Power distribution network fault positioning method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a power distribution network fault positioning method and device, electronic equipment and a storage medium. The method comprises the following steps: obtaining first graph structure data of a target power distribution network, the first graph structure data comprising first zero sequence current data of each test node in the target power distribution network and a first adjacent matrix corresponding to the target power distribution network, the first adjacency matrix is obtained based on the connection relation of the test nodes; inputting the first graph structure data into a pre-trained fault positioning model to obtain a fault probability of each test node; and obtaining a fault section of the target power distribution network according to each fault probability. According to the technical scheme provided by the embodiment of the invention, the fault section in the power distribution network can be quickly and accurately positioned.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of power grid engineering data processing, and in particular to a distribution network fault locating method, device, electronic equipment and storage medium. Background Art

[0002] Fault location in distribution networks is one of the keys to ensuring power distribution safety. Fault location plays an important role in timely fault inspection and repair, power supply restoration, and improving the stability of distribution networks.

[0003] In the related art, there are two methods for locating distribution network faults: one is to locate distribution network faults based on traveling waves. This fault location method is often easily affected by the terrain and the structure of the distribution network line, and has a technical problem of low distribution network fault location accuracy. Summary of the invention

[0004] In view of the above technical problems, the present invention provides a distribution network fault location method, device, electronic device and storage medium to achieve relatively fast and accurate location of the fault section in the distribution network.

[0005] According to one aspect of the present invention, a method for locating a distribution network fault is provided, the method comprising:

[0006] Acquire first graph structure data of a target distribution network, wherein the first graph structure data includes first zero-sequence current data of each test node in the target distribution network and a first adjacency matrix corresponding to the target distribution network, and the first adjacency matrix is ​​obtained based on a connection relationship between each of the test nodes;

[0007] Inputting the first graph structure data into a pre-trained fault location model to obtain the fault probability of each test node;

[0008] According to each of the fault probabilities, a fault section of the target distribution network is obtained.

[0009] According to another aspect of the present invention, a distribution network fault location device is provided. The device comprises:

[0010] A data acquisition module, configured to acquire first graph structure data of a target distribution network, wherein the first graph structure data includes first zero-sequence current data of each test node in the target distribution network and a first adjacency matrix corresponding to the target distribution network, wherein the first adjacency matrix is ​​obtained based on a connection relationship between each of the test nodes;

[0011] A fault prediction module, used for inputting the first graph structure data into a pre-trained fault location model to obtain the fault probability of each test node;

[0012] A fault location module is used to obtain the fault section of the target distribution network according to each of the fault probabilities.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the distribution network fault location method described in any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the distribution network fault location method described in any embodiment of the present invention when executed.

[0018] The technical solution of the embodiment of the present invention is to obtain the first graph structure data of the target distribution network, wherein the first graph structure data includes the first zero-sequence current data of each test node in the target distribution network and the first adjacency matrix corresponding to the target distribution network, and the first adjacency matrix is ​​obtained based on the connection relationship of each test node. The first graph structure data is input into a pre-trained fault location model to obtain the fault probability of each test node; based on each fault probability, the fault section of the target distribution network is obtained. The technical solution of the embodiment of the present invention realizes relatively fast and accurate positioning of the fault section in the distribution network.

[0019] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 A schematic diagram of a flow chart of a distribution network fault location method provided in Embodiment 1 of the present invention;

[0022] Figure 2 A schematic diagram of a flow chart of a distribution network fault location method provided in Embodiment 2 of the present invention;

[0023] Figure 3 A model architecture diagram of an initial network model suitable for locating distribution network faults provided in the third embodiment of the present invention;

[0024] Figure 4 A schematic flow chart of an optional embodiment of a method for locating a distribution network fault provided in Embodiment 3 of the present invention;

[0025] Figure 5 A schematic diagram of the structure of a distribution network fault location device provided in Embodiment 4 of the present invention;

[0026] Figure 6 This is a schematic diagram of the structure of an electronic device provided in Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.

[0030] Embodiment 1

[0031] Figure 1A flow chart of a distribution network fault locating method provided in Embodiment 1 of the present invention is provided. This embodiment is applicable to the situation of locating a fault in a distribution network. The method can be executed by a distribution network fault locating device. The distribution network fault locating device can be implemented in the form of hardware and / or software. The distribution network fault locating device can be configured in an electronic device such as a computer or a server.

[0032] like Figure 1 As shown, the method of this embodiment includes:

[0033] S110. Obtain first graph structure data of a target distribution network, wherein the first graph structure data includes first zero-sequence current data of each test node in the target distribution network and a first adjacency matrix corresponding to the target distribution network, and the first adjacency matrix is ​​obtained based on a connection relationship between each of the test nodes.

[0034] Among them, the target distribution network can be understood as a distribution network that needs to be fault located. The first graph structure data can be understood as the graph structure data of the target distribution network. In an embodiment of the present invention, the first graph structure data may include the first zero-sequence current data of each test node in the target distribution network and the first adjacency matrix corresponding to the target distribution network. Among them, the test node can be understood as a node in the target distribution network that needs to be fault detected. The first zero-sequence current data can be understood as the zero-sequence current data of the test node. In an embodiment of the present invention, the zero-sequence current data of the test node can be determined based on the zero-sequence current of the branch where the test node is located. Exemplarily, the zero-sequence current data of the test node located at the head end of the branch can be the zero-sequence current of the branch. The first adjacency matrix can be understood as an adjacency matrix used to represent the connection relationship between the test nodes in the target distribution network. The first adjacency matrix can be obtained based on the connection relationship between each of the test nodes.

[0035] In the embodiment of the present invention, the first graph structure data may be represented as G=(X, R), wherein R may be represented as the first adjacency matrix corresponding to the target power distribution network, and X may be represented as the first zero-sequence current data of each test node.

[0036]

[0037] In the embodiment of the present invention, X=[I 1ZERO , I 2ZERO ,...,I NZERO ],X∈R N×M , where N can be expressed as the total number of test nodes in the target distribution network. NZEROIt can be represented as the time series sample data of M sampling points of the Nth test node. In the embodiment of the present invention, the value of M can be set according to actual needs, which is not specifically limited here. For example, M can be 1024. Among them, Among them, I A It can be expressed as the A-phase current measured by the three-phase current transformer. B It can be expressed as the B-phase current measured by the three-phase current transformer. C It can be expressed as the C-phase current measured by the three-phase current transformer.

[0038] It should be noted that in the power system, zero-sequence current refers to the current generated when the vector sum of the three-phase currents is not zero in a three-phase system. Zero-sequence current is usually associated with ground faults. In actual scenarios, for cable lines, zero-sequence current transformers are devices used to detect zero-sequence currents. They are usually installed at the neutral point of the three-phase power system to detect zero-sequence currents caused by faults or unbalanced loads. Zero-sequence current transformers are generally used to collect zero-sequence currents. Current measurement of overhead lines usually uses three-phase current transformers, which are installed on each of the three-phase lines. Therefore, the zero-sequence current can be calculated by measuring the three-phase current.

[0039] In an embodiment of the invention, the first zero-sequence current data in the first graph structure data may be normalized zero-sequence current data. Optionally, the first zero-sequence current data may be preprocessed according to the following formula to obtain normalized zero-sequence current data:

[0040]

[0041] Among them, I norm It can be expressed as the normalized first zero-sequence current data. min It can be expressed as the minimum value among the first zero-sequence current data. max It can be expressed as the maximum value among the first zero-sequence current data. I can be expressed as the first zero-sequence current data.

[0042] In an embodiment of the present invention, a normalization method is used to preprocess the first zero-sequence current data, so that the minimum value in the first zero-sequence current data can be mapped to 0, the maximum value in the first zero-sequence current data can be mapped to 1, and other values ​​can be linearly mapped between 0 and 1, which can help reduce the impact of outliers on the model.

[0043] In an embodiment of the invention, the obtaining of the first graph structure data of the target distribution network may include: obtaining a topological structure diagram of the target distribution network, so that a first adjacency matrix corresponding to the target distribution network can be obtained based on the topological structure diagram. Obtaining the first zero-sequence current data of each test node in the target distribution network. Further, the first graph structure data of the target distribution network can be obtained based on the first adjacency matrix and each of the first zero-sequence current data.

[0044] S120: Input the first graph structure data into a pre-trained fault location model to obtain the fault probability of each test node.

[0045] Among them, the pre-trained fault location model can be used to perform fault detection on the test nodes in the target power distribution to obtain the fault probability of the test nodes. Specifically, after obtaining the first graph structure data of the target power distribution network, the first graph structure data can be input into the pre-trained fault location model. Thereby, the output result of the fault location model can be obtained, that is, the fault probability of each test node can be obtained.

[0046] In an embodiment of the present invention, a method for obtaining a trained fault location model may include: obtaining a sample data set; wherein the sample data set may include sample data and an expected output result corresponding to the sample data, the sample data may be a second graph structure data of a reference distribution network, the second graph structure data may include the second zero-sequence current data of each reference node in the reference distribution network and a second adjacency matrix corresponding to the reference distribution network, the expected output result may be a classification label of each reference node, and the classification label is used to mark whether each reference node is a faulty node. Exemplarily, the label of a faulty node may be set to 1, and the label of a non-faulty node may be set to 0.

[0047] Furthermore, the sample data can be input into a pre-built initial network model, so that an actual output result can be obtained. Then, the initial network model can be trained based on the actual output result and the expected output result. Thus, the trained fault location model can be obtained. In an embodiment of the present invention, the network structure of the pre-built initial network model can include a graph attention layer, a long short-term memory layer, a splicing layer, a fully connected layer, and a sigmoid classification layer.

[0048] Alternatively, the formula of the softmax function can be:

[0049]

[0050] Among them, x is the input value of the sigmoid classification layer, and e is the base of the natural logarithm.

[0051] In an embodiment of the present invention, the training of the initial network model based on the actual output result and the expected output result may include: calculating a loss value based on a preset loss function according to the actual output result and the expected output result; thereby, the network parameters of the initial network model may be iteratively updated using a back propagation algorithm according to the loss value to obtain a trained fault location model. The preset loss function may be understood as a pre-set loss function. In an embodiment of the present invention, the preset loss function may be a Focal Loss loss function. Using the FocalLoss loss function as the loss function of an embodiment of the invention may improve the performance of the model on an unbalanced data set, thereby improving the accuracy of the fault location model.

[0052] S130. Obtain a fault section of the target distribution network according to each of the fault probabilities.

[0053] In an embodiment of the present invention, obtaining the fault section of the target distribution network according to each of the fault probabilities may include: determining a faulty node in the target distribution network based on each of the fault probabilities, and obtaining the fault section of the target distribution network according to node location information of the faulty node. The node location information may be understood as location information of the faulty node.

[0054] Specifically, the faulty node in the target distribution network can be determined based on each of the fault probabilities, thereby obtaining the location information of the faulty node, that is, the node location information, and further obtaining the faulty section of the target distribution network based on the node location information.

[0055] In an embodiment of the present invention, based on each of the fault probabilities, the fault node in the target distribution network is determined. Specifically, a threshold value, i.e., a preset threshold value, can be pre-set. Thus, each of the fault probabilities can be compared with the preset threshold value to obtain a comparison result. Further, based on the comparison result, a test node corresponding to the fault probability exceeding the preset threshold value can be determined. Further, the test node corresponding to the fault probability exceeding the preset threshold value can be used as a fault node.

[0056] Exemplarily, when there is one faulty node and no other test nodes exist in the branch where the faulty node is located, the branch where the faulty node is located can be used as the faulty section of the target distribution network. When there is one faulty node and the branch where the faulty node is located includes other test nodes, the test nodes adjacent to the faulty node on the branch can be determined. If the number of test nodes adjacent to the faulty node is one, the line between the faulty node and the test nodes adjacent to the faulty node can be used as the faulty section of the target distribution network; if the number of test nodes adjacent to the faulty node is two, the line between the two test nodes adjacent to the faulty node can be used as the faulty section of the target distribution network. In an embodiment of the present invention, when there are multiple faulty nodes, the line including each of the faulty nodes can be used as the faulty section of the target distribution network.

[0057] The technical solution of the embodiment of the present invention is to obtain the first graph structure data of the target distribution network, wherein the first graph structure data includes the first zero-sequence current data of each test node in the target distribution network and the first adjacency matrix corresponding to the target distribution network, and the first adjacency matrix is ​​obtained based on the connection relationship of each test node. The first graph structure data is input into a pre-trained fault location model to obtain the fault probability of each test node; based on each fault probability, the fault section of the target distribution network is obtained. The technical solution of the embodiment of the present invention realizes relatively fast and accurate positioning of the fault section in the distribution network.

[0058] Embodiment 2

[0059] Figure 2 A flow chart of a distribution network fault location method provided for the second embodiment of the present invention, based on the above-mentioned embodiment, optionally, the pre-trained fault location model includes a graph attention layer, a long-short memory layer, a splicing layer, a fully connected layer and a sigmod classification layer. On this basis, optionally, the first graph structure data is input into the pre-trained fault location model to obtain the fault probability of each test node, including: inputting the first graph structure data into the graph attention layer to obtain the spatial feature data of each test node; inputting the first graph structure data into the long-short memory layer to obtain the time series feature data of each test node; inputting the spatial feature data and the time series feature data of each test node into the splicing layer to obtain the node feature data of each test node; inputting each node feature data into the fully connected layer to obtain the fully connected feature data of each test node, and inputting each fully connected feature data into the sigmod classification layer to obtain the fault probability of each test node. Among them, the technical features that are the same or similar to the above-mentioned embodiments are not repeated here.

[0060] like Figure 2 As shown, the method of this embodiment specifically includes:

[0061] S210, obtaining first graph structure data of a target distribution network, wherein the first graph structure data includes first zero-sequence current data of each test node in the target distribution network and a first adjacency matrix corresponding to the target distribution network, wherein the first adjacency matrix is ​​obtained based on a connection relationship between each of the test nodes.

[0062] S220. Input the first graph structure data into the graph attention layer to obtain the spatial feature data of each of the test nodes.

[0063] Specifically, the first graph structure data is input into the graph attention layer, so that the spatial features of the first graph structure data can be extracted through the graph attention layer, thereby obtaining the spatial feature data of each test node.

[0064] Exemplarily, the input of the graph attention layer is a set of feature vectors of the test nodes, that is, a set of the first zero-sequence current data X = {x1, x2, ..., x N},x i ∈R F . Where N is the number of test nodes in the target distribution network, xi is the feature vector of node i, and F is the number of features of the node. The output of the graph attention layer is a new feature set, that is, the set of spatial feature data of each test node X′={x′1,x′2,…,x′ N},x i ′∈R F′ , it should be noted that x and x′ have different dimensions.

[0065] Specifically, the attention coefficient between each test node can be calculated according to the following formula:

[0066] e ij =a([Wx i ||Wx j ]) j∈N i

[0067] Among them, e ij It can be expressed as the attention coefficient of test node j to test node i, where test node j is the neighbor node of test node i. a is the attention function, which maps the concatenated high-dimensional features to a real number; || is the feature concatenation operation; x i It can be expressed as the feature vector of test node i. j can be the feature vector of test node j; W is the shared parameter matrix of linear transformation; N i is the total number of first-order neighbor nodes of test node i.

[0068] Afterwards, the calculated attention coefficient can be nonlinearly transformed and normalized according to the following formula to obtain the attention weight:

[0069]

[0070] Finally, the feature vector of the test node i can be updated by weighted summing of the adjacent node vectors based on the attention weights according to the following formula:

[0071]

[0072] It should be noted that the performance of the model is improved by using a multi-head attention mechanism. In an embodiment of the present invention, the number of attention heads can be set to 4. That is, the attention weights are calculated and the node features are updated respectively by 4 independently working attention mechanisms, and then the vectors updated by each attention mechanism are spliced.

[0073] It should be noted that when the topological structure of the distribution network changes, such as the addition or removal of nodes or branches, that is, the adjacency matrix and node feature vector of the distribution network change accordingly. The attention weights between nodes can be automatically adjusted through the graph attention network layer to adapt to the changes in the topological structure of the distribution network, thereby improving the stability and accuracy of the model.

[0074] S230: Input the first graph structure data into the long short-term memory layer to obtain the timing feature data of each of the test nodes.

[0075] Specifically, the first graph structure data can be input into the long short-term memory layer, so that the time series features of the first graph structure data can be extracted through the long short-term memory layer, thereby obtaining the time series feature data of each test node. In an embodiment of the present invention, a plurality of gate structures can be stacked to form an LSTM hidden layer, wherein each gate structure can include an input gate, an output gate, a forget gate, a sigmoid function, and a tanh function.

[0076] S240, inputting the spatial feature data and the temporal feature data of each of the test nodes into a splicing layer to obtain node feature data of each of the test nodes.

[0077] Specifically, after obtaining the spatial feature data and the temporal feature data of each of the test nodes, the spatial feature data and the temporal feature data of each of the test nodes can be input into the splicing layer, so that the spatial feature data and the temporal feature data of each of the test nodes are spliced ​​through the splicing layer. Then, the processing result after the feature data of each of the test nodes are spliced ​​can be obtained, that is, the node feature data of each of the test nodes.

[0078] S250, inputting the feature data of each node into the fully connected layer to obtain the fully connected feature data of each test node, and inputting each fully connected feature data into the sigmod classification layer to obtain the failure probability of each test node.

[0079] Specifically, the feature data of each node is input into the fully connected layer, so that the output result of the fully connected layer, that is, the fully connected feature data of each test node, can be obtained. Then, the fully connected feature data can be input into the sigmod classification layer. Thus, the output result of the sigmod classification layer can be obtained, that is, the failure probability of each test node can be obtained.

[0080] S260. Obtain a fault section of the target distribution network according to each of the fault probabilities.

[0081] The technical solution of the embodiment of the present invention, the pre-trained fault location model includes a graph attention layer, a long-short memory layer, a splicing layer, a fully connected layer and a sigmoid classification layer. By inputting the first graph structure data into the graph attention layer, the spatial feature data of each of the test nodes is obtained; the first graph structure data is input into the long-short memory layer to obtain the time series feature data of each of the test nodes; the spatial feature data and the time series feature data of each of the test nodes are input into the splicing layer to obtain the node feature data of each of the test nodes; each of the node feature data is input into the fully connected layer to obtain the fully connected feature data of each of the test nodes, and each of the fully connected feature data is input into the sigmoid classification layer to obtain the fault probability of each of the test nodes, so as to achieve a relatively fast and accurate positioning of the faulty nodes in the distribution network.

[0082] Embodiment 3

[0083] Embodiment 3 of the present invention provides an optional embodiment of a distribution network fault location method, such as Figure 3 The architecture of the pre-built initial network model shown includes graph attention layer, LSTM (long short memory layer), concatenation layer, fully connected layer, and sigmoid classification layer. Figure 4 Based on the architecture of the initial network model, the method of locating the fault of the distribution network specifically includes the following steps:

[0084] 1. Produce graph data, that is, obtain first graph structure data and second graph structure data of the distribution network.

[0085] The first graph structure data may include the first zero-sequence current data of each test node in the target distribution network and the first adjacency matrix corresponding to the target distribution network, and the first adjacency matrix may be obtained based on the connection relationship of each test node. The second graph structure data may include the second zero-sequence current data of each reference node in the reference distribution network and the second adjacency matrix corresponding to the reference distribution network, and the expected output result may be a classification label of each reference node, and the classification label is used to mark whether each reference node is a fault node.

[0086] 2. Input the second graph structure data into the pre-built initial network model (GAT-LSTM model), extract the spatial features of each node in the target power distribution through graph convolution, and extract the spatial features of each node in the target power distribution through the gate structure unit, that is, obtain the spatial feature data and time feature data of each reference node in the target distribution network.

[0087] 3. Perform feature splicing on the spatial feature data and the temporal feature data of each test node to obtain the spliced ​​result, that is, the node feature data of each reference node.

[0088] 4. Input the node feature data of each reference node into the fully connected layer to obtain the output result of the fully connected layer, that is, to obtain the fully connected feature data of each reference node. Input each fully connected feature data into the sigmoid classification layer to obtain the failure probability of each reference node.

[0089] 5. Based on the failure probability of each reference node output by the model and the classification label of each reference node, loss calculation is performed.

[0090] 6. When the number of loss calculations reaches the number of iterations, stop model training and obtain the pre-trained model, that is, obtain the trained fault location model.

[0091] 7 After the trained fault location model is obtained, the first graph structure data can be input into the trained fault location model to obtain an output location result, that is, the fault section of the output target distribution network.

[0092] The technical solution of the embodiment of the present invention constructs an initial network model and trains the constructed initial network model to obtain a trained fault location model. Then, the fault location of the distribution network is performed based on the trained fault location model, thereby achieving a relatively fast and accurate location of the fault section in the distribution network.

[0093] Embodiment 4

[0094] Figure 5 This is a schematic diagram of the structure of a distribution network fault location device provided by the third embodiment of the present invention. Figure 5 As shown, the device includes: a data acquisition module 310 , a fault prediction module 320 and a fault location module 330 .

[0095] Among them, the data acquisition module 310 is used to obtain the first graph structure data of the target distribution network, wherein the first graph structure data includes the first zero-sequence current data of each test node in the target distribution network and the first adjacency matrix corresponding to the target distribution network, and the first adjacency matrix is ​​obtained based on the connection relationship between each of the test nodes; the fault prediction module 320 is used to input the first graph structure data into a pre-trained fault location model to obtain the fault probability of each of the test nodes; the fault location module 330 is used to obtain the fault section of the target distribution network according to each of the fault probabilities.

[0096] The technical solution of the embodiment of the present invention is to obtain the first graph structure data of the target distribution network through the data acquisition module, wherein the first graph structure data includes the first zero-sequence current data of each test node in the target distribution network and the first adjacency matrix corresponding to the target distribution network, and the first adjacency matrix is ​​obtained based on the connection relationship of each test node; the first graph structure data is input into the pre-trained fault location model through the fault prediction module to obtain the fault probability of each test node; the fault location module is used to obtain the fault section of the target distribution network according to each fault probability. The technical solution of the embodiment of the present invention realizes the relatively fast and accurate positioning of the fault section in the distribution network.

[0097] Optionally, the pre-trained fault location model includes a graph attention layer, a long short-term memory layer, a splicing layer, a fully connected layer and a sigmoid classification layer.

[0098] Optionally, the fault prediction module 320 is specifically used to: input the first graph structure data into the graph attention layer to obtain the spatial feature data of each of the test nodes; input the first graph structure data into the long short memory layer to obtain the temporal feature data of each of the test nodes; input the spatial feature data and the temporal feature data of each of the test nodes into the splicing layer to obtain the node feature data of each of the test nodes; input each of the node feature data into the fully connected layer to obtain the fully connected feature data of each of the test nodes, and input each of the fully connected feature data into the sigmoid classification layer to obtain the fault probability of each of the test nodes.

[0099] Optionally, the device further includes a model training module, which includes a sample acquisition unit, a model training unit, and a model acquisition unit; wherein,

[0100] A sample acquisition unit, used to: acquire a sample data set; wherein the sample data set includes sample data and an expected output result corresponding to the sample data, the sample data is second graph structure data of a reference distribution network, the second graph structure data includes second zero-sequence current data of each reference node in the reference distribution network and a second adjacency matrix corresponding to the reference distribution network, the expected output result is a classification label of each reference node, and the classification label is used to mark whether each reference node is a fault node;

[0101] A model training unit is used to: input the sample data into a pre-built initial network model to obtain an actual output result;

[0102] The model obtaining unit is used to: train the initial network model based on the actual output result and the expected output result to obtain the trained fault location model.

[0103] Optionally, the model acquisition unit is used to calculate a loss value based on a preset loss function according to the actual output result and the expected output result; and according to the loss value, a back propagation algorithm is used to iteratively update the network parameters of the initial network model to obtain a trained fault location model.

[0104] Optionally, the fault location module 330 is used to determine a faulty node in the target distribution network based on each of the fault probabilities, and obtain a faulty section of the target distribution network according to node location information of the faulty node.

[0105] Optionally, the data acquisition module 310 is used to obtain a topological structure diagram of the target distribution network, and obtain a first adjacency matrix corresponding to the target distribution network based on the topological structure diagram; obtain first zero-sequence current data of each test node in the target distribution network; and obtain first graph structure data of the target distribution network based on the first adjacency matrix and each of the first zero-sequence current data.

[0106] The distribution network fault locating device provided in the embodiment of the present invention can execute the distribution network fault locating method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0107] It is worth noting that the various units and modules included in the above-mentioned distribution network fault locating device are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the embodiments of the present invention.

[0108] Embodiment 5

[0109] Figure 6A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0110] like Figure 6 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0111] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0112] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a distribution network fault location method.

[0113] In some embodiments, the distribution network fault location method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the distribution network fault location method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the distribution network fault location method in any other appropriate manner (e.g., by means of firmware).

[0114] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0115] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0116] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0117] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0118] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0119] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0120] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0121] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A distribution network fault location method, characterized in that: include: Acquire first graph structure data of a target distribution network, wherein the first graph structure data includes first zero-sequence current data of each test node in the target distribution network and a first adjacency matrix corresponding to the target distribution network, and the first adjacency matrix is ​​obtained based on a connection relationship between each of the test nodes; Inputting the first graph structure data into a pre-trained fault location model to obtain the fault probability of each test node; According to each of the fault probabilities, a fault section of the target distribution network is obtained.

2. The method according to claim 1, characterized in that The pre-trained fault location model includes a graph attention layer, a long short-term memory layer, a splicing layer, a fully connected layer, and a sigmoid classification layer.

3. The method according to claim 2, characterized in that The step of inputting the first graph structure data into a pre-trained fault location model to obtain the fault probability of each test node includes: Inputting the first graph structure data into the graph attention layer to obtain spatial feature data of each test node; Inputting the first graph structure data into the long short-term memory layer to obtain the time series feature data of each test node; Inputting the spatial feature data and the temporal feature data of each of the test nodes into the splicing layer to obtain node feature data of each of the test nodes; The feature data of each node is input into the fully connected layer to obtain the fully connected feature data of each test node, and the fully connected feature data is input into the sigmod classification layer to obtain the failure probability of each test node.

4. The method according to claim 1, characterized in that: The method further comprises: Acquire a sample data set; wherein the sample data set includes sample data and expected output results corresponding to the sample data, the sample data is second graph structure data of a reference distribution network, the second graph structure data includes second zero-sequence current data of each reference node in the reference distribution network and a second adjacency matrix corresponding to the reference distribution network, the expected output result is a classification label of each reference node, and the classification label is used to mark whether each reference node is a fault node; Inputting the sample data into a pre-built initial network model to obtain an actual output result; Based on the actual output result and the expected output result, the initial network model is trained to obtain the trained fault location model.

5. The method according to claim 4, characterized in that The training of the initial network model based on the actual output result and the expected output result includes: Calculate the loss value according to the actual output result and the expected output result based on a preset loss function; According to the loss value, a back propagation algorithm is used to iteratively update the network parameters of the initial network model to obtain a trained fault location model.

6. The method according to claim 1, characterized in that The obtaining of the fault section of the target distribution network according to each of the fault probabilities includes: A faulty node in the target distribution network is determined based on each of the fault probabilities, and a faulty section of the target distribution network is obtained according to node location information of the faulty node.

7. The method according to claim 1, characterized in that The obtaining of first graph structure data of the target distribution network includes: Acquire a topological structure diagram of a target distribution network, and obtain a first adjacency matrix corresponding to the target distribution network based on the topological structure diagram; Acquire first zero-sequence current data of each test node in the target distribution network; The first graph structure data of the target distribution network is obtained based on the first adjacency matrix and each of the first zero-sequence current data.

8. A distribution network fault location device, characterized in that: include: A data acquisition module, configured to acquire first graph structure data of a target distribution network, wherein the first graph structure data includes first zero-sequence current data of each test node in the target distribution network and a first adjacency matrix corresponding to the target distribution network, wherein the first adjacency matrix is ​​obtained based on a connection relationship between each of the test nodes; A fault prediction module, used for inputting the first graph structure data into a pre-trained fault location model to obtain the fault probability of each test node; A fault location module is used to obtain the fault section of the target distribution network according to each of the fault probabilities.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the distribution network fault location method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the distribution network fault location method according to any one of claims 1 to 7 when executed.

Images