Channel Cable Fault Location Method and Related Equipment Based on GIM Parametric Model

Abstract

The present application relates to the technical field of power fault troubleshooting, and provides a method and related equipment for channel cable fault location based on a GIM parametric model. The method of the present application can perform GIM parametric processing on the geometric parameter set, topological connection matrix, and equipment space coordinates of the channel cable to construct a model; based on this model, signal observation points are deployed to obtain forward and backward traveling wave signals, and the corresponding energy entropy feature vectors are obtained through wavelet packet decomposition and reconstruction; a preset regression mapping model is used to determine the fault distance, and then the fault location is obtained in combination with the position of the signal observation point; among them, the present application can combine the physical layout of the cable system and the quantitative analysis of fault characteristics, can effectively overcome difficulties such as complex non-linear characteristics and significant noise interference, and can achieve high-precision location of the fault location of the channel cable in complex laying scenarios.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of power failure troubleshooting, and in particular to a method for locating channel cable faults based on a GIM parametric model and related devices. Background Art

[0002] During the construction of the power grid, it is usually necessary to lay concealed cables. However, due to the complex operating environment of the concealed cables, the traditional method of manual survey cannot quickly and effectively locate cable faults. Therefore, there is currently a solution that analyzes the traveling wave signals of the cable system and diagnoses cable faults based on the analysis results, which can effectively improve the fault location efficiency compared to the traditional manual method.

[0003] However, the traditional traveling wave analysis method can only perform simple feature extraction on time-domain or frequency-domain signals, and is generally only applicable to concealed cable laying scenarios with linear characteristics and no obvious noise interference. When facing complex laying scenarios with non-linear characteristics and significant noise interference, the traditional traveling wave analysis method has certain limitations in the extraction and modeling accuracy of fault signals, and cannot meet the high-precision fault location requirements in engineering practice. Summary of the Invention

[0004] The embodiments of the present application provide a method for locating channel cable faults based on a GIM parametric model, which can collect traveling wave signals by constructing a GIM parametric model, perform wavelet packet decomposition and reconstruction on the traveling wave signals to obtain energy entropy features, and then determine the fault distance based on the energy entropy features, and combine with the GIM parametric model to obtain the fault location, so as to achieve high-precision fault location.

[0005] To achieve the above object, in a first aspect, the present application provides a method for locating channel cable faults based on a GIM parametric model, including: performing GIM parametric processing on the geometric parameter set, topological connection matrix, and device space coordinates corresponding to the channel cable to construct a GIM parametric model; deploying a plurality of signal observation points based on the GIM parametric model, and obtaining the forward traveling wave signal and reverse traveling wave signal of the voltage during cable fault through the signal observation points; respectively performing wavelet packet decomposition and reconstruction processing on the forward traveling wave signal and the reverse traveling wave signal to obtain a forward energy entropy feature vector and a reverse energy entropy feature vector; determining the fault distance corresponding to the forward energy entropy feature vector and the reverse energy entropy feature vector based on a preset regression mapping model; and obtaining the fault location according to the fault distance and the deployment position of the corresponding signal observation point in the GIM parametric model.

[0006] In some embodiments, performing GIM parameterization processing on the geometric parameter set, topological connection matrix, and device space coordinates corresponding to the channel cable to construct a GIM parameterization model includes: converting the cross-sectional dimensions, bending radius, and elevation data of the cable channel into a geometric parameter set, where the geometric parameter set characterizes the spatial form of the cable channel and the three-dimensional structure of the laying path; defining the logical connection relationship between the cable branch nodes and devices in the cable channel to generate a topological connection matrix, where the topological connection matrix characterizes the electrical connectivity of each node in the cable network; integrating the three-dimensional space coordinates and electrical parameters of the auxiliary devices in the cable channel to obtain device space coordinates, where the device space coordinates characterize the physical positions of the devices in the channel; and generating a GIM parameterization model including the cable path, device positions, and topological relationships based on the geometric parameter set, topological connection matrix, and device space coordinates.

[0007] In some embodiments, performing wavelet packet decomposition and reconstruction processing on the forward traveling wave signal and the reverse traveling wave signal respectively to obtain a forward energy entropy feature vector and a reverse energy entropy feature vector includes: performing three-layer wavelet packet decomposition on the forward traveling wave signal and the reverse traveling wave signal respectively, generating low-frequency and high-frequency subbands for each layer of decomposition to obtain time-domain waveforms of multiple frequency bands; calculating energy entropy eigenvalues for the reconstructed signals of each frequency band, where the energy entropy eigenvalue is the product of the frequency band energy ratio based on Parseval's theorem and the Shannon entropy; and arranging the energy entropy eigenvalues of multiple frequency bands in frequency order to form a forward energy entropy feature vector and a reverse energy entropy feature vector.

[0008] In some embodiments, performing three-layer wavelet packet decomposition on the forward traveling wave signal and the reverse traveling wave signal respectively, generating low-frequency and high-frequency subbands for each layer of decomposition to obtain time-domain waveforms of multiple frequency bands includes: decomposing the forward traveling wave signal and the reverse traveling wave signal based on a preset wavelet basis function to generate wavelet packet decomposition coefficients for each layer; and performing threshold filtering processing on the wavelet packet decomposition coefficients to reconstruct the low-frequency subband and high-frequency subband signals layer by layer to obtain time-domain waveforms of multiple frequency bands.

[0009] In some embodiments, determining the fault distances corresponding to the forward energy entropy feature vector and the reverse energy entropy feature vector based on a preset regression mapping model includes: obtaining a training data set, where the training data set contains the forward and reverse energy entropy feature vectors of historical fault cases and the corresponding true fault distances; training a regression mapping model using a support vector regression algorithm based on the training data set, where the regression mapping model takes the energy entropy feature vector as the input and the fault distance as the output; optimizing the hyperparameters of the regression mapping model through cross-validation until the prediction error converges to a preset threshold; and inputting the forward energy entropy feature vector and the reverse energy entropy feature vector into the optimized regression mapping model to obtain the fault distance.

[0010] In some embodiments, obtaining the fault location according to the fault distance and the deployment position of the corresponding signal observation point in the GIM parametric model includes: determining the propagation paths corresponding to the forward traveling wave signal and the backward traveling wave signal according to the GIM parametric model and the deployment position of the corresponding signal observation point in the GIM parametric model; marking the fault location in the GIM parametric model according to the propagation paths, the fault distance, and the deployment position.

[0011] In some embodiments, deploying a plurality of signal observation points based on the GIM parametric model and obtaining the forward traveling wave signal and the backward traveling wave signal of the voltage during a cable fault through the signal observation points includes: determining the cable position according to the geometric parameter set and the topological connection matrix of the cable trench, and deploying a plurality of signal observation points based on the cable position; synchronously collecting the forward traveling wave signal and the backward traveling wave signal of the voltage within a preset time window during a cable fault through the signal observation points.

[0012] To achieve the above object, in a second aspect, an embodiment of the present application provides a channel cable fault location device based on a GIM parametric model, including: a model construction module, configured to perform GIM parametric processing on the geometric parameter set, the topological connection matrix, and the device space coordinates corresponding to the channel cable to construct a GIM parametric model; a signal acquisition module, configured to deploy a plurality of signal observation points based on the GIM parametric model and obtain the forward traveling wave signal and the backward traveling wave signal of the voltage during a cable fault through the signal observation points; a feature extraction module, configured to perform wavelet packet decomposition and reconstruction processing on the forward traveling wave signal and the backward traveling wave signal respectively to obtain a forward energy entropy feature vector and a backward energy entropy feature vector; a distance determination module, configured to determine the fault distance corresponding to the forward energy entropy feature vector and the backward energy entropy feature vector based on a preset regression mapping model; a position location module, configured to obtain the fault location according to the fault distance and the deployment position of the corresponding signal observation point in the GIM parametric model.

[0013] To achieve the above object, in a third aspect, an embodiment of the present application provides an electronic device, including: at least one processor; at least one memory, configured to store at least one program; when at least one of the programs is executed by at least one of the processors, implementing the channel cable fault location method based on the GIM parametric model as described in any item of the first aspect.

[0014] To achieve the above object, in a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, storing computer-executable instructions, where the computer-executable instructions are used to execute the channel cable fault location method based on the GIM parametric model as described in any item of the first aspect.

[0015] The embodiment of the present application provides a method and related devices for locating channel cable faults based on a GIM parametric model. The geometric parameter set, topological connection matrix, and device spatial coordinates corresponding to the channel cable can be subjected to GIM parametric processing to construct a GIM parametric model. Based on the GIM parametric model, multiple signal observation points are deployed, and the forward traveling wave signal and reverse traveling wave signal of the voltage during cable faults are obtained through the signal observation points. The forward traveling wave signal and the reverse traveling wave signal are respectively subjected to wavelet packet decomposition and reconstruction processing to obtain a forward energy entropy feature vector and a reverse energy entropy feature vector. Based on a preset regression mapping model, the fault distance corresponding to the forward energy entropy feature vector and the reverse energy entropy feature vector is determined. According to the fault distance and the deployment position of the corresponding signal observation point in the GIM parametric model, the fault position is obtained. Among them, it can be understood that, compared with traditional modeling methods, the GIM parametric model can accurately present the complex geometric features, topological structure, and device position information of the cable environment, covering many key factors such as the cable laying path, channel shape, and distribution of auxiliary equipment. Based on the GIM parametric model, multiple signal observation points are deployed, and the accurate information provided by the model can be fully utilized to effectively collect the traveling wave signals during faults, and the forward traveling wave signal and the reverse traveling wave signal are subjected to wavelet packet decomposition and reconstruction processing, and the forward energy entropy feature vector and the reverse energy entropy feature vector are effectively extracted for the calculation of the fault distance. Based on the calculated fault distance and the deployment position of the signal observation point in the GIM parametric model, the fault position is determined, so as to combine the physical layout of the cable system with the quantitative analysis results of the fault characteristics, and high-precision positioning of the fault position can be achieved in complex laying scenarios with complex non-linear characteristics and significant noise interference. Description of the Drawings

[0016] Figure 1 It is a flowchart of the method for locating channel cable faults based on a GIM parametric model provided by an embodiment of the present application;

[0017] Figure 2 It is a flowchart of the method for constructing a GIM parametric model in the method for locating channel cable faults based on a GIM parametric model provided by an embodiment of the present application;

[0018] Figure 3 It is a flowchart of the method for obtaining a forward energy entropy feature vector and a reverse energy entropy feature vector in the method for locating channel cable faults based on a GIM parametric model provided by an embodiment of the present application;

[0019] Figure 4 It is a schematic diagram of three-layer wavelet packet decomposition in the method for locating channel cable faults based on a GIM parametric model provided by an embodiment of the present application;

[0020] Figure 5 In the channel cable fault location method based on the GIM parametric model provided by an embodiment of the present application, the voltage traveling wave diagram under normal conditions;

[0021] Figure 6 In the channel cable fault location method based on the GIM parametric model provided by an embodiment of the present application, the voltage traveling wave diagram under single-phase grounding fault;

[0022] Figure 7 In the channel cable fault location method based on the GIM parametric model provided by an embodiment of the present application, the method flow chart for obtaining the time-domain waveforms of multiple frequency bands;

[0023] Figure 8 In the channel cable fault location method based on the GIM parametric model provided by an embodiment of the present application, the time-domain diagram of the forward traveling wave wavelet decomposition and reconstruction;

[0024] Figure 9 In the channel cable fault location method based on the GIM parametric model provided by an embodiment of the present application, the time-domain diagram of the reverse traveling wave wavelet decomposition and reconstruction;

[0025] Figure 10 The structural schematic diagram of the electronic device provided by an embodiment of the present application. Detailed implementation manners

[0026] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0027] In some embodiments, although the functional modules are divided in the system schematic diagram and the logical sequence is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the system or the sequence in the flowchart. Terms such as first and second in the description, claims and the above-mentioned drawings are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence.

[0028] In addition, unless otherwise clearly specified and limited, the term "connected / linked" should be understood in a broad sense. For example, it can be a fixed connection or a movable connection, or a detachable connection or a non-detachable connection, or an integral connection; it can be a mechanical connection, an electrical connection or can communicate with each other; it can be directly connected or indirectly connected through an intermediate medium.

[0029] In the description of the embodiments of the present application, the descriptions referring to terms such as "one embodiment / implementation", "another embodiment / implementation" or "certain embodiments / implementations", "in the above embodiments / implementations", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiments or examples are included in at least two embodiments or implementations disclosed in the present application. In the disclosure of the present application, the schematic expressions of the above terms do not necessarily refer to the same embodiment or implementation. It should be noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that in the flowchart.

[0030] The solution of the present application will be described below with reference to the accompanying drawings.

[0031] Currently, during the construction of the power grid, it is usually necessary to lay concealed cables. However, due to the complex operating environment of the concealed cables, the traditional scheme of manual survey cannot quickly and effectively locate cable faults. Therefore, there is currently a scheme for analyzing the traveling wave signals of the cable system and diagnosing cable faults based on the analysis results, which can effectively improve the fault location efficiency compared with the traditional manual scheme. However, the traditional traveling wave analysis scheme can only perform simple feature extraction on time-domain or frequency-domain signals, and is generally only applicable to the concealed cable laying scenarios with linear characteristics and no obvious noise interference. When facing complex laying scenarios with complex non-linear characteristics and significant noise interference, the traditional traveling wave analysis scheme has certain limitations in the extraction and modeling accuracy of fault signals, and cannot meet the high-precision positioning requirements for fault locations in engineering practice.

[0032] Based on this, the embodiments of the present application provide a method for locating channel cable faults based on a GIM parametric model. It can collect traveling wave signals by constructing a GIM parametric model, perform wavelet packet decomposition and reconstruction on the traveling wave signals to obtain energy entropy features, and then determine the fault distance through the energy entropy features, and combine the GIM parametric model to obtain the fault location, so as to achieve high-precision positioning of the fault location.

[0033] Refer to Figure 1 , Figure 1 is the flowchart of the method for locating channel cable faults based on the GIM parametric model provided by an embodiment of the present application; To achieve the above object, in a first aspect, the present application provides a method for locating channel cable faults based on a GIM parametric model, including but not limited to the following steps:

[0034] Step S110, perform GIM parametric processing on the geometric parameter set, topological connection matrix, and device space coordinates corresponding to the channel cable to construct a GIM parametric model;

[0035] Step S120: Deploy multiple signal observation points based on the GIM parametric model, and obtain the forward traveling wave signal and the backward traveling wave signal of the voltage during cable faults through the signal observation points;

[0036] Step S130: Perform wavelet packet decomposition and reconstruction processing on the forward traveling wave signal and the backward traveling wave signal respectively to obtain the forward energy entropy feature vector and the backward energy entropy feature vector;

[0037] Step S140: Based on the preset regression mapping model, determine the fault distances corresponding to the forward energy entropy feature vector and the backward energy entropy feature vector;

[0038] Step S150: Obtain the fault location according to the fault distance and the deployment positions of the corresponding signal observation points in the GIM parametric model.

[0039] Among them, the method of this application first performs GIM parametric processing on the geometric parameter set, topological connection matrix and equipment space coordinates of the channel cable to construct a model; based on this model, signal observation points are deployed to obtain the forward and backward traveling wave signals, and the corresponding energy entropy feature vectors are obtained through wavelet packet decomposition and reconstruction; the preset regression mapping model is used to determine the fault distance, and then the fault location is obtained by combining the position of the signal observation point; among them, this application can combine the physical layout of the cable system and the quantitative analysis of fault characteristics, and can effectively overcome difficulties such as complex non-linear characteristics and significant noise interference, and can achieve high-precision positioning of the fault location of the channel cable in complex laying scenarios.

[0040] It can be understood that compared with traditional modeling methods, the GIM parametric model can accurately present the complex geometric characteristics, topological structure and equipment position information of the cable environment, covering many key factors such as the cable laying path, channel shape and distribution of auxiliary equipment. Deploying multiple signal observation points based on the GIM parametric model can make full use of the accurate information provided by the model to effectively collect the traveling wave signals during faults, and perform wavelet packet decomposition and reconstruction processing on the forward traveling wave signal and the backward traveling wave signal, effectively extracting the forward energy entropy feature vector and the backward energy entropy feature vector for calculating the fault distance, and determining the fault location according to the calculated fault distance and the deployment position of the signal observation point in the GIM parametric model, so as to combine the physical layout of the cable system with the quantitative analysis results of fault characteristics, and can achieve high-precision positioning of the fault location in complex laying scenarios with complex non-linear characteristics and significant noise interference.

[0041] In some embodiments, it can be understood that multiple signal observation points are deployed based on the GIM parametric model. This deployment method makes full use of the accurate information provided by the model, can scientifically and reasonably select signal acquisition positions in the cable system, and improves the effectiveness of the signal observation point layout compared with random or empirical deployment. These signal observation points can accurately obtain the forward traveling wave signal and the backward traveling wave signal of the voltage during cable faults. Further, by performing wavelet packet decomposition and reconstruction processing on the forward traveling wave signal and the backward traveling wave signal respectively, the present application can deeply explore the characteristics of the traveling wave signal in different frequency bands, effectively extract the forward energy entropy feature vector and the backward energy entropy feature vector. Compared with the simple signal processing means in the traditional scheme, it can better adapt to the complexity and variability of cable fault signals, decompose the signal into multiple frequency bands, highlight the essential characteristics of the fault signal, and improve the accuracy of fault feature extraction. Further, based on a preset regression mapping model, the fault distance corresponding to the forward energy entropy feature vector and the backward energy entropy feature vector is determined. This regression mapping model can be obtained through the training and optimization of sample data, and can accurately map the energy entropy feature vector to the fault distance. Compared with the traditional fault distance calculation method, it has higher accuracy and adaptability. Furthermore, the fault location can be determined based on the calculated fault distance and the deployment position of the signal observation point in the GIM parametric model, combining the physical layout of the cable system with the quantitative analysis results of fault characteristics, overcoming the limitations of relying solely on signal processing or only on geographical information for positioning, and can also accurately map the fault distance information to the actual cable position, quickly and accurately locate the fault point in the channel cable, facilitating the timely implementation of repair and maintenance work, effectively shortening the fault repair time, and reducing the losses caused by cable faults.

[0042] It should be noted that the method of the present application takes the GIM parametric model as the core, facilitating flexible modeling and signal acquisition scheme adjustment for channel cable systems with different specifications and layouts. Whether it is a simple small cable network or a complex large industrial cable system, efficient and accurate fault location can be achieved by appropriately adjusting the model parameters and the deployment of signal observation points. At the same time, the preset regression mapping model can also be optimized and trained according to the characteristics of the specific cable system, further enhancing the adaptability and scalability of the entire fault location method, and being able to exhibit good fault location performance in a variety of actual scenarios.

[0043] Refer to Figure 2 , Figure 2 FIG. is a flowchart of a method for constructing a GIM parametric model in a channel cable fault location method based on a GIM parametric model provided by an embodiment of the present application. In some embodiments, GIM parametric processing is performed on the geometric parameter set, topological connection matrix, and device space coordinates corresponding to the channel cable to construct a GIM parametric model, including but not limited to the following steps:

[0044] Step S210: Convert the cross-sectional dimensions, bending radius, and elevation data of the cable trench into a set of geometric parameters, which characterize the three-dimensional structure of the spatial form of the cable trench and the laying path.

[0045] Step S220: Define the logical connection relationship between the cable branch nodes and equipment in the cable trench to generate a topological connection matrix, which characterizes the electrical connectivity of each node in the cable network.

[0046] Step S230: Integrate the three-dimensional spatial coordinates and electrical parameters of the auxiliary equipment in the cable trench to obtain equipment spatial coordinates, which characterize the physical positions of the equipment in the trench.

[0047] Step S240: Generate a GIM parametric model that includes the cable path, equipment positions, and topological relationships based on the set of geometric parameters, topological connection matrix, and equipment spatial coordinates.

[0048] In some embodiments, perform GIM parametric processing on the set of geometric parameters, topological connection matrix, and equipment spatial coordinates corresponding to the trench cables to construct a GIM parametric model. The specific steps are as follows: First, convert the cross-sectional dimensions, bending radius, elevation data, etc. of the cable trench into a set of geometric parameters, which are used to characterize the three-dimensional structure of the spatial form of the cable trench and the laying path. Through these parameters, the shape, size, and spatial orientation of the trench can be accurately described, providing a basis for subsequent modeling and analysis. Further, define the logical connection relationship between the cable branch nodes and equipment in the cable trench to generate a topological connection matrix, which is used to characterize the electrical connectivity of each node in the cable network, that is, the connection sequence and method between each section of the cable and its association with the equipment, which helps to accurately reflect the connection structure of the cable network in the model. Further, integrate the three-dimensional spatial coordinates and electrical parameters of the auxiliary equipment in the cable trench to obtain equipment spatial coordinates, which are used to characterize the physical positions of the equipment in the trench. By integrating the three-dimensional spatial coordinates and electrical parameters of the equipment into the model, the equipment distribution in the trench can be more comprehensively described. Further, generate a GIM parametric model that includes the cable path, equipment positions, and topological relationships based on the set of geometric parameters, topological connection matrix, and equipment spatial coordinates. This model combines the geometric, connection, and equipment information of the trench, providing detailed and accurate basic data support for subsequent cable fault location, enabling the rapid and accurate location of the fault position when a fault occurs, and improving the operation and maintenance efficiency and safety of the power system.

[0049] In some embodiments, GIM parametric modeling can parameterize the cable laying environment and geometric features, integrating information such as engineering terrain, cable trenches, and equipment foundation locations into a unified model. It mainly involves trench parameterization, cable layout parameterization, equipment parameterization, and topological relationship parameterization. It can be understood that the trench not only includes the cable body but also involves support structures, equipment connection components, and other auxiliary facilities, and it is necessary to extract its geometric characteristics, such as trench width, height, length, and bending radius, as parametric model information.

[0050] Specifically, trench parameterization is a mathematical description of the geometric features and spatial distribution of cable laying trenches to support their flexibility and accuracy in digital modeling and engineering analysis. The core of its modeling lies in parameterizing the geometric parameter set, topological connection matrix, and equipment spatial coordinates, and assigning the parameterized data to a 3D model.

[0051] The geometric parameter set of the trench can be represented by the set J:

[0052] (1)

[0053] In the formula, W is the available width of the trench cross-section, in meters; H is the net clearance height inside the trench from the bottom to the top; L is the total extended length of the trench along its centerline from the starting point to the ending point; R is the curvature radius of the trench in the curved section, used to represent the smoothness of the curved part; θ is the inclination angle of the trench in the ramp section, used to calibrate the vertical change of the trench.

[0054] Among them, the bending radius R can be expressed as:

[0055] (2)

[0056] (3)

[0057] In the formula, k is the curvature, , and , are the first and second derivatives of the coordinates, respectively.

[0058] It can be understood that spatial topology parameterization includes the topological connection matrix and equipment spatial coordinates. The spatial topology parameterization expression is the core part of cable laying and trench geometric feature modeling. By establishing the spatial relationship between the 3D geometric models of cables, support structures, and trenches, a high-precision description of cable laying is achieved; among them, the node coordinates describe the geometric positions of key points of cables or trenches, such as cable starting points, turning points, and intersection points, etc.; through node coordinate topology, a 3D coordinate terrain information model of the trench can be constructed.

[0059] The node geometric coordinates are calculated as shown in Equation (4):

[0060] (4)

[0061] In the formula, is the three-dimensional coordinate of the th node; ( ) is the three-dimensional coordinate of the th sampling point in the point cloud data; M is the total number of sampling points.

[0062] In summary, in this application, the key geometric parameters of the channel can be extracted first, the key geometric parameters can be standardized using a mathematical model to form a channel parameterization model, and a set of geometric parameters can be obtained; further, a three-dimensional digital model of the buried pipe in the equipment area, the cable body, and the channel foundation can be established and input into the channel parameterization model, and the cable layout and the channel foundation geometric model can be integrated through splicing and fusion to obtain a topological connection matrix; further, engineering terrain data can be introduced to form three-dimensional coordinate data, the coordinates of each node can be defined, and the nodes can be connected to auxiliary facilities such as cables and channel supports using topological relationships to obtain the equipment space coordinates. After combining with the channel parameterization model, a three-dimensional digital model of cable laying can be finally generated

[0063] Reference Figure 3 , Figure 3 is the method flow chart for obtaining the forward energy entropy eigenvector and the reverse energy entropy eigenvector in the channel cable fault location method based on the GIM parameterization model provided by an embodiment of this application; in some embodiments, the forward traveling wave signal and the reverse traveling wave signal are respectively subjected to wavelet packet decomposition and reconstruction processing to obtain the forward energy entropy eigenvector and the reverse energy entropy eigenvector, including but not limited to the following steps:

[0064] Step S310, perform three-layer wavelet packet decomposition on the forward traveling wave signal and the reverse traveling wave signal respectively. Each layer of decomposition generates low-frequency and high-frequency sub-bands to obtain the time-domain waveforms of multiple frequency bands;

[0065] Step S320, calculate the energy entropy eigenvalue for the reconstructed signal of each frequency band. The energy entropy eigenvalue is the product of the frequency band energy ratio based on Parseval's theorem and the Shannon entropy;

[0066] Step S330, arrange the energy entropy eigenvalues of multiple frequency bands in the order of frequency to form the forward energy entropy eigenvector and the reverse energy entropy eigenvector.

[0067] In some embodiments, the forward traveling wave signal and the reverse traveling wave signal are respectively subjected to wavelet packet decomposition and reconstruction processing to obtain a forward energy entropy feature vector and a reverse energy entropy feature vector. The specific steps are as follows: First, the forward traveling wave signal and the reverse traveling wave signal are respectively subjected to three-layer wavelet packet decomposition. In each layer of decomposition, the signal is decomposed into low-frequency and high-frequency sub-bands, thereby obtaining time-domain waveforms in multiple frequency bands. Through this decomposition, the characteristics of the signal in different frequency bands can be captured, especially the fault characteristic information in the high-frequency components. Further, the energy entropy eigenvalue is calculated for the reconstructed signal of each frequency band. The energy entropy eigenvalue is the product of the frequency band energy ratio based on Parseval's theorem and the Shannon entropy. Specifically, according to Parseval's theorem, the total energy of the signal is equal to the sum of the energies of each frequency band, and the Shannon entropy reflects the complexity and uncertainty of the signal. By calculating the product of the energy ratio of each frequency band and the Shannon entropy, the energy entropy eigenvalue of this frequency band can be obtained, which helps to more accurately reflect the change characteristics of the fault signal. Further, the energy entropy eigenvalues of multiple frequency bands are arranged in frequency order to form a forward energy entropy feature vector and a reverse energy entropy feature vector, which not only retains the characteristic information of the signal in different frequency bands but also provides an ordered feature input for subsequent fault location analysis, enabling the fault location model to more effectively utilize these features for predicting the fault distance and determining the fault location.

[0068] It can be understood that the wavelet transform used in this application is a mathematical method that can perform time localization on different frequency components in a given signal. Compared with the short-time Fourier transform, the wavelet transform has an adjustable window and can obtain a localization function between time, space, and the frequency domain. Among them, each layer of decomposition of the discrete wavelet transform can only decompose the low-frequency part and cannot predict the micro-changes of high-frequency components due to the incomplete decomposition of the signal. As an advanced signal processing tool, the wavelet packet transform can provide fine frequency resolution in the high-frequency domain by completing the decomposition and analyzing the low-frequency and high-frequency sub-bands of the signal, thereby being able to predict the tiny changes of the signal.

[0069] Reference Figure 4 , Figure 4 is a schematic diagram of three-layer wavelet packet decomposition in the channel cable fault location method based on the GIM parametric model provided in an embodiment of this application. In some embodiments, Figure 4 shows the three-layer wavelet packet decomposition process in the channel cable fault location method based on the GIM parametric model in this application. The following is a detailed explanation:

[0070] The original signal (ROOT) is the voltage traveling wave signal when a cable fault occurs, which contains the forward traveling wave and the backward traveling wave signals at the moment of fault occurrence, and can be a time-series signal containing information with different frequency components; further, in the first-level decomposition, the original signal is decomposed into a low-frequency part (A) and a high-frequency part (D). The low-frequency part (A) retains the low-frequency components in the signal, usually corresponding to the slow-changing part of the signal, and the high-frequency part (D) contains the high-frequency components in the signal, usually corresponding to the fast-changing part of the signal, such as the mutation at the moment of fault occurrence.

[0071] Further, in the second-level decomposition, the low-frequency part (A) obtained from the first-level decomposition is further decomposed into a low-frequency part (AA) and a high-frequency part (AD); the low-frequency part (A) obtained from the first-level decomposition is decomposed into a high-frequency part (AD) to capture the high-frequency changes in the low-frequency part; the high-frequency part (D) obtained from the first-level decomposition is decomposed into a low-frequency part (DA) to capture the low-frequency components in the high-frequency part; the high-frequency part (D) obtained from the first-level decomposition is decomposed into a high-frequency part (DD) to capture the high-frequency components in the high-frequency part.

[0072] Further, in the third-level decomposition, the low-frequency part (AA) obtained from the second-level decomposition is further decomposed into a low-frequency part (AAA) and a high-frequency part (AAD); the low-frequency part (AA) obtained from the second-level decomposition is decomposed into a high-frequency part (AAD); the high-frequency part (AD) obtained from the second-level decomposition is decomposed into a low-frequency part (ADA); the high-frequency part (AD) obtained from the second-level decomposition is decomposed into a high-frequency part (ADD); the low-frequency part (DA) obtained from the second-level decomposition is decomposed into a low-frequency part (DAA); the low-frequency part (DA) obtained from the second-level decomposition is decomposed into a high-frequency part (DAD); the high-frequency part (DD) obtained from the second-level decomposition is decomposed into a low-frequency part (DDA); the high-frequency part (DD) obtained from the second-level decomposition is decomposed into a high-frequency part (DDD).

[0073] It can be understood that through three-level wavelet packet decomposition, the original signal is decomposed into time-domain waveforms in multiple different frequency bands. The signals in each frequency band can be analyzed separately, the energy entropy eigenvalue in that frequency band is extracted, the energy entropy eigenvalue of the reconstructed signal in each frequency band is calculated, and they are arranged in the order of frequency to form a forward energy entropy eigenvector and a backward energy entropy eigenvector. These eigenvectors will be used as inputs for subsequent fault distance prediction.

[0074] In some embodiments, the signal of wavelet packet decomposition can be expressed as Equation (5):

[0075] (5)

[0076] In the formula: Denoted as the \(j\)-th decomposition layer of the \(i\)-th sub-band, it can be represented by a linear combination of wavelet packet functions.

[0077] (6)

[0078] In the formula: is the wavelet packet function; is the wavelet packet coefficient.

[0079] Among them, the relationship between the \((j + 1)\)-th decomposition layer and the \(j\)-th decomposition layer can be expressed as:

[0080] (7)

[0081] (8)

[0082] (9)

[0083] In the formula: \(h(k)\) is the low-pass filtering coefficient; \(g(k)\) is the high-pass filtering coefficient; \(k\) represents the conversion parameter.

[0084] Through the orthogonal conjugate filter bank, the signal is decomposed into some non-overlapping frequency bands. For a signal with an original length of \(N\), when it is processed by wavelet packet decomposition to the \(j\)-th layer, the signal length of the sub-band is reduced to \(N / 2^j\).

[0085] (2) Wavelet packet energy and entropy

[0086] For the reconstructed sub-band , the signal energy of each frequency band can be calculated by the Euclidean norm method:

[0087] (10)

[0088] In the formula: \(m = 0, 1,... 2^j - 1\).

[0089] According to Parseval's theorem, the total signal energy \(E_{total}\) is the sum of the energies of each band in a decomposition layer.

[0090] (11)

[0091] Therefore, the relative wavelet energy of each sub-band is:

[0092] (12)

[0093] In addition, according to the classical Shannon information theory, entropy can reflect the direct relationship between the size of information and its uncertainty. This parameter has become a suitable complexity measure for the study of non-linear time series in many fields. The calculation formula of entropy is:

[0094] (13)

[0095] In some cases, the time-domain signal or FFT spectrum cannot directly or significantly detect signal changes, but has a very obvious impact on the entropy value. The more stable a system is, the smaller the entropy; conversely, the larger the entropy value, the greater the possibility of system changes. Therefore, using the wavelet packet energy entropy as a characteristic value for fault diagnosis can accurately show the occurrence of faults.

[0096] Reference Figure 5 and Figure 6 , Figure 5 In the channel cable fault location method based on the GIM parametric model provided by an embodiment of the present application, under normal circumstances, the voltage traveling wave diagram Figure 6 In the channel cable fault location method based on the GIM parametric model provided by an embodiment of the present application, the voltage traveling wave diagram under single-phase grounding fault; among them, when selecting the fault voltage signal with a fault time of 0.01 s and a fault duration of 0.015 s for wavelet packet decomposition, as Figure 5 shown is the cable voltage traveling wave signal from 0 s to 0.015 s under normal conditions, and the waveform shows stable periodic oscillations without obvious amplitude changes or mutations; Figure 6 Shown is the voltage traveling wave diagram under the selected single-phase grounding fault. Among them, the dotted part in the figure is the forward traveling wave signal, and the solid line is the reverse traveling wave signal. The cable fault condition and fault distance are judged by extracting the forward and reverse traveling wave signals of the voltage during the fault.

[0097] Reference Figure 7 , Figure 7 In the channel cable fault location method based on the GIM parametric model provided by an embodiment of the present application, it is the flowchart of the method for obtaining the time-domain waveforms of multiple frequency bands; in some embodiments, the forward traveling wave signal and the reverse traveling wave signal are respectively decomposed by three layers of wavelet packet decomposition, and each layer of decomposition generates low-frequency and high-frequency sub-bands to obtain the time-domain waveforms of multiple frequency bands, including but not limited to the following steps:

[0098] Step S710, based on a preset wavelet basis function, decompose the forward traveling wave signal and the reverse traveling wave signal to generate the wavelet packet decomposition coefficients of each layer;

[0099] Step S720, perform threshold filtering on the wavelet packet decomposition coefficients to reconstruct the low-frequency and high-frequency sub-band signals layer by layer to obtain the time-domain waveforms of multiple frequency bands.

[0100] In some embodiments, the forward traveling wave signal and the reverse traveling wave signal are respectively subjected to three-layer wavelet packet decomposition. Each layer of decomposition generates low-frequency and high-frequency sub-bands, and time-domain waveforms in multiple frequency bands are obtained. The specific steps are as follows: First, based on a preset wavelet basis function, the forward traveling wave signal and the reverse traveling wave signal are decomposed to generate wavelet packet decomposition coefficients for each layer. Different wavelet basis functions are applicable to different types of signals, and the wavelet basis function can be selected based on the material characteristics of the cable. Through wavelet packet decomposition, the signal can be decomposed into sub-bands in different frequency bands, thereby capturing the characteristics of the signal in different frequency ranges. Further, threshold filtering is performed on the wavelet packet decomposition coefficients to reconstruct the low-frequency and high-frequency sub-band signals layer by layer, obtaining time-domain waveforms in multiple frequency bands. Threshold filtering can effectively remove noise interference and retain the useful information in the signal. Through layer-by-layer reconstruction, time-domain waveforms in each frequency band can be obtained.

[0101] It can be understood that when a cable fault occurs, the amplitude of the voltage signal will suddenly decrease or disappear. Especially in the case of a short-circuit fault, the fault will cause a sudden interruption or short circuit of the current path, resulting in abnormal changes in the voltage signal, and the harmonic components in the voltage signal will increase abnormally. Therefore, in this application, a voltage characteristic signal with high sensitivity to cable faults can be extracted, subjected to wavelet packet decomposition and reconstruction, and the energy entropy of each frequency band can be obtained as a characteristic value of the cable traveling wave fault.

[0102] Refer to Figure 4 , according to Figure 4 the three-layer wavelet packet decomposition schematic diagram in , actually, wavelet packet decomposition is to convolve the wavelet packet decomposition coefficients with low-pass and high-pass filters to obtain the decomposition coefficients of the next-layer low-frequency and high-frequency components, and so on; the original signal function is the voltage signal

[0103] (14)

[0104] (15)

[0105] (16)

[0106] (17)

[0107] In the formula: and are the wavelet packet decomposition coefficients of the low-pass filter and the high-pass filter; is the wavelet packet decomposition coefficient of node n in the (m - 1)th layer; and are respectively the low-frequency decomposition coefficient of node 2n in the mth layer and the high-frequency decomposition coefficient of node 2n + 1 in the mth layer; is the decomposition level;

[0108] Wavelet packet reconstruction and decomposition are opposite processes. Reconstruction can obtain the wavelet packet coefficients of the corresponding nodes in the upper layer:

[0109] (18)

[0110] In the formula: and are the coefficients of the low-pass filter and high-pass filter of the wavelet packet reconstruction algorithm;

[0111] Each node represents a frequency band. The wavelet packet reconstruction coefficients reflect the time-domain signal characteristics of the corresponding frequency band. To quantify the signal characteristics after wavelet packet decomposition and reconstruction, energy analysis can be performed on the reconstruction coefficients as follows:

[0112] (19)

[0113] In the formula: represents the wavelet packet reconstruction energy of node n in the (m - 1)-th layer;

[0114] As can be seen from the above calculations, the three-phase voltage signal can be decomposed and reconstructed by the wavelet packet algorithm, and the wavelet packet reconstruction coefficients in each frequency band are the instantaneous values of the phase voltage components. In addition, the wavelet packet reconstruction energy can be used to quantify the phase voltage characteristics in the time-frequency domain;

[0115] Next, calculate the wavelet packet reconstruction energy entropy as the frequency-domain characteristic parameter of the voltage fault traveling wave signal;

[0116] Its energy entropy can be expressed as:

[0117] (20)

[0118] (21)

[0119] In the formula: represents the wavelet packet reconstruction energy entropy ratio of node n in the (m - 1)-th layer.

[0120] In summary, according to formulas (14) to (21), threshold filtering processing can be performed on the wavelet packet decomposition coefficients to reconstruct the low-frequency sub-band and high-frequency sub-band signals layer by layer, obtain the time-domain waveforms of multiple frequency bands, and complete the extraction of cable traveling wave fault characteristics.

[0121] Reference Figure 8 and Figure 9 , Figure 8 is the time-domain diagram of forward traveling wave wavelet decomposition and reconstruction in the channel cable fault location method based on the GIM parametric model provided by an embodiment of this application, Figure 9In the method for locating channel cable faults based on the GIM parametric model provided by an embodiment of the present application, the reverse traveling wave wavelet decomposition reconstructs the time-domain diagram; wherein, the positive wave and the reverse wave of the fault voltage traveling wave are subjected to three-layer wavelet decomposition, and the original traveling wave and the reconstructed high-frequency wavelet decomposition are as Figure 8 and Figure 9 shown.

[0122] In some embodiments, based on a preset regression mapping model, determining the fault distances corresponding to the forward energy entropy feature vector and the reverse energy entropy feature vector includes: obtaining a training data set, where the training data set contains the forward and reverse energy entropy feature vectors of historical fault cases and the corresponding true fault distances; based on the training data set, using the support vector regression algorithm to train the regression mapping model, and the regression mapping model takes the energy entropy feature vector as the input and the fault distance as the output; optimizing the hyperparameters of the regression mapping model through cross-validation until the prediction error converges to a preset threshold; inputting the forward energy entropy feature vector and the reverse energy entropy feature vector into the optimized regression mapping model to obtain the fault distance.

[0123] Among them, it can be understood that the training data set contains the forward and reverse energy entropy feature vectors of historical fault cases and the corresponding true fault distances. These data are used to train the regression mapping model so that it can accurately map the energy entropy feature vector to the fault distance. By using the support vector regression algorithm to train the regression mapping model, support vector regression is a regression algorithm based on support vector machines and is suitable for nonlinear regression problems. The regression mapping model takes the energy entropy feature vector as the input and the fault distance as the output, and learns the mapping relationship between the input features and the output results through the training data set.

[0124] Furthermore, the hyperparameters of the regression mapping model are optimized through cross-validation. Cross-validation is a method for evaluating the performance of a model and optimizing the model parameters. By dividing the training data set into multiple subsets, taking turns using a part of them as the validation set and the other parts as the training set, training and validating the model multiple times to find the optimal combination of hyperparameters, so that the prediction error converges to a preset threshold. Furthermore, after the model training and optimization are completed, the forward and reverse energy entropy feature vectors in the actual fault cases can be input into the regression mapping model, and the model will output the corresponding fault distance prediction value.

[0125] It can be understood that in the normal operating state, the cable traveling wave signal exhibits obvious periodic characteristics. Its frequency domain characteristics are mainly manifested in that the energy is concentrated in the low frequency band and the signal amplitude remains stable. However, when a single-phase ground fault occurs, the voltage traveling wave signal will be affected by the shock wave at the fault point, and significant amplitude changes will occur during the fault time period (0.01s - 0.015s), especially the energy of the high-frequency component will be significantly enhanced. Therefore, by extracting the forward and backward voltage traveling wave signals during the fault, the specific situation and fault distance of the cable fault can be effectively judged. The forward traveling wave signal corresponds to the propagation path from the fault point to the observation point, while the backward traveling wave signal is the path characteristic reflected from the fault point. Figure 5 It shows that the amplitude of the forward traveling wave signal (dashed line) increases sharply when the fault occurs, while the backward traveling wave signal (solid line) shows a delayed disturbance. The characteristic differences between the two in the time domain and frequency domain provide a strong basis for fault location.

[0126] Furthermore, the forward and backward traveling wave signals are respectively decomposed by three-layer wavelet packet decomposition. The first layer decomposition obtains low-frequency and high-frequency components. Subsequently, each component is further decomposed, and the signals containing fault characteristics in the high-frequency components are extracted and reconstructed to obtain the reconstructed high-frequency wavelet signals as shown in Figure 5 and Figure 6 The energy of the high-frequency component of the forward traveling wave signal increases significantly near the fault point, and the amplitude fluctuation range is large, indicating that the fault shock wave reaches the observation point first. The energy of the high-frequency component of the backward traveling wave is weak and there is a certain time delay.

[0127] Furthermore, by analyzing the energy distribution of the components in each frequency band, the energy entropy of the key frequency band is extracted as the fault diagnosis feature. Table (1) lists the sample data of the energy entropy of the forward and backward traveling waves and the output fault distance under various fault types:

[0128] Table (1)

[0129]

[0130] From the data in Table (1), it can be seen that under normal conditions, the energy entropy of the forward and backward traveling waves (0.0834 and 0.0825) is similar and low. As the fault type becomes more complex, from single-phase ground fault to two-phase ground fault, the energy entropy increases significantly. The forward and backward energy entropy extracted by wavelet packet decomposition accurately reflects the change characteristics of the fault signal, and the non-linear change of the energy entropy with the fault type and distance is significant, providing a reliable basis for fault location.

[0131] In order to establish a non-linear mapping model between the energy entropy and the fault distance, support vector regression SVR can be used. The expression of the SVR model is:

[0132] Fault distance (22)

[0133] Wherein, and are the input eigenvalue, representing the energy entropy of the forward traveling wave and the energy entropy of the backward traveling wave respectively; N is the number of support vectors in the training set; and are the Lagrange multipliers of the support vectors, used for weight allocation; is the parameter of the radial basis function, controlling the amplitude of the non - linear mapping; and are respectively the energy entropy feature vector and the input feature vector of the

[0134] In some embodiments, according to the fault distance and the deployment position of the corresponding signal observation point in the GIM parametric model, the fault location is obtained, including: determining the propagation paths corresponding to the forward traveling wave signal and the backward traveling wave signal according to the GIM parametric model and the deployment position of the corresponding signal observation point in the GIM parametric model; marking the fault location in the GIM parametric model according to the propagation paths, the fault distance and the deployment position.

[0135] Wherein, the GIM parametric model includes the geometric parameter set, topological connection relationship and device space coordinates of the cable trench. Through these information, the position of the signal observation point and the propagation path of the signal in the cable can be determined. Therefore, using the known fault distance and signal propagation path, the fault point can be located from the signal observation point along the cable path towards the fault direction, and the position of the fault point on the cable path can be determined and marked in the GIM parametric model, so as to facilitate the maintenance personnel to quickly and accurately find the fault point.

[0136] In some embodiments, multiple signal observation points are deployed based on the GIM parametric model, and the forward traveling wave signal and the backward traveling wave signal of the voltage during cable fault are obtained through the signal observation points, including: determining the cable position according to the geometric parameter set and topological connection matrix of the cable trench, and deploying multiple signal observation points based on the cable position; synchronously collecting the forward traveling wave signal and the backward traveling wave signal of the voltage within a preset time window during cable fault through the signal observation points.

[0137] Among them, the specific position of the cable in the channel is determined by using the geometric parameter set and topological connection matrix in the GIM parametric model, and then signal observation points are deployed at key nodes and characteristic sections to cover the main part and fault-prone areas of the cable. Further, high-frequency sampling sensors are set at the signal observation points. When a cable fault occurs, these sensors can monitor and record the forward traveling wave signal and reverse traveling wave signal at the moment of the fault, as well as characteristic information such as the time when the fault occurs, providing data support for subsequent fault location analysis.

[0138] In some embodiments, the explanations of all the professional names designed in this application are as follows;

[0139] GIM parametric model: The GridInformationModel parametric model can perform digital modeling for concealed cables in channels. By parametrically describing the cable laying environment and geometric features, information such as engineering terrain, cable channels, and equipment foundation positions is integrated into a unified model to support cable fault location and operation and maintenance management.

[0140] Forward traveling wave signal and reverse traveling wave signal: When a cable fault occurs, the voltage traveling wave will propagate from the fault point to both ends, forming a forward traveling wave and a reverse traveling wave. The forward traveling wave signal corresponds to the propagation path from the fault point to the observation point, and the reverse traveling wave signal is the path feature reflected from the fault point. By analyzing these two signals, the specific situation and fault distance of the cable fault can be judged.

[0141] Energy entropy feature vector: A vector formed by arranging the energy entropy eigenvalues obtained by performing wavelet packet decomposition and reconstruction on the traveling wave signal and calculating the signal energy entropy of each frequency band in the order of frequency. Energy entropy reflects the energy distribution and complexity of the signal in different frequency bands and is an important parameter for cable fault feature extraction.

[0142] Support vector regression algorithm: A regression algorithm based on statistical learning theory. By finding an optimal hyperplane, the prediction error on the training data is minimized, and at the same time, the complexity of the model is also minimized. SVR is suitable for small-sample and non-linear regression problems and is used in this application to establish a non-linear mapping relationship between the energy entropy feature vector and the fault distance.

[0143] Cross-validation: A method for evaluating model performance and optimizing model parameters. By dividing the data set into multiple subsets, taking turns using a part of them as the validation set and the other parts as the training set, training and validating the model multiple times to evaluate the generalization ability of the model and select the optimal model parameters.

[0144] Signal observation point: A point deployed in a cable trench for monitoring the operating status of cables. Usually, high-frequency sampling sensors are set up to collect voltage traveling wave signals during cable faults in real time, including forward traveling wave signals and reverse traveling wave signals, providing data support for fault location.

[0145] To achieve the above object, in a second aspect, an embodiment of the present application provides a trench cable fault location device based on a GIM parametric model, including: a model construction module for performing GIM parametric processing on the geometric parameter set, topological connection matrix, and device spatial coordinates corresponding to the trench cable to construct a GIM parametric model; a signal acquisition module for deploying a plurality of signal observation points based on the GIM parametric model and obtaining forward traveling wave signals and reverse traveling wave signals of the voltage during cable faults through the signal observation points; a feature extraction module for performing wavelet packet decomposition and reconstruction processing on the forward traveling wave signal and the reverse traveling wave signal respectively to obtain a forward energy entropy feature vector and a reverse energy entropy feature vector; a distance determination module for determining the fault distance corresponding to the forward energy entropy feature vector and the reverse energy entropy feature vector based on a preset regression mapping model; a location positioning module for obtaining the fault location according to the fault distance and the deployment position of the corresponding signal observation point in the GIM parametric model.

[0146] Reference Figure 10 , Figure 10 is a schematic structural diagram of an electronic device provided in an embodiment of the present application. The electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the trench cable fault location method or traffic prediction method based on the GIM parametric model in any one of the above embodiments. For example, when executing the method steps S110 to S150 described above Figure 1 in Figure 2 the method steps S210 to S240 in Figure 3 the method steps S310 to S330 in Figure 7 the method steps S710 to S720 in

[0147] The electronic device 1000 in the embodiment of the present application includes one or more processors 1010 and a memory 1020. Figure 10 Taking one processor 1010 and one memory 1020 as an example in

[0148] The processor 1010 and the memory 1020 can be connected through a bus or other means. Figure 10 Taking the connection through a bus as an example in

[0149] The memory 1020, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory 1020 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory 1020 may optionally include a memory 1020 that is remotely located relative to the processor 1010, and these remote memories can be connected to the electronic device 1000 through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0150] In some embodiments, when the processor executes the computer program, it executes the channel cable fault location method based on the GIM parametric model in any one of the above embodiments at a preset interval.

[0151] Those skilled in the art can understand that Figure 10 the device structure shown in does not constitute a limitation on the electronic device 1000, and it may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0152] In Figure 10 the electronic device 1000 shown, the processor 1010 can be used to call the channel cable fault location method based on the GIM parametric model stored in the memory 1020, so as to implement the channel cable fault location method based on the GIM parametric model.

[0153] Based on the hardware structure of the above electronic device 1000, various embodiments of the channel cable fault location device based on the GIM parametric model of the present application are proposed. At the same time, the non-transitory software programs and instructions required to implement the channel cable fault location method based on the GIM parametric model in the above embodiments are stored in the memory, and when executed by the processor, they execute the channel cable fault location method based on the GIM parametric model in the above embodiments.

[0154] The embodiments of the present application also provide a computer-readable storage medium, which stores computer-executable instructions for executing the above channel cable fault location method based on the GIM parametric model, and can enable the above one or more processors to execute the channel cable fault location method based on the GIM parametric model in any one of the above embodiments or the traffic prediction method. For example, execute the method steps S110 to step S150 described above Figure 1 in, Figure 2 the method steps S210 to step S240 in, Figure 3 the method steps S310 to step S330 in, Figure 7Method steps S710 to S720 therein.

[0155] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed to multiple network nodes. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0156] Those of ordinary skill in the art can understand that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and their appropriate combinations. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer-readable storage medium (or non-transitory medium) and a communication medium (or transitory medium). As is well known to those of ordinary skill in the art, the term computer-readable storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer-readable storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory, or other memory technologies, CD-ROM, digital versatile disk (DVD), or other optical disk storage, magnetic cassette, tape, magnetic disk storage, or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, as is well known to those of ordinary skill in the art, a communication medium typically includes computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and can include any information delivery medium.

[0157] The above is a specific description of the preferred embodiment of this application, but this application is not limited to the above embodiments. Those skilled in the art can also make various equivalent deformations or substitutions without departing from the spirit of this application, and these equivalent deformations or substitutions are all included within the scope defined by the claims of this application.

Claims

1. A channel cable fault location method based on a GIM parameterized model, characterized in that: include: Perform GIM parameterization on the geometric parameter set, topological connection matrix and equipment space coordinates corresponding to the channel cable to construct a GIM parameterized model; Deploy multiple signal observation points based on the GIM parameterized model, and obtain forward traveling wave signals and reverse traveling wave signals of voltage when a cable fault occurs through the signal observation points; Performing wavelet packet decomposition and reconstruction processing on the forward traveling wave signal and the reverse traveling wave signal respectively to obtain a forward energy entropy feature vector and a reverse energy entropy feature vector; Based on a preset regression mapping model, determining the fault distances corresponding to the forward energy entropy feature vector and the reverse energy entropy feature vector; The fault location is obtained according to the fault distance and the deployment location of the corresponding signal observation point in the GIM parameterized model.

2. The method according to claim 1, characterized in that The GIM parameterization processing is performed on the geometric parameter set, the topological connection matrix and the device space coordinates corresponding to the channel cable to construct the GIM parameterized model, including: Converting the cross-sectional dimensions, bending radius, and elevation data of the cable trench into a geometric parameter set, wherein the geometric parameter set represents the spatial form of the cable trench and the three-dimensional structure of the laying path; Defining the logical connection relationship between the cable branch nodes and the equipment of the cable channel, and generating a topological connection matrix, wherein the topological connection matrix represents the electrical connectivity of each node in the cable network; Integrate the three-dimensional spatial coordinates and electrical parameters of the auxiliary equipment in the cable trench to obtain the equipment spatial coordinates, wherein the equipment spatial coordinates represent the physical position of the equipment in the trench; Based on the geometric parameter set, the topological connection matrix and the equipment space coordinates, a GIM parameterized model including the cable path, equipment location and topological relationship is generated.

3. The method according to claim 1, characterized in that The step of performing wavelet packet decomposition and reconstruction processing on the forward traveling wave signal and the reverse traveling wave signal to obtain a forward energy entropy feature vector and a reverse energy entropy feature vector comprises: Performing three-layer wavelet packet decomposition on the forward traveling wave signal and the reverse traveling wave signal, respectively, generating low-frequency and high-frequency sub-bands in each layer of decomposition, and obtaining time domain waveforms of multiple frequency bands; Calculating an energy entropy eigenvalue for the reconstructed signal of each frequency band, where the energy entropy eigenvalue is the product of the frequency band energy proportion based on Parsval's theorem and Shannon entropy; The energy entropy eigenvalues ​​of multiple frequency bands are arranged in frequency order to form a forward energy entropy eigenvector and a reverse energy entropy eigenvector.

4. The method according to claim 3, characterized in that The forward traveling wave signal and the reverse traveling wave signal are respectively subjected to three-layer wavelet packet decomposition, and each layer of decomposition generates low-frequency and high-frequency sub-bands to obtain time domain waveforms of multiple frequency bands, including: Based on a preset wavelet basis function, the forward traveling wave signal and the reverse traveling wave signal are decomposed to generate wavelet packet decomposition coefficients of each layer; Threshold filtering is performed on the wavelet packet decomposition coefficients to reconstruct low-frequency sub-band and high-frequency sub-band signals layer by layer to obtain time domain waveforms of multiple frequency bands.

5. The method according to claim 1, characterized in that The determining, based on a preset regression mapping model, the fault distances corresponding to the forward energy entropy feature vector and the reverse energy entropy feature vector comprises: Acquire a training data set, wherein the training data set includes forward and reverse energy entropy feature vectors of historical fault cases and corresponding real fault distances; Based on the training data set, a support vector regression algorithm is used to train a regression mapping model, wherein the regression mapping model takes an energy entropy feature vector as input and a fault distance as output; Optimizing the hyperparameters of the regression mapping model through cross-validation until the prediction error converges to a preset threshold; The forward energy entropy feature vector and the reverse energy entropy feature vector are input into the optimized regression mapping model to obtain the fault distance.

6. The method according to claim 1, characterized in that The obtaining of the fault location according to the fault distance and the deployment location of the corresponding signal observation point in the GIM parameterized model includes: Determine the propagation paths corresponding to the forward traveling wave signal and the reverse traveling wave signal according to the GIM parameterized model and the deployment positions of the corresponding signal observation points in the GIM parameterized model; The fault location is marked in the GIM parameterized model according to the propagation path, the fault distance and the deployment location.

7. The method according to claim 2, characterized in that The method of deploying a plurality of signal observation points based on the GIM parameterized model and obtaining a forward traveling wave signal and a reverse traveling wave signal of a voltage when a cable fault occurs through the signal observation points includes: Determine the cable position according to the geometric parameter set and the topological connection matrix of the cable channel, and deploy multiple signal observation points based on the cable position; The forward traveling wave signal and the reverse traveling wave signal of the voltage within a preset time window when the cable fails are synchronously collected through the signal observation point.

8. A channel cable fault location device based on a GIM parameterized model, characterized in that: include: The model building module is used to perform GIM parameterization processing on the geometric parameter set, topological connection matrix and equipment space coordinates corresponding to the channel cable to build a GIM parameterized model; A signal acquisition module, used to deploy multiple signal observation points based on the GIM parameterized model, and obtain a forward traveling wave signal and a reverse traveling wave signal of the voltage when the cable fails through the signal observation points; A feature extraction module, used to perform wavelet packet decomposition and reconstruction processing on the forward traveling wave signal and the reverse traveling wave signal respectively, to obtain a forward energy entropy feature vector and a reverse energy entropy feature vector; A distance determination module, used to determine the fault distance corresponding to the forward energy entropy feature vector and the reverse energy entropy feature vector based on a preset regression mapping model; The position positioning module is used to obtain the fault location according to the fault distance and the deployment position of the corresponding signal observation point in the GIM parameterized model.

9. An electronic device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When at least one of the programs is executed by at least one of the processors, the channel cable fault location method based on the GIM parameterized model as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing computer-executable instructions, characterized in that: The computer executable instructions are used to execute the channel cable fault location method based on the GIM parameterized model as described in any one of claims 1 to 7.

Images