Cable accessory fault determination method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of cable accessory detection, and discloses a cable accessory fault determination method and device, electronic equipment and a storage medium, and the method comprises the steps: collecting a sound signal of a cable accessory in an operation process; constructing a sound feature matrix based on the sound signals; a pre-trained classification model is adopted to classify the sound feature matrix, and a fault classification result is obtained; the fault type of the cable accessory is determined according to the fault classification result, the fault early warning information is generated, physical contact and potential damage to the cable accessory can be avoided, the anti-interference capability is high, the method is not sensitive to electromagnetic interference, stable performance is kept in a complex electric power environment, and therefore the accuracy of fault detection is improved.

Description

Technical Field

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

[0002] Cable accessories are an indispensable and important part of the power system. Their main function is to connect and protect cables to ensure the reliability and safety of power transmission. However, during long-term operation, cable accessories may fail due to environmental impact, mechanical damage, aging, etc. If these faults are not detected and handled in time, they may cause serious accidents in the power system. Therefore, cable accessory fault detection has become an important means to ensure the stable operation of the power system.

[0003] At present, the existing cable accessory fault detection method mainly judges the fault by detecting the electrical parameters of the cable accessories (such as voltage, current, insulation resistance, etc.). Although it can effectively detect certain types of faults, it is limited by external electromagnetic interference in actual applications, the detection results are easily affected and the accuracy is not high. Summary of the invention

[0004] The present application proposes a cable accessory fault determination method, device, electronic device and storage medium to improve the fault detection accuracy of the cable accessories.

[0005] In a first aspect, a method for determining a cable accessory fault is provided, comprising:

[0006] Collect the sound signals of cable accessories during operation;

[0007] constructing a sound feature matrix based on the sound signal;

[0008] Using a pre-trained classification model to classify the sound feature matrix to obtain a fault classification result;

[0009] The fault type of the cable accessory is determined according to the fault classification result, and fault warning information is generated.

[0010] In a second aspect, a cable accessory fault determination device is provided, comprising:

[0011] A collection module is used to collect sound signals of cable accessories during operation;

[0012] A construction module, used for constructing a sound feature matrix based on the sound signal;

[0013] A classification module, used to classify the sound feature matrix using a pre-trained classification model to obtain a fault classification result;

[0014] A determination module is used to determine the fault type of the cable accessory according to the fault classification result and generate fault warning information.

[0015] Optionally, in some embodiments of the present application, the building blocks include:

[0016] An extraction submodule, used for extracting features of the sound signal using Mel-frequency cepstral coefficients to obtain multi-frame feature vectors;

[0017] A calculation submodule, used for calculating the scatter entropy value of the feature vector of each frame based on the scatter entropy;

[0018] A construction submodule is used to construct a sound feature matrix based on each of the feature vectors and each of the spread entropy values.

[0019] Optionally, in some embodiments of the present application, the cable accessory fault determination device further includes a dimensionality reduction processing module, and the dimensionality reduction processing module is specifically used to:

[0020] A dimension reduction process is performed on each of the feature vectors.

[0021] Optionally, in some embodiments of the present application, the dimension reduction processing module includes:

[0022] The dimensionality reduction processing submodule is used to perform dimensionality reduction processing on the sound feature matrix based on the principal component analysis method.

[0023] Optionally, in some embodiments of the present application, the cable accessory fault determination module further includes a preprocessing module, and the preprocessing module is specifically used to:

[0024] The sound signal is preprocessed, wherein the preprocessing operation includes filtering and denoising.

[0025] Optionally, in some embodiments of the present application, the cable accessory fault determination device further includes a training module, and the training module is specifically used to:

[0026] Acquire training data corresponding to the classification model to be trained, wherein the training data is historical sound signal data;

[0027] Dividing the training data into a training set and a test set according to a preset ratio;

[0028] The classification model to be trained is iteratively trained based on the training set and the classification model after the iterative training is verified using the test set to obtain the pre-trained classification model.

[0029] Optionally, in some embodiments of the present application, the training module includes:

[0030] A determination submodule, used to determine the model parameters of the classification model to be trained based on a bat optimization algorithm;

[0031] The training submodule is used to iteratively train the classification model to be trained based on the training set and the initial model parameters and optimize the model parameters, and use the test set to verify the classification model after the iterative training to obtain the pre-trained classification model.

[0032] According to a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned cable accessory fault determination method when executing the computer program.

[0033] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned cable accessory fault determination method are implemented.

[0034] The present application provides a cable accessory fault determination method, device, electronic device and storage medium, which collects the sound signal of the cable accessory during operation; constructs a sound feature matrix based on the sound signal; uses a pre-trained classification model to classify the sound feature matrix to obtain a fault classification result; determines the fault type of the cable accessory according to the fault classification result, and generates fault warning information. In the cable accessory fault determination scheme provided by the present application, the sound signal of the cable accessory during operation is analyzed to determine the fault type of the cable accessory, and generates fault warning information. It can monitor and diagnose the cable accessories in real time during operation, quickly identify faults, provide fault warnings, and ensure the safe operation of the power system. It has the advantages of high sensitivity and non-contact detection, avoiding physical contact and potential damage to the cable accessories, and has strong anti-interference ability, is insensitive to electromagnetic interference, and can maintain stable performance in a complex power environment, thereby improving the accuracy of fault detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 An application environment diagram of the cable accessory fault determination method provided in an embodiment of the present application;

[0037] Figure 2A flowchart of a cable accessory fault determination method provided in an embodiment of the present application;

[0038] Figure 3 A schematic diagram of a waveform of a sound signal of a cable accessory under normal conditions provided by an embodiment of the present application;

[0039] Figure 4 A schematic diagram of a waveform of a sound signal when a cable accessory is abnormal according to another embodiment of the present application;

[0040] Figure 5 A flowchart of a cable accessory determination method provided in yet another embodiment of the present application;

[0041] Figure 6 A structural block diagram of a cable accessory fault determination device provided in an embodiment of the present application;

[0042] Figure 7 A structural block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided so as to provide a full understanding of the embodiments of the present application. However, it will be appreciated by those skilled in the art that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the present application.

[0045] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0046] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0047] The cable accessory fault determination method provided by the embodiment of the present invention can be applied in the following aspects: Figure 1 In the application environment, the electronic device 120 communicates with the high-sensitivity microphone sensor 110 installed around the cable accessories through the network 130. The electronic device 120 can collect the sound signal of the cable accessories during operation through the high-sensitivity microphone sensor 110; construct a sound feature matrix based on the sound signal; classify the sound feature matrix using a pre-trained classification model to obtain a fault classification result; determine the fault type of the cable accessories according to the fault classification result, and generate fault warning information. The fault classification result and the fault warning information are displayed on the electronic device 120. In the present invention, by analyzing the sound signal of the cable accessories during operation, the fault type of the cable accessories is determined, and the fault warning information is generated, the cable accessories during operation can be monitored and diagnosed in real time, and the fault can be quickly identified, and a fault warning can be provided to ensure the safe operation of the power system. It has the advantages of high sensitivity and non-contact detection, avoiding physical contact and potential damage to the cable accessories, and has strong anti-interference ability, is insensitive to electromagnetic interference, and can maintain stable performance in a complex power environment, thereby improving the accuracy of fault detection.

[0048] See also Figure 2 As shown, Figure 2 A schematic diagram of a flow chart of a cable accessory fault determination method provided by an embodiment of the present invention. The method can be applied to a server. This embodiment is illustrated by applying it to a server. The cable accessory fault determination method includes the following steps:

[0049] 101: Collect the sound signals of cable accessories during operation.

[0050] The sound signal can be an analog signal or a discrete signal after digitization. The sound signal can be the electrical noise caused by the current when the cable accessories are in operation, especially under high voltage conditions, accompanied by a slight arc discharge sound. The electrical noise can be caused by faults such as arc discharge, partial discharge or current instability.

[0051] The sound signals of the cable accessories during operation can be collected by a high-sensitivity microphone.

[0052] For example, key monitoring areas of cable accessories can be determined. Key monitoring areas may include cable joints, terminals, insulators, etc. Multiple high-sensitivity microphones can be evenly distributed around the cable accessories to ensure coverage of key monitoring areas, so that sound signals of cable accessories during operation can be collected through multiple high-sensitivity microphones.

[0053] 102: Construct a sound feature matrix based on the sound signal.

[0054] Determine the total number of frames of the sound signal, perform feature extraction on each frame of the sound signal, obtain multiple features of each frame of the sound signal, and construct a feature vector corresponding to each frame of the sound signal. Then, construct a sound feature matrix based on the total number of frames and each feature vector.

[0055] For example, assuming that the total number of sound signal frames is 1000 frames, the frame length of each sound signal frame is 25ms, and the frame shift is 10ms, then the number of features of each sound signal frame is 13, that is, the feature vector corresponding to each sound signal frame is 13 dimensions, so that a sound feature matrix of size 1000*13 can be obtained.

[0056] 103: Classify the sound feature matrix using a pre-trained classification model to obtain a fault classification result.

[0057] The pre-trained classification model can be different types of classifiers such as support vector machine (SVM), decision tree, random forest, neural network, etc.

[0058] The fault classification result may include multiple fault types corresponding to the sound feature matrix and the predicted probability corresponding to each fault type. The fault classification result may be a vector consisting of multiple fault types and the predicted probability of each fault type. The sum of the probabilities of the vector may not be 1; for example, [partial discharge: 0.8 insulation aging: 0.6 mechanical damage: 0.4]; or the sum of the probabilities of the vector is 1; for example, [partial discharge: 0.6 insulation aging: 0.3 mechanical damage: 0.1].

[0059] The sound feature matrix is ​​used as the input of the pre-trained classification model, so that the pre-trained classification model performs classification prediction on the sound feature matrix to obtain a fault classification result.

[0060] 104: Determine the fault type of the cable accessory according to the fault classification result, and generate fault warning information.

[0061] Fault warning information can include the fault type and the corresponding handling strategy for the fault type. For example, fault type: partial discharge; predicted probability: 0.85; impact: may cause cable performance degradation, and may cause larger-scale power failures in the long term; handling strategy: immediately arrange professional technicians to inspect and prepare necessary maintenance work; urgency: high.

[0062] The sizes of various predicted probabilities in the fault classification results can be compared, and the fault type corresponding to the largest predicted probability can be selected as the fault type of the cable accessory. Fault warning information can be generated based on the fault type, and the fault warning information can be sent to relevant personnel so that the relevant personnel can deal with the fault type in a timely manner.

[0063] A preset threshold can be set, and the fault type corresponding to the predicted probability greater than the preset threshold is used as the fault type of the cable accessory, and fault warning information is generated. For example, if the preset threshold is 0.5 and the fault classification result is [partial discharge: 0.8 insulation aging: 0.6 mechanical damage: 0.4], the fault type can be partial discharge and insulation aging.

[0064] The fault diagnosis results and fault warning information can also be sent to the monitoring center, so that managers can understand the operating status of cable accessories in a timely manner.

[0065] It is also possible to locate the fault location of the cable accessories by analyzing information such as the source and acquisition time of the sound signal, that is, the sound sensor mainly collects the sound signal of the regional location of the cable accessories.

[0066] In one embodiment, constructing a sound feature matrix based on the sound signal includes:

[0067] Using Mel-frequency cepstrum coefficients to extract features from the sound signal to obtain multi-frame feature vectors;

[0068] Calculate the scatter entropy value of the feature vector of each frame based on the scatter entropy;

[0069] A sound feature matrix is ​​constructed based on each of the feature vectors and each of the spread entropy values.

[0070] Mel-Frequency Cepstral Coefficients (MFCCs) are linear transformations of the logarithmic energy spectrum based on the nonlinear Mel scale of sound frequency, and are used to extract the features of sound signals. The spread entropy value is used to represent the feature distribution of the feature vector of each frame. A higher spread entropy value indicates that the feature distribution is more dispersed, while a lower spread entropy value indicates that the feature distribution is more concentrated.

[0071] In one embodiment, the parameter settings of the Mel cepstral coefficient and the spread entropy can be set according to the actual application scenario, and are not specifically limited here. For example, the Mel cepstral coefficient frame length is set to 25ms, the frame shift is set to 10ms, the dimension is set to 13, and the cumulative contribution rate is set to 0.85. The embedding dimension of the spread entropy is set to 3, the delay is 1, and the number of feature categories is set to 6.

[0072] For example, assume that a sound sensor is used to collect sound signals. The sound sensor can be a high-sensitivity microphone array. High-sensitivity microphone sensors are arranged at different locations of the cable accessories, and the collected sound signals are stored as raw data. Assume that 5 sensors are arranged at different locations, the sampling frequency of each sensor is 44.1kHz, and the length of the collected signal is 10 seconds. The amount of data collected by each sensor for the sound signal is: 441,000 data points. Each frame of the sound signal includes 441 data points, and the total number of frames of the sound signal is 1,000 frames:

[0073]

[0074] Assume that the frame length of the sound signal is 25ms and the frame shift is 10ms. The number of features extracted from each frame through the Mel-frequency cepstral coefficient is 13, that is, the feature vector of each frame is 13 dimensions, and the total size is 1000 frames * 13 dimensions = 13000 feature elements; then, according to the dimension of the feature vector, the number of feature categories m can be determined to be 13, and the category probability of each eigenvalue in the feature vector is predicted, so as to calculate the scatter entropy value corresponding to the feature vector based on the number of feature categories and the category probability of each eigenvalue in the feature vector, and then concatenate the scatter entropy value and the feature vector to obtain a new feature vector, so as to form a sound feature matrix according to each new feature vector.

[0075] The scatter entropy value of each frame feature vector can be calculated according to the following formula:

[0076]

[0077] In the formula, DE is the diffusion entropy value, p i is the class probability of the i-th eigenvalue in the eigenvector.

[0078] In one embodiment, before calculating the scatter entropy value of the feature vector of each frame based on the scatter entropy, the method further includes:

[0079] A dimension reduction process is performed on each of the feature vectors.

[0080] The feature vector can be reduced in dimension by using a preset dimensionality reduction processing method, thereby retaining the key features in the feature vector, reducing the complexity of the data, avoiding the dimensionality disaster, and improving the computational efficiency. The preset dimensionality reduction processing method may be linear discriminant analysis, singular value decomposition, autoencoder, etc.

[0081] For example, assuming that the total number of frames of the sound signal is 1000, the dimension of the feature vector is 13, and after the dimensionality reduction of the feature vector, the feature vector retains 5 eigenvalues, then the dimension of the feature vector is 5. Therefore, the size of the matrix formed by the dimensionality reduced feature vector is 1000 frames * 5 dimensions = 5000 elements.

[0082] In one embodiment, the dimensionality reduction processing of each of the feature vectors includes:

[0083] The sound feature matrix is ​​subjected to dimensionality reduction processing based on a principal component analysis method.

[0084] The principal component analysis method transforms the eigenvectors into a new coordinate system through an orthogonal transformation so that the variance of any projection of the eigenvectors is maximized, thereby retaining the most important eigenvalues.

[0085] For example, assuming that the total number of frames of the sound signal is 1000 frames, the dimension of the feature vector is 13 dimensions, and the principal component analysis method is used to reduce the 13-dimensional feature vector to obtain a 5-dimensional feature vector. The size of the matrix after dimensionality reduction is 1000 frames * 5 dimensions = 5000 elements.

[0086] For example, if there is no abnormality in the cable accessories, the collected sound signal is subjected to feature extraction through the Mel cepstral coefficients, and the obtained feature vector is as follows:

[0087] Frame 1: [12.34,-1.56,2.45,-0.34,0.78,-1.23,0.56,-0.67,0.45,-0.89,0.23,-0.34,0.67];

[0088] Frame 2: [12.45,-1.67,2.56,-0.45,0.89,-1.34,0.67,-0.78,0.56,-1.00,0.34,-0.45,0.78].

[0089] The feature vector after dimensionality reduction of a frame is:

[0090] PCA 正常 =[3.45,2.34,1.23,0.89,0.56]

[0091] The scatter entropy (DE) value of the feature vector after dimensionality reduction is:

[0092] DE 正常 =[0.56,0.45,0.34,0.67,0.78]

[0093] If a fault occurs in the cable accessories, such as a partial discharge fault, the collected acoustic signal is subjected to feature extraction using the Mel cepstral coefficients, and the resulting feature vector is as follows:

[0094] Frame 1: [15.34,-3.56,5.45,-2.34,3.78,-4.23,3.56,-3.67,3.45,-3.89,3.23,-3.34,3.67];

[0095] Frame 2: [15.45,-3.67,5.56,-2.45 3.89,-4.34,3.67,-3.78,3.56,-4.00,3.34,-3.45,3.78];

[0096] The feature vector after dimensionality reduction of a frame is:

[0097] PCA 故障 =[6.78,5.67,4.56,3.89,2.34]

[0098] The scatter entropy (DE) value of the feature vector after dimensionality reduction is:

[0099] DE 故障 =[0.89,1.00,0.78,1.11,1.22]

[0100] Therefore, the fault type of the cable accessory can be effectively identified based on the difference between the characteristic data of the sound signal when a fault occurs and the sound signal when normal.

[0101] In addition, according to Figures 4 to 5 As shown, according to the difference between the sound signal when a fault occurs and the sound signal when normal, the fault type of the cable accessory can be effectively identified.

[0102] In this application, the feature matrix is ​​reduced in dimension through principal component analysis to retain the main features, reduce data redundancy, and speed up subsequent calculations, which is beneficial to improving the accuracy and efficiency of fault result detection of cable accessories.

[0103] In one embodiment, before constructing the sound feature matrix based on the sound signal, the method further includes:

[0104] The sound signal is preprocessed, wherein the preprocessing operation includes filtering and denoising.

[0105] The sound signal can be filtered by a bandpass filter; the sound signal can be denoised by denoising techniques such as Wiener filtering, adaptive filtering and wavelet denoising.

[0106] In addition, the preprocessing operation also includes normalization processing, that is, the collected sound signals can also be normalized to reduce the impact of outliers on the feature extraction operation.

[0107] In this embodiment, by performing a preprocessing operation on the sound signal, the signal-to-noise ratio of the sound signal is improved, useful information is retained, and preparation is made for feature extraction and analysis, which is beneficial to improving the fault detection accuracy and stability of the cable accessories.

[0108] In one embodiment, before the pre-trained classification model is used to classify the sound feature matrix to obtain the fault classification result, the method further includes:

[0109] Acquire training data corresponding to the classification model to be trained, wherein the training data is historical sound signal data;

[0110] Dividing the training data into a training set and a test set according to a preset ratio;

[0111] The classification model to be trained is iteratively trained based on the training set and the classification model after the iterative training is verified using the test set to obtain the pre-trained classification model.

[0112] The training data is the sound signal sample data of the cable accessories in the operation process collected historically, and the fault type of each sound signal sample in the sound signal sample data. The training set is used to iteratively train the classification model to be trained, and the test set is used to verify the classification model after iterative training.

[0113] Assuming that the training data includes 1000 frames, the training data can be divided into a training set of 800 frames and a test set of 200 frames according to a preset ratio of 4:1. The classification model to be trained can be a support vector machine. During the iterative training process, the penalty coefficient and kernel function in the support vector machine are continuously updated until the training is completed.

[0114] In one embodiment, the iterative training of the classification model to be trained based on the training set and the use of the test set to verify the iteratively trained classification model to obtain the pre-trained classification model includes:

[0115] Determining model parameters of the classification model to be trained based on a bat optimization algorithm;

[0116] The classification model to be trained is iteratively trained based on the training set and the initial model parameters and the model parameters are optimized, and the classification model after the iterative training is verified using the test set to obtain the pre-trained classification model.

[0117] Set the default parameters of the bat optimization algorithm, including population size, number of iterations, frequency range, and dimension. Initialize the variation optimization algorithm to generate random particles and calculate the fitness value of the particles. When the fitness value of the particle is less than the set value or the number of iterations of the particle is greater than the set value, output the particles that meet the conditions. Substitute the penalty coefficient and kernel function optimized by the bat optimization algorithm into the support vector machine to complete the training of the classification model. Then input the test set into the trained support vector machine for classification verification.

[0118] During the training of the classification model, the model parameters of the support vector machine (such as the penalty coefficient and kernel function parameters) are optimized by the bat optimization algorithm to improve the classification performance of the support vector machine. Determine the parameters that need to be optimized in the support vector machine, such as the penalty coefficient and kernel function parameters; set the parameters of the bat optimization algorithm, including the number of bats, frequency range, volume, pulse rate, search space range, and maximum number of iterations; generate an initial population, that is, randomly generate a group of bat positions, each position represents a candidate solution for the model parameters of the support vector machine; use the parameters in the initial population (i.e., the penalty coefficient and kernel function parameters) to train the support vector machine. If the fitness of the new candidate solution is higher, update the model parameters of the support vector machine until the iteration termination condition is met, such as reaching the maximum number of iterations or the fitness improvement is lower than the preset threshold, select the bat position with the highest fitness from all bats as the optimal model parameters of the support vector machine, use the entire training set and the optimal model parameters to train the support vector machine, obtain the final classification model, and evaluate the performance of the support vector machine on the validation set to ensure the generalization ability of the support vector machine.

[0119] In one embodiment, if Figure 5 As shown, another cable accessory fault determination process is provided, which is as follows:

[0120] First, the acoustic signal of the cable accessories during operation is collected through the acoustic signal acquisition module. The acoustic signal acquisition module includes multiple high-sensitivity microphones arranged around the cable accessories; then, the collected acoustic signals are preprocessed by the signal processing module, including filtering, denoising and feature extraction; specifically, the Mel Frequency Cepstrum Coefficient (MFCC) and Dispersion Entropy (DE) are combined to extract the acoustic signal features and construct the feature matrix; then, the principal component analysis (PCA) is used to reduce the dimension of the feature matrix through the data analysis module to reduce the data dimension and retain the main features in the feature matrix. Finally, the penalty coefficient and kernel function parameters of the support vector machine can be optimized by the bat optimization algorithm (BA) to improve the classification accuracy, and then the support vector machine (SVM) is used to classify the processed acoustic signal features through the fault diagnosis module, diagnose the fault type of the cable accessories, and generate fault warning information. Finally, the fault classification results are sent to the monitoring center through the communication transmission module, so that the management personnel can understand the operation status of the cable accessories in time.

[0121] The above is the cable accessory fault determination process of this application.

[0122] As mentioned above, the present application provides a cable accessory fault determination method, device, electronic device and storage medium, by collecting the sound signal of the cable accessories during operation; constructing a sound feature matrix based on the sound signal; using a pre-trained classification model to classify the sound feature matrix to obtain a fault classification result; determining the fault type of the cable accessories according to the fault classification result, and generating fault warning information. In the cable accessory fault determination scheme provided by the present application, the sound signal of the cable accessories during operation is analyzed to determine the fault type of the cable accessories, and generate fault warning information. It can monitor and diagnose the cable accessories in real time during operation, quickly identify faults, provide fault warnings, and ensure the safe operation of the power system. It has the advantages of high sensitivity and non-contact detection, avoiding physical contact and potential damage to the cable accessories, and has strong anti-interference ability, is insensitive to electromagnetic interference, and can maintain stable performance in a complex power environment, thereby improving the accuracy of fault detection.

[0123] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0124] In one embodiment, a cable accessory fault determination device is provided, and the cable accessory fault determination device corresponds to the cable accessory fault determination method in the above embodiment. Figure 6 As shown, the cable accessory fault determination device includes a collection module 201, a construction module 202, a classification module 203 and a determination module 204. The detailed description of each functional module is as follows:

[0125] The acquisition module 201 is used to acquire the sound signal of the cable accessories during operation;

[0126] A construction module 202, configured to construct a sound feature matrix based on the sound signal;

[0127] A classification module 203 is used to classify the sound feature matrix using a pre-trained classification model to obtain a fault classification result;

[0128] The determination module 204 is used to determine the fault type of the cable accessory according to the fault classification result and generate fault warning information.

[0129] In this embodiment, the sound signal of the cable accessories during operation is collected; a sound feature matrix is ​​constructed based on the sound signal; a pre-trained classification model is used to classify the sound feature matrix to obtain a fault classification result; the fault type of the cable accessories is determined according to the fault classification result, and fault warning information is generated. In the cable accessory fault determination scheme provided in the present application, the sound signal of the cable accessories during operation is analyzed to determine the fault type of the cable accessories, and fault warning information is generated. The cable accessories during operation can be monitored and diagnosed in real time, and faults can be quickly identified, fault warnings can be provided, and the safe operation of the power system can be ensured. It has the advantages of high sensitivity and non-contact detection, avoiding physical contact and potential damage to the cable accessories, and has strong anti-interference ability, is insensitive to electromagnetic interference, and can maintain stable performance in a complex power environment, thereby improving the accuracy of fault detection.

[0130] Optionally, in some embodiments of the present application, the building blocks include:

[0131] An extraction submodule, used for extracting features of the sound signal using Mel-frequency cepstral coefficients to obtain multi-frame feature vectors;

[0132] A calculation submodule, used for calculating the scatter entropy value of the feature vector of each frame based on the scatter entropy;

[0133] A construction submodule is used to construct a sound feature matrix based on each of the feature vectors and each of the spread entropy values.

[0134] Optionally, in some embodiments of the present application, the cable accessory fault determination device further includes a dimensionality reduction processing module, and the dimensionality reduction processing module is specifically used to:

[0135] A dimension reduction process is performed on each of the feature vectors.

[0136] Optionally, in some embodiments of the present application, the dimension reduction processing module includes:

[0137] The dimensionality reduction processing submodule is used to perform dimensionality reduction processing on the sound feature matrix based on the principal component analysis method.

[0138] Optionally, in some embodiments of the present application, the cable accessory fault determination module further includes a preprocessing module, and the preprocessing module is specifically used to:

[0139] The sound signal is preprocessed, wherein the preprocessing operation includes filtering and denoising.

[0140] Optionally, in some embodiments of the present application, the cable accessory fault determination device further includes a training module, and the training module is specifically used to:

[0141] Acquire training data corresponding to the classification model to be trained, wherein the training data is historical sound signal data;

[0142] Dividing the training data into a training set and a test set according to a preset ratio;

[0143] The classification model to be trained is iteratively trained based on the training set and the classification model after the iterative training is verified using the test set to obtain the pre-trained classification model.

[0144] Optionally, in some embodiments of the present application, the training module includes:

[0145] A determination submodule, used to determine the model parameters of the classification model to be trained based on a bat optimization algorithm;

[0146] The training submodule is used to iteratively train the classification model to be trained based on the training set and the initial model parameters and optimize the model parameters, and use the test set to verify the classification model after the iterative training to obtain the pre-trained classification model.

[0147] In one embodiment, an electronic device is provided, the internal structure diagram of the electronic device can be as follows: Figure 7 As shown. The electronic device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external server through a network connection. When the computer program is executed by the processor, the functions or steps of a cable accessory fault determination method are implemented.

[0148] In one embodiment, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0149] Collect the sound signals of cable accessories during operation; construct a sound feature matrix based on the sound signals; use a pre-trained classification model to classify the sound feature matrix to obtain a fault classification result; determine the fault type of the cable accessories based on the fault classification result, and generate fault warning information.

[0150] In this embodiment, by analyzing the sound signals of the cable accessories during operation, the fault type of the cable accessories is determined and fault warning information is generated. The cable accessories during operation can be monitored and diagnosed in real time, and faults can be quickly identified. Fault warnings can be provided to ensure the safe operation of the power system. It has the advantages of high sensitivity and non-contact detection, avoiding physical contact and potential damage to the cable accessories, and has strong anti-interference ability and is insensitive to electromagnetic interference. It can maintain stable performance in a complex power environment, thereby improving the accuracy of fault detection.

[0151] In one embodiment, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0152] Collect the sound signals of cable accessories during operation; construct a sound feature matrix based on the sound signals; use a pre-trained classification model to classify the sound feature matrix to obtain a fault classification result; determine the fault type of the cable accessories based on the fault classification result, and generate fault warning information.

[0153] In this embodiment, by analyzing the sound signals of the cable accessories during operation, the fault type of the cable accessories is determined and fault warning information is generated. The cable accessories during operation can be monitored and diagnosed in real time, and faults can be quickly identified. Fault warnings can be provided to ensure the safe operation of the power system. It has the advantages of high sensitivity and non-contact detection, avoiding physical contact and potential damage to the cable accessories, and has strong anti-interference ability and is insensitive to electromagnetic interference. It can maintain stable performance in a complex power environment, thereby improving the accuracy of fault detection.

[0154] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or electronic device can refer to the relevant descriptions on the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0155] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0156] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0157] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A method for determining a cable accessory fault, characterized in that: include: Collect the sound signals of cable accessories during operation; constructing a sound feature matrix based on the sound signal; Using a pre-trained classification model to classify the sound feature matrix to obtain a fault classification result; The fault type of the cable accessory is determined according to the fault classification result, and fault warning information is generated.

2. The cable accessory fault determination method according to claim 1, characterized in that: The step of constructing a sound feature matrix based on the sound signal comprises: Using Mel frequency cepstrum coefficients to extract features of the sound signal to obtain multi-frame feature vectors; Calculate the scatter entropy value of the feature vector of each frame based on the scatter entropy; A sound feature matrix is ​​constructed based on each of the feature vectors and each of the spread entropy values.

3. The cable accessory fault determination method according to claim 2, characterized in that: Before calculating the scatter entropy value of the feature vector of each frame based on the scatter entropy, the method further includes: Perform dimensionality reduction processing on each of the feature vectors.

4. The cable accessory fault determination method according to claim 3, characterized in that: The dimensionality reduction processing of each of the feature vectors includes: The sound feature matrix is ​​subjected to dimensionality reduction processing based on a principal component analysis method.

5. The cable accessory fault determination method according to any one of claims 1 to 4, characterized in that: Before constructing the sound feature matrix based on the sound signal, the method further includes: The sound signal is preprocessed, wherein the preprocessing operation includes filtering and denoising.

6. The cable accessory fault determination method according to any one of claims 1 to 4, characterized in that: Before the method of using the pre-trained classification model to classify the sound feature matrix and obtain the fault classification result, the method further includes: Acquire training data corresponding to the classification model to be trained, wherein the training data is historical sound signal data; Dividing the training data into a training set and a test set according to a preset ratio; The classification model to be trained is iteratively trained based on the training set and the classification model after the iterative training is verified using the test set to obtain the pre-trained classification model.

7. The cable accessory fault determination method according to claim 6, characterized in that: The iterative training of the classification model to be trained based on the training set and the use of the test set to verify the iteratively trained classification model to obtain the pre-trained classification model includes: Determining model parameters of the classification model to be trained based on a bat optimization algorithm; The classification model to be trained is iteratively trained based on the training set and the initial model parameters and the model parameters are optimized, and the classification model after the iterative training is verified using the test set to obtain the pre-trained classification model.

8. A cable accessory fault determination device, characterized in that: include: A collection module is used to collect sound signals of cable accessories during operation; A construction module, used for constructing a sound feature matrix based on the sound signal; A classification module, used to classify the sound feature matrix using a pre-trained classification model to obtain a fault classification result; A determination module is used to determine the fault type of the cable accessory according to the fault classification result and generate fault warning information.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the cable accessory fault determination method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the cable accessory fault determination method according to any one of claims 1 to 7 are implemented.

Images