Power distribution network line fault detection method, device, equipment, medium and chip
By constructing a neural network model based on historical and topological data, the problem of relying on thresholds for fault judgment in existing distribution network lines was solved, enabling dynamic and accurate fault detection and early warning, and improving the intelligence level of the distribution network.
Patent Information
- Application Number
- CN202411772671.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-04
- Publication Date
- 2025-01-07
AI Technical Summary
Existing methods for fault diagnosis in distribution network lines rely on preset thresholds, which are difficult to adapt to changes in distribution network topology and user load, resulting in inaccurate fault diagnosis.
By acquiring historical data and network topology data of distribution network lines, labeling and constructing a sample set, training a neural network model for fault identification, and performing dynamic fault detection and early warning based on the model output.
It enables dynamic and efficient global and local fault diagnosis, improves the intelligence level of the distribution network, and enhances the accuracy of fault detection and early warning capabilities.
Smart Images
Figure CN119269974A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of fault detection, and in particular to a method, device, equipment, medium and chip for detecting faults in a distribution network line. Background Art
[0002] With the intelligentization of switches, switches installed in various sections of the distribution network can sense real-time parameters such as voltage and current at various locations of the branch lines. Some intelligent switches can even identify and judge faults that have occurred or are about to occur in the line. These capabilities greatly improve the level of control over the distribution network. The existing method of judging faults in distribution network lines is to compare the measured electrical parameters with the preset threshold. If the measured value exceeds the threshold, it is judged as a line fault, and the switch is used to perform protection actions or issue an alarm. Since the topology of the distribution network lines often changes, and the user's load in the current distribution network is changing all the time, with the access of new energy such as wind and light, the power of the power supply in the distribution network will also change continuously. These make it difficult to rely solely on a pre-set threshold as an indication of fault judgment. Summary of the invention
[0003] In order to solve the problems in the related art, the embodiments of the present disclosure provide a distribution network line fault detection method, device, equipment, medium and chip.
[0004] In a first aspect, the present disclosure provides a method for detecting a power distribution network line fault, comprising: Acquire historical data of distribution network line faults; the historical data includes electrical data measured by N switches in the distribution network and network topology data formed by each switch; wherein N is a natural number; Labeling the historical data, and constructing a sample set based on the labeled labels and the corresponding historical data; Using the sample set to train a fault identification model; Perform fault detection based on the trained fault identification model and execute preset operations according to the detection results.
[0005] In one implementation of the present disclosure, labeling the historical data includes: If a distribution network corresponding to a certain piece of historical data has a distribution network line fault at a corresponding time, and the distribution network line fault causes a first number of switches to perform a protection operation, then the first number of switches are marked with a first field, and the remaining number of switches are marked with a second field; The first field and the second field are combined into an N-dimensional vector as a label corresponding to the historical data.
[0006] In an implementation of the present disclosure, the distribution network line fault includes one or more of a switch failure, a distribution network branch line failure where the switch is located, and a distribution network entire network failure.
[0007] In one implementation of the present disclosure, the fault identification model is a neural network model; wherein the neural network model includes multiple layers of graph convolution layers, downsampling layers, multiple layers of fully connected layers and an output layer; the downsampling layer is arranged between every two layers of graph convolution layers.
[0008] In one implementation of the present disclosure, the performing of fault detection based on the trained fault identification model and executing a preset operation according to the detection result includes: Perform fault detection based on the trained fault identification model and output the probability value of the need for switch disconnection operation; If the probability value is in the first interval, the switch disconnection operation is not performed; If the probability value is in the second interval, a fault indication or fault warning operation is performed; If the probability value is in the third interval, the switch opening operation is performed.
[0009] In one implementation of the present disclosure, it further includes: Get real-time data on distribution network line faults; Label the real-time data, and add the labeled labels and the corresponding real-time data to the sample set.
[0010] In a second aspect, the present disclosure provides a distribution network line fault detection module, including: An acquisition module is configured to acquire historical data of distribution network line faults; the historical data includes electrical data measured by N switches in the distribution network and network topology data formed by each switch; wherein N is a natural number; A construction module is configured to label the historical data and construct a sample set based on the labeled labels and the corresponding historical data; A training module, configured to train a fault identification model using the sample set; The fault detection module is configured to perform fault detection based on the trained fault identification model and execute preset operations according to the detection results.
[0011] In an implementation of the present disclosure, the distribution network line fault detection device is configured in a switch of the distribution network, or in a distribution network device including a switch, or in a system including a distribution network device.
[0012] In an implementation of the present disclosure, the part of the construction module that labels the historical data is configured as follows: If a distribution network corresponding to a certain piece of historical data has a distribution network line fault at a corresponding time, and the distribution network line fault causes a first number of switches to perform a protection operation, then the first number of switches are marked with a first field, and the remaining number of switches are marked with a second field; The first field and the second field are combined into an N-dimensional vector as a label corresponding to the historical data.
[0013] In an implementation of the present disclosure, the distribution network line fault includes one or more of a switch failure, a distribution network branch line failure where the switch is located, and a distribution network entire network failure.
[0014] In one implementation of the present disclosure, the fault identification model is a neural network model; wherein the neural network model includes multiple layers of graph convolution layers, downsampling layers, multiple layers of fully connected layers and an output layer; the downsampling layer is provided between every two layers of graph convolution layers. In one implementation of the present disclosure, the fault detection module includes:
[0015] A detection unit is configured to perform fault detection based on a trained fault recognition model and output a probability value of a switch breaking operation being required; A first execution unit is configured to not execute the switch disconnection operation if the probability value is in a first interval; A second execution unit is configured to execute a fault indication or fault warning operation if the probability value is in a second interval; The third execution unit is configured to execute a switch opening operation if the probability value is in a third interval.
[0016] In a third aspect, an embodiment of the present disclosure provides an electronic device, comprising a memory and a processor, wherein the memory is used to store one or more computer instructions, and wherein the one or more computer instructions are executed by the processor to implement a method as described in any one of the first aspects.
[0017] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium on which computer instructions are stored. When the computer instructions are executed by a processor, the method described in any one of the first aspects is implemented.
[0018] In a fifth aspect, an embodiment of the present disclosure provides a chip, the chip including a processor, the processor being used to call a computer program in a memory to execute the above-mentioned distribution network line fault detection method.
[0019] The technical effects provided by the embodiments of the present disclosure may include the following beneficial effects: According to the technical solution provided by the embodiment of the present disclosure, a distribution network line fault detection method obtains historical data of distribution network line faults; the historical data includes electrical data measured by N switches in the distribution network and network topology data formed by each switch; wherein N is a natural number; labels are annotated for the historical data, and a sample set is constructed based on the annotated labels and the corresponding historical data; a fault identification model is trained using the sample set; fault detection is performed based on the trained fault identification model, and preset operations are performed based on the detection results. In the above technical solution, a fault identification model is trained based on electrical data, network topology data, and labels, and fault detection is performed based on the fault identification model, which can dynamically and efficiently realize global and local fault judgment and early warning, and improve the intelligence level of the distribution network.
[0020] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] Figure 1 A flow chart of a method for detecting a line fault in a power distribution network according to an embodiment of the present disclosure is shown.
[0022] Figure 2 A structural block diagram of a distribution network line fault detection device according to an embodiment of the present disclosure is shown.
[0023] Figure 3 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown.
[0024] Figure 4 A schematic diagram showing the structure of a computer system suitable for implementing the method according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0025] Hereinafter, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily implement them. In addition, for the sake of clarity, parts not related to the description of the exemplary embodiments are omitted in the accompanying drawings.
[0026] In the present disclosure, it should be understood that terms such as "include" or "have" are intended to indicate the presence of features, numbers, steps, behaviors, components, parts, or a combination thereof disclosed in the present specification, and are not intended to exclude the possibility that one or more other features, numbers, steps, behaviors, components, parts, or a combination thereof exist or are added.
[0027] It should also be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.
[0028] With the intelligentization of switches, switches installed in various sections of the distribution network can sense real-time parameters such as voltage and current at various locations of the branch lines. Some intelligent switches can even identify and judge faults that have occurred or are about to occur in the line. These capabilities greatly improve the level of control over the distribution network. The existing method of judging faults in distribution network lines is to compare the measured electrical parameters with the preset threshold. If the measured value exceeds the threshold, it is judged as a line fault, and the switch is used to perform protection actions or issue an alarm. Since the topology of the distribution network lines often changes, and the user's load in the current distribution network is changing all the time, with the access of new energy such as wind and light, the power of the power supply in the distribution network will also change continuously. These make it difficult to rely solely on a pre-set threshold as an indication of fault judgment.
[0029] Taking the above-mentioned defects into consideration, the distribution network line fault detection method provided by the present invention trains a fault identification model based on electrical data, network topology data and labels, and performs fault detection based on the fault identification model. It can dynamically and efficiently realize global and local fault judgment and early warning, thereby improving the intelligence level of the distribution network.
[0030] Figure 1 A flow chart of a method for detecting a line fault in a power distribution network according to an embodiment of the present disclosure is shown.
[0031] like Figure 1 As shown, the distribution network line fault detection method includes the following steps S110-S140: In step S110, historical data of distribution network line faults are obtained; the historical data includes electrical data measured by N switches in the distribution network and network topology data formed by each switch; wherein N is a natural number; In step S120, the historical data is labeled, and a sample set is constructed based on the labeled labels and the corresponding historical data; In step S130, the fault identification model is trained using the sample set; In step S140, fault detection is performed based on the trained fault identification model, and a preset operation is performed according to the detection result.
[0032] In the disclosed method, the sensing function of the intelligent switch can be used to obtain the electrical parameters of each switch. The electrical data may include current and voltage information, and may also include component information such as temperature, frequency, phase, power factor, harmonic content, content rate, etc. The electrical data of each switch is recorded as T i , 1<i<N, N is the total number of switches in the distribution network, each T i are represented as a vector.
[0033] In the disclosed method, the topology acquisition function of the intelligent switch can be used to obtain the network topology data formed by each switch offline or in real time, which is recorded as The network topology data in the present disclosure may be global network topology data of the distribution network, or network topology data of a branch line of the distribution network. In a specific implementation, the global network topology data of the distribution network is used as the network topology data formed by each switch.
[0034] In the disclosed method, a sample in the sample set is represented as .in, Represents the kth network topology data formed by each switch, Representation and The corresponding electrical data measured by the i-th switch, Representation and It should be noted that when the network topology where the switch is located changes, Take different values, otherwise The value of remains unchanged.
[0035] In the disclosed method, the samples in the sample set can also be embedded, and then the sample set is used to train the fault recognition model. The embedding process in the disclosed method can be selectively performed as needed, and the specific processing method can refer to the prior art and is not the focus of the disclosed method.
[0036] In one implementation of the present disclosure, the step of labeling the historical data in step S120 includes: If a distribution network corresponding to a certain piece of historical data has a distribution network line fault at a corresponding time, and the distribution network line fault causes a first number of switches to perform a protection operation, then the first number of switches are marked with a first field, and the remaining number of switches are marked with a second field; The first field and the second field are combined into an N-dimensional vector as a label corresponding to the historical data.
[0037] In the disclosed method, the distribution network line fault includes one or more of a switch failure, a distribution network branch line failure where the switch is located, and a distribution network entire network failure.
[0038] In a specific implementation, the distribution network line fault includes the switch itself fault, the distribution network branch line fault where the switch is located, and the distribution network fault. When labeling, the present disclosure does not distinguish the above fault types, but sets labels for samples based on whether the above distribution network line fault will cause the switch to open or close. In this way, when applying the subsequently trained fault recognition model for fault detection, the low-voltage switch can be used for distribution network line fault determination and protection action, or alarm can be issued for various fault types.
[0039] Specifically, a certain piece of historical data carries the label and time sequence of the distribution network. Labeling a certain piece of historical data means labeling N switches with the same distribution network label and the same time sequence with fault labels or non-fault labels. The label corresponding to a certain piece of historical data is an N-dimensional vector, and each switch corresponds to one component. If the distribution network corresponding to a certain piece of historical data has a distribution network line fault at the corresponding time, and the distribution network line fault causes a first number of switches to perform a disconnection protection operation, then the components corresponding to these first number of switches in this N-dimensional vector are marked as 1, and the remaining number of switches are marked as 0.
[0040] In one implementation of the present disclosure, the fault identification model is a neural network model; wherein the neural network model includes multiple layers of graph convolution layers, downsampling layers, multiple layers of fully connected layers and an output layer; the downsampling layer is arranged between every two layers of graph convolution layers.
[0041] In the disclosed method, 5 layers of graph convolution layers are used, and topk pooling is used between every two layers of graph convolution layers to extract all feature information from global to local levels, and the global information of 5 samples is comprehensively considered, for example, by adding operations, and then the addition result is used as the input of the fully connected layer. In the disclosed method, 3 layers of fully connected layers are used, and then the probability of each node involving a fault and requiring a protective action such as a switch disconnection operation is output. Such a technical solution enables the model to fully capture the normal and abnormal operation characteristics of the entire distribution network, the local distribution network, each part of the line, and each node at different levels, so that the model can make accurate judgments on whether a distribution network fault, a local fault, a line fault, or a node fault has occurred.
[0042] In one implementation of the present disclosure, the performing of fault detection based on the trained fault identification model and executing a preset operation according to the detection result includes: Perform fault detection based on the trained fault identification model and output the probability value of the need for switch disconnection operation; If the probability value is in the first interval, the switch disconnection operation is not performed; If the probability value is in the second interval, a fault indication or fault warning operation is performed; If the probability value is in the third interval, the switch opening operation is performed.
[0043] In the disclosed method, segmented operations are performed based on the output of the fault identification model, and specific operations include not performing a switch opening operation, performing a fault indication or fault warning operation, and performing a switch opening operation.
[0044] Specifically, if the output probability value is in the first interval, it indicates that there is a high probability that a fault will not occur, so the switch disconnection operation is not performed. If the probability value is in the second interval, it indicates that there is a possibility of a fault, so a fault indication or fault warning operation is performed. If the probability value is in the third interval, it indicates that there is a high probability that a fault will occur, so the switch disconnection operation is performed to protect the safety of the distribution network line. In a specific embodiment, the first interval is, for example, 0-0.75, the second interval is, for example, 0.76-0.85, and the third interval is, for example, above 0.85.
[0045] The settings of the first interval, the second interval and the third interval in the present disclosure can be adjusted according to actual conditions, and different fault identification models corresponding to different switches can set different first intervals, second intervals and third intervals, and the present disclosure does not impose any restrictions on this.
[0046] In one implementation of the present disclosure, it further includes: Get real-time data on distribution network line faults; Label the real-time data, and add the labeled labels and the corresponding real-time data to the sample set.
[0047] In the disclosed method, by continuously acquiring real-time data of distribution network line faults, the latest training data can be provided for the sample set, thereby continuously iterating the fault identification model, which is beneficial to improving the model's prediction accuracy for new data.
[0048] Figure 2 The structural block diagram of the power distribution network line fault detection device according to the embodiment of the present disclosure is shown. The device can be implemented as part or all of the electronic device through software, hardware or a combination of both.
[0049] like Figure 2 As shown, the distribution network line fault detection device 200 includes: The acquisition module 210 is configured to acquire historical data of the line fault of the distribution network; the historical data includes electrical data measured by N switches in the distribution network and network topology data formed by each switch; wherein N is a natural number; A construction module 220 is configured to label the historical data and construct a sample set based on the labeled labels and the corresponding historical data; A training module 230 is configured to train a fault identification model using the sample set; The fault detection module 240 is configured to perform fault detection based on the trained fault identification model and execute a preset operation according to the detection result.
[0050] The distribution network line fault detection device provided by the present invention trains a fault identification model based on electrical data, network topology data and labels, and performs fault detection based on the fault identification model. It can dynamically and efficiently realize global and local fault judgment and early warning, thereby improving the intelligence level of the distribution network.
[0051] In an implementation of the present disclosure, the distribution network line fault detection device is configured in a switch of the distribution network, or in a distribution network device including a switch, or in a system including a distribution network device.
[0052] In an implementation of the present disclosure, the part of the construction module that labels the historical data is configured as follows: If a distribution network corresponding to a certain piece of historical data has a distribution network line fault at a corresponding time, and the distribution network line fault causes a first number of switches to perform a protection operation, then the first number of switches are marked with a first field, and the remaining number of switches are marked with a second field; The first field and the second field are combined into an N-dimensional vector as a label corresponding to the historical data.
[0053] In an implementation of the present disclosure, the distribution network line fault includes one or more of a switch failure, a distribution network branch line failure where the switch is located, and a distribution network entire network failure.
[0054] In one implementation of the present disclosure, the fault identification model is a neural network model; wherein the neural network model includes multiple layers of graph convolution layers, downsampling layers, multiple layers of fully connected layers and an output layer; the downsampling layer is arranged between every two layers of graph convolution layers.
[0055] In one implementation of the present disclosure, the fault detection module includes: A detection unit is configured to perform fault detection based on a trained fault recognition model and output a probability value of a switch breaking operation being required; A first execution unit is configured to not execute the switch disconnection operation if the probability value is in a first interval; A second execution unit is configured to execute a fault indication or fault warning operation if the probability value is in a second interval; The third execution unit is configured to execute a switch opening operation if the probability value is in a third interval.
[0056] The present disclosure also discloses an electronic device, Figure 3 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown.
[0057] like Figure 3As shown, the electronic device includes a memory and a processor, wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method according to an embodiment of the present disclosure.
[0058] Acquire historical data of distribution network line faults; the historical data includes electrical data measured by N switches in the distribution network and network topology data formed by each switch; wherein N is a natural number; Labeling the historical data, and constructing a sample set based on the labeled labels and the corresponding historical data; Using the sample set to train a fault identification model; Perform fault detection based on the trained fault identification model and execute preset operations according to the detection results.
[0059] In one implementation of the present disclosure, labeling the historical data includes: If a distribution network corresponding to a certain piece of historical data has a distribution network line fault at a corresponding time, and the distribution network line fault causes a first number of switches to perform a protection operation, then the first number of switches are marked with a first field, and the remaining number of switches are marked with a second field; The first field and the second field are combined into an N-dimensional vector as a label corresponding to the historical data.
[0060] In an implementation of the present disclosure, the distribution network line fault includes one or more of a switch failure, a distribution network branch line failure where the switch is located, and a distribution network entire network failure.
[0061] In one implementation of the present disclosure, the fault identification model is a neural network model; wherein the neural network model includes multiple layers of graph convolution layers, downsampling layers, multiple layers of fully connected layers and an output layer; the downsampling layer is arranged between every two layers of graph convolution layers.
[0062] In one implementation of the present disclosure, the performing of fault detection based on the trained fault identification model and executing a preset operation according to the detection result includes: Perform fault detection based on the trained fault identification model and output the probability value of the need for switch disconnection operation; If the probability value is in the first interval, the switch disconnection operation is not performed; If the probability value is in the second interval, a fault indication or fault warning operation is performed; If the probability value is in the third interval, the switch opening operation is performed.
[0063] In one implementation of the present disclosure, it further includes: Get real-time data on distribution network line faults; Label the real-time data, and add the labeled labels and the corresponding real-time data to the sample set.
[0064] Figure 4 A schematic diagram showing the structure of a computer system suitable for implementing the method according to an embodiment of the present disclosure is shown.
[0065] like Figure 4 As shown, the computer system includes a processing unit, which can execute various methods in the above-mentioned embodiments according to a program stored in a read-only memory (ROM) or a program loaded from a storage part into a random access memory (RAM). In the RAM, various programs and data required for the operation of the computer system are also stored. The processing unit, ROM and RAM are connected to each other through a bus. An input / output (I / O) interface is also connected to the bus.
[0066] The following components are connected to the I / O interface: an input part including a keyboard, a mouse, etc.; an output part including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage part including a hard disk, etc.; and a communication part including a network interface card such as a LAN card, a modem, etc. The communication part performs a communication process via a network such as the Internet. The drive is also connected to the I / O interface as needed. Removable media, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on the drive as needed so that the computer program read therefrom can be installed into the storage part as needed. Among them, the processing unit can be implemented as a processing unit such as a CPU, a GPU, a TPU, an FPGA, an NPU, etc.
[0067] In particular, according to an embodiment of the present disclosure, the method described above can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program tangibly contained on a machine-readable medium, and the computer program includes a program code for executing the above method. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium.
[0068] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0069] The units or modules involved in the embodiments described in the present disclosure may be implemented by software or programmable hardware. The units or modules described may also be set in a processor, and the names of these units or modules do not constitute limitations on the units or modules themselves in some cases.
[0070] As another aspect, the present disclosure further provides a chip, which includes at least one processor and can be used to implement the method involved in the above system embodiment.
[0071] In one possible design, the chip also includes a memory, which is used to store program instructions and data, and the memory is located inside or outside the processor.
[0072] As another aspect, the present disclosure further provides a computer-readable storage medium, which may be a computer-readable storage medium included in the electronic device or computer system in the above embodiment; or a computer-readable storage medium that exists independently and is not assembled into a device. The computer-readable storage medium stores one or more programs, and the programs are used by one or more processors to execute the method described in the present disclosure.
[0073] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present disclosure is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other.
Claims
1. A method for detecting a distribution network line fault, characterized in that: include: Acquire historical data of distribution network line faults; the historical data includes electrical data measured by N switches in the distribution network and network topology data formed by each switch; wherein N is a natural number; Labeling the historical data, and constructing a sample set based on the labeled labels and the corresponding historical data; Using the sample set to train a fault identification model; Perform fault detection based on the trained fault identification model and execute preset operations according to the detection results.
2. The method for detecting a power distribution network line fault according to claim 1, characterized in that: The labeling of the historical data comprises: If a distribution network corresponding to a certain piece of historical data has a distribution network line fault at a corresponding time, and the distribution network line fault causes a first number of switches to perform a protection operation, then the first number of switches are marked with a first field, and the remaining number of switches are marked with a second field; The first field and the second field are combined into an N-dimensional vector as a label corresponding to the historical data.
3. The method for detecting a power distribution network line fault according to claim 2, characterized in that: The distribution network fault includes one or more of a switch fault, a distribution network branch line fault where the switch is located, and a distribution network fault.
4. The method for detecting a fault in a power distribution network according to claim 1, characterized in that: The fault identification model is a neural network model; wherein the neural network model includes multiple layers of graph convolution layers, downsampling layers, multiple layers of fully connected layers and an output layer; the downsampling layer is arranged between every two graph convolution layers.
5. The method for detecting a fault in a power distribution network according to claim 1, characterized in that: The fault detection based on the trained fault identification model and the execution of preset operations according to the detection results include: Perform fault detection based on the trained fault identification model and output the probability value of the need for switch disconnection operation; If the probability value is in the first interval, the switch disconnection operation is not performed; If the probability value is in the second interval, a fault indication or fault warning operation is performed; If the probability value is in the third interval, the switch opening operation is performed.
6. The method for detecting a fault in a power distribution network according to claim 1, characterized in that: Also includes: Get real-time data on distribution network line faults; Label the real-time data, and add the labeled labels and the corresponding real-time data to the sample set.
7. A distribution network line fault detection device, characterized in that: include: An acquisition module is configured to acquire historical data of distribution network line faults; the historical data includes electrical data measured by N switches in the distribution network and network topology data formed by each switch; wherein N is a natural number; A construction module is configured to label the historical data and construct a sample set based on the labeled labels and the corresponding historical data; A training module, configured to train a fault identification model using the sample set; The fault detection module is configured to perform fault detection based on the trained fault identification model and execute preset operations according to the detection results.
8. The distribution network line fault detection device according to claim 7, characterized in that: The distribution network line fault detection device is configured in a switch of the distribution network, or in a distribution network device including a switch, or in a system including a distribution network device.
9. The distribution network line fault detection device according to claim 7, characterized in that: The part of the building module that labels the historical data is configured as follows: If a distribution network corresponding to a certain piece of historical data has a distribution network line fault at a corresponding time, and the distribution network line fault causes a first number of switches to perform a protection operation, then the first number of switches are marked with a first field, and the remaining number of switches are marked with a second field; The first field and the second field are combined into an N-dimensional vector as a label corresponding to the historical data.
10. The distribution network line fault detection device according to claim 9, characterized in that: The distribution network line fault includes one or more of a switch failure, a distribution network branch line failure where the switch is located, and a distribution network entire network failure.
11. The distribution network line fault detection device according to claim 7, characterized in that: The fault identification model is a neural network model; wherein the neural network model includes multiple layers of graph convolution layers, downsampling layers, multiple layers of fully connected layers and an output layer; the downsampling layer is arranged between every two graph convolution layers.
12. The distribution network line fault detection device according to claim 7, characterized in that: The fault detection module comprises: A detection unit is configured to perform fault detection based on a trained fault recognition model and output a probability value of a switch breaking operation being required; A first execution unit is configured to not execute the switch disconnection operation if the probability value is in a first interval; A second execution unit is configured to execute a fault indication or fault warning operation if the probability value is in a second interval; The third execution unit is configured to execute a switch opening operation if the probability value is in a third interval.
13. An electronic device, characterized in that: The method comprises a memory and a processor; wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method according to any one of claims 1 to 6.
14. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the computer instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.
15. A chip, characterized in that: include: At least one processor, configured to implement the method according to any one of claims 1 to 6.