratio of the level difference to the time difference is calculated to obtain the diffusion rate between the adjacent levels in that group. For example, if the abnormal time of abnormal path cluster 1 in level 2 is t=10s and the abnormal time in level 3 is t=15s, the level difference is 3-2=1, the time difference is 15-10=5s, and the diffusion rate is 1 / 5=0.2 levels / second. Next, the server finds the maximum diffusion rate among the diffusion rates between each adjacent level to obtain the vertical diffusion rate information of the abnormal path cluster between each level, which is also called the level diffusion degree.

[0137] Furthermore, in practical applications, after obtaining the diffusion speed between each group of adjacent layers, the server can also normalize the diffusion speed between each group of adjacent layers using a preset diffusion speed threshold as the denominator to obtain the target diffusion speed between each group of adjacent layers. Then, the server can find the maximum target diffusion speed among the target diffusion speeds between each group of adjacent layers as the vertical diffusion speed information.

[0138] In this embodiment, for each abnormal path cluster, firstly, the server calculates the average value of the lateral influence information of each abnormal path in the cluster at each level. Then, the server fuses the vertical diffusion velocity information of the abnormal path cluster between each level and the average value of the lateral influence information of each abnormal path in the cluster at each level to obtain the path cluster influence information of the abnormal path cluster. The server calculates the path cluster influence information of the abnormal path cluster based on the following formula 6:

[0139] (Formula 6)

[0140] in, Indicates the number of each abnormal path cluster. Indicates the first A cluster of abnormal paths, ; Indicates the first Path cluster impact information for an abnormal path cluster; This indicates the vertical diffusion rate information of abnormal path clusters between different levels. Indicates the weight of the longitudinal diffusion velocity; Indicates the first The average value of the horizontal impact information of each abnormal path in each abnormal path cluster at each level. This represents the average horizontal impact information weight.

[0141] In this embodiment, the server can evaluate the overall impact of abnormal path clusters on the power distribution network from multiple dimensions (horizontal and vertical) on the abnormal path clusters.

[0142] In one embodiment, in step S310 above, by fusing the path cluster impact information of each abnormal path cluster with the path penetration information and lateral impact information of each abnormal path at each level, the server calculates the current operating status of the distribution network using the following formula 7:

[0143] (Formula 7)

[0144] in, It indicates the current operating status of the power distribution network; This represents the first weight corresponding to the path cluster impact information of all abnormal path clusters; This represents the second weight corresponding to the path penetration information of all abnormal paths at each level. This represents the third weight corresponding to the horizontal influence information at each level.

[0145] In one embodiment, such as Figure 4 As shown, step S208 above, which predicts the operating status of the distribution network based on the current operating status, the current power grid hierarchy tree, and the current power grid knowledge graph, to obtain the predicted operating status result of the distribution network, includes the following steps:

[0146] Step S402: Extract the evolution strategy of the target entity associated with the abnormal path from the current power grid knowledge graph.

[0147] Step S404: According to the hierarchical relationship of the current power grid hierarchy tree and the evolution strategy of the target entity, the current operating status is taken as the initial operating state of the entity in the distribution network, and the operating state of the entity in the distribution network is deduced to obtain the predicted operating state of the entity in the distribution network.

[0148] Step S406: Based on the predicted operating status of entities in the distribution network, obtain the predicted operating status result of the distribution network.

[0149] In this embodiment, firstly, the server extracts the evolution strategy of the target entity associated with the abnormal path from the current power grid knowledge graph; then, according to the hierarchical relationship of the current power grid hierarchy tree and the evolution strategy of the target entity, the server deduces the operating state of the entity in the distribution network with the current operating status as the initial operating state of the entity in the distribution network, and obtains the predicted operating state of the entity in the distribution network; finally, the server obtains the predicted operating status result of the distribution network based on the predicted operating state of the entity in the distribution network.

[0150] Among them, the evolution direction of the target entity's evolution strategy state and the conditions for the generation and extinction of the association relationship.

[0151] In this embodiment, the problem of "disconnect between static model and real-time state" in traditional simulation is solved by extrapolation and prediction. The results not only fit the actual hierarchical structure of the distribution network, but also reflect the dynamic interaction between entities, providing a forward-looking basis for real-time scheduling and fault handling of the distribution network.

[0152] In one embodiment, step S208 above, which predicts the operating status of the distribution network based on the current operating status, the current power grid hierarchy tree, and the current power grid knowledge graph to obtain the distribution network operating status prediction result, further includes the following steps:

[0153] Based on the combination of the current operating status and the current power grid hierarchy tree, abnormal paths are mapped to the corresponding levels of the current power grid hierarchy tree, and the topological features and entity interaction features of the abnormal paths at each level after mapping are extracted to obtain the abnormal path features at each level. The topological features include the node level span and node connection density covered by the path, and the entity interaction features include the state transition frequency and the rate of change of association strength of entities in the path.

[0154] From the entity evolution strategies in the current power grid knowledge graph, target entity evolution strategies for path adaptation based on abnormal path characteristics are extracted; the entity evolution strategy includes the state evolution direction of the entity under the influence of abnormal paths and the conditions for the generation and destruction of association relationships.

[0155] Based on the target entity evolution strategy, the current state of each entity in the abnormal path is deduced and initialized to determine the initial state transition direction of the entity under the action of the evolution strategy. Based on the hierarchical logic of the current power grid hierarchy tree and the initial state transition direction, the dynamic changes of the relationship between adjacent hierarchical entities are deduced to obtain the dynamic relationship evolution result. Based on the dynamic relationship evolution result, the entity evolution strategy of the current power grid knowledge graph and the hierarchical logic of the current power grid hierarchy tree, the operation status prediction result is determined. For example, based on the dynamic correlation evolution results, paths spanning multiple levels in abnormal paths are fused and deduced to obtain composite abnormal paths; the composite abnormal paths are combined with the entity evolution strategy of the current power grid knowledge graph to construct path constraints for entity evolution; based on the dynamic correlation evolution results and path constraints, the states of entities at each level of the current power grid hierarchical tree are collaboratively deduced to obtain the deduced states of entities at each level; the deduced states of entities at each level are aggregated according to the hierarchical logic of the current power grid hierarchical tree to form a complete deduction path network; and based on the entity evolution strategy of the current power grid knowledge graph, the evolution trend of entity correlation relationships in the deduction path network is analyzed to obtain the operational status prediction results.

[0156] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0157] Based on the same inventive concept, this application also provides a distribution network operation status prediction device for implementing the above-mentioned distribution network operation status prediction method. The solution provided by this device is similar to the implementation scheme described in the above method. Therefore, the specific limitations of one or more distribution network operation status prediction device embodiments provided below can be found in the limitations of the distribution network operation status prediction method described above, and will not be repeated here.

[0158] In one embodiment, such as Figure 5 As shown, a power distribution network operation status prediction device is provided, including: a power grid information acquisition module 502, an abnormal path extraction module 504, an operation status determination module 506, and an operation status prediction module 508, wherein:

[0159] The power grid information acquisition module 502 is used to reconstruct the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network based on the current operating data of the distribution network, so as to obtain the current power grid hierarchy tree and current power grid knowledge graph of the distribution network. The power grid hierarchy tree of the distribution network is used to represent the hierarchical relationship between topological nodes in the physical topology of the distribution network. The power grid hierarchy tree includes multiple levels, each level includes at least one tree node, and each tree node represents a topological node. The power grid knowledge graph of the distribution network is used to represent the association relationship between entities in the distribution network. The entities in the distribution network include topological nodes and corresponding fault information and operation and maintenance information. Each graph node in the power grid knowledge graph represents an entity.

[0160] The abnormal path extraction module 504 is used to extract abnormal paths from the current power grid knowledge graph by matching the tree nodes in the current power grid hierarchy tree with the graph nodes in the current power grid knowledge graph; the graph nodes included in the abnormal path are abnormal graph nodes.

[0161] The operational status determination module 506 is used to determine the current operational status of the distribution network based on the distribution of matching tree nodes of abnormal graph nodes in the abnormal path in the current power grid hierarchical tree; the matching tree nodes of the abnormal graph nodes are the tree nodes in the current power grid hierarchical tree that match the abnormal graph nodes; the current operational status is used to characterize the operational status of entities in the distribution network.

[0162] The operation status prediction module 508 is used to predict the operation status of the distribution network based on the current operation status, the current power grid hierarchy tree, and the current power grid knowledge graph, and obtain the operation status prediction result of the distribution network.

[0163] In one embodiment, the power grid information acquisition module 502 is further configured to update the hierarchical relationship between tree nodes in the historical power grid hierarchical tree based on the operating data to obtain the current power grid hierarchical tree; and update the association relationship between graph nodes in the historical power grid knowledge graph based on the matching relationship between tree nodes in the current power grid hierarchical tree and graph nodes in the historical power grid knowledge graph to obtain the current power grid knowledge graph.

[0164] In one embodiment, the abnormal path extraction module 504 is further configured to extract the association relationships between matching graph nodes of tree nodes in the current power grid hierarchical tree from the current power grid knowledge graph to obtain a current association relationship set; the matching graph nodes of tree nodes are graph nodes in the current power grid knowledge graph that match the tree nodes; in the current association relationship set, target association relationships that differ from the historical association relationship set are determined; the historical association relationship set is obtained based on the association relationships between matching graph nodes of tree nodes in the historical power grid hierarchical tree in the historical power grid knowledge graph; and the abnormal path is obtained according to the target association relationship.

[0165] In one embodiment, there are multiple abnormal paths.

[0166] The operational status determination module 506 is also used to, for each abnormal path, determine the path penetration information, lateral impact information, and vertical impact information of the abnormal path at each level based on the distribution of the matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree; for each level, it fuses the lateral impact information and vertical impact information of the abnormal path at the level to determine the path impact information of the abnormal path at the level; it clusters each abnormal path according to the path impact information of each abnormal path at each level to obtain at least one abnormal path cluster; and for each abnormal path cluster, it fuses the abnormal path information... The path cluster impact information of abnormal path clusters is obtained by combining the longitudinal diffusion speed information of path clusters between different levels and the lateral impact information of path clusters at different levels. The longitudinal diffusion speed information is determined based on the abnormal time of abnormal path clusters at different levels. The abnormal time of abnormal path clusters at different levels is the time when the target graph node of the abnormal path cluster at the different levels becomes abnormal. The target graph node is the abnormal graph node that matches the matching tree node in the different levels. The current operating status of the distribution network is obtained by fusing the path cluster impact information of each abnormal path cluster with the path penetration information and lateral impact information of each abnormal path at different levels.

[0167] In one embodiment, the operational status determination module 506 is further configured to, for each level, obtain path penetration information of the abnormal path under the level based on the number of abnormal graph nodes of the abnormal path, the number of each level, and the number of matching tree nodes of the abnormal path under the level; obtain lateral impact information of the abnormal path under the level based on the number of matching tree nodes of the abnormal path under the level and the number of tree nodes in the level; obtain hierarchical transmission information of the abnormal path between the upper level and the level above the level based on the lateral impact information of the abnormal path under the level and the structural similarity information between the upper level and the level; obtain a target level set according to the level and the level before the level; and obtain vertical impact information of the abnormal path under the level based on the hierarchical transmission information between the levels in the target level set.

[0168] In one embodiment, the operational status determination module 506 is further configured to obtain the diffusion speed between adjacent levels based on the time difference between the abnormal times corresponding to each adjacent level and the level difference between each adjacent level; obtain the longitudinal diffusion speed information of the abnormal path cluster between each level based on the maximum diffusion speed among the diffusion speeds between adjacent levels; and fuse the longitudinal diffusion speed information of the abnormal path cluster between each level and the lateral influence information of each abnormal path in the abnormal path cluster under each level to obtain the path cluster influence information of the abnormal path cluster, including: fusing the average value of the longitudinal diffusion speed information of the abnormal path cluster between each level and the lateral influence information of each abnormal path in the abnormal path cluster under each level to obtain the path cluster influence information of the abnormal path cluster.

[0169] In one embodiment, the operation status prediction module 508 is further configured to extract the target entity evolution strategy associated with the abnormal path from the current power grid knowledge graph; according to the hierarchical relationship of the current power grid hierarchy tree and the target entity evolution strategy, taking the current operation status as the initial operation state of the entity in the distribution network, the operation state of the entity in the distribution network is deduced to obtain the predicted operation state of the entity in the distribution network; and based on the predicted operation state of the entity in the distribution network, the operation status prediction result of the distribution network is obtained.

[0170] Each module in the aforementioned power distribution network operation status prediction device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0171] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows. Figure 6As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores operational data of the power distribution network. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for predicting the operational status of a power distribution network.

[0172] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0173] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0174] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above-described method embodiments.

[0175] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0176] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0177] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0178] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for predicting the operational status of a power distribution network, characterized in that, The method includes: Based on the current operating data of the distribution network, the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network are reconstructed to obtain the current power grid hierarchy tree and current power grid knowledge graph of the distribution network. The power grid hierarchy tree of the distribution network is used to represent the hierarchical relationship between topological nodes in the physical topology of the distribution network. The power grid hierarchy tree includes multiple levels, each level includes at least one tree node, and each tree node represents a topological node. The power grid knowledge graph of the distribution network is used to represent the association relationship between entities in the distribution network. The entities in the distribution network include the topological nodes and corresponding fault information and operation and maintenance information. Each graph node in the power grid knowledge graph represents an entity. By matching the tree nodes in the current power grid hierarchical tree with the graph nodes in the current power grid knowledge graph, abnormal paths are extracted from the current power grid knowledge graph; the graph nodes included in the abnormal paths are abnormal graph nodes. Based on the distribution of matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree, the current operating status of the distribution network is determined; the matching tree nodes of the abnormal graph nodes are the tree nodes in the current power grid hierarchical tree that match the abnormal graph nodes; the current operating status is used to characterize the operating status of entities in the distribution network. Based on the current operating status, the current power grid hierarchy tree, and the current power grid knowledge graph, the operating status of the distribution network is predicted, and the predicted operating status result of the distribution network is obtained.

2. The method according to claim 1, characterized in that, The process involves reconstructing the historical power grid hierarchy tree and historical power grid knowledge graph of the power distribution network based on its current operational data, resulting in the current power grid hierarchy tree and current power grid knowledge graph of the power distribution network. This includes: Based on the operational data, the hierarchical relationship between tree nodes in the historical power grid hierarchy tree is updated to obtain the current power grid hierarchy tree; Based on the matching relationship between the tree nodes in the current power grid hierarchical tree and the graph nodes in the historical power grid knowledge graph, the association relationship between the graph nodes in the historical power grid knowledge graph is updated to obtain the current power grid knowledge graph.

3. The method according to claim 1, characterized in that, The step of extracting abnormal paths from the current power grid knowledge graph by matching the tree nodes in the current power grid hierarchical tree with the graph nodes in the current power grid knowledge graph includes: From the current power grid knowledge graph, the association relationships between matching graph nodes of tree nodes in the current power grid hierarchical tree are extracted to obtain the current association relationship set; the matching graph nodes of the tree nodes are the graph nodes in the current power grid knowledge graph that match the tree nodes; In the current set of associations, target associations that differ from the historical set of associations are identified; the historical set of associations is obtained based on the associations between tree nodes in the historical power grid hierarchical tree and matching graph nodes in the historical power grid knowledge graph. The abnormal path is obtained based on the target association relationship.

4. The method according to claim 1, characterized in that, There are multiple abnormal paths; The determination of the current operating status of the distribution network based on the distribution of matching tree nodes of the abnormal graph nodes in the abnormal path within the current power grid hierarchical tree includes: For each abnormal path, based on the distribution of the matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree, the path penetration information, lateral impact information and vertical impact information of the abnormal path at each level are determined. For each level, the horizontal and vertical impact information of the abnormal path at that level are fused to determine the path impact information of the abnormal path at that level; By using the path influence information of each abnormal path at each level, each abnormal path is clustered to obtain at least one abnormal path cluster. For each abnormal path cluster, the vertical diffusion speed information of the abnormal path cluster between each level and the lateral influence information of each abnormal path in the abnormal path cluster under each level are fused to obtain the path cluster influence information of the abnormal path cluster; the vertical diffusion speed information is determined based on the abnormal time of the abnormal path cluster under each level, the abnormal time of the abnormal path cluster under the level is the time when the target graph node of the abnormal path cluster under the level becomes abnormal, and the target graph node is the abnormal graph node that matches the matching tree node in the level; By integrating the path cluster impact information of each abnormal path cluster, as well as the path penetration information and lateral impact information of each abnormal path at each level, the current operating status of the power distribution network is obtained.

5. The method according to claim 4, characterized in that, The distribution of matching tree nodes based on the abnormal graph nodes in the abnormal path within the current power grid hierarchical tree, determining the path penetration information, lateral impact information, and vertical impact information of the abnormal path at each level, includes: For each level, based on the number of abnormal graph nodes of the abnormal path, the number of each level, and the number of matching tree nodes of the abnormal path under the level, the path penetration information of the abnormal path under the level is obtained. Based on the number of matching tree nodes of the abnormal path at the level and the number of tree nodes in the level, the lateral impact information of the abnormal path at the level is obtained. Based on the lateral influence information of the abnormal path at the level and the level above the level, as well as the structural similarity information between the level above the level and the level, the hierarchical transmission information of the abnormal path between the level above the level and the level is obtained. Based on the stated level and the levels preceding it, a target level set is obtained. Based on the hierarchical transfer information between the levels in the target level set, the vertical impact information of the abnormal path under the stated level is obtained.

6. The method according to claim 4, characterized in that, Before fusing the vertical diffusion velocity information of the abnormal path cluster between each level and the lateral influence information of each abnormal path in the abnormal path cluster at each level to obtain the path cluster influence information of the abnormal path cluster, the method further includes: Based on the time difference between the abnormal times corresponding to each adjacent level and the level difference between each adjacent level, the diffusion speed between each adjacent level is obtained. Based on the maximum diffusion speed among the diffusion speeds between each adjacent level, the longitudinal diffusion speed information of the abnormal path cluster between each level is obtained. The path cluster influence information of the abnormal path cluster is obtained by fusing the vertical diffusion velocity information of the abnormal path cluster between each level and the lateral influence information of each abnormal path in the abnormal path cluster at each level, including: The path cluster influence information of the abnormal path cluster is obtained by fusing the longitudinal diffusion speed information of the abnormal path cluster between each level and the average value of the lateral influence information of each abnormal path in the abnormal path cluster at each level.

7. The method according to any one of claims 1 to 6, characterized in that, The prediction of the distribution network's operational status based on the current operational status, the current power grid hierarchy tree, and the current power grid knowledge graph, to obtain the distribution network's operational status prediction result, includes: Extract the target entity evolution strategy associated with the abnormal path from the current power grid knowledge graph; Based on the hierarchical relationship of the current power grid hierarchy tree and the target entity evolution strategy, taking the current operating status as the initial operating state of the entities in the distribution network, the operating state of the entities in the distribution network is deduced to obtain the predicted operating state of the entities in the distribution network. Based on the predicted operating status of the entities in the power distribution network, the predicted operating status of the power distribution network is obtained.

8. A device for predicting the operational status of a power distribution network, characterized in that, The device includes: The power grid information acquisition module is used to reconstruct the historical power grid hierarchy tree and historical power grid knowledge graph of the distribution network based on the current operating data of the distribution network, so as to obtain the current power grid hierarchy tree and current power grid knowledge graph of the distribution network. The power grid hierarchy tree of the distribution network is used to represent the hierarchical relationship between topological nodes in the physical topology structure of the distribution network. The power grid hierarchy tree includes multiple levels, each level includes at least one tree node, and each tree node represents a topological node. The power grid knowledge graph of the distribution network is used to represent the association relationship between entities in the distribution network. The entities in the distribution network include the topological nodes and corresponding fault information and operation and maintenance information. Each graph node in the power grid knowledge graph represents an entity. An abnormal path extraction module is used to extract abnormal paths from the current power grid knowledge graph by matching the tree nodes in the current power grid hierarchical tree with the graph nodes in the current power grid knowledge graph; the graph nodes included in the abnormal path are abnormal graph nodes. The operational status determination module is used to determine the current operational status of the distribution network based on the distribution of matching tree nodes of the abnormal graph nodes in the abnormal path in the current power grid hierarchical tree; the matching tree nodes of the abnormal graph nodes are the tree nodes in the current power grid hierarchical tree that match the abnormal graph nodes; the current operational status is used to characterize the operational status of entities in the distribution network; The operation status prediction module is used to predict the operation status of the distribution network based on the current operation status, the current power grid hierarchy tree, and the current power grid knowledge graph, and obtain the operation status prediction result of the distribution network.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images