Fault arc detection method, device and equipment and storage medium

By combining current and sound signals through wavelet packet decomposition and a deep residual shrinkage network, the method improves arc fault detection accuracy in power systems by leveraging multi-source signal monitoring and intelligent analysis.

CN119064731APending Publication Date: 2024-12-03UNIV OF SCI & TECH OF CHINA
View PDF 0 Cites 4 Cited by

Patent Information

Application Number
CN202411250683.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-06
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Current arc fault detection technologies in power systems are limited by electrical noise interference, leading to reduced accuracy and difficulty in detecting arc faults, especially in complex power systems with varying frequencies and atypical conditions.

Method used

A method that combines current and sound signals using wavelet packet decomposition to extract feature vectors, fuses them into a joint feature vector, and utilizes a deep residual shrinkage network model for intelligent analysis to improve detection accuracy.

Benefits of technology

Enhances the accuracy of arc fault detection by integrating current and sound signals, providing a comprehensive data basis for precise identification of potential arc faults in low-voltage distribution systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119064731A_ABST
    Figure CN119064731A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of electric power safety monitoring, in particular to a fault arc detection method, device and equipment and a storage medium, the method is applied to a low-voltage power distribution system, the low-voltage power distribution system is electrically connected with a load, and the method comprises the steps that a current signal and a sound signal of the low-voltage power distribution system at the current moment are acquired; performing wavelet packet decomposition according to the current signal and the sound signal to obtain a current feature vector of the current signal and a sound feature vector of the sound signal, and determining a joint feature vector according to the current feature vector and the sound feature vector; and determining a fault arc detection result of the low-voltage power distribution system according to a preset fault arc detection model and the joint feature vector. According to the invention, the accuracy of fault arc detection is obviously improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of power safety monitoring, and in particular to a fault arc detection method, device, equipment and storage medium. Background Art

[0002] With the rapid development of the power industry and the popularization of smart grids, the safe and stable operation of power systems has become increasingly important. Arc faults are one of the common safety hazards in power systems. Their rapid and accurate detection and location are of great significance for preventing fires, protecting equipment, and ensuring power supply reliability.

[0003] However, the current mainstream arc fault detection technology mainly relies on monitoring the changes in current and voltage signals. Although this method has achieved preliminary identification of arc faults to a certain extent, it faces many limitations and challenges in practical applications. Specifically, during the operation of the power system, it will inevitably be affected by electrical noise generated by various factors such as the internal equipment, the external environment, and operational activities. These noise signals are often similar to the signal characteristics generated by arc faults, which can easily introduce interference during the detection process, resulting in reduced accuracy of the detection results.

[0004] Therefore, how to improve the accuracy of fault arc detection is a technical problem that needs to be solved urgently. Summary of the invention

[0005] The main purpose of the present application is to provide a method, device, equipment and storage medium for fault arc detection, aiming to improve the accuracy of fault arc detection.

[0006] To achieve the above object, the present application provides a fault arc detection method, which is applied to a low-voltage power distribution system, the low-voltage power distribution system is electrically connected to a load, and the fault arc detection method includes:

[0007] Obtaining the current signal and sound signal of the low-voltage power distribution system at the current moment;

[0008] Performing wavelet packet decomposition on the current signal and the sound signal to obtain a current feature vector of the current signal and a sound feature vector of the sound signal, and determining a joint feature vector based on the current feature vector and the sound feature vector;

[0009] The arc fault detection result of the low-voltage power distribution system is determined according to a preset arc fault detection model and the joint feature vector.

[0010] In one embodiment, the step of performing wavelet packet decomposition on the current signal and the sound signal to obtain a current feature vector of the current signal and a sound feature vector of the sound signal comprises:

[0011] The current signal or the sound signal is used as the signal to be decomposed, and the signal to be decomposed is subjected to wavelet packet decomposition according to a preset energy feature algorithm to obtain a signal energy feature of the signal to be decomposed, wherein the energy feature is the current energy feature of the current signal or the sound energy feature of the sound signal, and the energy feature algorithm is:

[0012]

[0013] in, is the coefficient of the kth frequency band of the jth layer at the nth sampling point after the wavelet packet decomposition of the signal to be decomposed, is the signal energy feature of the kth frequency band of the jth layer after the wavelet packet decomposition of the signal to be decomposed, is the kth frequency band coefficient of the jth layer after the wavelet packet decomposition of the signal to be decomposed;

[0014] Determine the sampling frequency of each sampling point and the total number of all the sampling points, and perform frequency feature processing on the signal to be decomposed according to the sampling frequency and the total number to obtain the signal frequency feature of the signal to be decomposed, wherein the signal frequency feature is the current frequency feature of the current signal or the sound frequency feature of the sound signal;

[0015] The current energy feature and the current frequency feature are used as a current feature vector, and the sound energy feature and the sound frequency feature are used as a sound feature vector.

[0016] In one embodiment, the step of determining a joint feature vector based on the current feature vector and the sound feature vector comprises:

[0017] The current feature vector and the sound feature vector are spliced ​​according to a preset vector splicing algorithm to obtain a spliced ​​feature vector. The vector splicing algorithm is:

[0018]

[0019] Among them, F fused is the concatenated feature vector, F i is the current eigenvector, F s is the sound feature vector;

[0020] Determine a feature selection matrix corresponding to the concatenated feature vectors, and perform weighted processing on each important arc feature in the feature selection matrix to obtain a weighted matrix;

[0021] The product of the concatenated feature vector and the feature selection matrix is ​​multiplied by the weighting matrix to obtain a joint feature vector.

[0022] In one embodiment, the step of determining the arc fault detection result of the low voltage power distribution system according to the preset arc fault detection model and the joint feature vector comprises:

[0023] Inputting the joint feature vector into a preset arc fault detection model;

[0024] After determining that the convolution layer in the fault arc detection model receives the joint feature vector, the convolution layer performs a convolution operation on the joint feature vector in a preset residual shrinkage module to obtain an initial arc feature;

[0025] The fault arc detection result of the low-voltage power distribution system is determined according to the initial characteristics of the arc.

[0026] In one embodiment, the step of determining the fault arc detection result of the low-voltage power distribution system according to the initial characteristics of the arc includes:

[0027] The arc initial features are transmitted from the convolutional layer to the fully connected layer in the fault arc detection model, and the arc initial features are fully connected according to the fully connected algorithm corresponding to the fully connected layer to obtain arc splicing features, and the fully connected algorithm is:

[0028] F fc =σ(W fc F res +b fc )

[0029] Among them, F fc is the arc splicing feature, σ is the activation function of the fault arc detection model, W fc is the weight matrix of the fully connected layer, F res is the initial characteristic of the arc, b fc is the bias term of the fully connected layer;

[0030] The fault arc detection result of the low-voltage power distribution system is determined according to the arc splicing characteristics.

[0031] In one embodiment, the step of determining the fault arc detection result of the low-voltage power distribution system according to the arc splicing feature includes:

[0032] The arc initial features are transmitted from the fully connected layer to the output layer in the fault arc detection model, and the arc splicing features are subjected to fault detection according to the fault detection algorithm corresponding to the output layer to obtain the fault arc detection result of the low-voltage power distribution system, wherein the fault detection algorithm is:

[0033]

[0034] in, is the fault arc detection result of the low-voltage power distribution system, softmax is the probability function of the fault arc occurring in the low-voltage power distribution system, W out is the weight matrix of the output layer, F fc is the arc splicing feature, b out is the bias term of the output layer.

[0035] In one embodiment, after the step of determining the arc fault detection result of the low voltage power distribution system according to the preset arc fault detection model and the joint feature vector, the arc fault detection method includes:

[0036] Detecting whether a result threshold of the arc fault detection result exceeds a preset arc fault threshold;

[0037] If so, the number of arc half-cycles when the low-voltage power distribution system supplies power to the load within the preset counting time is accumulated, until it is determined that the accumulated number of arc half-cycles matches the preset arc count number, then it is determined that a fault arc occurs in the low-voltage power distribution system;

[0038] If not, the next moment after the current moment is taken as the next current moment, and the process returns to the step of obtaining the current signal and the sound signal of the low-voltage power distribution system at the current moment.

[0039] In addition, to achieve the above-mentioned purpose, the present application also provides a fault arc detection device, the fault arc detection device comprising:

[0040] An acquisition module, used to acquire the current signal and the sound signal of the low-voltage power distribution system at the current moment;

[0041] a decomposition module, configured to perform wavelet packet decomposition on the current signal and the sound signal to obtain a current feature vector of the current signal and a sound feature vector of the sound signal, and determine a joint feature vector based on the current feature vector and the sound feature vector;

[0042] The detection module is used to determine the arc fault detection result of the low-voltage power distribution system according to a preset arc fault detection model and the joint feature vector.

[0043] Each functional module of the arc fault detection device of the present application implements the steps of the arc fault detection method of the present application as described above during operation.

[0044] In addition, to achieve the above-mentioned purpose, the present application also provides a device, which includes a memory, a processor, and a fault arc detection program stored in the memory and executable on the processor, and the fault arc detection program implements the steps of the above-mentioned fault arc detection method when executed by the processor.

[0045] In addition, to achieve the above-mentioned purpose, the present application also provides a storage medium, which is a computer-readable storage medium, and a fault arc detection program is stored on the computer-readable storage medium. When the fault arc detection program is executed by the processor, the steps of the above-mentioned fault arc detection method are implemented.

[0046] The present application provides a fault arc detection method that optimizes the traditional arc fault detection technology in a targeted manner, significantly improving the accuracy of fault arc detection. Specifically, the fault arc detection method set in the present application can timely obtain the current signal and sound signal of the low-voltage distribution system at the current moment, thereby realizing multi-source signal monitoring when the low-voltage distribution system is powered, and providing rich and reliable multi-source signal support for subsequent arc fault detection; next, the wavelet packet decomposition technology is used to extract highly representative current feature vectors and sound feature vectors from the current signal and the sound signal respectively; then, the complementarity of the current mutation and the accompanying acoustic characteristics when the arc occurs is fully utilized, and the current feature vector and the sound feature vector are fused into a joint feature vector, thereby further enhancing the dimension and accuracy of fault arc identification. Finally, based on the joint feature vector, the preset fault arc detection model is used for intelligent analysis, and the fault arc detection results of the low-voltage distribution system can be accurately obtained, so that the potential fault arc in the low-voltage distribution system can be more accurately identified, significantly improving the accuracy of fault arc detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flow chart of the first embodiment of the fault arc detection method of the present application;

[0048] Figure 2 It is a schematic diagram of an arc experiment platform involved in an embodiment of a fault arc detection method of the present application;

[0049] Figure 3 It is a comparison diagram of the current waveform of an incandescent lamp when there is a fault arc and when there is no fault arc involved in the embodiment of the present application;

[0050] Figure 4 It is a comparison diagram of sound signals when there is a fault arc and when there is no fault arc involved in the embodiment of the present application;

[0051] Figure 5 It is the original audio waveform diagram involved in the embodiment of the present application and the comparison diagram after preprocessing;

[0052] Figure 6 It is another schematic diagram of a polymerization operation involved in the embodiment of the present application;

[0053] Figure 7 It is a flow chart of the arc fault detection strategy involved in the embodiment of the present application;

[0054] Figure 8 It is a visualized confusion matrix diagram of the fault arc detection result involved in the embodiment of the present application;

[0055] Fig. 9 It is a structural schematic diagram of a fault arc detection device involved in an embodiment of the present application;

[0056] Fig.10 This is a schematic diagram of the structure of the arc fault detection device involved in the embodiment of the present application.

[0057] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0058] The present application embodiment provides a fault arc detection method, referring to Figure 1 As shown, Figure 1 It is a flow chart of the first embodiment of the fault arc detection method of the present application.

[0059] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0060] With the development of social economy, fire incidents occur frequently, among which the number of fires caused by electricity accounts for about 30% of all fires. In the low-voltage power distribution system, when the equipment is aged, the insulation layer is damaged, the electrical line connection is loose, and the connection point is poorly contacted, a fault arc will be generated. However, the series fault arc in the low-voltage power distribution system is highly concealed, and the current value is often less than the threshold of the traditional circuit breaker, so it cannot be effectively identified. When the arc continues to burn, the temperature gradually rises to thousands of degrees, which is enough to ignite the surrounding combustibles, and then cause electrical fires, seriously threatening the safety of life and property of people. Therefore, studying an efficient AC arc fault detection method is of great significance to ensure electrical safety.

[0061] The current mainstream arc fault detection technology mainly relies on monitoring the changes in current and voltage signals. The limitation of this method is that it is affected by electrical noise and system load changes, which reduces the accuracy of detection. Especially in large-scale power systems, accurate detection of signals becomes more difficult. At the same time, some existing technologies are limited to a specific frequency range and have difficulty in effectively detecting changing frequencies or atypical arc faults. This makes these technologies perform poorly when dealing with different frequency faults in complex power systems. Traditional arc fault detection technology has limited adaptability under these conditions, which may lead to missed detection or false alarms. However, traditional arc fault detection technology ignores the sound signals generated by the arc, which is a potentially important source of information. The sound characteristics of arc faults can provide an additional detection dimension, but current technologies fail to fully utilize this potential advantage.

[0062] Therefore, in order to solve the above-mentioned defects of the traditional arc fault detection technology, the present application provides a fault arc detection method, device, equipment and storage medium.

[0063] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, a database system, etc., or a device capable of realizing the above functions, such as a fault arc detection device, etc. The following takes the fault arc detection device as an example to illustrate this embodiment and the following embodiments.

[0064] The arc fault detection method of the present application includes the following implementation steps S10 to S30.

[0065] Step S10: Acquire the current signal and sound signal of the low-voltage power distribution system at the current moment.

[0066] In this embodiment, the present application is provided with an acoustic signal acquisition device and a current transformer on the circuit in which the low-voltage power distribution system supplies power to the load. When the low-voltage power distribution system supplies power to the load, the current transformer is triggered to collect the original current signal of the low-voltage power distribution system at the current moment, and the acoustic signal acquisition device is synchronously triggered to collect the original sound signal of the low-voltage power distribution system at the current moment. That is to say, when the low-voltage power distribution system supplies power to the load, the current sensor is triggered to synchronize the clock signal with the acoustic signal acquisition device (i.e., the sound sensor) to ensure that the current sensor and the acoustic signal acquisition device are collected at the same time, and at the same time, a timestamp mark is introduced when the sensor collects the signal, the exact collection time of each sampling point is recorded, and stored in the metadata of the data file, further ensuring the synchronization of the collected electrical original signal and the original sound signal; subsequently, the original current signal and the original sound signal are preprocessed, and the preprocessed original current signal (i.e., the current signal) and the preprocessed original sound signal (i.e., the sound signal) can be accurately obtained.

[0067] It should be noted that the signal preprocessing at least includes removing the DC component, high-frequency noise filtering and signal normalization.

[0068] Exemplarily, removing the DC component means eliminating the DC component by subtracting the average value of the sensor signal. Specifically, assuming that the collected sensor signal is x(t), T is the total signal duration of the sensor signal, and the algorithm corresponding to removing the DC component is:

[0069]

[0070] Among them, x dc (t) represents the sensor signal after the DC component is removed (i.e., the signal to be filtered), x(t) is the sensor signal, x(t)dt is the average value of the sensor signal, and the sensor signal is an original electrical signal or an original sound signal.

[0071] High-frequency noise filtering refers to removing high-frequency noise from the signal to be filtered by a low-pass filter, and the transfer function of the low-pass filter can be expressed as:

[0072]

[0073] Where f is the input signal frequency of the signal to be filtered, f c is the cutoff frequency of the low-pass filter, H(f) is the low-frequency component of the signal to be filtered, and m is the order of the low-pass filter. In addition, the transfer function of the low-pass filter shows that as the frequency of the input signal increases, the gain of the filter gradually decreases. c Before, the filter gain was close to 1; at the cutoff frequency f cAfter that, the gain of the filter decreases rapidly and eventually tends to 0. This means that by selecting a suitable cutoff frequency, the low-pass filter can attenuate or filter out the high-frequency components of the input signal, thereby retaining or amplifying the low-frequency components.

[0074] Signal normalization refers to normalizing the sensor signal after filtering (i.e., the signal to be filtered that contains only low-frequency components) through a normalization algorithm, so that the amplitude of the sensor signal after filtering is within the range of [0,1]. The normalization algorithm can be expressed as:

[0075]

[0076] Among them, x norm (t) is the sensor signal after preprocessing, and the preprocessed sensor signal is a current signal or a sound signal; x′(t) is the sensor signal after filtering.

[0077] In the specific embodiment, please refer to Figure 2 , Figure 2 It is a schematic diagram of an arc experiment platform involved in an embodiment of the arc fault detection method of the present application. The low-voltage power distribution system provided in the present application may include a 220V 50Hz AC power supply. Specifically, the arc experiment platform provided in the present application is connected to a load by a 220V 50Hz AC power supply, an air switch, an acoustic signal acquisition device, a current transformer (i.e., a current sensor), and a fault arc generation module. Among them, the 220V 50Hz AC power supply is used to power the load, the air switch is used for circuit protection, the current sensor and the acoustic signal acquisition device are used to collect the original current signal and the original sound signal when the low-voltage power distribution system supplies power to the load, and the fault arc generation module is used to simulate a fault arc, and the fault arc generation module can be understood as a device for simulating the generation of a fault arc provided in the present application. For example, please refer to Figure 3 , Figure 3 is a comparison diagram of the current waveform of an incandescent lamp when there is a fault arc and when there is no fault arc, according to the embodiment of the present application, wherein: Figure 3 The horizontal axis shows the number of sampling points, and the vertical axis shows the corresponding amplitude. Figure 3 (a) is the current signal waveform collected when the low-voltage distribution system supplies power to the load under normal circumstances. Figure 3 (b) is the current signal waveform collected when a fault arc occurs when the low-voltage distribution system supplies power to the load.

[0078] Please refer to Figure 4 , Figure 4 It is a comparison diagram of sound signals when there is a fault arc and when there is no fault arc involved in the embodiment scheme of the present application. Figure 4 (a) is the waveform of the sound signal when there is no fault arc. Figure 4(b) is the waveform of the sound signal when there is no fault arc. Figure 4 (a) and Figure 4 (b) Under the same time length, the amplitude of the load in normal state is stable below 0.05, while the characteristic of the fault arc becomes dozens of times larger. In addition, please refer to Figure 5 , Figure 5 It is the original audio waveform diagram involved in the embodiment scheme of the present application and the comparison diagram after preprocessing.

[0079] Step S20: performing wavelet packet decomposition on the current signal and the sound signal to obtain a current feature vector of the current signal and a sound feature vector of the sound signal, and determining a joint feature vector based on the current feature vector and the sound feature vector.

[0080] In this embodiment, the present application implements the wavelet packet decomposition technology on the current signal and the sound signal synchronously, so that it can accurately extract highly representative feature vectors from these two dimensions (i.e., the current dimension and the sound position) - that is, the current feature vector and the sound feature vector, which not only retains the key information in each signal, but also realizes the effective analysis of the multi-level and multi-frequency characteristics of complex signals. Subsequently, the current feature vector and the sound feature vector are feature fused to form a joint feature vector, which integrates the characteristics of the current and sound signals, providing a more comprehensive and accurate data basis for subsequent fault arc detection, and significantly improving the accuracy and reliability of fault arc detection.

[0081] Step S30: determining the arc fault detection result of the low voltage power distribution system according to the preset arc fault detection model and the joint feature vector.

[0082] In this embodiment, based on the joint feature vector, intelligent analysis is performed using a preset arc fault detection model to accurately obtain the arc fault detection results of the low-voltage power distribution system, thereby more accurately identifying potential arc faults in the low-voltage power distribution system, significantly improving the accuracy of arc fault detection.

[0083] It should be noted that the arc fault detection model set in this application involves the following Figure 6 As shown, Figure 6 It is a schematic diagram of the structure of the fault arc detection model involved in the embodiment of the present application. Figure 6The Conv shown in the figure is the convolution layer in the fault arc detection model, the moving step size of the convolution kernel in the convolution layer is set to 2, the number of channels of the output feature map in the convolution layer is K=A, DRSBU is the residual shrinkage module set in the fault arc detection model, BN is the batch normalization set in the fault arc detection model, ReLU is the activation function set in the fault arc detection model, GAP is the global mean pooling layer set in the fault arc detection model, and FC is the fully connected output layer in the fault arc detection model.

[0084] In summary, the present application provides a fault arc detection method that optimizes the traditional arc fault detection technology in a targeted manner, significantly improving the accuracy of fault arc detection. Specifically, the fault arc detection method provided in the present application can timely obtain the current signal and sound signal of the low-voltage distribution system at the current moment, thereby realizing multi-source signal monitoring when the low-voltage distribution system is powered on, and providing rich and reliable multi-source signal support for subsequent arc fault detection; next, the wavelet packet decomposition technology is used to extract highly representative current feature vectors and sound feature vectors from the current signal and the sound signal respectively; then, the complementarity of the current mutation and the accompanying acoustic characteristics when the arc occurs is fully utilized, and the current feature vector and the sound feature vector are fused into a joint feature vector, thereby further enhancing the dimension and accuracy of fault arc identification. Finally, based on the joint feature vector, the preset fault arc detection model is used for intelligent analysis, and the fault arc detection results of the low-voltage distribution system can be accurately obtained, so that the potential fault arc in the low-voltage distribution system can be more accurately identified, significantly improving the accuracy of fault arc detection.

[0085] Further, based on the first embodiment of the arc fault detection method of the present application, a second embodiment of the arc fault detection method of the present application is proposed. In some feasible embodiments, the above step S20: performing wavelet packet decomposition based on the current signal and the sound signal to obtain the current feature vector of the current signal and the sound feature vector of the sound signal, also includes the following implementation steps S201 to S203.

[0086] Step S201: taking the current signal or the sound signal as the signal to be decomposed, and performing wavelet packet decomposition on the signal to be decomposed according to a preset energy feature algorithm to obtain a signal energy feature of the signal to be decomposed, wherein the energy feature is the current energy feature of the current signal or the sound energy feature of the sound signal, and the energy feature algorithm is:

[0087]

[0088] in, is the coefficient of the kth frequency band of the jth layer at the nth sampling point after the wavelet packet decomposition of the signal to be decomposed, is the signal energy feature of the kth frequency band of the jth layer after the wavelet packet decomposition of the signal to be decomposed, is the kth frequency band coefficient of the jth layer after the wavelet packet decomposition of the signal to be decomposed.

[0089] In this embodiment, after the preprocessed current initial signal and the preprocessed sound initial signal are subjected to wavelet packet decomposition (i.e., after the current signal and the sound signal are subjected to wavelet packet decomposition), the current energy characteristics and the sound energy characteristics can be respectively extracted from the current signal and the sound signal after the wavelet packet decomposition according to a preset energy characteristic algorithm, thereby providing an accurate and reliable characteristic data basis for the subsequent fault arc identification.

[0090] It should be noted that the current energy feature is used to describe the energy distribution of the current signal in each frequency band, and the sound energy feature is used to describe the energy distribution of the sound signal in each frequency band.

[0091] Step S202: Determine the sampling frequency of each of the sampling points and the total number of all the sampling points, and perform frequency feature processing on the signal to be decomposed according to the sampling frequency and the total number to obtain the signal frequency feature of the signal to be decomposed, wherein the signal frequency feature is the current frequency feature of the current signal or the sound frequency feature of the sound signal.

[0092] In this embodiment, after performing wavelet packet decomposition on the current signal and the sound signal, the sampling frequency of the sampling points corresponding to each current signal and the sound signal, as well as the total number of all sampling points can also be determined; then, the frequency characteristics of the current signal and the sound signal are processed according to the sampling frequency and the total number, respectively, so that the current frequency characteristics of the current signal and the sound frequency characteristics of the sound signal can be accurately obtained.

[0093] It should be noted that the current frequency characteristic is used to describe the frequency components of the current signal in each frequency band, and the sound frequency characteristic is used to describe the frequency components of the sound signal in each frequency band. The current frequency characteristic is the center frequency and bandwidth of the current signal in each frequency band, and the sound frequency characteristic is the center frequency and bandwidth of the sound signal in each frequency band. In addition, the calculation algorithm of the center frequency can be expressed as:

[0094]

[0095] in, represents the center frequency of the current signal or sound signal in each frequency band, that is, the center frequency of the kth frequency band of the jth layer after the wavelet packet decomposition of the current signal or sound signal. N is the total number of all sampling points, and f[n] represents the sampling frequency of the nth sampling point. The coefficient of the kth frequency band of the jth layer at the nth sampling point after the wavelet packet decomposition of the current signal or sound signal.

[0096] The bandwidth calculation algorithm can be expressed as:

[0097]

[0098] in, represents the bandwidth of the current signal or sound signal in each frequency band, that is, the bandwidth of the kth frequency band of the jth layer after the wavelet packet decomposition of the current signal or sound signal. N is the total number of all sampling points, and f[n] represents the sampling frequency of the nth sampling point. It represents the center frequency of the kth frequency band of the jth layer after the wavelet packet decomposition of the current signal or sound signal. The coefficient of the kth frequency band of the jth layer at the nth sampling point after the wavelet packet decomposition of the current signal or sound signal.

[0099] Step S203: taking the current energy feature and the current frequency feature as a current feature vector, and taking the sound energy feature and the sound frequency feature as a sound feature vector.

[0100] In this embodiment, the present application performs feature fusion on the current energy characteristics and current frequency characteristics of different modes in the current signal to form a current feature vector; and performs feature fusion on the current energy characteristics and sound frequency characteristics of different modes in the sound signal to form a sound feature vector.

[0101] Furthermore, in some feasible embodiments, the above step S20: determining a joint feature vector according to the current feature vector and the sound feature vector, further includes the following implementation steps A10 to A30:

[0102] Step A10: splicing the current feature vector and the sound feature vector according to a preset vector splicing algorithm to obtain a spliced ​​feature vector, wherein the vector splicing algorithm is:

[0103]

[0104] Among them, F fused is the concatenated feature vector, F i is the current eigenvector, F s is the sound feature vector.

[0105] In this embodiment, the current characteristic vector F i and the sound feature vector F s Splice to form the spliced ​​feature vector in, Represents the feature concatenation operation, which concatenates two vectors by column to form a new feature vector.

[0106] Step A20: determining a feature selection matrix corresponding to the concatenated feature vectors, and performing weighted processing on each important arc feature in the feature selection matrix to obtain a weighted matrix;

[0107] Step A30: multiply the product of the concatenated feature vector and the feature selection matrix by the weighting matrix to obtain a joint feature vector.

[0108] In this embodiment, in order to improve the effectiveness of the fusion features, the feature selection method principal component analysis PCA is used to select the first k important arc features from the spliced ​​feature vector, and then a feature selection matrix is ​​constructed based on these k important arc features, and these important arc features are weighted to form a k-dimensional diagonal weighted matrix (i.e., a weighted matrix) to obtain the final joint feature vector.

[0109] It should be noted that the calculation algorithm of the joint eigenvector can be expressed as:

[0110] F selected =W*(S*Ff used )

[0111] Among them, F selected represents the joint eigenvector, W represents the k-dimensional diagonal weighted matrix, and S represents the feature selection matrix.

[0112] Furthermore, in some feasible embodiments, the above step S30: determining the arc fault detection result of the low voltage power distribution system according to the preset arc fault detection model and the joint feature vector, also includes the following implementation steps S301 to S303.

[0113] Step S301: inputting the joint feature vector into a preset arc fault detection model.

[0114] In this embodiment, the fault arc detection model preset in the present application includes at least an input layer, a convolution layer, a residual shrinkage module, a fully connected layer and an output layer. The input layer set in the present application responds to the input joint feature vector and outputs the joint feature vector to the convolution layer in the fault arc detection model.

[0115] It should be noted that the fault arc detection model preset in this application can be understood as a deep residual shrinkage network model. This application trains the joint feature vector and identifies faults through the deep residual shrinkage network model. The deep residual shrinkage network model is a deep learning model that combines the residual network and the attention mechanism, which can reduce the impact of noise on feature extraction while maintaining the depth of the model. Its main feature is to effectively suppress noise through residual connections and shrinkage functions, and improve the accuracy of fault detection.

[0116] Step S302: after determining that the convolution layer in the fault arc detection model receives the joint feature vector, the convolution layer performs a convolution operation on the joint feature vector in a preset residual shrinkage module to obtain an initial arc feature.

[0117] In this embodiment, after determining that the convolution layer in the fault arc detection model receives the joint feature vector, the convolution layer performs a convolution operation on the joint feature vector in a preset residual shrinkage module to obtain the initial arc feature.

[0118] In a specific embodiment, there are multiple convolutional layers, and each convolutional layer is arranged in sequence. The convolutional layer with the smallest sequence number is used as the target convolutional layer. A convolution operation is performed on the joint feature vector received by the input layer based on the target convolutional layer to obtain an initial feature vector output by the target convolutional layer. The next convolutional layer of the target convolutional layer is used as the next target convolutional layer, and the initial feature vector is returned as the next joint feature vector to perform a step of performing a convolution operation on the joint feature vector received by the input layer based on the target convolutional layer until all convolutional layers have completed the convolution operation, and an output feature map of the convolutional layer (i.e., the arc feature obtained by the convolution operation of each convolutional layer) is obtained; next, when the residual shrinkage module set in the convolutional layer receives the output feature map of the convolutional layer, residual processing is performed on the output feature map according to a preset residual algorithm to obtain the initial feature of the arc.

[0119] It should be noted that the algorithm of the convolution operation can be expressed as:

[0120] F conv =σ(W conv *F fused +b conv )

[0121] Among them, F conv is the output feature map of the convolutional layer; σ is the activation function, which is used to introduce nonlinearity. Common activation functions include ReLU, Sigmoid or Tanh functions; W conv is the weight matrix of the convolutional layer; F fused is the joint eigenvector; b conv is the bias term of the convolutional layer.

[0122] The default residual algorithm can be expressed as

[0123]

[0124] Among them, θ refers to the shrinkage threshold of the residual shrinkage module, which can be customized according to application requirements. W represents the residual function of the residual shrinkage module; res represents the weight matrix of the residual shrinkage module; F conv is the output feature map of the convolutional layer; F res Indicates the initial characteristics of the arc.

[0125] The initial characteristics of the arc can be understood as the initial characteristics of the arc generated when the low-voltage distribution system supplies power to the load, such as sudden changes in the current waveform, abnormal patterns of voltage drop, and sudden changes in the sound signal waveform.

[0126] Step S303: determining the fault arc detection result of the low-voltage power distribution system according to the initial characteristics of the arc.

[0127] In this embodiment, the arc fault detection result of the low-voltage power distribution system can be quickly and accurately obtained based on the initial characteristics of the arc.

[0128] Furthermore, in some other feasible embodiments, the above step S303: determining the fault arc detection result of the low-voltage power distribution system according to the initial characteristics of the arc, may also include the following implementation steps S3021 to S3022.

[0129] Step S3021: transferring the initial arc features from the convolutional layer to the fully connected layer in the fault arc detection model, and performing a fully connected operation on the initial arc features according to the fully connected algorithm corresponding to the fully connected layer to obtain arc splicing features, wherein the fully connected algorithm is:

[0130] F fc =σ(W fc F res +b fc )

[0131] Among them, F fc is the arc splicing feature, σ is the activation function of the fault arc detection model, W fc is the weight matrix of the fully connected layer, F res is the initial characteristic of the arc, b fc is the bias term of the fully connected layer.

[0132] Step S3022: Determine the fault arc detection result of the low-voltage power distribution system according to the arc splicing characteristics.

[0133] In this embodiment, the initial arc features are efficiently transferred from the convolutional layer to the fully connected layer in the fault arc detection model, and the fully connected algorithm unique to the fully connected layer is applied to deeply integrate and splice the initial arc features, so as to form a highly generalized and information-rich arc splicing feature, thereby significantly enhancing the ability of the fault arc detection model to capture arc fault features; next, based on the arc splicing feature, the fault arc detection result of the low-voltage power distribution system can be accurately improved.

[0134] Furthermore, in some feasible embodiments, the above step S3022: determining the fault arc detection result of the low-voltage power distribution system according to the arc splicing characteristics, may also include the following implementation step S30221.

[0135] Step S30221: transmitting the initial arc feature from the fully connected layer to the output layer in the fault arc detection model, and performing fault detection on the arc splicing feature according to the fault detection algorithm corresponding to the output layer to obtain the fault arc detection result of the low-voltage power distribution system, wherein the fault detection algorithm:

[0136]

[0137] in, is the fault arc detection result of the low-voltage power distribution system, softmax is the probability function of the fault arc occurring in the low-voltage power distribution system, W out is the weight matrix of the output layer, F fc is the arc splicing feature, b out is the bias term of the output layer.

[0138] In this embodiment, the arc splicing features obtained after deep processing of the arc initial features by the fully connected layer are further transmitted to the output layer of the fault arc detection model, and the arc splicing features are subjected to fault detection according to the fault detection algorithm corresponding to the output layer, so that the fault arc detection results can be accurately obtained, thereby ensuring the accuracy and reliability of the fault arc detection results.

[0139] Further, in some feasible embodiments, after the above step S30: determining the arc fault detection result of the low voltage power distribution system according to the preset arc fault detection model and the joint feature vector, the arc fault detection method may further include the following implementation steps B10 to B30.

[0140] Step B10: Detect whether a result threshold of the arc fault detection result exceeds a preset arc fault threshold.

[0141] In this embodiment, after the above steps S10 to S30, referring to Figure 7 , Figure 7 This is a flow chart of the arc fault detection strategy involved in the embodiment of the present application. After performing three-layer wavelet packet decomposition on the collected current data (i.e., the initial current signal and the initial sound signal), the current feature vector and the sound feature vector are extracted from the initial current signal and the initial sound signal after the wavelet packet decomposition, and the current feature vector and the sound feature vector are feature fused to obtain a joint feature vector; then the joint feature vector is used as the input of the preset arc fault detection model for model classification calculation to obtain a calculation result (i.e., the arc fault detection result); next, the result threshold of the arc fault detection result is compared with the preset arc fault threshold, thereby further improving the accuracy and reliability of arc fault detection.

[0142] It should be noted that the preset arc fault threshold of 0.5 is only one implementation of the present application, that is, the preset arc fault threshold can be customized according to application requirements, and the present application does not impose any limitation thereto.

[0143] Step B20: If so, accumulate the number of arc half-cycles when the low-voltage power distribution system supplies power to the load within the preset counting time, until it is determined that the accumulated number of arc half-cycles matches the preset arc count number, then determine that a fault arc occurs in the low-voltage power distribution system.

[0144] In this embodiment, referring to 7, if the result threshold of the fault arc detection result is greater than the preset arc fault threshold, the number of arc half-cycles when the low-voltage power distribution system supplies power to the load within the preset counting time is accumulated. If it is detected that the accumulated number of arc half-cycles within the preset counting time exceeds the preset arc count number (that is, the accumulated number of arc half-cycles matches the preset arc count number), it is determined that a fault arc has occurred in the low-voltage power distribution system.

[0145] Step B30: If not, then the next moment after the current moment is taken as the next current moment, and the process returns to the step of obtaining the current signal and the sound signal of the low-voltage power distribution system at the current moment.

[0146] In this embodiment, refer to Figure 7 If the result threshold of the arc fault detection result is less than or equal to the preset arc fault threshold, return to the step of obtaining the current signal and sound signal of the low-voltage power distribution system at the current moment.

[0147] It should be noted that, refer to Figure 8 , Figure 8This is a visualization confusion matrix diagram of the arc fault detection result involved in the embodiment of the present application. After preprocessing the initial current signal and the initial sound signal when the low-voltage power distribution supplies power to the load, the present application performs wavelet packet decomposition to obtain the wavelet coefficients of each layer, and then performs corresponding feature extraction and fusion to obtain the joint feature vector; next, the joint feature vector is input into the deep residual shrinkage network model for model classification calculation to obtain the arc fault detection result. And the arc fault detection result is referenced Figure 8 Experiments have shown that the joint feature vector of the present application has a good detection effect on series fault arcs in low-voltage power distribution systems after detection and classification by a deep residual shrinkage network model.

[0148] In summary, the present application provides a fault arc detection method that optimizes the traditional arc fault detection technology in a targeted manner, significantly improving the accuracy of fault arc detection. Specifically, the fault arc detection method provided in the present application can timely obtain the current signal and sound signal of the low-voltage distribution system at the current moment, thereby realizing multi-source signal monitoring when the low-voltage distribution system is powered on, and providing rich and reliable multi-source signal support for subsequent arc fault detection; next, the wavelet packet decomposition technology is used to extract highly representative current feature vectors and sound feature vectors from the current signal and the sound signal respectively; then, the complementarity of the current mutation and the accompanying acoustic characteristics when the arc occurs is fully utilized, and the current feature vector and the sound feature vector are fused into a joint feature vector, thereby further enhancing the dimension and accuracy of fault arc identification. Finally, based on the joint feature vector, the preset fault arc detection model is used for intelligent analysis, and the fault arc detection results of the low-voltage distribution system can be accurately obtained, so that the potential fault arc in the low-voltage distribution system can be more accurately identified, significantly improving the accuracy of fault arc detection.

[0149] In addition, the present application also provides a fault arc detection device, please refer to Fig. 9 , Fig. 9 Schematic diagram of the structure of the arc fault detection device involved in the embodiment of the present application. The arc fault detection device provided in the present application includes:

[0150] An acquisition module H01 is used to acquire the current signal and the sound signal of the low-voltage power distribution system at the current moment;

[0151] A decomposition module H02, configured to perform wavelet packet decomposition on the current signal and the sound signal to obtain a current feature vector of the current signal and a sound feature vector of the sound signal, and determine a joint feature vector based on the current feature vector and the sound feature vector;

[0152] The detection module H03 is used to determine the arc fault detection result of the low-voltage power distribution system according to a preset arc fault detection model and the joint feature vector.

[0153] In addition, the present application also provides a fault arc detection device. Fig.10 , Fig.10 The schematic diagram of the structure of the arc fault detection device involved in the embodiment of the present application is shown in FIG. The device in the embodiment of the present application may be a device for locally running the arc fault detection method.

[0154] The present application provides a fault arc detection device, which includes: at least one processor; and a memory that is communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the fault arc detection device method in the above-mentioned embodiment 1.

[0155] Reference below Fig.10 , which shows a schematic diagram of the structure of a fault arc detection device suitable for implementing the embodiment of the present application. The fault arc detection device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The arc fault detection device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0156] like Fig.10As shown, the arc fault detection device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the arc fault detection device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following devices can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the arc fault detection device to communicate with other devices wirelessly or by wire to exchange data. Although the arc fault detection device with various devices is shown in the figure, it should be understood that it is not required to implement or have all the devices shown. More or fewer devices can be implemented or have alternatively.

[0157] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0158] The arc fault detection device provided by the present application adopts the arc fault detection device method in the above embodiment, which can solve the technical problem of low reliability of the arc fault detection device. Compared with the prior art, the beneficial effects of the arc fault detection device provided by the present application are the same as the beneficial effects of the arc fault detection device method provided by the above embodiment, and other technical features in the arc fault detection device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0159] In addition, the present application provides a computer-readable storage medium on which a fault arc detection program is stored, and when the fault arc detection program is executed by a processor, the steps of the above-mentioned fault arc detection method are implemented.

[0160] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0161] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0162] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a computer-readable storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0163] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A fault arc detection method, characterized in that: The arc fault detection method is applied to a low voltage power distribution system, the low voltage power distribution system is electrically connected to a load, and the arc fault detection method comprises: Obtaining the current signal and sound signal of the low-voltage power distribution system at the current moment; Performing wavelet packet decomposition on the current signal and the sound signal to obtain a current feature vector of the current signal and a sound feature vector of the sound signal, and determining a joint feature vector based on the current feature vector and the sound feature vector; The arc fault detection result of the low-voltage power distribution system is determined according to a preset arc fault detection model and the joint feature vector.

2. The arc fault detection method according to claim 1, characterized in that: The step of performing wavelet packet decomposition on the current signal and the sound signal to obtain a current feature vector of the current signal and a sound feature vector of the sound signal comprises: The current signal or the sound signal is used as the signal to be decomposed, and the signal to be decomposed is subjected to wavelet packet decomposition according to a preset energy feature algorithm to obtain a signal energy feature of the signal to be decomposed, wherein the energy feature is the current energy feature of the current signal or the sound energy feature of the sound signal, and the energy feature algorithm is: in, is the coefficient of the kth frequency band of the jth layer at the nth sampling point after the wavelet packet decomposition of the signal to be decomposed, is the signal energy feature of the kth frequency band of the jth layer after the wavelet packet decomposition of the signal to be decomposed, is the kth frequency band coefficient of the jth layer after the wavelet packet decomposition of the signal to be decomposed; Determine the sampling frequency of each sampling point and the total number of all the sampling points, and perform frequency feature processing on the signal to be decomposed according to the sampling frequency and the total number to obtain the signal frequency feature of the signal to be decomposed, wherein the signal frequency feature is the current frequency feature of the current signal or the sound frequency feature of the sound signal; The current energy feature and the current frequency feature are used as a current feature vector, and the sound energy feature and the sound frequency feature are used as a sound feature vector.

3. The arc fault detection method according to claim 1, characterized in that: The step of determining a joint feature vector based on the current feature vector and the sound feature vector comprises: The current feature vector and the sound feature vector are spliced ​​according to a preset vector splicing algorithm to obtain a spliced ​​feature vector. The vector splicing algorithm is: F fused = i ⊕F s Among them, F fused is the concatenated feature vector, F i is the current eigenvector, F s is the sound feature vector; Determine a feature selection matrix corresponding to the concatenated feature vectors, and perform weighted processing on each important arc feature in the feature selection matrix to obtain a weighted matrix; The product of the concatenated feature vector and the feature selection matrix is ​​multiplied by the weighting matrix to obtain a joint feature vector.

4. The arc fault detection method according to claim 1, characterized in that: The step of determining the arc fault detection result of the low voltage power distribution system according to the preset arc fault detection model and the joint feature vector comprises: Inputting the joint feature vector into a preset arc fault detection model; After determining that the convolution layer in the fault arc detection model receives the joint feature vector, the convolution layer performs a convolution operation on the joint feature vector in a preset residual shrinkage module to obtain an initial arc feature; The fault arc detection result of the low-voltage power distribution system is determined according to the initial characteristics of the arc.

5. The arc fault detection method according to claim 4, characterized in that: The step of determining the fault arc detection result of the low-voltage power distribution system according to the initial characteristics of the arc comprises: The arc initial features are transmitted from the convolutional layer to the fully connected layer in the fault arc detection model, and the arc initial features are fully connected according to the fully connected algorithm corresponding to the fully connected layer to obtain arc splicing features, and the fully connected algorithm is: F fc =(W fc F res + fc ) Among them, F fc is the arc splicing feature, σ is the activation function of the fault arc detection model, W fc is the weight matrix of the fully connected layer, F res is the initial characteristic of the arc, b fc is the bias term of the fully connected layer; The fault arc detection result of the low-voltage power distribution system is determined according to the arc splicing characteristics.

6. The arc fault detection method according to claim 5, characterized in that: The step of determining the fault arc detection result of the low-voltage power distribution system according to the arc splicing feature comprises: The arc initial features are transmitted from the fully connected layer to the output layer in the fault arc detection model, and the arc splicing features are subjected to fault detection according to the fault detection algorithm corresponding to the output layer to obtain the fault arc detection result of the low-voltage power distribution system, wherein the fault detection algorithm is: in, is the fault arc detection result of the low-voltage power distribution system, softmax is the probability function of the fault arc occurring in the low-voltage power distribution system, W out is the weight matrix of the output layer, F fc is the arc splicing feature, b out is the bias term of the output layer.

7. The arc fault detection method according to claim 1, characterized in that: After the step of determining the arc fault detection result of the low voltage power distribution system according to the preset arc fault detection model and the joint feature vector, the arc fault detection method comprises: Detecting whether a result threshold of the arc fault detection result exceeds a preset arc fault threshold; If so, the number of arc half-cycles when the low-voltage power distribution system supplies power to the load within the preset counting time is accumulated, until it is determined that the accumulated number of arc half-cycles matches the preset arc count number, then it is determined that a fault arc occurs in the low-voltage power distribution system; If not, the next moment after the current moment is taken as the next current moment, and the process returns to the step of obtaining the current signal and the sound signal of the low-voltage power distribution system at the current moment.

8. A fault arc detection device, characterized in that: The arc fault detection device comprises: An acquisition module, used to acquire the current signal and the sound signal of the low-voltage power distribution system at the current moment; a decomposition module, configured to perform wavelet packet decomposition on the current signal and the sound signal to obtain a current feature vector of the current signal and a sound feature vector of the sound signal, and determine a joint feature vector based on the current feature vector and the sound feature vector; The detection module is used to determine the arc fault detection result of the low-voltage power distribution system according to a preset arc fault detection model and the joint feature vector.

9. A fault arc detection device, characterized in that: The arc fault detection device includes a memory, a processor, and an arc fault detection program stored in the memory and executable on the processor. When the processor executes the arc fault detection program, the steps of the arc fault detection method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an arc fault detection program, and when the arc fault detection program is executed by a processor, the steps of the arc fault detection method according to any one of claims 1 to 7 are implemented.

Citation Information

Cited By

  • Arcing detection method between series energy storage battery packs and related product

    CN120254508A

  • Arcing detection method between energy storage battery packs in stacked energy storage system and related product

    CN120254510A

  • Power distribution box arc fault early warning system

    CN120955559A

  • An electric arc fault early warning system for a distribution box

    CN120955559B