Switch cabinet fault diagnosis method and device, electronic equipment and storage medium

Abstract

The invention provides a switch cabinet fault diagnosis method and device, electronic equipment and a storage medium, and belongs to the technical field of fault diagnosis, and the method comprises the steps: collecting a current time sequence signal of a switch cabinet, and building an observation matrix of the current time sequence signal; constructing a null space projection operator orthogonal to the healthy current time sequence signal; projecting the observation matrix to a null space by using a null space projection operator, and separating a residual matrix representing an abnormal current time sequence signal; according to the residual matrix, whether the switch cabinet has a fault is judged, and if yes, a fault dictionary matrix describing the fault category of the switch cabinet is constructed; otherwise, returning the current time sequence signal, and establishing an observation matrix of the current time sequence signal; the residual matrix and the fault dictionary matrix are combined, and the fault similarity of the current time sequence signals and different switch cabinet fault categories is calculated; and outputting the switch cabinet fault category corresponding to the maximum fault similarity as a switch cabinet fault diagnosis result. The accuracy and efficiency of fault diagnosis of the switch cabinet can be improved.

Description

Technical Field

[0001] This application belongs to the field of fault diagnosis technology, and more specifically, relates to a method and device for fault diagnosis of switchgear, electronic equipment, and storage medium. Background Technology

[0002] A switchgear is a piece of equipment used in power systems to control, protect, and monitor the operation of electrical equipment. It connects and disconnects circuits via switching devices, facilitating load switching and fault isolation in electrical systems. Switchgear is typically installed in distribution rooms or substations and is an important component of power equipment systems.

[0003] Switchgear is a core piece of equipment in a power system responsible for controlling and protecting circuits. A fault in switchgear can lead to power outages, equipment damage, production stoppages, and even personnel accidents. Therefore, timely detection and diagnosis of switchgear faults can effectively prevent power system shutdowns and ensure stable production operations. Simultaneously, accurate fault diagnosis helps maintenance personnel locate and repair faults, reducing maintenance costs and time, improving system reliability and safety, and preventing wider-ranging power accidents caused by fault escalation.

[0004] However, existing switchgear fault diagnosis methods typically determine whether the switchgear is abnormal by judging the original current signal based on a fixed threshold. This makes it difficult to separate normal signals from abnormal signals, and they are easily affected by environmental interference, resulting in low diagnostic accuracy, a high false judgment rate, and an inability to efficiently handle complex fault modes. Summary of the Invention

[0005] The purpose of this application is to provide a method, device, electronic equipment, and storage medium for fault diagnosis of switchgear, in order to solve the problems that existing fault diagnosis methods for switchgear usually determine whether the switchgear is abnormal by judging the original current signal based on a fixed threshold. This method is difficult to separate normal signals from abnormal signals, is easily affected by environmental interference, resulting in low diagnostic accuracy, high false judgment rate, and inability to efficiently handle complex fault modes.

[0006] A first aspect of this application provides a method for diagnosing switchgear faults, the method comprising: Acquire the current timing signal of the switchgear and establish an observation matrix for the current timing signal; The observation matrix is ​​projected onto the null space using a null projection operator orthogonal to the healthy current time series signal, and the residual matrix representing the abnormal current time series signal is separated. Based on the residual matrix and the Marchenko-Pastul law, determine whether there is a fault in the switch cabinet. If so, proceed to the next step; otherwise, return to the first step. Construct a fault dictionary matrix describing the fault categories of the switchgear; By combining the residual matrix and the fault dictionary matrix, the fault similarity between the current time-series signal and different switchgear fault categories is calculated; The switchgear fault category corresponding to the maximum fault similarity is output as the switchgear fault diagnosis result.

[0007] A second aspect of this application provides a switchgear fault diagnosis device, the device comprising: The acquisition module is used to acquire the current timing signal of the switchgear and establish the observation matrix of the current timing signal; The separation module is used to project the observation matrix to the null space using a null space projection operator orthogonal to the healthy current time series signal, and separate the residual matrix representing the abnormal current time series signal. The judgment module is used to determine whether there is a fault in the switch cabinet based on the residual matrix and the Marchenko-Pastur theorem. If so, the second construction module is called; otherwise, the acquisition module is called. The module is used to build a fault dictionary matrix that describes the fault categories of the switchgear; The calculation module is used to combine the residual matrix and the fault dictionary matrix to calculate the fault similarity between the current time-series signal and different switchgear fault categories; The output module is used to output the switchgear fault category corresponding to the maximum fault similarity as the switchgear fault diagnosis result.

[0008] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described switchgear fault diagnosis method.

[0009] In a fourth aspect of this application, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described switchgear fault diagnosis method.

[0010] The beneficial effects of the switchgear fault diagnosis method and device, electronic equipment, and storage medium provided in this application are as follows: In this embodiment, based on the characteristics of the switchgear's healthy operating current signal, the method first constructs a mathematical projection operator that can accurately filter out healthy signal components. After acquiring the current timing signal to be detected, the projection operator maps it to the null space. At this point, healthy signal components are effectively suppressed, while any abnormal signals deviating from the healthy pattern are significantly separated, forming a residual matrix. By applying the Marchenko-Pastur theorem in random matrix theory to analyze the statistical properties of the residual matrix, the presence of a fault can be sensitively and reliably determined, avoiding the limitations of fixed thresholds and significantly improving the ability to resist environmental interference. Once a fault is detected, the similarity between the residual signal and various known fault modes is calculated using a fault dictionary matrix. Finally, the fault category with the highest matching degree is output as the diagnostic result. This method adaptively separates abnormal signals through mathematical projection, eliminating the dependence on fixed empirical thresholds and greatly reducing the false positive rate. Using random matrix theory for fault presence detection provides robustness against environmental noise and signal fluctuations. Combining a fault mode dictionary for similarity matching enables efficient identification of various complex fault types, thereby improving the accuracy and efficiency of switchgear fault diagnosis. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart illustrating a switchgear fault diagnosis method provided in an embodiment of this application; Figure 2 This is a structural block diagram of a switchgear fault diagnosis device provided in one embodiment of the present application; Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0013] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0014] It is understood that in the embodiments of this application, data such as user information are involved. When the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with relevant laws, regulations and standards.

[0015] It should be noted that the terms "first," "second," etc., used in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein.

[0016] Before providing a further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application will be explained, and the nouns and terms involved in the embodiments of this application shall be interpreted as follows.

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0018] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a switchgear fault diagnosis method provided in an embodiment of this application.

[0019] like Figure 1 As shown, the switchgear fault diagnosis method provided in this application embodiment may include: Acquire the current timing signal of the switchgear and establish an observation matrix for the current timing signal.

[0020] Among them, the current time-series signal refers to the signal of current change over time collected during the operation of the switchgear, which can reflect the operating status and health condition of the switchgear. The observation matrix is ​​constructed by sampling and organizing the current time-series signal to create a matrix containing current signal values ​​at multiple time points. It serves as the foundational data for subsequent fault diagnosis and analysis.

[0021] In one possible implementation, the current timing signal of the switchgear is acquired, and an observation matrix of the current timing signal is established, specifically including: Acquire current timing signals.

[0022] The current timing signal is divided into preset time windows, where each preset time window includes multiple consecutive current values.

[0023] It should be noted that the length of the acquired current time-series signal can be preset. Optionally, it can be set to an integer multiple of the number of current values ​​within a preset time window to ensure that the entire observation matrix is ​​filled. For example, if the current time-series signal is acquired at 9 time points, it can be split into [I1,I2,I3], [I2,I3,I4], [I3,I4,I5]—[I7,I8,I9] by a preset time window of 3. Each of these split time windows is then filled into the observation matrix as a row to obtain the observation matrix.

[0024] The split current time-series signals are sequentially filled into the observation matrix. Each row of the observation matrix represents the current value at different times within a preset time window, and each column of the observation matrix represents the continuous current value at different times.

[0025] Specifically, the process first collects the current timing signals generated by the switchgear during operation. These signals reflect the switchgear's operating status and health condition. By dividing these signals into preset time windows, each containing multiple consecutive current values, a matrix structure is formed. This structure helps the system capture the timing characteristics of current changes. After division, the observation matrix can comprehensively represent the changing patterns of the current timing signals, organizing the timing information of the current signals into rows and columns in an orderly manner, facilitating subsequent analysis and processing. Each row represents the current value within a time window, while each column represents the current change at different time points. This method effectively extracts regular information from the time series, making the signal fluctuation characteristics more obvious. Converting the raw current signal into a standardized matrix form makes the timing characteristics of the signal easier to extract and identify, enhancing the accuracy and reliability of subsequent fault detection.

[0026] The observation matrix is ​​projected onto the null space using a null projection operator orthogonal to the healthy current time series signal, thus separating the residual matrix representing the abnormal current time series signal.

[0027] The null-space projection operator is used to separate healthy current timing signals from other signals. Its function is to project the healthy signal components into the null space, meaning that healthy signal components are eliminated or suppressed, while any abnormal signals deviating from the healthy pattern remain unaffected and can be significantly separated from the observed signal. The null-space projection operator can accurately filter out healthy signal components from the current timing signal, thereby effectively highlighting abnormal signals. This process helps to more accurately identify whether a switchgear fault exists.

[0028] In one possible implementation, the method for constructing the null space projection operator specifically includes: Obtain the health status observation matrix of the switchgear.

[0029] Specifically, the current timing signal of the switchgear in a healthy state (such as after factory calibration) can be collected first, and then converted into a health status observation matrix.

[0030] Singular value decomposition is performed on the health status observation matrix to obtain the right singular vector matrix.

[0031] Determine the truncation rank of the right singular vector matrix.

[0032] Optionally, the truncated rank can be determined using the singular value difference (SVD) method. The specific process is as follows: First, obtain the singular value matrix obtained from the singular value decomposition of the health state observation matrix. Then, perform difference operations on the singular values ​​of the diagonal elements of the singular value matrix to obtain a difference sequence. Finally, select the index variable of the singular value of the diagonal element corresponding to the peak value of the difference sequence as the truncated rank output.

[0033] Select multiple right singular vectors corresponding to singular values ​​from the right singular vector matrix in descending order of singular values. The number of right singular vectors is equal to the truncation rank.

[0034] Specifically, right singular vectors corresponding to multiple singular values ​​are selected from the right singular vector matrix in descending order of singular values; that is, the first column and the second column of the right singular vector matrix are selected. k List, k This indicates a truncated rank.

[0035] Arrange the selected right singular vectors column by column to obtain the orthogonal basis matrix.

[0036] The null space projection operator is obtained by subtracting the product of the orthogonal basis matrix and the transpose of the orthogonal basis matrix from the identity matrix.

[0037] The formula for the null projection operator is as follows: ; in, Represents the null space projection operator. This indicates the truncated rank, and the subscript T indicates the transpose. Represents the identity matrix. Describes an orthogonal basis matrix. Denotes the first orthogonal basis matrix. i Singular vectors on the right.

[0038] Specifically, the process first constructs a health status observation matrix by acquiring the current timing signal of the switchgear in a healthy state. Singular value decomposition is then performed on the health status observation matrix to extract the right singular vectors. The truncation rank is determined based on the difference of the singular values, thus selecting the singular vectors most relevant to the health signal. Next, these selected right singular vectors are arranged into an orthogonal basis matrix, and a null space projection operator is constructed by subtracting the product of this matrix and its transpose from the identity matrix. This projection operator accurately projects the components of the health signal into the null space, effectively suppressing the health signal while preserving abnormal signals that differ from the healthy pattern. In this way, abnormal signals can be significantly separated from the observation matrix. This avoids the limitations of directly using a fixed threshold, greatly improving the subsequent ability to identify abnormal signals and the accuracy of fault diagnosis.

[0039] In this context, the null space refers to the result of the projection operator in mathematics. The healthy signal components in the observation matrix are mapped to the null space, meaning these signals are eliminated or suppressed, and only the abnormal signal components are retained. The residual matrix is ​​the matrix obtained after applying the null space projection operator, containing the projected abnormal current time-series signal portion, which is the abnormal component after removing the healthy signal.

[0040] In one possible implementation, the residual matrix is ​​specifically the product of the observation matrix and the null space projection operator.

[0041] The formula for calculating the residual matrix is ​​as follows: ; in, Represents the residual matrix. Represents the observation matrix. This represents the null space projection operator.

[0042] Specifically, the residual matrix is ​​obtained by multiplying the observation matrix by the null space projection operator. This process uses the null space projection operator to suppress the healthy current signal component from the observation matrix, leaving only the abnormal signal portion that differs from the healthy mode. This method effectively separates the signal components deviating from the healthy state, forming the residual matrix and providing a clear representation of the abnormal signal. By eliminating the influence of the healthy signal through precise mathematical operations and highlighting the abnormal portion, subsequent fault diagnosis becomes more accurate and exhibits stronger resistance to environmental noise.

[0043] Based on the residual matrix and the Marchenko-Pastur theorem, determine whether there is a fault in the switchgear. If so, proceed to the next step; otherwise, return to the first step.

[0044] The Marchenko-Pastor law is a statistical theory primarily used to analyze the eigenvalue distribution of random matrices. According to this law, when the dimension of a matrix is ​​high, its eigenvalues ​​will exhibit a specific distribution, which can be used to determine if anomaly patterns exist within the matrix. In switchgear fault diagnosis, analyzing the eigenvalue distribution of the residual matrix can determine if any fault-related abnormal signals exist. By combining the residual matrix with the Marchenko-Pastor law, its statistical properties are analyzed to determine if a switchgear fault exists. If the eigenvalues ​​of the residual matrix exhibit an abnormal pattern, a fault is indicated, and subsequent steps are initiated; otherwise, data is collected again for analysis.

[0045] In one possible implementation, the presence of a fault in the switchgear is determined based on the residual matrix and the Marchenko-Pastul law, specifically including: Calculate the estimated variance and aspect ratio of the residual matrix, where the aspect ratio is the ratio of the number of rows to the number of columns in the residual matrix.

[0046] By combining the estimated variance and the aspect ratio of the residual matrix, the theoretical noise upper bound of the residual matrix is ​​calculated using the Marchenko-Pastur law.

[0047] The specific formula for calculating the theoretical noise upper bound is as follows: ; ; in, Indicates the upper bound of theoretical noise. This indicates the estimated variance. This represents the aspect ratio of the residual matrix. m and n These represent the number of rows and columns of the residual matrix, respectively.

[0048] It should be noted that calculating the theoretical noise upper bound by combining the estimated variance and the aspect ratio of the residual matrix provides a dynamic, data-statistic-based benchmark for subsequent fault diagnosis. It does not rely on a fixed threshold but adaptively adjusts the noise upper bound based on the statistical properties and dimensionality of the residual matrix. This approach effectively separates normal signals from fault signals, avoiding misjudgments caused by environmental noise or data fluctuations. This calculation method exhibits strong robustness, self-adjusting under different environmental conditions, thus improving the accuracy and reliability of fault detection.

[0049] Extract the largest eigenvalue of the residual matrix.

[0050] Understandably, the process of extracting the largest eigenvalue involves first performing eigenvalue decomposition on the residual matrix and calculating all its eigenvalues. Then, the largest eigenvalue is selected from the calculated eigenvalues; this largest eigenvalue reflects the main energy component of the residual matrix. By extracting the largest eigenvalue, the difference between the current signal and the health state can be effectively quantified.

[0051] If the maximum eigenvalue is greater than the theoretical noise upper bound, the switchgear is determined to be faulty; otherwise, the switchgear is determined not to be faulty.

[0052] Specifically, by combining the residual matrix and the Marchenko-Pastur theorem, the estimated variance and aspect ratio of the residual matrix are first calculated, and then this information is used to calculate the theoretical noise upper bound. Next, the largest eigenvalue of the residual matrix is ​​extracted through eigenvalue decomposition. If the largest eigenvalue is greater than the theoretical noise upper bound, it indicates the presence of an abnormal pattern in the signal, thus indicating a fault in the switchgear. Statistical analysis accurately distinguishes between normal and abnormal signals and avoids the subjectivity of manually set thresholds, exhibiting strong robustness and adaptability, and accurately detecting faults even under conditions of significant noise interference.

[0053] Furthermore, the pre-emptive statistical fault detection process effectively filters the health status data stream, initiating subsequent fault classification only for abnormal signals exceeding the upper limit of random noise statistics. An adaptive noise upper bound constructed based on the Marchenko-Pastur theorem accurately distinguishes between random environmental noise and deterministic fault characteristics. When the maximum eigenvalue does not exceed the threshold, the system directly returns to the data acquisition step, avoiding redundant fault mode matching calculations on health data and significantly improving system efficiency. Resource-intensive fault dictionary matching is activated only when a genuine fault is confirmed, ensuring that computational resources are focused on critical anomaly analysis. This hierarchical processing mechanism reduces the overall computational load and avoids the risk of misclassification when there are no faults, significantly optimizing the real-time performance and reliability of the diagnostic system.

[0054] Construct a fault dictionary matrix that describes the fault categories of the switchgear.

[0055] The fault dictionary matrix is ​​a matrix containing features of multiple known fault categories. These features are obtained by analyzing historical fault data and help the system match current abnormal signals with known fault types, thereby achieving accurate fault classification.

[0056] In one possible implementation, the switchgear fault categories include arc faults, short-circuit faults, insulation faults, contact faults, and overheating faults. A fault dictionary matrix describing the switchgear fault categories is constructed, specifically including: Extract multiple current signals from switchgear fault categories of the same type.

[0057] Optionally, offline experiments or simulations can be conducted to identify potential faults such as arcing, short-circuit, insulation, contact, and overheating. Multiple current signals belonging to different fault conditions can then be acquired.

[0058] Extract the main feature vector representing the corresponding switchgear fault category from the obtained current signal.

[0059] Specifically, the extraction method is as follows: First, the current signals are assembled into a current signal matrix with the same length and width as the number of current signals. Then, singular value decomposition is performed on it to obtain a singular vector matrix. The first column of the singular vector matrix is ​​the principal singular vector corresponding to the largest singular value. This principal singular vector is the principal feature vector (i.e., the feature direction that best represents the corresponding switchgear fault category).

[0060] By normalizing the principal feature vectors using the L2 norm, unit principal feature vectors for each switchgear fault category are obtained.

[0061] The specific formula for calculating the unit principal eigenvector is as follows: ; in, Indicates the first q Unit principal feature vector of switchgear fault categories Indicates the first q The L2 norm of the principal feature vector corresponding to the fault category of the switch cabinet.

[0062] By combining the unit principal feature vectors of each switchgear fault category, a fault dictionary matrix is ​​obtained.

[0063] The formula for the fault dictionary matrix is ​​as follows:

[0064] in, Represents the fault dictionary matrix. Indicates the first q Unit principal feature vector of switchgear fault categories , M This indicates the total number of switchgear fault categories.

[0065] Specifically, by extracting current signals from different fault categories, singular value decomposition is used to extract the principal feature vector for each fault type. Then, these feature vectors are normalized using the L2 norm to obtain the unit principal feature vector for each fault category. A fault dictionary matrix is ​​constructed by combining the unit principal feature vectors of all categories. This process extracts and standardizes the key features of each fault category to ensure accurate matching of the corresponding fault type during fault identification. This method can efficiently extract representative features from complex current signals, making subsequent fault diagnosis more accurate and avoiding redundant processing of complex signals, thereby improving the accuracy and efficiency of fault classification.

[0066] By combining the residual matrix and the fault dictionary matrix, the fault similarity between the current timing signal and different switchgear fault categories is calculated.

[0067] Understandably, by calculating the similarity between the current timing signal and each fault category, the degree of matching between the current signal and different fault modes can be determined. This process, through quantitative comparison, enables the system to accurately identify the fault type most similar to the current fault signal, thereby quickly locating the specific fault category and improving diagnostic efficiency.

[0068] In one possible implementation, the residual matrix and fault dictionary matrix are combined to calculate the fault similarity between the current timing signal and different switchgear fault categories, specifically including: Extract the left singular vector of the residual matrix.

[0069] Specifically, the residual matrix is ​​first decomposed into singular values, and the left singular vector corresponding to the largest singular value in the decomposition result is the desired result.

[0070] Calculate the product of the left singular vector and the transpose of the unit principal eigenvector of each switch cabinet fault category.

[0071] The absolute value of the product is taken as the fault similarity output.

[0072] The specific formula for calculating fault similarity is as follows: ; in, Indicates the first i The transpose of the unit principal feature vector corresponding to each switch cabinet fault category This represents the left singular vector of the residual matrix.

[0073] Specifically, the fault similarity is obtained by extracting the left singular vector of the residual matrix and multiplying it with the transpose of the unit principal eigenvector of each fault category. This process helps the system accurately determine the similarity between the current signal and each fault mode and quickly identify the most likely fault type by quantifying the degree of matching between the current signal and each fault category.

[0074] The switchgear fault category corresponding to the maximum fault similarity is output as the switchgear fault diagnosis result.

[0075] In practical applications, the current time-series signal is first acquired and split into an observation matrix through a preset time window. Next, a null-space projection operator is constructed to separate healthy and abnormal signals, making the abnormal signals more prominent and easier to identify. Then, the Marchenko-Pastur law is used to calculate the theoretical noise upper bound, and eigenvalue decomposition is combined with analysis of the residual matrix to determine whether a fault exists in the signal. By constructing a fault dictionary matrix, the feature information of different fault modes is extracted and standardized, and the similarity between the current signal and the fault type is further calculated, ultimately outputting the fault type. This scheme, based on dynamic modeling and statistical analysis of healthy signals, effectively avoids the limitations of fixed thresholds and can adaptively handle noise and complex fault modes, thereby improving the accuracy of fault diagnosis.

[0076] In this embodiment, based on the characteristics of the switchgear's healthy operating current signal, the method first constructs a mathematical projection operator that can accurately filter out healthy signal components. After acquiring the current timing signal to be detected, the projection operator maps it to the null space. At this point, healthy signal components are effectively suppressed, while any abnormal signals deviating from the healthy pattern are significantly separated, forming a residual matrix. By applying the Marchenko-Pastur theorem in random matrix theory to analyze the statistical properties of the residual matrix, the presence of a fault can be sensitively and reliably determined, avoiding the limitations of fixed thresholds and significantly improving the ability to resist environmental interference. Once a fault is detected, the similarity between the residual signal and various known fault modes is calculated using a fault dictionary matrix. Finally, the fault category with the highest matching degree is output as the diagnostic result. This method adaptively separates abnormal signals through mathematical projection, eliminating the dependence on fixed empirical thresholds and greatly reducing the false positive rate. Using random matrix theory for fault presence detection is robust to environmental noise and signal fluctuations. Combining similarity matching with a fault mode dictionary can efficiently identify various complex fault types, thereby improving the accuracy and efficiency of switchgear fault diagnosis.

[0077] Based on the same inventive concept, this application also provides a switchgear fault diagnosis device for implementing the switchgear fault diagnosis method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more switchgear fault diagnosis device embodiments provided below can be found in the limitations of the switchgear fault diagnosis method described above, and will not be repeated here.

[0078] This application provides a switchgear fault diagnosis device, such as... Figure 2 As shown, the switchgear fault diagnosis device 20 includes: The acquisition module 201 is used to acquire the current timing signal of the switchgear and establish the observation matrix of the current timing signal.

[0079] The separation module 202 is used to project the observation matrix to the null space using a null space projection operator orthogonal to the healthy current time series signal, and separate the residual matrix representing the abnormal current time series signal.

[0080] The judgment module 203 is used to determine whether there is a fault in the switch cabinet based on the residual matrix and the Marchenko-Pastur law. If so, the second construction module 204 is called; otherwise, the acquisition module 201 is called.

[0081] Module 204 is used to construct a fault dictionary matrix that describes the fault categories of the switchgear.

[0082] The calculation module 205 is used to combine the residual matrix and the fault dictionary matrix to calculate the fault similarity between the current timing signal and different switchgear fault categories.

[0083] The output module 206 is used to output the switch cabinet fault category corresponding to the maximum fault similarity as the switch cabinet fault diagnosis result.

[0084] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of each module / unit in the above-described device embodiments, for example... Figure 3The functions of the acquisition module 201, separation module 202, judgment module 203, construction module 204, calculation module 205, and output module 206 are shown.

[0085] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0086] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0087] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store device type information.

[0088] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation method described in the switch cabinet fault diagnosis method provided in the embodiments of this application, or they can execute the implementation method of the electronic device described in the embodiments of this application, which will not be repeated here.

[0089] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0090] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., provided on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0091] Those skilled in the art will recognize that the modules / units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0092] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0093] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules, units, or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or modules / units, or it may be an electrical, mechanical, or other form of connection.

[0094] The modules / units described as separate components may or may not be physically separate. Similarly, the components shown as modules / units may or may not be physical modules / units; they may be located in one place or distributed across multiple network modules / units. Some or all of the modules / units can be selected to achieve the purpose of the embodiments in this application, depending on actual needs.

[0095] Furthermore, the functional modules / units in the various embodiments of this application can be integrated into one processing module / unit, or each module / unit can exist physically separately, or two or more modules / units can be integrated into one module / unit. The integrated modules / units described above can be implemented in hardware or as software functional modules / units.

[0096] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for diagnosing switchgear faults, characterized in that, The method includes: The current timing signal of the switchgear is collected, and an observation matrix of the current timing signal is established. The observation matrix is ​​projected onto the null space using a null projection operator orthogonal to the healthy current time series signal, and the residual matrix representing the abnormal current time series signal is separated. Based on the residual matrix, and in conjunction with the Marchenko-Pastul law, determine whether the switch cabinet is faulty. If so, proceed to the next step; otherwise, return to the first step. Construct a fault dictionary matrix describing the fault categories of the switchgear; By combining the residual matrix and the fault dictionary matrix, the fault similarity between the current timing signal and different fault categories of the switchgear is calculated; The switchgear fault category corresponding to the maximum fault similarity is output as the switchgear fault diagnosis result.

2. The switchgear fault diagnosis method as described in claim 1, characterized in that, The process of acquiring the current timing signal of the switchgear and establishing the observation matrix of the current timing signal specifically includes: Acquire the current timing signal; The current timing signal is divided into preset time windows, wherein each preset time window includes multiple consecutive current values; The split current timing signals are sequentially filled into the observation matrix, wherein each row of the observation matrix represents the current value at different times within a preset time window, and each column of the observation matrix represents the continuous current value at different times.

3. The switchgear fault diagnosis method as described in claim 1, characterized in that, The method for constructing the null space projection operator specifically includes: Obtain the health status observation matrix of the switchgear; Singular value decomposition is performed on the health status observation matrix to obtain the right singular vector matrix; Determine the truncated rank of the right singular vector matrix; From the right singular vector matrix, select multiple right singular vectors corresponding to singular values ​​in descending order of singular values, wherein the number of right singular vectors is equal to the truncated rank; Arrange the selected right singular vectors column by column to obtain the orthogonal basis matrix; The null space projection operator is obtained by subtracting the product of the orthogonal basis matrix and the transpose of the orthogonal basis matrix from the identity matrix.

4. The switchgear fault diagnosis method as described in claim 1, characterized in that, The residual matrix is ​​specifically the product of the observation matrix and the null space projection operator.

5. The switchgear fault diagnosis method as described in claim 1, characterized in that, The step of determining whether the switchgear is faulty based on the residual matrix and the Marchenko-Pastur theorem specifically includes: Calculate the estimated variance and aspect ratio of the residual matrix, wherein the aspect ratio is the ratio of the number of rows to the number of columns of the residual matrix; Combining the estimated variance and the aspect ratio of the residual matrix, the theoretical noise upper bound of the residual matrix is ​​calculated using the Marchenko-Pastur theorem. Extract the maximum eigenvalue of the residual matrix; If the maximum eigenvalue is greater than the theoretical noise upper bound, the switchgear is determined to be faulty; otherwise, the switchgear is determined not to be faulty.

6. The switchgear fault diagnosis method as described in claim 1, characterized in that, The switchgear fault categories include arc faults, short circuit faults, insulation faults, contact faults, and overheating faults; the construction of the fault dictionary matrix describing the switchgear fault categories specifically includes: Extract multiple current signals of the same type of switchgear fault category; Extract the main feature vector representing the corresponding switchgear fault category from the obtained current signal; The principal feature vectors are normalized using the L2 norm to obtain the unit principal feature vectors for each of the switchgear fault categories. The fault dictionary matrix is ​​obtained by combining the unit principal feature vectors of each of the switchgear fault categories.

7. The switchgear fault diagnosis method as described in claim 6, characterized in that, The step of combining the residual matrix and the fault dictionary matrix to calculate the fault similarity between the current time-series signal and different fault categories of the switchgear specifically includes: Extract the left singular vector of the residual matrix; Calculate the product of the left singular vector and the transpose of the unit principal eigenvector of each of the switchgear fault categories; The absolute value of the product is taken as the fault similarity output.

8. A switchgear fault diagnosis device, characterized in that, The device includes: The acquisition module is used to acquire the current timing signal of the switchgear and establish the observation matrix of the current timing signal; The separation module is used to project the observation matrix to the null space using a null space projection operator orthogonal to the healthy current time series signal, and separate out the residual matrix representing the abnormal current time series signal. The judgment module is used to determine whether the switch cabinet has a fault based on the residual matrix and the Marchenko-Pastur law. If it does, the second construction module is called; otherwise, the acquisition module is called. The module is used to build a fault dictionary matrix that describes the fault categories of the switchgear; The calculation module is used to combine the residual matrix and the fault dictionary matrix to calculate the fault similarity between the current timing signal and different fault categories of the switchgear; The output module is used to output the switchgear fault category corresponding to the maximum fault similarity as the switchgear fault diagnosis result.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images