Fault positioning method of power distribution network, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a fault positioning method of a power distribution network, electronic equipment and a storage medium. The method comprises the following steps: performing fault detection based on regional operation data and state data of a line region in the power distribution network to obtain a detection result; when the detection result is abnormal, obtaining the topological structure of the line area and the operation data of the detection node; based on the topological structure and the operation data of the detection nodes, performing node classification on the detection nodes to obtain fault nodes; selecting a candidate line associated with the fault node in the line area according to the topological structure; and determining the position of a fault point according to the fault detection signal of the candidate line. The method is used for achieving the effect of improving the fault positioning efficiency.

Description

Technical Field

[0001] The present application relates to the fields of power grid fault location technology and artificial intelligence technology, and in particular to a distribution network fault location method, electronic equipment, and storage medium. Background Art

[0002] Failures in the distribution network will affect users' electricity consumption. In order to ensure stable power supply, it is necessary to accurately locate the fault in the distribution network when it fails, so as to facilitate timely repairs and restore power supply.

[0003] In the related art, when a fault occurs in a distribution network, only the line area where the fault occurs can be determined, and each device or line in the line area needs to be checked one by one to determine the fault point of the distribution network. This method is inefficient. Summary of the invention

[0004] The embodiments of the present application provide a fault location method, an electronic device and a storage medium for a distribution network, so as to achieve the effect of improving the efficiency of fault location.

[0005] In a first aspect, an embodiment of the present application provides a fault location method for a distribution network, comprising: performing fault detection based on regional operation data and status data of a line area in the distribution network to obtain a detection result; when the detection result is abnormal, obtaining a topological structure of the line area and operation data of a detection node; based on the topological structure and operation data of the detection node, performing node classification on the detection node to obtain a faulty node; selecting a candidate line associated with the faulty node in the line area according to the topological structure; and determining the location of the fault point according to a fault detection signal of the candidate line.

[0006] In a possible implementation, fault detection is performed based on regional operation data and status data of a line area in a distribution network to obtain a detection result, including: extracting features from the regional operation data of the line area in the distribution network to obtain regional data features; determining state features based on a first state of an environment in the line area and a second state of line equipment in the line area; fusing the regional data features and the state features to obtain fused features; and performing fault detection on the fused features to obtain a detection result.

[0007] In one possible implementation, based on the topological structure and the operating data of the detection nodes, the detection nodes are classified to obtain faulty nodes, including: determining an adjacency matrix according to the topological structure; performing feature extraction on the operating data of the detection nodes to obtain node data features; and performing node classification on the node data features and the adjacency matrix through a graph neural network to obtain faulty nodes.

[0008] In a possible implementation, the fault location method of the distribution network also includes: determining the operating data samples of the node samples in the line area samples based on the operating data big data pool; determining the adjacency matrix samples based on the topological structure samples of the line area samples; performing feature extraction on the operating data samples of the node samples to obtain the node data feature samples; inputting the node data feature samples and the adjacency matrix samples into the initial graph neural network to obtain the node failure probability; calculating the loss value based on the node failure probability and the node label of the operating data sample; and training the initial graph neural network based on the loss value to obtain the graph neural network.

[0009] In a possible implementation, selecting a candidate line in a line area according to a fault node includes: for each fault node, if the targeted fault node is an isolated fault node, taking a first line in the line area connected to the targeted fault node as a candidate line; if there is a second line between the targeted fault node and another fault node, taking the second line as a candidate line.

[0010] In a possible implementation, the fault detection signal is a traveling wave signal; determining the fault point location according to the fault detection signal of the candidate line includes: acquiring the traveling wave signal through a traveling wave sensor of the candidate line; and determining the fault point location on the candidate line according to the traveling wave signal.

[0011] In a second aspect, an embodiment of the present application provides a fault location device for a distribution network, comprising:

[0012] A regional detection module is used to perform fault detection based on regional operation data and status data of the line area in the distribution network to obtain a detection result;

[0013] An acquisition module, used to acquire the topological structure of the line area and the operation data of the detection node when the detection result is abnormal;

[0014] A node detection module is used to classify detection nodes based on the topological structure and the operation data of the detection nodes to obtain faulty nodes;

[0015] A selection module, used for selecting a candidate line associated with a faulty node in a line area according to a topological structure;

[0016] The fault location module is used to determine the location of the fault point according to the fault detection signal of the candidate line.

[0017] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory, a processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory, so that the processor executes the first aspect above and / or various possible implementations of the first aspect.

[0018] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementations of the first aspect.

[0019] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.

[0020] The fault location method, electronic device and storage medium of the distribution network provided in the embodiment of the present application perform fault detection based on the regional operation data and status data of the line area in the distribution network to obtain a detection result. When the detection result is abnormal, the topological structure of the line area and the operation data of the detection node are obtained, and the detection node is classified according to the topological structure and the operation data of the detection node to obtain the fault node; that is, when there is a fault in the line area, the fault range is narrowed from the regional level to the node level through the topological structure and the operation data of the detection node, and the candidate line associated with the fault node is selected in the line area through the topological structure, and the fault range is limited to the line level, and then the fault point position is determined through the fault detection signal of the candidate line, and the fault point position is determined based on the candidate line with possible faults. The present application gradually narrows the fault range, so that the amount of data involved in the calculation each time is small, and the fault range can be quickly reduced. In addition, the fault point position can be quickly and accurately determined on the candidate line through the fault detection signal, thereby improving the efficiency and accuracy of fault location. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0022] Figure 1 A schematic diagram of a scenario of a method for locating a fault in a distribution network provided in this application;

[0023] Figure 2 Schematic diagram of the process of the fault location method of the distribution network provided in this application Figure 1 ;

[0024] Figure 3 A schematic diagram of the structure of the fault location device provided for this application;

[0025] Figure 4 A schematic diagram showing the topology and fault point on the fault location device provided for this application;

[0026] Figure 5 Schematic diagram of the process of the fault location method of the distribution network provided in this application Figure 2 ;

[0027] Figure 6 A schematic diagram of the structure of a fault location device for a distribution network provided in this application;

[0028] Figure 7 A schematic diagram of the structure of the electronic device provided in this application.

[0029] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0030] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0031] The fault location method of the distribution network provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown in the figure, the data acquisition device 102 of the distribution network and the fault location device 104 communicate through the network; the data storage system can store the data that the fault location device 104 needs to process. The data storage system can be integrated on the fault location device 104, or it can be placed on the cloud or other network servers; the fault location device 104 can be implemented with an independent server or a server cluster composed of multiple servers.

[0032] The regional operation data and status data of the line area can be obtained through the data acquisition device 102, and the fault location device 102 performs fault detection based on the regional operation data and status data to obtain a detection result. When the detection result is abnormal, the topological structure of the line area and the operation data of the detection node are obtained through the data acquisition device 102, and the topological structure and the operation data of the detection node are processed by the fault location device 104 to obtain the fault node, and the candidate line is obtained in the line area through the fault location device 104, and the fault detection signal of the candidate line is obtained through the data acquisition device 102. The fault detection signal is processed by the fault location device 104 to obtain the fault point position.

[0033] In practical applications, the data acquisition device 102 and the fault location device 104 may be implemented separately, or the data acquisition device 102 may be integrated into the fault location device 104 .

[0034] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0035] Figure 2 Schematic diagram of the process of the fault location method of the distribution network provided in this application Figure 1 , the fault location method of the distribution network can be applied to electronic devices, which can be Figure 1 Fault location equipment in Figure 2 As shown, the fault location method of the distribution network includes:

[0036] S201. Perform fault detection based on regional operation data and status data of a line area in a distribution network to obtain a detection result.

[0037] Among them, the distribution network is used to transmit electric energy from the transmission system or power plant to the user end. Its main functions include: power distribution, voltage conversion and power quality control, fault isolation, etc. The line area is a component of the distribution network. The distribution network includes multiple line areas, and different line areas may correspond to different power supply ranges. Usually, a line area is composed of a group of power lines and related equipment.

[0038] It should be noted that, before S201, the step also includes dividing the distribution network lines into a plurality of line areas.

[0039] The regional operation data refers to the total operation data of the line area, which may include regional total voltage data, regional total current data and regional power factor data.

[0040] It should be noted that the regional total voltage data and the regional total current data can represent the power load situation and voltage stability of the entire line area. If there is a fault in the line area, the regional total voltage data and the regional total current data may be abnormal.

[0041] The regional power factor data is the ratio of active power to apparent power in the line area. If there are problems such as excessive reactive power, unbalanced load or aging equipment in the line area, the regional power factor data may be abnormal.

[0042] Among them, the status data may include the environmental status of the line area and the operating status of the line equipment in the line area; under different environmental conditions, the regional operating data has different performances, and extreme environments may cause line area failures; the aging status of the line equipment may cause line area failures.

[0043] Optionally, the fault location device obtains regional operation data of the line area, obtains environmental status and operation status of the line area, performs fault detection on the regional operation data, environmental status and operation status through a first fault detection model, and obtains normal or abnormal detection results.

[0044] Optionally, the fault locating device obtains regional operating data of the line area, obtains the environmental status of the line area and reference operating data under the environmental status, determines difference data based on the reference operating data and the regional operating data, and obtains the operating status of each device in the line area; performs fault detection on the operating status and the difference data through a second fault detection model to obtain a normal or abnormal detection result.

[0045] It should be noted that the detection result is used to indicate whether there is a fault in the line area. When the detection result is normal, it indicates that there is no fault in the line area. When the detection result is abnormal, it indicates that there is a fault in the line area.

[0046] If the detection result is abnormal, the subsequent fault location steps are executed. If the detection result is normal, there is no need to perform subsequent fault location. In the related art, a large amount of data in the line area is directly processed to determine whether there is a fault in the line area, or to determine whether there is a fault and obtain the fault location at the same time. In this way, even if there is no fault in the line area, a large amount of data processing is required. The embodiment of the present application first processes the regional operation data, and only needs to introduce more data for subsequent fault location when the detection result is abnormal. This reduces the overall data processing volume, reduces the burden on hardware resources, and improves the efficiency of the fault location device.

[0047] S202: When the detection result is abnormal, obtain the topological structure of the line area and the operation data of the detection node.

[0048] The lines in the line area are connected through connection points, such as connection points used to connect the main line and the branch line; the topological structure is used to reflect the relationship between the lines connected through the connection points.

[0049] The number of detection nodes can be one or more, and the detection nodes can be at least part of the connection points in the line area; optionally, the detection nodes can be important connection points in the line area, such as the detection nodes are part of the connection points with the most lines connected in the line area; optionally, each connection point in the line area can be a detection node.

[0050] The operation data of the node may be voltage data, current data and power factor data of the node.

[0051] Specifically, when the detection result is abnormal, the fault location device obtains the topology structure and voltage data, current data, and power factor data of each detection node.

[0052] S203: Based on the topological structure and the operation data of the detection nodes, the detection nodes are classified to obtain faulty nodes.

[0053] Among them, some nodes in the detection nodes are among the faulty nodes.

[0054] Specifically, a graph neural network can be used to classify the topological structure and the running data of the detection nodes to obtain the faulty nodes. A graph neural network (GNN) is a neural network used to process graph structure data. Lines and connection points can be naturally represented as a graph structure (topological structure), where lines are edges and connection points are nodes.

[0055] Through graph neural networks, the topological structure of the line area and the operating data of the detection nodes can be processed, and fault detection of the detection nodes can be performed using the reasoning ability of graph neural networks.

[0056] In a possible implementation, based on the topological structure and the operating data of the detection nodes, the detection nodes are classified to obtain faulty nodes, including: determining an adjacency matrix according to the topological structure; performing feature extraction on the operating data of the detection nodes to obtain node data features; and performing node classification on the node data features and the adjacency matrix through a graph neural network to obtain faulty nodes.

[0057] Among them, the adjacency matrix is ​​a two-dimensional array used to describe the connection relationship between connection points in the topological structure. For example, if there are n connection points in the line area (some or all of which are detection nodes), then the adjacency matrix is ​​an n*n matrix. The element Aij in the matrix represents the connection relationship between connection point i and connection point j. If there is a line between connection point i and connection point j, Aij is 1. If there is no line between connection point i and connection point j, Aij is 0.

[0058] Among them, the faulty node is a node that may have a fault among the detection nodes; it should be noted that when a fault exists in a line area, it may affect multiple nodes and multiple lines to become abnormal. Therefore, by detecting the faulty node, the scope of the fault impact can be quickly narrowed.

[0059] Specifically, the fault location device processes the topological structure and the operating data of the detection node through the node fault detection model to determine the faulty node; the node fault detection model includes: a structure conversion network, a first feature extraction network and a graph neural network; the topological structure is processed through the structure conversion network to obtain an adjacency matrix, the operating data is input into the first feature extraction network to obtain node data features, the node data features and the adjacency matrix are input into the graph neural network, the abnormal probability of each detection node is determined through the graph neural network, the faulty node is determined according to the abnormal probability of each detection node, and the faulty node is obtained according to the probability value of the faulty node.

[0060] Among them, the operating data of the detection node includes the voltage data, current data and power factor data of the detection node; feature extraction is performed on the operating data of the detection node to obtain node data features, including: feature extraction is performed on the voltage data of the detection node through a first feature extraction network to obtain node voltage features, feature extraction is performed on the current data of the detection node to obtain node current features, feature extraction is performed on the power factor data of the detection node to obtain node power factor features, and node voltage features, node current features and node power factor features are spliced ​​to obtain node data features.

[0061] In the above embodiment, by performing node classification on the topological structure and the operating data of the detection nodes through the graph neural network, the faulty nodes in the line area can be quickly obtained. In addition, in the process of determining the faulty nodes, only the operating data of the topological structure and the detection nodes are required, and all the data in the line area is not required, which reduces the amount of data processed and improves the efficiency of fault location.

[0062] S204: Select a candidate line associated with the faulty node in the line area according to the topological structure.

[0063] Among them, in the line area, the candidate lines associated with the faulty node are connected with the faulty node.

[0064] Optionally, for each faulty node, the lines in the line area connected to the faulty node may be determined according to the topological structure, thereby obtaining candidate lines.

[0065] Optionally, S204 includes: for each fault node, determining whether the targeted fault node is an isolated fault node based on the topological structure; if the targeted fault node is an isolated fault node, taking the first line connected to the targeted fault node in the line area as a candidate line; if the targeted fault node is not an isolated fault node, taking the second line between the targeted fault node and another fault node as a candidate line.

[0066] An isolated faulty node means that all nodes directly connected to it (connected through a line) are not faulty nodes; if at least one of the nodes directly connected to the faulty node is a faulty node, the faulty node is not an isolated faulty node.

[0067] Specifically, for each faulty node, other nodes directly connected to the faulty node are determined. If other nodes are not faulty nodes, the faulty node is regarded as an isolated faulty node, and all first lines connected to the faulty node are regarded as candidate lines.

[0068] If there is a faulty node among other nodes directly connected to the targeted faulty node, the second line between the targeted faulty node and the faulty node directly connected thereto is used as a candidate line.

[0069] In the above embodiment, if both nodes are faulty nodes, the line between the two nodes is used as a candidate line, which can quickly narrow the range of candidate lines that may be faulty and improve the efficiency of fault location.

[0070] S205: Determine the location of the fault point according to the fault detection signal of the candidate line.

[0071] Among them, the fault detection signal can be a zero-sequence current and zero-sequence voltage signal; under normal circumstances, the three-phase current in the distribution network is symmetrical, so the zero-sequence current (that is, the sum of the three-phase currents) is zero. But when a fault occurs, such as a single-phase grounding fault, the balance of the three-phase current is broken, and the zero-sequence current and voltage will change. By analyzing these changes (such as the amplitude and direction of the zero-sequence current and the amplitude of the zero-sequence voltage, etc.), the location of the fault can be inferred.

[0072] Among them, the fault detection signal can be a traveling wave signal; when a fault occurs, the current and voltage at the fault point will change suddenly, and this change will propagate along the line in the form of electromagnetic waves; these electromagnetic waves are traveling wave signals, which carry information such as the location of the fault point and the fault type; by analyzing the traveling wave signal, the location of the fault point can be determined.

[0073] In a possible implementation, determining the fault point location according to the fault detection signal of the candidate line includes: acquiring a traveling wave signal through a traveling wave sensor of the candidate line; and determining the fault point location on the candidate line according to the traveling wave signal.

[0074] The traveling wave signal includes a voltage traveling wave signal and a current traveling wave signal.

[0075] Specifically, for each candidate line, the fault location device can obtain the traveling wave signal of the candidate line through the traveling wave sensor on the candidate line; if the traveling wave signal of the candidate line is not obtained, it means that there is no fault on the candidate line, and the candidate line is eliminated.

[0076] It should be noted that the traveling wave sensor can detect high-frequency transient traveling waves that occur when a line fault occurs, and obtain traveling wave signals after interference filtering, signal amplification, and analog-to-digital conversion of the high-frequency transient traveling waves.

[0077] The fault location device determines the first propagation time of the traveling wave signal to one end of the candidate line and the second propagation time to the other end of the candidate line for the obtained traveling wave signal of the candidate line, and determines the fault point location according to the first propagation time, the second propagation time and the length of the candidate line.

[0078] It should be noted that if real-time traveling wave signal detection is performed on all lines in the line area to further detect the fault location of the line, that is, the traveling wave signal needs to be detected even when there is no fault in the line, the detection of the traveling wave signal will cause more hardware resources to be occupied, and when a line fault occurs, the fault may affect multiple lines. Therefore, locating the fault point based on the traveling wave signal also involves a large amount of calculation, resulting in low efficiency in fault location. After determining the candidate lines where faults may exist, the fault detection range of the traveling wave signal is narrowed, which greatly reduces the amount of data and improves the efficiency of fault location.

[0079] The fault location method of the distribution network provided by the present application performs fault detection based on the regional operation data and status data of the line area in the distribution network to obtain a detection result. When the detection result is abnormal, the topological structure of the line area and the operation data of the detection node are obtained, and the detection node is classified according to the topological structure and the operation data of the detection node to obtain the fault node; that is, when there is a fault in the line area, the fault range is narrowed from the regional level to the node level through the topological structure and the operation data of the detection node, and the candidate line associated with the fault node is selected in the line area through the topological structure, and the fault range is limited to the line level, and then the fault point position is determined through the fault detection signal of the candidate line, and the fault point position is determined based on the candidate line that may be faulty. The present application gradually narrows the fault range, so that the amount of data involved in the calculation each time is small, and the fault range can be quickly narrowed, and the fault point position can be quickly and accurately determined on the candidate line through the fault detection signal, thereby improving the efficiency and accuracy of fault location.

[0080] In a possible implementation, fault detection is performed based on regional operation data and status data of a line area in a distribution network to obtain a detection result, including: extracting features from the regional operation data of the line area in the distribution network to obtain regional data features; determining state features based on a first state of an environment in the line area and a second state of line equipment in the line area; fusing the regional data features and the state features to obtain fused features; and performing fault detection on the fused features to obtain a detection result.

[0081] Among them, the regional operation data may include regional voltage data, regional current data and regional power factor data; the first state may be one of a lightning state, a strong wind state, a heavy rainfall state, a high temperature state and a normal state; the second state may be one of a healthy state and an old state.

[0082] Specifically, the fault detection model can be used to perform fault detection on regional operation data of the line area in the distribution network to obtain a detection result; the fault detection model includes: a first feature extraction module, an encoder, a feature fusion module and a first classification module.

[0083] The regional operation data of the line area in the distribution network is input into the first feature extraction module to obtain the regional data features; the regional data features can be in the form of feature vectors; the first state of the environment in the line area and the second state of the line equipment in the line area are encoded by an encoder to obtain state features; the state features can be in the form of feature vectors; the regional data features and the state features are input into the feature fusion module to obtain fused features, and the fused features are input into the first classification module to obtain detection results.

[0084] Among them, the regional operation data of the line area in the distribution network is input into the first feature extraction module to obtain the regional data features. The regional voltage data, the regional current data and the regional power factor data can be respectively input into the first feature extraction module, and the feature extraction is performed by the first feature extraction module to obtain the regional voltage features, the regional current features and the regional power factor features respectively, and the regional voltage features, the regional current features and the regional power factor features are spliced ​​to obtain the regional data features.

[0085] In some embodiments, the fault detection model is obtained by training the initial detection model with a line area fault big data pool and a line area normal big data pool; specifically, a negative sample set is determined by the line area fault big data pool, and a positive sample set is determined by the area normal big data pool; the negative sample set includes: a first area operation data sample, a first state data sample and a fault label; the positive sample set includes: a second area operation data sample, a second state data sample and a normal label; the initial detection model is trained with the negative sample set and the positive sample set, including: inputting the first area operation data sample and the first state data sample into the initial detection model to obtain a first prediction result; determining a first loss value through the first prediction result and the fault label, and adjusting the parameters of the initial detection model using the first loss value to obtain a trained initial detection model; inputting the second area operation data sample and the second state data sample into the initial detection model to obtain a second prediction result, determining a second loss value through the second prediction result and the normal label, and adjusting the parameters of the trained initial detection model using the second loss value to obtain a trained initial detection model, until the trained initial detection model converges to obtain a fault detection model.

[0086] In the above embodiment, the amount of data involved is small through the regional operation data, the first state of the environment of the line area and the second state of the line equipment, and the detection result of the line area can be obtained quickly and accurately, thereby improving the efficiency of fault location.

[0087] In a possible implementation, the fault location method of the distribution network also includes: determining the operating data samples of the node samples in the line area samples based on the operating data big data pool; determining the adjacency matrix samples based on the topological structure samples of the line area samples; performing feature extraction on the operating data samples of the node samples to obtain node data feature samples; inputting the node data feature samples and the adjacency matrix samples into the initial graph neural network to obtain the predicted fault node; calculating the loss value based on the predicted fault node and the node label of the operating data sample; and training the initial graph neural network based on the loss value to obtain the graph neural network.

[0088] Among them, the big data pool of operation data includes a large amount of historical operation data of each node in the line area, and the node label corresponding to each historical operation data; the historical operation data is used as the operation data sample; the node label of the operation data sample includes the node label of each node in the line area sample, and each node label can indicate normal or abnormal.

[0089] Specifically, a topological structure sample of the line area sample and a plurality of node samples included in the line area sample are obtained; an adjacency matrix sample is determined according to the topological structure sample; for each node sample of the line area sample, a node data feature sample of the node sample is extracted by a feature extractor, and the adjacency matrix sample of the line area sample and the plurality of node data feature samples under the line area sample are input into the initial graph neural network, and the node failure probabilities corresponding to each of the plurality of node samples are determined by the initial graph neural network, and the loss value is calculated according to the node failure probabilities corresponding to each of the plurality of node samples and the corresponding node labels, and the parameters of the initial graph neural network are adjusted according to the loss value, and the initial graph neural network is iteratively trained according to the above process until the initial graph neural network converges, and the converged initial graph neural network is used as the graph neural network.

[0090] In the above embodiment, the operation data samples and node labels of the node samples in the line area samples are determined according to the operation data big data pool, so that rich sample data for training can be obtained, thereby improving the quality of the trained graph neural network and the accuracy of node fault detection.

[0091] Optionally, after the fault point is located, the topology structure can be displayed and the fault point can be highlighted in the topology result. For example, the fault point in the topology structure is displayed in red and other locations are displayed in green. The fault point in the line area can be determined more intuitively to facilitate emergency repairs by staff.

[0092] like Figure 3 As shown, the fault location device includes: a central control module, an analog input module connected to the central control module, a digital input module, a digital output module, an analog output module, a data storage module, a data communication transmission module, a power module and a data exchange module, the data exchange module is connected to a host computer through a network, the host computer is used to control the fault location device, the digital input module is connected to 3 external interfaces, the digital output module is connected to 3 external interfaces; the analog output module includes a display screen, the display screen is used to display the topological structure of the line area, and highlight the fault point in the topological structure; as shown Figure 4 As shown, the topology is displayed on a display screen of the fault locating device, and the fault point is highlighted in the topology.

[0093] Figure 5 Schematic diagram of the process of the fault location method of the distribution network provided in this application Figure 2 ,like Figure 5 As shown, in this embodiment Figure 2 Based on the embodiment, a method for locating a fault in a distribution network is described in detail. The method includes:

[0094] S501. Determine the operation data samples of the node samples in the line area samples according to the operation data big data pool; determine the adjacency matrix samples according to the topological structure samples of the line area samples; extract features of the operation data samples of the node samples to obtain node data feature samples; input the node data feature samples and the adjacency matrix samples into the initial graph neural network to obtain the node failure probability; calculate the loss value according to the node failure probability and the node label of the operation data sample; train the initial graph neural network based on the loss value to obtain the graph neural network;

[0095] S502, extracting features from the regional operation data of the line area in the distribution network to obtain regional data features; determining state features according to a first state of the environment in the line area and a second state of the line equipment in the line area; fusing the regional data features and the state features to obtain fused features; performing fault detection on the fused features to obtain detection results; when the detection result is abnormal, obtaining the topological structure of the line area and the operation data of the detection node;

[0096] S503, determining an adjacency matrix according to the topological structure; extracting features from the running data of the detection node to obtain node data features; classifying nodes based on the node data features and the adjacency matrix through a graph neural network to obtain faulty nodes;

[0097] S504: for each faulty node, determine whether the faulty node is an isolated faulty node based on the topological structure; if the faulty node is an isolated faulty node, use the first line connected to the faulty node in the line area as a candidate line; if the faulty node is not an isolated faulty node, use the second line between the faulty node and another faulty node as a candidate line;

[0098] S505 . Acquire a traveling wave signal through a traveling wave sensor of the candidate line; and determine the location of a fault point on the candidate line according to the traveling wave signal.

[0099] The fault location method of the distribution network provided by the present application performs fault detection based on the regional operation data and status data of the line area in the distribution network to obtain a detection result. When the detection result is abnormal, the topological structure of the line area and the operation data of the detection node are obtained, and the detection node is classified according to the topological structure and the operation data of the detection node to obtain the fault node; that is, when there is a fault in the line area, the fault range is narrowed from the regional level to the node level through the topological structure and the operation data of the detection node, and the candidate line associated with the fault node is selected in the line area through the topological structure, and the fault range is limited to the line level, and then the fault point position is determined through the fault detection signal of the candidate line, and the fault point position is determined based on the candidate line that may be faulty. The present application gradually narrows the fault range, so that the amount of data involved in the calculation each time is small, and the fault range can be quickly narrowed, and the fault point position can be quickly and accurately determined on the candidate line through the fault detection signal, thereby improving the efficiency and accuracy of fault location.

[0100] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0101] Figure 6 A schematic diagram of the structure of the fault location device for the distribution network provided in this application, such as Figure 6 As shown, the fault location device 60 for the distribution network provided in this embodiment includes:

[0102] The regional detection module 601 is used to perform fault detection based on regional operation data and status data of the line area in the distribution network to obtain a detection result;

[0103] An acquisition module 602 is used to acquire the topological structure of the line area and the operation data of the detection node when the detection result is abnormal;

[0104] The node detection module 603 is used to classify the detection nodes based on the topological structure and the operation data of the detection nodes to obtain the faulty nodes;

[0105] A selection module 604, configured to select a candidate line associated with the faulty node in the line area according to the topological structure;

[0106] The fault location module 605 is used to determine the location of the fault point according to the fault detection signal of the candidate line.

[0107] In one possible implementation, the regional detection module 601 is used to extract features of regional operation data of a line area in a distribution network to obtain regional data features; determine state features based on a first state of an environment in the line area and a second state of line equipment in the line area; fuse the regional data features and the state features to obtain fused features; and perform fault detection on the fused features to obtain detection results.

[0108] The node detection module 603 is used to determine the adjacency matrix according to the topological structure; extract the features of the running data of the detection node to obtain the node data features; classify the nodes according to the node data features and the adjacency matrix through the graph neural network to obtain the faulty nodes.

[0109] In a possible implementation, the fault location device of the distribution network also includes: a graph neural network training module, which is used to determine the operating data samples of the node samples in the line area samples based on the operating data big data pool; determine the adjacency matrix samples based on the topological structure samples of the line area samples; perform feature extraction on the operating data samples of the node samples to obtain node data feature samples; input the node data feature samples and the adjacency matrix samples into the initial graph neural network to obtain the node failure probability; calculate the loss value based on the node failure probability and the node label of the operating data sample; and train the initial graph neural network based on the loss value to obtain the graph neural network.

[0110] In a possible implementation, the selection module 604 is further used to, for each fault node, if the fault node is an isolated fault node, use the first line in the line area connected to the fault node as a candidate line; if there is a second line between the fault node and another fault node, use the second line as a candidate line.

[0111] In a possible implementation, the fault detection signal is a traveling wave signal; the fault location module 605 is further configured to obtain the traveling wave signal through a traveling wave sensor of the candidate line; and determine the location of the fault point on the candidate line according to the traveling wave signal.

[0112] The fault location device for the distribution network provided in this embodiment can execute the fault location method for the distribution network provided in the above method embodiment. The implementation principle and technical effect are similar, and will not be described in detail in this embodiment.

[0113] Figure 7 This is a schematic diagram of the structure of the electronic device provided in this application. Figure 7 As shown, the electronic device 70 provided in this embodiment includes: at least one processor 701 and a memory 702. Optionally, the device 70 further includes a communication component 703. The processor 701, the memory 702 and the communication component 703 are connected via a bus.

[0114] In a specific implementation process, at least one processor 701 executes the computer-executable instructions stored in the memory 702, so that at least one processor 701 executes the above method.

[0115] The specific implementation process of the processor 701 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0116] In the above embodiments, it should be understood that the processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the invention can be directly implemented as a hardware processor, or can be implemented by a combination of hardware and software modules in the processor.

[0117] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (NVM), such as at least one disk storage.

[0118] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.

[0119] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0120] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0121] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0122] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0123] The division of units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0124] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0125] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0126] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0127] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0128] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for locating a fault in a distribution network, characterized in that: include: Perform fault detection based on regional operation data and status data of the line area in the distribution network to obtain detection results; When the detection result is abnormal, obtaining the topological structure of the line area and the operation data of the detection node; Based on the topological structure and the operation data of the detection nodes, the detection nodes are classified to obtain faulty nodes; Selecting a candidate line associated with the faulty node in the line area according to the topological structure; The fault point location is determined according to the fault detection signal of the candidate line.

2. The method according to claim 1, characterized in that The fault detection is performed based on the regional operation data and status data of the line area in the distribution network to obtain the detection result, including: Extract features of regional operation data of the line area in the distribution network to obtain regional data features; Determining a state feature according to a first state of an environment in which the line area is located and a second state of a line device in the line area; Fusing the regional data feature and the state feature to obtain a fused feature; Fault detection is performed on the fused features to obtain a detection result.

3. The method according to claim 1, characterized in that The performing node classification on the detection node based on the topological structure and the operation data of the detection node to obtain the faulty node includes: Determining an adjacency matrix according to the topological structure; Extracting features from the running data of the detection node to obtain node data features; The node data features and the adjacency matrix are classified by using a graph neural network to obtain faulty nodes.

4. The method according to claim 3, characterized in that The method further comprises: Determine the operation data samples of the node samples in the line area samples according to the operation data big data pool; Determine an adjacency matrix sample according to the topological structure sample of the line area sample; Extracting features from the running data samples of the node samples to obtain node data feature samples; Inputting the node data feature sample and the adjacency matrix sample into the initial graph neural network to obtain the node failure probability; Calculate the loss value according to the node failure probability and the node label of the running data sample; The initial graph neural network is trained based on the loss value to obtain a graph neural network.

5. The method according to claim 1, characterized in that The selecting, in the line area according to the topological structure, a candidate line associated with the faulty node comprises: For each faulty node, determining whether the faulty node is an isolated faulty node based on the topological structure; If the targeted fault node is an isolated fault node, taking the first line in the line area connected to the targeted fault node as a candidate line; If the targeted faulty node is not an isolated faulty node, a second line between the targeted faulty node and another faulty node is used as a candidate line.

6. The method according to any one of claims 1 to 5, characterized in that The fault detection signal is a traveling wave signal; The step of determining the location of the fault point according to the fault detection signal of the candidate line comprises: Acquire the traveling wave signal through the traveling wave sensor of the candidate line; The location of the fault point on the candidate line is determined according to the traveling wave signal.

7. A fault location device for a distribution network, characterized in that: The device comprises: A regional detection module is used to perform fault detection based on regional operation data and status data of the line area in the distribution network to obtain a detection result; An acquisition module, used for acquiring the topological structure of the line area and the operation data of the detection node when the detection result is abnormal; A node detection module, used for classifying the detection nodes based on the topological structure and the operation data of the detection nodes to obtain faulty nodes; A selection module, configured to select a candidate line associated with the faulty node in the line area according to the topological structure; The fault location module is used to determine the location of the fault point according to the fault detection signal of the candidate line.

8. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

10. A computer program product, characterized in that The method comprises computer-executable instructions, which implement the method according to any one of claims 1 to 6 when executed by a processor.

Images