Cable fault identification and positioning method, device, equipment and storage medium

Abstract

The embodiment of the present invention discloses a cable fault identification and positioning method and device, equipment and storage medium, the method includes: using the acquired multi-source measured data to train the preset BP neural network to obtain a neural network model; and inputting the acquired voltage traveling wave signal into the neural network model after signal processing to obtain cable fault information; constructing a three-dimensional digital logic model of cable laying based on the cable parameters of the substation; locating in the three-dimensional digital logic model according to the cable fault information to determine the cable information at the fault. By performing a refined three-dimensional digital simulation of the cable laying of the substation, the cable laying can be accurately described so that the fault location can be found more accurately; secondly, by fusing the multi-source measured data, more abundant and accurate training data is obtained, and the neural network model is obtained using the training data to achieve a more accurate output result of the neural network model.

Description

Technical Field

[0001] The present invention relates to the technical field of fault location, and in particular to a cable fault identification and location method and device, equipment and storage medium. Background Art

[0002] In the installation of large-scale electromechanical projects, cables are one of the key infrastructures for modern communications and data transmission, and their laying quality and reliability are crucial to the normal operation of electromechanical projects. However, during the laying and use of cables, various faults are inevitable, such as cable breakage, insulation damage, joint failure, etc. These faults may lead to serious consequences such as power transmission interruption, communication signal loss, and equipment failure.

[0003] Therefore, in order to realize the cable fault location and identification, there are some fault location and identification methods, which are used to locate and maintain after the fault occurs. However, the existing fault location and identification methods have the problem of insufficient precision, resulting in inaccurate fault location and identification results. Summary of the invention

[0004] Based on this, it is necessary to propose a cable fault identification and positioning method and device, equipment and storage medium to accurately locate the fault location in response to the above problems.

[0005] To achieve the above objectives, the present application provides a first aspect of a cable fault identification and location method, the method comprising:

[0006] Using the acquired multi-source measured data to train the preset BP neural network, a neural network model for fault identification and location is obtained, wherein the multi-source measured data is test data measured in different ways;

[0007] Acquire a voltage traveling wave signal of a substation cable, perform signal processing on the voltage traveling wave signal, and input the signal into the neural network model to obtain cable fault information output by the neural network model;

[0008] Constructing a three-dimensional digital logic model of cable laying based on the cable parameters of the substation;

[0009] The cable fault information is located in the three-dimensional digital logic model to determine the cable information at the fault location.

[0010] Furthermore, the three-dimensional digital logic model of cable laying is constructed based on the cable parameters of the substation, specifically including:

[0011] Obtaining cable laying parameters and cable logic parameters of the substation;

[0012] Modeling the cable laying according to the cable laying parameters to obtain a three-dimensional digital model;

[0013] The three-dimensional digital model is subjected to information fusion according to the cable logic parameters to obtain a three-dimensional digital logic model.

[0014] Furthermore, the cable laying parameters include at least cable configuration parameters, cable location parameters and cable pipe burial parameters;

[0015] Then, modeling the cable laying according to the cable laying parameters to obtain a three-dimensional digital model specifically includes:

[0016] Modeling is performed according to the cable configuration parameters to obtain a cable channel model;

[0017] Modeling is performed according to the cable position parameters to obtain a cable body model;

[0018] Modeling is performed according to the cable buried pipe parameters to obtain an electrical buried pipe model;

[0019] A comprehensive analysis is performed based on the spatial positions and interactions of the cable trench model, the cable body model and the electrical buried pipe model to determine a three-dimensional digital model of the transformer cable laying.

[0020] Further, the three-dimensional digital model is subjected to information fusion according to the cable logic parameters to obtain a three-dimensional digital logic model, which specifically includes:

[0021] Based on a preset configuration file, information fusion is performed on the cable logic parameters to obtain a configuration description file;

[0022] The configuration description file is stored in the three-dimensional digital model in the form of metadata to obtain a three-dimensional digital logical model.

[0023] Furthermore, the method of using the acquired multi-source measured data to train a preset BP neural network to obtain a neural network model for fault identification and location specifically includes:

[0024] Acquiring multi-source measured data of the substation cable;

[0025] Performing data processing according to the multi-source measured data to obtain a training sample set for training, wherein each training sample in the training sample set contains an energy value of a voltage traveling wave signal of the transformer cable and cable fault information;

[0026] The BP neural network is trained according to the training sample set to obtain a neural network model, wherein the neural network model takes the relative energy value as input and the cable fault information as output.

[0027] Furthermore, the voltage traveling wave signal is processed and then input into the neural network model to obtain the cable fault information output by the neural network model, which specifically includes:

[0028] Performing wavelet decomposition on the voltage traveling wave signal to obtain a target energy value;

[0029] The target relative energy value is input into the neural network model to obtain target cable fault information output by the neural network model.

[0030] Further, the cable fault information includes at least the fault type and the fault distance;

[0031] Then, positioning the cable fault information in the three-dimensional digital logic model to determine the cable information at the fault location specifically includes:

[0032] Cable identification and coordinate positioning are performed in the three-dimensional digital logic model according to the fault type and fault distance to determine the wiring information of the faulty cable.

[0033] To achieve the above-mentioned purpose, the second aspect of the present application provides a cable fault identification and location device, the device comprising: a simulation unit, a model training unit, an information acquisition unit and a fault location unit;

[0034] A model training unit, used to train a preset BP neural network using the acquired multi-source measured data to obtain a neural network model for fault identification and location, wherein the multi-source measured data are test data measured in different ways;

[0035] An information acquisition unit, used for acquiring a voltage traveling wave signal of a substation cable, and inputting the voltage traveling wave signal into the neural network model after performing signal processing to obtain cable fault information output by the neural network model;

[0036] A simulation unit, used to build a three-dimensional digital logic model of cable laying based on the cable parameters of the substation;

[0037] The fault location unit is used to locate the fault in the three-dimensional digital logic model according to the cable fault information, and determine the cable information at the fault location.

[0038] To achieve the above-mentioned purpose, the third aspect of the present application provides a computer-readable storage medium storing a computer program, characterized in that when the computer program is executed by a processor, the processor executes the steps of the method described in the first aspect.

[0039] To achieve the above-mentioned purpose, the fourth aspect of the present application provides a computer device, including a memory and a processor, characterized in that the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method described in the first aspect.

[0040] The embodiments of the present invention have the following beneficial effects:

[0041] The embodiment of the present invention proposes a cable fault identification and positioning method, which includes: using the obtained multi-source measured data to train the preset BP neural network to obtain a neural network model for fault identification and positioning, wherein the multi-source measured data is the test data measured by different methods; obtaining the voltage traveling wave signal of the substation cable, and inputting the voltage traveling wave signal into the neural network model after signal processing, to obtain the cable fault information output by the neural network model; constructing a three-dimensional digital logic model of cable laying based on the cable parameters of the substation; locating in the three-dimensional digital logic model according to the cable fault information, and determining the cable information at the fault. By constructing a three-dimensional digital logic model for cable laying, that is, performing a refined three-dimensional digital simulation of the cable laying of the substation, the geometric and topological information of the cable can be described, so that the cable laying can be accurately described, so that the fault location can be found more accurately; secondly, by fusing the multi-source measured data, more abundant and practical training data is obtained, and the training data is used to train the BP neural network to obtain the neural network model, so as to achieve a more accurate result of the output of the neural network model. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative work.

[0043] in:

[0044] Figure 1 A schematic diagram of a process of a cable fault identification and location method according to an embodiment of the present invention;

[0045] Figure 2 A hierarchical structure diagram of a physical circuit model in an embodiment of the present invention;

[0046] Figure 3 A voltage traveling wave diagram under a ground fault according to an embodiment of the present invention;

[0047] Figure 4 This is a structural block diagram of a cable fault identification and locating device according to an embodiment of the present invention;

[0048] Figure 5 Schematic diagram of the internal structure of a computer device in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 are within the scope of protection of the present invention.

[0050] Current cable laying fault location technology usually relies on a single data source, such as cable local temperature, current and other monitoring data. This single data source may not be able to provide all-round and comprehensive fault diagnosis information, resulting in limited accurate description of fault location and morphology, and low accuracy and reliability of fault judgment. At the same time, data analysis and processing may also be affected by noise interference, data loss and other factors, reducing the efficiency and accuracy of fault diagnosis.

[0051] To this end, the present invention proposes a three-dimensional digital cable laying fault location and identification technology based on multi-source data fusion to improve the accuracy and reliability of cable laying fault location technology, reduce the losses caused by faults, and ensure the stable operation of the power system.

[0052] Based on this, in the embodiment of the present application, a cable fault identification and positioning method is provided. Figure 1 , Figure 1 1 is a flow chart of a cable fault identification and location method according to an embodiment of the present invention, which specifically includes:

[0053] Step 100, using the acquired multi-source measured data to train a preset BP neural network to obtain a neural network model for fault identification and location, wherein the multi-source measured data is test data measured in different ways.

[0054] In the embodiment of the present invention, the multi-source measured data is test data obtained by measuring in different ways, and the test data includes fault information data of the transformer cable.

[0055] In a feasible embodiment, the multi-source measured data at least includes: online monitoring data, routine test data and diagnostic test data. Among them, the online monitoring data is the monitoring of the equipment connected to the cable when the transformer cable is in normal operation, monitoring the disconnection signal and fault signal of the cable, etc. The monitoring frequency and volume of the data are very high, and complete evaluation data can be obtained; routine test data is a regular test carried out by the operation and inspection department to evaluate the status of the equipment. The data volume is small, but the data is accurate and the frequency is low; diagnostic test data is often an evaluation test of the equipment carried out when the substation has an adverse situation. It is mainly a manual measurement test carried out after there is defect information or an alarm of an adverse working condition. The data is often entered manually, the data volume is small, and the data is scarce.

[0056] By acquiring different monitoring data in different ways, more abundant fault information data can be obtained. If the BP neural network is trained using multi-source measured data, the accuracy and reliability of BP neural network training can be improved.

[0057] Since the voltage traveling wave signal of the cable can fully reflect the state of the cable, in the embodiment of the present invention, the processed data of the voltage traveling wave signal in the multi-source measured data is used as input, and the cable fault information is used as output for training to obtain a neural network model that outputs the cable fault information after training.

[0058] Step 200, obtaining the voltage traveling wave signal of the substation cable, and inputting the voltage traveling wave signal into the neural network model after performing signal processing, so as to obtain the cable fault information output by the neural network model.

[0059] When a fault location request is received, the voltage traveling wave signal of the substation cable needs to be obtained first, and the voltage traveling wave signal is processed as the input value of the neural network model to obtain the cable fault information output by the neural network model.

[0060] Step 300: construct a three-dimensional digital logic model of cable laying based on the cable parameters of the substation.

[0061] In order to better locate the fault location and facilitate maintenance personnel to find the fault location faster and more accurately, the cable parameters of the substation are obtained, and professional 3D modeling software is used to model according to the cable parameters to obtain a 3D digital logical model of the substation cable laying. The 3D digital logical model contains the internal detailed data and logical relationships of the substation cable laying.

[0062] Step 400: locate the fault in the three-dimensional digital logic model according to the cable fault information to determine the cable information at the fault location.

[0063] Since the three-dimensional digital logical model can describe the geometric and topological information of the cable, the cable laying can be accurately described. Therefore, after obtaining the cable fault information, the fault location can be accurately located based on the cable fault information.

[0064] The embodiment of the present invention obtains richer and more practical training data by fusing multi-source measured data, and uses the training data to train the BP neural network to obtain a neural network model, so as to achieve a more accurate result output by the neural network model; secondly, by constructing a three-dimensional digital logic model for cable laying, that is, performing a refined three-dimensional digital simulation of the cable laying of the substation, the geometric and topological information of the cable can be described, so that the cable laying can be accurately described, so that the fault location can be located more accurately.

[0065] In order to obtain more comprehensive information about transformer cable laying and obtain a more accurate three-dimensional digital logic model, step 300 of the embodiment of the present invention constructs a three-dimensional digital logic model of cable laying based on the cable parameters of the substation, specifically including:

[0066] Step 310, obtain the cable laying parameters and cable logic parameters of the substation.

[0067] Step 320, modeling the cable laying according to the cable laying parameters to obtain a three-dimensional digital model.

[0068] Cable laying parameters are actual parameters of the cable body, carrier, position, buried pipe, etc. Cable laying parameters take into account the spatial position, geometry, material properties and other comprehensive information of cable buried pipe, cable body and cable trench, so as to ensure the accuracy and completeness of the three-dimensional digital model constructed according to the cable laying parameters. These data may come from multiple sources such as field measurements, geographic information systems (GIS), engineering drawings, etc.

[0069] In a feasible embodiment of the present invention, the cable laying parameters at least include cable configuration parameters, cable location parameters and cable buried pipe parameters. Based on this Step 320, the cable laying is modeled according to the cable laying parameters to obtain a three-dimensional digital model, which specifically includes:

[0070] Step 321, modeling is performed according to the cable configuration parameters to obtain the cable channel model.

[0071] The detailed modeling of the cable laying trench and the brackets in the trench not only needs to consider the width and depth of the overall trench, but also needs to configure and model its ancillary facilities. By dividing the cable channel into multiple parts such as the net width and net height of the cable trench, the relative position of the bracket, the bracket column, and the bracket arm, the cable configuration parameters are obtained, and the cable trench model is obtained based on the cable configuration parameters to achieve the purpose of detailed modeling.

[0072] Step 322, modeling is performed according to the cable position parameters to obtain the cable body model.

[0073] In cable laying, the cable arrangement of each layer of the bracket is different, and there are different requirements for whether different types of cables can be stacked in multiple layers. Under normal circumstances, except for the same circuit of single-core power cables for AC systems, which can be stacked in a herringbone shape, multiple power cables of the same important circuit are generally not suitable for stacking. Control and signal cables can be close or stacked in multiple layers. There are also relevant regulations on the arrangement of cables on the layer of the bracket. For example, the dual power circuit power cables of important circuits such as fire pumps or main transformer water spray, UPS power cables, and important protection power circuit control cables all need to be laid in layers and sides. Generally speaking, there are four layers of brackets, and the numbers 1-4 indicate different lengths and types of cables. The most typical configuration is shown in Table 1.

[0074] Table 1 Cable laying arrangement position table

[0075]

[0076] Therefore, the embodiment of the present invention obtains the cable position parameters and performs modeling based on the parameters to obtain a cable body model including information such as the cable direction, size, and different cable types, thereby achieving the purpose of refined modeling.

[0077] Step 323, modeling is performed according to the cable buried pipe parameters to obtain an electrical buried pipe model.

[0078] In cable laying, cable laying in pipes is usually required for indoor wall threading, cable laying through roads, equipment mechanism box to trenches, and short routes. Therefore, obtaining cable pipe laying parameters for modeling to obtain an electrical pipe laying model containing start and end point information, coordinates, pipe diameter, and other information can achieve the purpose of refined modeling.

[0079] Step 324, conduct a comprehensive analysis based on the spatial positions and interactions of the cable trench model, cable body model and electrical buried pipe model to determine the three-dimensional digital model of the transformer cable laying.

[0080] By analyzing the spatial position and interaction of the cable trench model, cable body model and electrical buried pipe model, the connection and influence relationship between them can be determined. For example, the contact between the cable buried pipe and the cable body, the influence of the topological structure of the cable trench on the cable, etc., a more accurate three-dimensional digital model can be obtained.

[0081] After the 3D digital model is established based on the cable trench model, cable body model and electrical buried pipe model, it needs to be verified and optimized. This can be done by comparing it with the actual scene, using ground measurement data or other reliable data sources for verification, and adjusting and optimizing the model based on the verification results to improve the accuracy and reliability of the model.

[0082] By comprehensively utilizing the information of cable buried pipes, cable bodies and cable configurations, a detailed three-dimensional digital model is established, providing a reliable foundation for digital cable laying fault location and identification technology.

[0083] Step 330, information fusion is performed on the three-dimensional digital model according to the cable logic parameters to obtain a three-dimensional digital logic model.

[0084] In order to obtain a scenario closer to the actual cable laying, the embodiment of the present invention also uses cable logic parameters to fuse information of the three-dimensional digital model. The cable logic parameters may include: basic equipment information, digital three-dimensional information and manually added information.

[0085] Basic equipment information mainly includes the data of cables and carriers, which can be represented by multiple structured data tables. Please refer to Table 2.

[0086] Table 2 Basic equipment information structured data table.

[0087]

[0088]

[0089] Digital three-dimensional information mainly includes digital elevation model information and refined modeling model data information. Digital elevation model information mainly includes contour lines, discrete elevation points, grid data, etc. in the three-dimensional digital modeling of substations; refined modeling model data information mainly includes the center line of cable trenches and channels, node information (number and coordinates) of cable laying, cable laying position information (number of layers of the bracket side shelves where the cable is laid on the cable channel and the offset on the shelf, etc.), digital information of equipment at both ends (equipment number, name, placement point coordinates, wiring point coordinates, etc.), and other auxiliary carrier digital modeling information.

[0090] Manually added information mainly includes some standard guidelines, management regulations and related requirements, as well as inspection records.

[0091] After obtaining the cable logical parameters composed of the above-mentioned multi-source data, step 330 performs information fusion on the three-dimensional digital model according to the cable logical parameters to obtain the three-dimensional digital logical model, specifically including: performing information fusion on the cable logical parameters based on a preset configuration file to obtain a configuration description file; storing the configuration description file in the form of metadata in the three-dimensional digital model to obtain the three-dimensional digital logical model.

[0092] Integrate logical data into the three-dimensional digital model, strengthen the connection between models, and more clearly describe the actual substation cable laying situation.

[0093] Specifically, the carrier information of the XML configuration file can be integrated. Using the modeling method based on the XML configuration file, the multi-source data information of the cable logic parameters is integrated into the configuration description file, and stored in the three-dimensional digital model in the form of metadata to obtain a three-dimensional digital logical model.

[0094] The XML configuration file format is used to describe the multi-source data information of the cable. The following takes a substation project as an example. Construct a SLCL (the root element of the substation secondary system physical circuit model XML file). For details, please refer to Figure 2 , Figure 2 It is a hierarchical structure diagram of the physical loop model in an embodiment of the present invention, with the substation as the main node, and the interval node region and cable node cable below, wherein the interval node region includes multiple cubicles, and the cubicle includes device device, unit unit, and in-cabinet wiring node information inwire.

[0095] To facilitate understanding of the technology of the present invention, the following specifically introduces how to obtain a neural network model with accurate fault identification and positioning based on multi-source measured data. Specifically, step 100 uses the acquired multi-source measured data to train a preset BP neural network to obtain a neural network model for fault identification and positioning, which specifically includes:

[0096] Step 110, obtain multi-source measured data of substation cables.

[0097] Step 120, performing data processing based on multi-source measured data to obtain a training sample set for training, wherein each training sample in the training sample set includes an energy value of a voltage traveling wave signal of a transformer cable and cable fault information.

[0098] Step 130, training the BP neural network according to the training sample set to obtain a neural network model, wherein the neural network model takes the relative energy value as input and the cable fault information as output.

[0099] In the embodiment of the present invention, the BP neural network architecture can have three or more layers, mainly including an input layer, one or more hidden layers and an output layer, and each layer is composed of a number of nodes. The nodes of each layer are usually connected by a full connection method, and the neurons in a layer are independent of each other.

[0100] The training and learning process of BP neural network is the process of constantly adjusting the connection weights. After initializing the weights, BP neural network is trained according to the training sample set. The error function is constructed and the weights are adjusted according to the gradient descent method of the learning rate so that the error between the output result and the expected output reaches the specified minimum value. The general error function uses the root mean square, which propagates the error to the weights layer by layer, and stops the iteration when the error reaches the minimum.

[0101] When considering the application of BP neural network modeling, the error function is modeled as:

[0102]

[0103] in, is the output value of the BP neural network, j is any one of the output neurons, and l is the number of preset output neurons, which can be expressed in matrix form as is the expected output value in the training sample set.

[0104] There are two main stages in BP neural network learning, namely the forward stage and the back propagation stage. In the forward propagation process, the gradient descent method with a learning rate is used to adjust the parameters in the negative gradient direction of the target. Given a learning rate η, the gradient descent rate of the weights between the input layer and the hidden layer neurons is:

[0105]

[0106] Where η is the learning rate of the given gradient descent method; E k is the error; v ih is the connection weight between the input layer neurons and the hidden neurons; i refers to any input layer neuron; h refers to any hidden neuron; j is any output layer neuron.

[0107] During the back propagation process, the gradient descent rate of the weights between the hidden layer and the output layer neurons is:

[0108]

[0109] In the formula, w hj is the connection weight between the hidden neurons and the output layer neurons.

[0110] In the training process of BP neural network, the gradient descent rate is used to adjust the weights of neurons connecting the input layer and hidden layer and neurons connecting the hidden layer and output layer. By fine-tuning the weights in the opposite direction of the gradient, the output of the network can be closer to the expected output, thereby reducing the error and improving the performance of the network. After a large number of experiments, it is known that the convergence speed is fastest in 0-100 iterations, and the rate is stable, dropping below 0.1; starting from 500 times, the error convergence rate tends to be horizontal, and finally the global optimal value is obtained around 1000 times, and the optimal error is 0.0013473, which meets the expected accuracy requirements.

[0111] After the trained neural network model is obtained, when a fault location request is received, the fault information can be obtained through the neural network model. Specifically, step 200 processes the voltage traveling wave signal and inputs it into the neural network model to obtain the cable fault information output by the neural network model, which specifically includes:

[0112] Step 210, perform wavelet decomposition on the voltage traveling wave signal to obtain the target energy value.

[0113] Step 220, input the target relative energy value into the neural network model to obtain the target cable fault information output by the neural network model.

[0114] When conducting diagnostic tests on-site at electromechanical engineering sites, analysis is generally performed based on voltage traveling wave signals. Figure 3 As shown, Figure 3 : is a voltage traveling wave diagram under a ground fault according to an embodiment of the present invention, wherein the forward traveling wave signal is a dotted line curve, and the reverse traveling wave signal is a solid line curve. Figure 3 It can be seen that the relative energy value of the voltage traveling wave signal shows different states under different states. Therefore, the cable fault condition and fault distance can be judged by extracting the forward and reverse traveling wave signals of the voltage at the time of the fault.

[0115] In order to effectively extract the traveling wave component in the traveling wave signal, this embodiment uses three-layer wavelet decomposition for the voltage traveling wave signal to remove high-frequency noise and low-frequency trends from the voltage traveling wave signal while maintaining a balance point, which can provide sufficient signal information and maintain calculation efficiency. The voltage traveling wave signal is decomposed by three-layer wavelet decomposition to form relative energy values ​​of forward traveling waves and relative energy values ​​of reverse traveling waves. These relative energy values ​​can describe the signal energy distribution under different faults and are used to characterize the fault signal characteristics. Under different fault conditions, the relative energy value is different. Therefore, it can be used as the input value of the BP neural network. After calculation by the neural network model, the cable fault information corresponding to the relative energy value is obtained.

[0116] In an embodiment of the present invention, the cable fault information includes at least the fault type and the fault distance. Then, step 400 is to locate the cable fault in the three-dimensional digital logic model according to the cable fault information and determine the cable information at the fault location. Specifically, it includes: according to the fault type and the fault distance, the cable is identified and the coordinates are located in the three-dimensional digital logic model to determine the wiring information of the faulty cable.

[0117] For details, please refer to Table 3, which is a table showing the relationship between relative energy values ​​and fault information samples in the neural network model. In Table 3, different vectors output by the neural network model represent different fault types. The fault feature vector is a vector representation used to characterize different fault states or normal states. For example, the normal state is 000, the A-phase grounding fault is 100, and the vector corresponding to the B-phase grounding fault is 010. The output fault distance is the distance from a single end to the fault point.

[0118] Based on this, after obtaining the fault type and fault distance, accurate coordinate positioning and cable identification can be performed in the three-dimensional digital logical model, and the wiring information of the faulty cable can be found, such as the cable category, the corresponding positions and names of the cable starting and ending devices, the function and model of the cable, and the loops involved in the cable, and the cable fault information and the wiring information involved in the cable can be displayed in the three-dimensional digital logical model.

[0119] Table 3 Relationship between relative energy value and fault information samples in the neural network model

[0120]

[0121] The present invention also provides a cable fault identification and location device, see Figure 4 , Figure 4 4 is a structural block diagram of a cable fault identification and locating device according to an embodiment of the present invention. The device includes: a simulation unit 401 , a model training unit 402 , an information acquisition unit 403 and a fault locating unit 404 .

[0122] The model training unit 401 is used to train a preset BP neural network using the acquired multi-source measured data to obtain a neural network model for fault identification and location, wherein the multi-source measured data is test data measured in different ways.

[0123] The information acquisition unit 402 is used to acquire the voltage traveling wave signal of the substation cable, and input the voltage traveling wave signal into the neural network model after performing signal processing to obtain the cable fault information output by the neural network model.

[0124] The simulation unit 403 is used to construct a three-dimensional digital logic model of cable laying based on the cable parameters of the substation.

[0125] The fault location unit 404 is used to locate the fault in the three-dimensional digital logic model according to the cable fault information, and determine the cable information at the fault location.

[0126] The cable fault identification and locating device proposed in the embodiment of the present invention constructs a three-dimensional digital logical model for cable laying, that is, performs a refined three-dimensional digital simulation of the cable laying of the substation, which can describe the geometric and topological information of the cable, so that the cable laying can be accurately described, so that the fault location can be found more accurately; secondly, by fusing multi-source measured data, a richer and more practical training sample set is obtained, and the training samples are used to train the BP neural network to obtain a neural network model, so as to achieve a more accurate output result of the neural network model.

[0127] Figure 5 FIG. 1 shows an internal structure diagram of a computer device in one embodiment of the present invention. The computer device may be a terminal or a system. Figure 5 As shown, the computer device includes a processor, a memory and a network interface connected via a system bus. The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor may implement each step in the above method embodiment. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor may implement each step in the above method embodiment. Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0128] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes each step in the above method embodiment.

[0129] In one embodiment, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the processor executes each step in the above method embodiment.

[0130] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0131] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0132] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A cable fault identification and location method, characterized in that: The method comprises: Using the acquired multi-source measured data to train the preset BP neural network, a neural network model for fault identification and location is obtained, wherein the multi-source measured data is test data measured in different ways; Acquire a voltage traveling wave signal of a substation cable, perform signal processing on the voltage traveling wave signal, and input the signal into the neural network model to obtain cable fault information output by the neural network model; Constructing a three-dimensional digital logic model of cable laying based on the cable parameters of the substation; Positioning the cable fault in the three-dimensional digital logic model according to the cable fault information to determine the cable information at the fault location; The three-dimensional digital logic model of cable laying based on the cable parameters of the substation is constructed, specifically including: Obtaining cable laying parameters and cable logic parameters of the substation; Modeling the cable laying according to the cable laying parameters to obtain a three-dimensional digital model; Performing information fusion on the three-dimensional digital model according to the cable logic parameters to obtain a three-dimensional digital logic model; Wherein, the cable laying parameters at least include cable configuration parameters, cable location parameters and cable buried pipe parameters; Then, modeling the cable laying according to the cable laying parameters to obtain a three-dimensional digital model specifically includes: Modeling is performed according to the cable configuration parameters to obtain a cable channel model; Modeling is performed according to the cable position parameters to obtain a cable body model; Modeling is performed according to the cable buried pipe parameters to obtain an electrical buried pipe model; A three-dimensional digital model of the substation cable laying is determined by performing a comprehensive analysis based on the spatial positions and interactions of the cable channel model, the cable body model and the electrical buried pipe model; The step of fusing the three-dimensional digital model according to the cable logic parameters to obtain the three-dimensional digital logic model specifically includes: Based on a preset configuration file, information fusion is performed on the cable logic parameters to obtain a configuration description file; The configuration description file is stored in the three-dimensional digital model in the form of metadata to obtain a three-dimensional digital logical model.

2. The method according to claim 1, characterized in that The method of using the acquired multi-source measured data to train the preset BP neural network to obtain a neural network model for fault identification and location specifically includes: Acquiring multi-source measured data of the substation cable; Performing data processing according to the multi-source measured data to obtain a training sample set for training, wherein each training sample in the training sample set contains an energy value of a voltage traveling wave signal of the substation cable and cable fault information; The BP neural network is trained according to the training sample set to obtain a neural network model, wherein the neural network model takes the relative energy value as input and the cable fault information as output.

3. The method according to claim 2, characterized in that The step of processing the voltage traveling wave signal and inputting it into the neural network model to obtain the cable fault information output by the neural network model specifically includes: Performing wavelet decomposition on the voltage traveling wave signal to obtain a target relative energy value; The target relative energy value is input into the neural network model to obtain target cable fault information output by the neural network model.

4. The method according to claim 1, characterized in that: The cable fault information includes at least the fault type and the fault distance; Then, positioning the cable fault information in the three-dimensional digital logic model to determine the cable information at the fault location specifically includes: Cable identification and coordinate positioning are performed in the three-dimensional digital logic model according to the fault type and fault distance to determine the wiring information of the faulty cable.

5. A cable fault identification and positioning device, characterized in that: The device comprises: a simulation unit, a model training unit, an information acquisition unit and a fault location unit; A model training unit, used to train a preset BP neural network using the acquired multi-source measured data to obtain a neural network model for fault identification and location, wherein the multi-source measured data are test data measured in different ways; An information acquisition unit, used for acquiring a voltage traveling wave signal of a substation cable, and inputting the voltage traveling wave signal into the neural network model after performing signal processing to obtain cable fault information output by the neural network model; A simulation unit, used to build a three-dimensional digital logic model of cable laying based on the cable parameters of the substation; A fault location unit, used to locate the fault in the three-dimensional digital logic model according to the cable fault information, and determine the cable information at the fault location; The simulation unit is also used to obtain the cable laying parameters and cable logic parameters of the substation; Modeling the cable laying according to the cable laying parameters to obtain a three-dimensional digital model; Performing information fusion on the three-dimensional digital model according to the cable logic parameters to obtain a three-dimensional digital logic model; Wherein, the cable laying parameters at least include cable configuration parameters, cable location parameters and cable buried pipe parameters; Then the simulation unit is further used to perform modeling according to the cable configuration parameters to obtain a cable channel model; Modeling is performed according to the cable position parameters to obtain a cable body model; Modeling is performed according to the cable buried pipe parameters to obtain an electrical buried pipe model; A three-dimensional digital model of the substation cable laying is determined by performing a comprehensive analysis based on the spatial positions and interactions of the cable channel model, the cable body model and the electrical buried pipe model; The simulation unit is further used to perform information fusion on the cable logic parameters based on a preset configuration file to obtain a configuration description file; The configuration description file is stored in the three-dimensional digital model in the form of metadata to obtain a three-dimensional digital logical model.

6. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 4.

7. A computer device comprising a memory and a processor, characterized in that: The memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 4.

Images