Fault reason identification method, device and equipment based on primary and secondary fusion switch

Abstract

The invention relates to a fault cause identification method, device and equipment based on a primary and secondary fusion switch. The method comprises the following steps: controlling a primary fusion switch and a secondary fusion switch in a distribution line to acquire a secondary traveling wave signal generated by the distribution line in a working process; performing signal conversion on the secondary traveling wave signal to obtain a primary traveling wave signal of the distribution line; and extracting a plurality of traveling wave characteristics from the primary traveling wave signal of the distribution line, and inputting the plurality of traveling wave characteristics into a preset fault detection network for fault analysis to obtain a fault reason of the distribution line. According to the method, the primary traveling wave waveform is accurately inverted, so that the fault reason and / or the fault type of the distribution line can be accurately judged or identified according to a plurality of traveling wave characteristics in the accurate primary traveling wave signal, and the accuracy of judging the fault reason of the distribution line can be improved.

Description

Technical Field

[0001] This application relates to the field of power technology, and in particular to a method, apparatus and equipment for fault cause identification based on a primary and secondary integrated switch. Background Technology

[0002] The power distribution network is a crucial component of the power system, and its power supply reliability is closely related to the safety of users' production and daily life. However, the lines in the distribution network have complex structures and are prone to faults. Therefore, accurately determining the cause of faults in distribution lines is essential for the safe and reliable operation of the distribution network.

[0003] However, current methods for identifying the causes of power distribution line faults suffer from low accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide a fault cause identification method, device, and equipment based on a primary and secondary integrated switch to address the above-mentioned technical problems, which can improve the accuracy of the method for identifying the cause of power distribution line faults.

[0005] Firstly, this application provides a fault cause identification method based on a primary and secondary fusion switch, including:

[0006] The primary and secondary fusion switch in the control power distribution line collects the secondary traveling wave signal generated by the power distribution line during operation;

[0007] The secondary traveling wave signal is converted to obtain the primary traveling wave signal of the power distribution line;

[0008] Multiple traveling wave features are extracted from the primary traveling wave signal of the power distribution line, and the multiple traveling wave features are input into a preset fault detection network for fault analysis to obtain the cause of the fault in the power distribution line.

[0009] In one embodiment, the step of converting the secondary traveling wave signal to obtain the primary traveling wave signal of the power distribution line includes:

[0010] The secondary traveling wave signal is decomposed to obtain multiple first mode functions corresponding to the secondary traveling wave signal;

[0011] Based on the conversion relationship between the first mode function and the first traveling wave signal, an inversion model is constructed;

[0012] The multiple first mode functions are iteratively calculated based on the objective function of the inversion model to obtain the primary traveling wave signal of the power distribution line.

[0013] In one embodiment, the step of iteratively calculating the plurality of first mode functions based on the objective function of the inversion model to obtain the primary traveling wave signal of the power distribution line includes:

[0014] Obtain the initial number of iterations and the initial second mode function;

[0015] Based on the initial iteration number and the plurality of first mode functions, determine the gradient and initial step size parameters corresponding to the initial iteration number;

[0016] The next second mode function is determined based on the gradient corresponding to the initial iteration number, the initial step size parameter, the objective function, and the plurality of first mode functions;

[0017] The primary traveling wave signal of the power distribution line is determined based on the initial second mode function and the next second mode function.

[0018] In one embodiment, determining the primary traveling wave signal of the power distribution line based on the initial second mode function and the next second mode function includes:

[0019] If the initial second mode function and the next second mode function satisfy a preset termination condition, then the initial second mode function and the next second mode function are accumulated, and the accumulation result is determined as the first traveling wave signal;

[0020] If the initial second mode function and the next second mode function do not satisfy the preset termination condition, then the initial iteration number is updated, and the next second mode function is used as the new initial second mode function. Then, the process returns to the step of determining the gradient and initial step size parameters corresponding to the initial iteration number based on the initial iteration number and the plurality of first mode functions.

[0021] In one embodiment, the fault detection network includes a convolutional subnetwork and a classification subnetwork. The step of inputting the multiple traveling wave features into the preset fault detection network for fault analysis to obtain the cause of the fault in the power distribution line includes:

[0022] A feature matrix is ​​generated based on the multiple traveling wave features and the number of output channels of the convolutional sub-network; the multiple traveling wave features include amplitude, pulse width, and number of peaks;

[0023] The feature matrix is ​​input into the convolutional sub-network for four-dimensional convolution processing to obtain the output feature map;

[0024] The output feature map is input into the classification sub-network for fault analysis to obtain the fault cause of the power distribution line.

[0025] In one embodiment, the training method for the fault detection network includes:

[0026] Acquire multiple historical secondary traveling wave signals and the fault tags corresponding to each of the historical secondary traveling wave signals;

[0027] Extract multiple historical traveling wave features from each of the aforementioned historical secondary traveling wave signals;

[0028] A training dataset is constructed based on multiple historical traveling wave features and the fault labels corresponding to each of the historical secondary traveling wave signals.

[0029] The initial fault detection network is trained using the training dataset to obtain the fault detection network.

[0030] Secondly, this application also provides a fault cause identification device based on a primary and secondary fusion switch, comprising:

[0031] The acquisition module is used to control the primary and secondary fusion switch in the power distribution line to acquire the secondary traveling wave signal generated by the power distribution line during operation;

[0032] The signal conversion module is used to convert the secondary traveling wave signal to obtain the primary traveling wave signal of the power distribution line.

[0033] The fault analysis module is used to extract multiple traveling wave features from the primary traveling wave signal of the power distribution line, and input the multiple traveling wave features into a preset fault detection network for fault analysis to obtain the fault cause of the power distribution line.

[0034] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method in any of the embodiments of the first aspect described above.

[0035] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the method in any of the embodiments of the first aspect described above.

[0036] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the method in any of the embodiments of the first aspect described above.

[0037] The aforementioned method, apparatus, and equipment for fault cause identification based on a primary and secondary fusion switch control system in a power distribution line to collect secondary traveling wave signals generated during the operation of the power distribution line; perform signal conversion on the secondary traveling wave signals to obtain primary traveling wave signals of the power distribution line; extract multiple traveling wave features from the primary traveling wave signals of the power distribution line, and input the multiple traveling wave features into a preset fault detection network for fault analysis to obtain the fault cause of the power distribution line. Since this embodiment can perform signal conversion on the secondary traveling wave signals generated during the operation of the power distribution line to obtain the primary traveling wave signals of the power distribution line, the primary traveling wave waveform can be accurately reconstructed. Therefore, based on the multiple traveling wave features in the accurate primary traveling wave signals, the fault cause and / or fault type of the power distribution line can be accurately determined or identified, thus improving the accuracy of identifying the fault cause of the power distribution line. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is an application environment diagram of a fault cause identification method based on a primary and secondary fusion switch in one embodiment;

[0040] Figure 2 This is a flowchart illustrating a fault cause identification method based on a primary and secondary fusion switch in one embodiment.

[0041] Figure 3 This is a schematic diagram of a secondary traveling wave signal in one embodiment;

[0042] Figure 4 This is a flowchart illustrating the signal conversion steps in one embodiment;

[0043] Figure 5 This is a schematic diagram of multiple first mode functions in one embodiment;

[0044] Figure 6 This is a flowchart illustrating the iterative calculation steps in one embodiment;

[0045] Figure 7 This is a flowchart illustrating a fast iterative shrinkage threshold algorithm in one embodiment;

[0046] Figure 8 This is a comparative schematic diagram of multiple traveling wave signals in one embodiment;

[0047] Figure 9This is a flowchart illustrating a fault cause identification method based on a primary and secondary fusion switch in another embodiment.

[0048] Figure 10 This is a structural block diagram of a fault cause identification device based on a primary and secondary fusion switch in one embodiment. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0050] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this application.

[0052] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0053] The power distribution network is a crucial component of the power system, and its power supply reliability is closely related to the safety and well-being of users. However, the distribution network has a complex structure, including overhead lines, cables, and mixed lines, with numerous branch points, wide coverage, and long lengths. Operating under harsh conditions, it is prone to faults, such as grounding faults and short-circuit faults. Statistics show that annual power outage losses nationwide exceed hundreds of billions of yuan, with over 90% of power outages caused by distribution line faults. Therefore, accurately determining the cause of distribution line faults can improve the fault diagnosis and handling capabilities of the distribution network, quickly locate faults, and formulate line inspection strategies, thereby rapidly eliminating faults. This is an effective way to ensure the reliability of power supply, shorten power outage time, and improve the user's electricity experience. In short, accurately determining the cause of distribution line faults is crucial for the safe and reliable operation of the distribution network.

[0054] Because different traveling wave signals correspond to different fault causes in a power distribution network, these signals are currently used to identify different fault causes in distribution lines. However, traveling wave signals are broadband transient signals that appear in the form of current and voltage in a faulty power grid. They are typically detected using traveling wave sensors based on Rogowski coils, but due to the insufficient bandwidth of these sensors, some high-frequency data is lost, causing distortion of the primary traveling wave signal as it passes through the sensor. This makes it difficult to determine the fault cause based on the secondary traveling wave waveform characteristics. Therefore, the currently acquired traveling wave signals are not accurate enough, resulting in low accuracy in current methods for identifying the causes of power distribution line faults.

[0055] Having described the background technology of the fault cause identification method based on a primary and secondary integrated switch provided in the embodiments of this application, the implementation environment involved in the fault cause identification method based on a primary and secondary integrated switch provided in the embodiments of this application will be briefly described below. The fault cause identification method based on a primary and secondary integrated switch provided in the embodiments of this application can be applied to, for example... Figure 1The computer device shown is optionally integrated into a traveling wave positioning-based primary and secondary fusion switch, or it can be independently installed outside the traveling wave positioning-based primary and secondary fusion switch and communicatively connected to it. The traveling wave positioning-based primary and secondary fusion switch is a new type of intelligent device in power systems that deeply integrates primary equipment (main circuit high-voltage electrical equipment) and secondary equipment (control, monitoring, and protection units). It achieves distribution automation functions through hardware integration and software collaboration, improving the intelligence level and reliability of the distribution network. The primary equipment may include, but is not limited to, vacuum circuit breakers, disconnect switches, and instrument transformers, while the secondary equipment may include, but is not limited to, voltage / current sensors, traveling wave sensors, FTUs (Feeder Terminal Units), and capacitor power extraction units. The primary and secondary equipment are fully enclosed and integrated to reduce external wiring and adapt to live-line working scenarios.

[0056] Optionally, the computer device can be a terminal, a server, or an FTU. Of course, the embodiments of this application do not limit the specific structure of the computer device. For example, the internal structure diagram of the computer device can be as follows: Figure 1 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a fault cause identification method based on a primary and secondary fusion switch. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0057] Those skilled in the art will understand that Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0058] In one embodiment, such as Figure 2 As shown, a fault cause identification method based on a primary and secondary fusion switch is provided, which is then applied to... Figure 1 Taking a computer device as an example, the explanation includes the following steps:

[0059] S201 controls the primary and secondary fusion switch in the power distribution line to collect the secondary traveling wave signal generated by the power distribution line during operation.

[0060] In this context, "distribution line" refers to the lines in the power distribution network of a power system. The primary and secondary integrated switch is installed on the distribution line. This integrated switch is a new type of intelligent device in the power system that deeply integrates primary equipment (main circuit high-voltage electrical equipment) with secondary equipment (control, monitoring, and protection units). Through hardware integration and software collaboration, it achieves distribution automation functions, improving the intelligence level and reliability of the distribution network. The secondary traveling wave signal refers to the traveling wave signal acquired by the FTU. The traveling wave signal is a broadband transient signal appearing in the form of current and voltage in a faulty power grid. For example, the secondary traveling wave signal can be secondary traveling wave current data.

[0061] In this embodiment of the application, optionally, the computer device can control at least one FTU in the primary and secondary fusion switch of the power distribution line to collect at least one secondary traveling wave signal generated by the power distribution line in real time during operation; or, the computer device can also control at least one FTU in the primary and secondary fusion switch of the power distribution line to collect at least one secondary traveling wave signal generated by the power distribution line at regular intervals during operation. Of course, this embodiment of the application does not limit the specific implementation method for collecting the secondary traveling wave signal. For example, such as... Figure 3 As shown, Figure 3 This is a schematic diagram of a secondary traveling wave signal in one embodiment.

[0062] S202 performs signal conversion on the secondary traveling wave signal to obtain the primary traveling wave signal of the power distribution line.

[0063] In this embodiment, optionally, the computer device can directly perform signal conversion on the secondary traveling wave signal to obtain the primary traveling wave signal of the power distribution line; alternatively, the computer device can first decompose the secondary traveling wave signal, and then perform signal conversion on the decomposed secondary traveling wave signal to obtain the primary traveling wave signal of the power distribution line. Of course, this embodiment does not limit the specific implementation method of signal conversion. The primary traveling wave signal is the traveling wave signal obtained by inverting the secondary traveling wave signal.

[0064] S203 extracts multiple traveling wave features from the primary traveling wave signal of the power distribution line, and inputs the multiple traveling wave features into a preset fault detection network for fault analysis to obtain the cause of the power distribution line fault.

[0065] The traveling wave characteristics can include, but are not limited to, the amplitude, pulse width, and number of peaks of a single traveling wave. The preset fault detection network can be any type of deep learning network; for example, it can be a CNN (Convolutional Neural Networks) detection model. Fault causes can include, but are not limited to, grounding faults and short-circuit faults.

[0066] In this embodiment, the computer device can process the primary traveling wave signal retrieved from the power distribution line, thereby extracting multiple traveling wave features from the primary traveling wave signal. Optionally, the computer device can directly input the multiple traveling wave features into a preset fault detection network for fault analysis to obtain the cause of the power distribution line fault; alternatively, the computer device can first process the multiple traveling wave features to obtain processed traveling wave features, and then input the processed features into the preset fault detection network for fault analysis to obtain the cause of the power distribution line fault. Of course, this embodiment does not limit the specific implementation method of fault analysis.

[0067] In the aforementioned fault cause identification method based on a primary and secondary fusion switch, the primary and secondary fusion switch in the power distribution line is controlled to collect the secondary traveling wave signal generated during the operation of the power distribution line; the secondary traveling wave signal is converted to obtain the primary traveling wave signal of the power distribution line; multiple traveling wave features are extracted from the primary traveling wave signal of the power distribution line, and these features are input into a preset fault detection network for fault analysis to obtain the fault cause of the power distribution line. Since this embodiment can convert the secondary traveling wave signal generated during the operation of the power distribution line to obtain the primary traveling wave signal, the primary traveling wave waveform can be accurately reconstructed. Therefore, based on the multiple traveling wave features in the accurate primary traveling wave signal, the fault cause and / or fault type of the power distribution line can be accurately determined or identified, thus improving the accuracy of identifying the fault cause of the power distribution line.

[0068] In one embodiment, a signal conversion implementation method is provided, namely, "converting the secondary traveling wave signal to obtain the primary traveling wave signal of the power distribution line" in S202 above, such as... Figure 4 As shown, it includes:

[0069] S301, decompose the secondary traveling wave signal to obtain multiple first mode functions corresponding to the secondary traveling wave signal.

[0070] In this embodiment, a computer device can use the VMD (Variational Mode Decomposition) algorithm to decompose a quadratic traveling wave signal, obtaining multiple first mode functions corresponding to the quadratic traveling wave signal, namely, the traveling wave current IMFn components. The traveling wave current IMFn components are the nth order intrinsic mode functions (IMFs) extracted from the traveling wave current signal using signal decomposition methods, which may include, but are not limited to, EMD (Empirical Mode Decomposition) or Variational Mode Decomposition (VMD) algorithms. For example,... Figure 5 As shown, Figure 5 This is a schematic diagram of multiple first mode functions in one embodiment.

[0071] S302, construct an inversion model based on the conversion relationship between the first mode function and the first traveling wave signal.

[0072] In this embodiment, the computer device can construct an inversion model based on the transformation relationship between the first mode function and the primary traveling wave signal. The inversion model can be an L1 regularized inversion model. For example, the specific construction process of the inversion model is as follows:

[0073] The forward model of the traveling wave sensor can be expressed as follows (1):

[0074] (1)

[0075] Where: d is the forward modeling vector, i.e., multiple first mode functions; A is the forward modeling matrix, i.e., the matrix corresponding to the traveling wave sensor, which describes the propagation characteristics of the traveling wave in the sensor; x is the first traveling wave signal to be solved, i.e., the unknown quantity.

[0076] To reduce the influence of noisy and irrelevant features during the inversion process, L1 norm and regularization constraints are adopted to effectively reduce model complexity and highlight important features. The inversion model can be expressed as follows (2):

[0077] (2)

[0078] Where: d x It is the vector output by the sensor. The first term... This is the inversion objective function, ensuring that x approximates the first-order traveling wave signal as closely as possible to reconstruct the fault information. The second term... It is an L1 regularization constraint term. This represents the sum of the absolute values ​​of the elements in vector x. λ is a regularization parameter used to balance the relationship between data fit and model complexity.

[0079] S303, based on the objective function of the inversion model, iteratively calculates multiple first mode functions to obtain the primary traveling wave signal of the power distribution line.

[0080] In this embodiment of the application, the computer device can construct the objective function of the fast iterative shrinkage threshold algorithm based on the inversion model, as shown in the following equation (3):

[0081] (3)

[0082] in: = , = .

[0083] The objective function of the inversion is at the iteration point x. t A second-order Taylor expansion is performed in the vicinity, as shown in equation (4):

[0084] (4)

[0085] in: yes In x t gradient, yes In x t A square matrix composed of the second-order partial derivatives.

[0086] Substituting the Taylor expansion into the objective function and simplifying, we obtain the following equation (5):

[0087] (5)

[0088] Therefore, computer equipment can use a fast iterative shrinking threshold algorithm to iteratively calculate multiple first mode functions based on the objective function of the inversion model to obtain the primary traveling wave signal of the power distribution line.

[0089] In this embodiment, the secondary traveling wave signal can be decomposed to obtain multiple first mode functions corresponding to the secondary traveling wave signal. Based on the conversion relationship between the first mode functions and the primary traveling wave signal, an inversion model can be constructed. Thus, the multiple first mode functions can be accurately iteratively calculated according to the objective function of the inversion model to obtain an accurate primary traveling wave signal.

[0090] In one embodiment, an iterative calculation method is provided, namely, "according to the objective function of the inversion model, multiple first mode functions are iteratively calculated to obtain the primary traveling wave signal of the power distribution line" in S303 above. Figure 6 As shown, it includes:

[0091] S401, obtain the initial number of iterations and the initial second mode function.

[0092] In the embodiments of this application, such as Figure 7 As shown, Figure 7 This is a flowchart illustrating a fast iterative threshold shrinkage algorithm in one embodiment. S501, the computer device can be initialized, that is, the initial number of iterations and the initial second mode function can be set. For example, the initial number of iterations t=1 and the initial second mode function can be set to x0.

[0093] S402, based on the initial iteration number and multiple first mode functions, determine the gradient and initial step size parameters corresponding to the initial iteration number.

[0094] In this embodiment of the application, combined with Figure 7 As shown in S502, the computer device can determine the gradient corresponding to the initial iteration number based on the initial iteration number and multiple first mode functions. For example, the formula for calculating the gradient at the current point is shown in equation (6) below:

[0095] (6)

[0096] The computer device can determine the initial step size parameter corresponding to the initial iteration number based on the initial iteration number and multiple first mode functions. Optionally, the computer device can set the initial regularization parameter λ. t And select the initial step size parameter α t and β t , where α t =1 / λ t ,β t Take a constant of 0.5. Alternatively, combine... Figure 7 As shown in S503, the computer device can also use the adaptive regularization parameter strategy (CMD, Clustering Modification Directions) to calculate the regularization parameter λ, and thus determine the initial step size parameter based on the regularization parameter λ, as shown in the following formula (7):

[0097] (7)

[0098] It should be noted that the introduced regularization parameter λ plays a role in balancing the data fitting terms and model constraint terms, and has a significant impact on the inversion results of fault traveling waves with different noise levels, specifically as follows: 1) When the value of λ is too large, the first-order traveling wave signal x to be solved can be guaranteed to be sufficiently sparse, but the forward response value may not fit the observed data well; 2) When the value of λ is too small or equal to zero, the objective function approximates a least squares problem, and the ill-posedness of the model (i.e., the solution does not have uniqueness or stability) makes it difficult to obtain the optimal solution through inversion. Therefore, using the CMD scheme to calculate the value of λ before the start of each iteration can effectively improve the convergence speed of the model while ensuring inversion accuracy.

[0099] S403, determine the next second mode function based on the gradient corresponding to the initial iteration number, the initial step size parameter, the objective function, and multiple first mode functions.

[0100] In this embodiment, the computer device can determine the next second mode function based on the gradient corresponding to the initial iteration number, the initial step size parameter, the objective function, and multiple first mode functions. The specific formulas are shown in equations (8)-(10) below:

[0101] Update momentum: (8)

[0102] Threshold shrinkage: (9)

[0103] After simplification, the objective function is substituted into the threshold to perform the primal dual solution, yielding the next second mode function x. t+1 for: (10)

[0104] S404, determine the primary traveling wave signal of the power distribution line based on the initial second mode function and the next second mode function.

[0105] In this embodiment, the computer device can determine the primary traveling wave signal of the power distribution line based on the initial second mode function and the next second mode function. In one embodiment, S404 includes:

[0106] If the initial second mode function and the next second mode function satisfy the preset termination condition, then the initial second mode function and the next second mode function are accumulated, and the accumulated result is determined as a traveling wave signal.

[0107] If the initial second mode function and the next second mode function do not meet the preset termination condition, then update the initial iteration number, use the next second mode function as the new initial second mode function, and return to execute the step of determining the gradient and initial step size parameters corresponding to the initial iteration number based on the initial iteration number and multiple first mode functions.

[0108] In this embodiment of the application, combined with Figure 7 As shown in S504, the computer device can determine that the initial second mode function and the next second mode function satisfy a preset termination condition. The preset termination condition is as follows (11):

[0109] (11)

[0110] Where: t k This represents the number of iterations. It is a very small constant to prevent the denominator from being zero.

[0111] Combination Figure 7 As shown in step S505, if the initial second mode function and the next second mode function satisfy a preset termination condition, the computer device can accumulate the initial second mode function and the next second mode function, and determine the accumulation result as a first-order traveling wave signal. That is, the computer device can linearly superimpose each second mode function obtained from the inversion, and determine the superposition result as the final first-order traveling wave signal.

[0112] Combination Figure 7 As shown in step S506, if the initial second mode function and the next second mode function do not meet the preset termination condition, the computer device can update the initial iteration count. For example, it can increment the initial iteration count by one, use the next second mode function as the new initial second mode function, and return to execute the step of determining the gradient and initial step size parameters corresponding to the initial iteration count based on the initial iteration count and multiple first mode functions. In other words, the computer device can determine the gradient and initial step size parameters corresponding to the new initial iteration count based on the new initial iteration count and multiple first mode functions; thus, it can determine the new next second mode function based on the gradient, the initial step size parameters, the objective function, and multiple first mode functions; subsequently, it determines the primary traveling wave signal of the power distribution line based on the new initial second mode function and the new next second second mode function.

[0113] For example, such as Figure 8 As shown, Figure 8 This is a comparative schematic diagram of multiple traveling wave signals in one embodiment. Figure 8 It can be seen that the traveling wave signal (i.e., ...) retrieved by the method of this application embodiment is... Figure 8 The inverted traveling wave (in the process) and the directly acquired primary traveling wave signal (i.e. Figure 8 The waveform of the first traveling wave in the present application is similar to that of the first traveling wave, which proves that the method in the present application is effective.

[0114] In this embodiment, the initial number of iterations and the initial second mode function can be obtained in advance. Based on the initial number of iterations and multiple first mode functions, the gradient and initial step size parameters corresponding to the initial number of iterations are determined. Therefore, based on the gradient corresponding to the initial number of iterations, the initial step size parameters, the objective function, and multiple first mode functions, the next second mode function is determined. Consequently, based on the initial second mode function and the next second mode function, the primary traveling wave signal of the power distribution line can be accurately determined.

[0115] In one embodiment, the fault detection network includes a convolutional subnetwork and a classification subnetwork. Based on this, a fault analysis implementation method is provided, namely, the "inputting multiple traveling wave features into a preset fault detection network for fault analysis to obtain the fault cause of the power distribution line" in S203 above, includes:

[0116] A feature matrix is ​​generated based on multiple traveling wave features and the number of output channels of the convolutional subnetwork; the multiple traveling wave features include amplitude, pulse width, and number of peaks.

[0117] The feature matrix is ​​input into the convolutional sub-network for four-dimensional convolution processing to obtain the output feature map.

[0118] The output feature map is input into the classification subnetwork for fault analysis to obtain the cause of the power distribution line fault.

[0119] In this embodiment, since the fault detection network includes a convolutional subnetwork, the computer device can generate a feature matrix based on multiple traveling wave features and the number of output channels of the convolutional subnetwork. The multiple traveling wave features include amplitude, pulse width, and number of peaks. For example, the traveling wave image input to the model is the feature matrix. Where m represents the amplitude of the traveling wave, n represents the pulse width of the traveling wave, c represents the number of peaks of the traveling wave, and d is the number of output channels.

[0120] Thus, a computer device can input the feature matrix into a convolutional sub-network for four-dimensional convolution processing to obtain an output feature map. For example, in a convolutional layer, a CNN performs convolution operations on the input image using convolution kernels (filters), denoted as K. R m×n×c Output feature map after convolution It can be expressed as the following formula (12):

[0121] (12)

[0122] Where: k1, k2, k3, and k4 are the sizes of the convolutional kernels in the four dimensions; i and j are the spatial indices of the output feature map, l is the index of the output channel, and o is the index in the fourth dimension. It is the bias term for the l-th output channel.

[0123] The size of the output feature map is calculated using the following formulas (13)-(16):

[0124] Output dimension 1 = (13)

[0125] Output dimension 2 = (14)

[0126] Output dimension 3 = (15)

[0127] Output dimension 4 = (16)

[0128] Among them, p1, p2, p3, and p4 are the fills in the four dimensions, and s1, s2, s3, and s4 are the step sizes in the four dimensions.

[0129] Since the aforementioned fault detection network includes a convolutional subnetwork and a classification subnetwork, the computer device can input the output feature map into the classification subnetwork for fault analysis to determine the cause of the power distribution line fault. For example, the specific process is as follows:

[0130] First, an activation function is used to introduce non-linearity into the convolution result (i.e., the output feature map), as shown in equation (17):

[0131] (17)

[0132] Then, perform a pooling operation on the result obtained in (17), using max pooling as shown in equation (18):

[0133] (18)

[0134] Where s is the size of the pooling window.

[0135] Thus, high-dimensional feature maps can be obtained through convolution and pooling operations. These high-dimensional feature maps can then be flattened and input into a fully connected layer for classification. For example, the feature vector I is first linearly transformed using the weight matrix Q and the bias vector n, and then an activation function is applied to obtain the final classification result, i.e., the cause of the power distribution line fault. The corresponding calculation formula is: Where σ is the activation function used to convert the scores of each category into probability values, and the sum of these probability values ​​is 1.

[0136] In this embodiment, a feature matrix can be generated based on multiple traveling wave features and the number of output channels of the convolutional subnetwork. This feature matrix is ​​then input into the convolutional subnetwork for four-dimensional convolution processing, resulting in a more accurate output feature map. This output feature map can then be input into the classification subnetwork for fault analysis, accurately identifying the cause of faults in the power distribution line.

[0137] In one embodiment, a training method for a fault detection network is provided, namely, the training method for the aforementioned fault detection network, comprising:

[0138] Acquire multiple historical secondary traveling wave signals and the corresponding fault tags for each historical secondary traveling wave signal.

[0139] Multiple historical traveling wave features are extracted from each historical secondary traveling wave signal.

[0140] A training dataset is constructed based on multiple historical traveling wave features and the fault labels corresponding to each historical secondary traveling wave signal.

[0141] The initial fault detection network is trained using the training dataset to obtain the fault detection network.

[0142] In this embodiment, the computer device can obtain multiple historical secondary traveling wave signals and corresponding fault labels from a database and / or public platform and / or public webpage. Therefore, the computer device can perform signal conversion on each historical secondary traveling wave signal to invert and obtain the corresponding historical primary traveling wave signals, and extract multiple historical traveling wave features from each historical primary traveling wave signal. These historical traveling wave features may include, but are not limited to, the amplitude, pulse width, and number of peaks of each historical primary traveling wave signal. Subsequently, the computer device can construct a training dataset based on the multiple historical traveling wave features and the corresponding fault labels of each historical secondary traveling wave signal, and divide the training dataset into a training set and a test set. Furthermore, the computer device can pre-construct an initial fault detection network. Thus, the computer device can train the initial fault detection network using the training set in the training dataset and test the initial fault detection network using the test set in the training dataset to obtain the final fault detection network.

[0143] In this embodiment, by acquiring multiple historical secondary traveling wave signals and the corresponding fault labels for each historical secondary traveling wave signal, and extracting multiple historical traveling wave features from each historical secondary traveling wave signal, a rich training dataset can be constructed based on these features and the corresponding fault labels. Therefore, an accurate fault detection network can be obtained by training the initial fault detection network using this rich training dataset.

[0144] In an optional embodiment, such as Figure 9 As shown, a fault cause identification method based on a primary and secondary fusion switch is provided, applied to computer equipment, including:

[0145] S601 controls the primary and secondary fusion switch in the power distribution line to collect the secondary traveling wave signal generated during the operation of the power distribution line;

[0146] S602, decompose the secondary traveling wave signal to obtain multiple first mode functions corresponding to the secondary traveling wave signal;

[0147] S603, construct an inversion model based on the conversion relationship between the first mode function and the first traveling wave signal;

[0148] S604 uses a fast iterative shrinking threshold algorithm to iteratively calculate multiple first mode functions based on the objective function of the inversion model to obtain the primary traveling wave signal of the power distribution line;

[0149] Specifically, S604 includes:

[0150] Obtain the initial number of iterations and the initial second mode function;

[0151] Based on the initial number of iterations and multiple first mode functions, determine the gradient and initial step size parameters corresponding to the initial number of iterations;

[0152] Based on the gradient corresponding to the initial iteration number, the initial step size parameter, the objective function, and multiple first mode functions, determine the next second mode function;

[0153] If the initial second mode function and the next second mode function satisfy the preset termination condition, then the initial second mode function and the next second mode function are accumulated, and the accumulated result is determined as a traveling wave signal;

[0154] If the initial second mode function and the next second mode function do not meet the preset termination condition, then update the initial iteration number, use the next second mode function as the new initial second mode function, and return to execute the step of determining the gradient and initial step size parameters corresponding to the initial iteration number based on the initial iteration number and multiple first mode functions;

[0155] S605 extracts multiple traveling wave features from the primary traveling wave signal of a power distribution line;

[0156] S606 inputs multiple traveling wave characteristics into a preset fault detection network for fault analysis to obtain the cause of the power distribution line fault.

[0157] Specifically, S606 includes:

[0158] A feature matrix is ​​generated based on multiple traveling wave features and the number of output channels of the convolutional sub-network; the multiple traveling wave features include amplitude, pulse width, and number of peaks;

[0159] The feature matrix is ​​input into the convolutional sub-network for four-dimensional convolution processing to obtain the output feature map;

[0160] The output feature map is input into the classification subnetwork for fault analysis to obtain the cause of the power distribution line fault.

[0161] The training method for the fault detection network can be referred to in the above embodiments, and will not be repeated here.

[0162] In the aforementioned fault cause identification method based on a primary and secondary fusion switch, the primary and secondary fusion switch in the power distribution line is controlled to collect the secondary traveling wave signal generated during the operation of the power distribution line; the secondary traveling wave signal is converted to obtain the primary traveling wave signal of the power distribution line; multiple traveling wave features are extracted from the primary traveling wave signal of the power distribution line, and these features are input into a preset fault detection network for fault analysis to obtain the fault cause of the power distribution line. Since this embodiment can convert the secondary traveling wave signal generated during the operation of the power distribution line to obtain the primary traveling wave signal, the primary traveling wave waveform can be accurately reconstructed. Therefore, based on the multiple traveling wave features in the accurate primary traveling wave signal, the fault cause and / or fault type of the power distribution line can be accurately determined or identified, thus improving the accuracy of identifying the fault cause of the power distribution line.

[0163] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0164] Based on the same inventive concept, this application also provides a device for identifying the cause of a fault based on a primary and secondary fusion switch, used to implement the aforementioned method for identifying the cause of a fault based on a primary and secondary fusion switch. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the device for identifying the cause of a fault based on a primary and secondary fusion switch provided below can be found in the limitations of the method for identifying the cause of a fault based on a primary and secondary fusion switch described above, and will not be repeated here.

[0165] In one exemplary embodiment, such as Figure 10 As shown, a fault cause identification device based on a primary and secondary fusion switch is provided, including: a data acquisition module 31, a signal conversion module 32, and a fault analysis module 33, wherein:

[0166] The acquisition module 31 is used to control the primary and secondary fusion switch in the power distribution line to acquire the secondary traveling wave signal generated by the power distribution line during operation.

[0167] The signal conversion module 32 is used to convert the secondary traveling wave signal to obtain the primary traveling wave signal of the power distribution line.

[0168] The fault analysis module 33 is used to extract multiple traveling wave features from the primary traveling wave signal of the power distribution line, and input the multiple traveling wave features into a preset fault detection network for fault analysis to obtain the fault cause of the power distribution line.

[0169] In one embodiment, the signal conversion module 32 includes:

[0170] The decomposition unit is used to decompose the secondary traveling wave signal to obtain multiple first mode functions corresponding to the secondary traveling wave signal;

[0171] The building unit is used to construct an inversion model based on the conversion relationship between the first mode function and the first traveling wave signal;

[0172] The calculation unit is used to iteratively calculate multiple first mode functions based on the objective function of the inversion model to obtain the primary traveling wave signal of the power distribution line.

[0173] In one embodiment, the computing unit includes:

[0174] Obtain sub-units to obtain the initial iteration count and the initial second mode function;

[0175] The first determining sub-unit is used to determine the gradient and initial step size parameters corresponding to the initial iteration number based on the initial iteration number and multiple first mode functions;

[0176] The second determining subunit is used to determine the next second mode function based on the gradient corresponding to the initial iteration number, the initial step size parameter, the objective function, and multiple first mode functions;

[0177] The third determining subunit is used to determine the primary traveling wave signal of the power distribution line based on the initial second mode function and the next second mode function.

[0178] In one embodiment, the third determining subunit is specifically used for:

[0179] If the initial second mode function and the next second mode function satisfy the preset termination condition, then the initial second mode function and the next second mode function are accumulated, and the accumulated result is determined as a traveling wave signal;

[0180] If the initial second mode function and the next second mode function do not meet the preset termination condition, then update the initial iteration number, use the next second mode function as the new initial second mode function, and return to execute the step of determining the gradient and initial step size parameters corresponding to the initial iteration number based on the initial iteration number and multiple first mode functions.

[0181] In one embodiment, the fault detection network includes a convolutional subnetwork and a classification subnetwork, and the fault analysis module 33 includes:

[0182] The feature matrix generation unit is used to generate a feature matrix based on multiple traveling wave features and the number of output channels of the convolutional sub-network; the multiple traveling wave features include amplitude, pulse width, and number of peaks.

[0183] The convolution processing unit is used to input the feature matrix into the convolutional sub-network for four-dimensional convolution processing to obtain the output feature map;

[0184] The fault analysis unit is used to input the output feature map into the classification subnetwork for fault analysis to obtain the cause of the power distribution line fault.

[0185] In one embodiment, the above-mentioned fault cause identification device based on primary and secondary fusion switches further includes:

[0186] The historical secondary traveling wave signal acquisition module is used to acquire multiple historical secondary traveling wave signals and the fault tags corresponding to each historical secondary traveling wave signal;

[0187] The historical traveling wave feature extraction module is used to extract multiple historical traveling wave features from each historical secondary traveling wave signal;

[0188] The training dataset construction module is used to construct a training dataset based on multiple historical traveling wave features and the fault labels corresponding to each historical secondary traveling wave signal.

[0189] The training module is used to train the initial fault detection network based on the training dataset to obtain the fault detection network.

[0190] Each module in the aforementioned fault cause identification device based on a primary and secondary fusion switch can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0191] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 1 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a fault cause identification method based on a primary and secondary fusion switch. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0192] Those skilled in the art will understand that Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0193] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0194] The primary and secondary integrated switch in the control power distribution line collects the secondary traveling wave signal generated by the power distribution line during operation;

[0195] The secondary traveling wave signal is converted to obtain the primary traveling wave signal of the power distribution line;

[0196] Multiple traveling wave features are extracted from the primary traveling wave signal of the power distribution line, and these features are input into a preset fault detection network for fault analysis to obtain the cause of the power distribution line fault.

[0197] In one embodiment, the secondary traveling wave signal is converted to obtain the primary traveling wave signal of the power distribution line. When the processor executes the computer program, it also performs the following steps:

[0198] The secondary traveling wave signal is decomposed to obtain multiple first mode functions corresponding to the secondary traveling wave signal;

[0199] Based on the conversion relationship between the first mode function and the first traveling wave signal, an inversion model is constructed;

[0200] The primary traveling wave signal of the power distribution line is obtained by iteratively calculating multiple first-mode functions based on the objective function of the inversion model.

[0201] In one embodiment, multiple first mode functions are iteratively calculated based on the objective function of the inversion model to obtain the primary traveling wave signal of the power distribution line. When the processor executes the computer program, it also performs the following steps:

[0202] Obtain the initial number of iterations and the initial second mode function;

[0203] Based on the initial number of iterations and multiple first mode functions, determine the gradient and initial step size parameters corresponding to the initial number of iterations;

[0204] Based on the gradient corresponding to the initial iteration number, the initial step size parameter, the objective function, and multiple first mode functions, determine the next second mode function;

[0205] The primary traveling wave signal of the power distribution line is determined based on the initial second mode function and the next second mode function.

[0206] In one embodiment, based on the initial second mode function and the next second mode function, the primary traveling wave signal of the power distribution line is determined, and the processor, when executing the computer program, further implements the following steps:

[0207] If the initial second mode function and the next second mode function satisfy the preset termination condition, then the initial second mode function and the next second mode function are accumulated, and the accumulated result is determined as a traveling wave signal;

[0208] If the initial second mode function and the next second mode function do not meet the preset termination condition, then update the initial iteration number, use the next second mode function as the new initial second mode function, and return to execute the step of determining the gradient and initial step size parameters corresponding to the initial iteration number based on the initial iteration number and multiple first mode functions.

[0209] In one embodiment, the fault detection network includes a convolutional subnetwork and a classification subnetwork. Multiple traveling wave features are input into the preset fault detection network for fault analysis to obtain the cause of the power distribution line fault. When the processor executes the computer program, it also performs the following steps:

[0210] A feature matrix is ​​generated based on multiple traveling wave features and the number of output channels of the convolutional sub-network; the multiple traveling wave features include amplitude, pulse width, and number of peaks;

[0211] The feature matrix is ​​input into the convolutional sub-network for four-dimensional convolution processing to obtain the output feature map;

[0212] The output feature map is input into the classification subnetwork for fault analysis to obtain the cause of the power distribution line fault.

[0213] In one embodiment, the training method for the fault detection network further includes the following steps when the processor executes the computer program:

[0214] Acquire multiple historical secondary traveling wave signals and the fault tags corresponding to each historical secondary traveling wave signal;

[0215] Extract multiple historical traveling wave features from each historical secondary traveling wave signal;

[0216] A training dataset is constructed based on multiple historical traveling wave features and the fault labels corresponding to each historical secondary traveling wave signal;

[0217] The initial fault detection network is trained using the training dataset to obtain the fault detection network.

[0218] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0219] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0220] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0221] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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 application.

[0222] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for fault cause identification based on a primary and secondary fusion switch, characterized in that, The method includes: The primary and secondary fusion switch in the control power distribution line collects the secondary traveling wave signal generated by the power distribution line during operation; The secondary traveling wave signal is converted to obtain the primary traveling wave signal of the power distribution line; Multiple traveling wave features are extracted from the primary traveling wave signal of the power distribution line, and the multiple traveling wave features are input into a preset fault detection network for fault analysis to obtain the cause of the fault in the power distribution line.

2. The method according to claim 1, characterized in that, The step of converting the secondary traveling wave signal to obtain the primary traveling wave signal of the power distribution line includes: The secondary traveling wave signal is decomposed to obtain multiple first mode functions corresponding to the secondary traveling wave signal; Based on the conversion relationship between the first mode function and the first traveling wave signal, an inversion model is constructed; The multiple first mode functions are iteratively calculated based on the objective function of the inversion model to obtain the primary traveling wave signal of the power distribution line.

3. The method according to claim 2, characterized in that, The step of iteratively calculating the plurality of first mode functions based on the objective function of the inversion model to obtain the primary traveling wave signal of the power distribution line includes: Obtain the initial number of iterations and the initial second mode function; Based on the initial iteration number and the plurality of first mode functions, determine the gradient and initial step size parameters corresponding to the initial iteration number; The next second mode function is determined based on the gradient corresponding to the initial iteration number, the initial step size parameter, the objective function, and the plurality of first mode functions; The primary traveling wave signal of the power distribution line is determined based on the initial second mode function and the next second mode function.

4. The method according to claim 3, characterized in that, Determining the primary traveling wave signal of the power distribution line based on the initial second mode function and the next second mode function includes: If the initial second mode function and the next second mode function satisfy a preset termination condition, then the initial second mode function and the next second mode function are accumulated, and the accumulation result is determined as the first traveling wave signal; If the initial second mode function and the next second mode function do not satisfy the preset termination condition, then the initial iteration number is updated, and the next second mode function is used as the new initial second mode function. Then, the process returns to the step of determining the gradient and initial step size parameters corresponding to the initial iteration number based on the initial iteration number and the plurality of first mode functions.

5. The method according to any one of claims 1-4, characterized in that, The fault detection network includes a convolutional subnetwork and a classification subnetwork. The process of inputting the multiple traveling wave features into the preset fault detection network for fault analysis to obtain the cause of the power distribution line fault includes: A feature matrix is ​​generated based on the multiple traveling wave features and the number of output channels of the convolutional sub-network; the multiple traveling wave features include amplitude, pulse width, and number of peaks; The feature matrix is ​​input into the convolutional sub-network for four-dimensional convolution processing to obtain the output feature map; The output feature map is input into the classification sub-network for fault analysis to obtain the fault cause of the power distribution line.

6. The method according to any one of claims 1-4, characterized in that, The training method for the fault detection network includes: Acquire multiple historical secondary traveling wave signals and the fault tags corresponding to each of the historical secondary traveling wave signals; Extract multiple historical traveling wave features from each of the aforementioned historical secondary traveling wave signals; A training dataset is constructed based on multiple historical traveling wave features and the fault labels corresponding to each of the historical secondary traveling wave signals. The initial fault detection network is trained using the training dataset to obtain the fault detection network.

7. A fault cause identification device based on a primary and secondary fusion switch, characterized in that, The device includes: The acquisition module is used to control the primary and secondary fusion switch in the power distribution line to acquire the secondary traveling wave signal generated by the power distribution line during operation; The signal conversion module is used to convert the secondary traveling wave signal to obtain the primary traveling wave signal of the power distribution line. The fault analysis module is used to extract multiple traveling wave features from the primary traveling wave signal of the power distribution line, and input the multiple traveling wave features into a preset fault detection network for fault analysis to obtain the fault cause of the power distribution line.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images