A method, apparatus, electrical arc detection device, and medium for identifying an electrical arc signal
By using a high sampling rate and a multiple decision strategy to identify electric arc signals, the problem of misjudgment and missed judgment caused by low sampling rate in the existing technology is solved, and higher recognition accuracy and reliability are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- QINGDAO TOPSCOMM COMM
- Filing Date
- 2022-11-23
- Publication Date
- 2026-06-02
AI Technical Summary
Existing arc detection equipment has a low sampling rate and is susceptible to environmental interference, leading to misjudgments and missed detections, making it difficult to accurately identify arc signals and increasing the risk of fire.
An analog-to-digital converter is used to acquire AC circuit signals at a sampling rate of 1 Mbps or higher. Bandpass filtering and high-frequency signal splitting are performed, and arc signals are identified by combining frequency eigenvalues and similarity decision strategies.
By employing a high sampling rate and a multiple decision strategy, the false positive rate was significantly reduced, the accuracy of arc signal identification was improved, and the impact of environmental interference was reduced.
Smart Images
Figure CN115902550B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of fault arc detection, and in particular to a method, apparatus, arc detection equipment and medium for identifying arc signals. Background Technology
[0002] Electrical fires account for a significant proportion of all fire accidents, and arcing is a major cause of electrical fires. When an arc occurs, it can easily ignite a fire. Because of the diverse types of loads in a circuit, arc data and normal data are not significantly different in the time domain, even at low frequencies. Therefore, traditional circuit protection devices cannot effectively detect arcing faults in circuits, making them more prone to causing fires.
[0003] Traditional arc detection equipment, limited by the development of electronic technology, has a low sampling rate for current signals and examines arc characteristics solely through low-frequency current waveforms. This makes it highly susceptible to interference from low-frequency signals in the environment, leading to misjudgments and missed detections of arc signals. Furthermore, even detection methods that utilize model discrimination largely rely on training models using low-frequency features. These models are limited by computational power and are also prone to misjudgments due to environmental factors, resulting in frequent false tripping and hindering widespread application. Misjudging arc signals can easily cause fires.
[0004] Therefore, accurately identifying electric arc signals is a technical problem that urgently needs to be solved by those in this field. Summary of the Invention
[0005] The purpose of this application is to provide a method, apparatus, arc detection equipment and medium for identifying arc signals, so as to more accurately distinguish between arc signals and interference signals in AC circuits.
[0006] To address the aforementioned technical problems, this application provides a method for identifying electric arc signals, comprising:
[0007] The target signal in the AC circuit is acquired by an analog-to-digital converter at a acquisition rate of 1 Mbps or higher.
[0008] The target signal is bandpass filtered to obtain the target frequency channel signal;
[0009] The target frequency channel signal is split into multiple single-frequency channel frequency signals;
[0010] Each frequency signal is determined to be an arc signal or a non-arc signal according to a pre-set frequency feature decision strategy; wherein, the frequency feature decision strategy includes at least a frequency feature value decision strategy and a frequency feature similarity decision strategy.
[0011] Preferably, a half-wave is used as the smallest discrimination unit; determining whether the current half-wave is the arc signal or the non-arc signal according to the pre-set frequency characteristic value decision strategy includes:
[0012] Obtain the magnitude of each feature quantity of the current frequency channel and the threshold corresponding to each feature quantity; wherein, the plurality of feature quantities include at least the mean of the half-wave, the variance of the half-wave, and the range of the half-wave;
[0013] The magnitude of each of the characteristic quantities of the current frequency channel is compared with the corresponding threshold.
[0014] If the magnitude of all the feature values of the current channel is greater than or equal to the corresponding threshold, then the decision result of determining the frequency feature value of the current frequency channel is passed; return to the step of obtaining the magnitude of each feature value of the current frequency channel and the threshold value corresponding to each feature value, until all the frequency channels have been compared;
[0015] The decision result of obtaining the frequency feature value is the first number of the current frequency channels that have passed;
[0016] If the first quantity meets the first preset requirement, the current half-wave is determined to be the arc signal;
[0017] Conversely, the current half-wave is determined to be the non-arc signal.
[0018] Preferably, determining whether the current half-wave is the arc signal or the non-arc signal according to the pre-set frequency feature similarity decision strategy includes:
[0019] Obtain the similarity between the current frequency channel and each of the frequency channels in the series.
[0020] Obtain the similarity threshold between the frequency channel and each frequency channel in all the frequency channels;
[0021] The similarity of each frequency channel is compared with the corresponding similarity threshold;
[0022] If the similarity of the frequency channel is greater than or equal to the corresponding similarity threshold, then the decision result of the frequency feature similarity of the current channel is passed; return to the step of obtaining the similarity between the current frequency channel and each of the frequency channels; until all the frequency channels have been compared.
[0023] The decision result of obtaining the frequency feature similarity is the second number of the current frequency channels that have passed;
[0024] If the second quantity meets the second preset requirement, the current half-wave is determined to be the arc signal;
[0025] Conversely, the current half-wave is determined to be the non-arc signal.
[0026] Preferably, after determining that the current half-wave is the arc signal, the method further includes:
[0027] Based on the first quantity, the quantity of the feature quantity, the judgment result of the frequency feature value, the magnitude of each of the feature quantities in the current frequency channel that is passed, and the corresponding threshold, the first over-threshold coefficient corresponding to the frequency feature value is determined;
[0028] The second threshold coefficient corresponding to the frequency feature similarity is determined based on the similarity of the current frequency channel that has passed the judgment result of the second quantity and the frequency feature similarity, and the corresponding similarity threshold.
[0029] The decision result of the current half-wave is determined based on the first overthreshold coefficient and the second overthreshold coefficient;
[0030] If the judgment result meets the third preset requirement, the current half-wave is finally determined to be the arc signal.
[0031] Preferably, obtaining the similarity between the current frequency channel and each of the frequency channels includes:
[0032] The similarity index values between the current frequency channel and each of the frequency channels are obtained, as well as the weight values of each similarity index; wherein, the similarity index includes at least mean similarity, variance similarity, and data point similarity;
[0033] The similarity between the current frequency channel and each of the frequency channels is determined based on the values of each similarity index and the weight values of each similarity index.
[0034] Preferably, before performing bandpass filtering on the signal in the target frequency band to obtain the signal of each frequency channel, the method further includes:
[0035] The signal in the target frequency band is low-pass filtered, gain adjusted, and then converted by the analog-to-digital converter.
[0036] Preferably, the acquisition of the target frequency band signal in the AC circuit via the analog-to-digital converter includes:
[0037] The signal in the AC circuit is acquired by the analog-to-digital converter.
[0038] The signal is filtered by an LC filter circuit to obtain the signal of the target frequency band.
[0039] To address the aforementioned technical problems, this application also provides a device for identifying electric arc signals, comprising:
[0040] The acquisition module is used to acquire target signals in AC circuits at an acquisition rate of greater than or equal to 1 Mbps using an analog-to-digital converter.
[0041] The filtering module is used to perform bandpass filtering on the target signal in order to obtain the target frequency channel signal;
[0042] A splitting module is used to split the target frequency channel signal into multiple single-frequency channel frequency signals;
[0043] The determining module is used to determine whether each frequency signal is an arc signal or a non-arc signal according to a pre-set frequency feature decision strategy; wherein, the frequency feature decision strategy includes at least a frequency feature value decision strategy and a frequency feature similarity decision strategy.
[0044] To address the aforementioned technical problems, this application also provides an arc detection device, comprising:
[0045] Memory, used to store computer programs;
[0046] A processor is configured to implement the steps of the method for identifying electric arc signals described above when executing the computer program.
[0047] To address the aforementioned technical problems, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method for identifying electric arc signals described above.
[0048] The method for identifying electric arc signals provided in this application includes: acquiring a target signal in an AC circuit at an acquisition rate greater than or equal to 1 MHz using an analog-to-digital converter; performing bandpass filtering on the target signal to obtain a target frequency channel signal; splitting the target frequency channel signal into multiple single-frequency channel frequency signals; and determining whether each frequency signal is an electric arc signal or a non-electric arc signal according to a pre-set frequency feature decision strategy; wherein the frequency feature decision strategy includes at least a frequency feature value decision strategy and a frequency feature similarity decision strategy. Previous methods examined arc characteristics based on low-frequency current waveforms, which are easily affected by low-frequency signals in the environment, leading to errors in arc signal identification. In contrast, the method in this application uses an analog-to-digital converter to acquire the target signal in the AC circuit at a sampling rate greater than or equal to 1 MHz, i.e., acquiring a high-frequency signal. Arc signals and non-arc signals are distinguished based on the characteristics of the high-frequency signal in the AC circuit. Identifying arc signals based on the characteristics of high-frequency signals reduces the impact of low-frequency interference in the environment on the discrimination, significantly lowering the false positive rate. Furthermore, the frequency feature decision strategy used in this application for arc signal identification includes at least a frequency feature value decision strategy and a frequency feature similarity decision strategy. Compared to using only one decision strategy, the method in this application can improve the accuracy of identification.
[0049] In addition, this application also provides a device for identifying electric arc signals, an electric arc detection device, and a computer-readable storage medium, which have the same or corresponding technical features as the aforementioned method for identifying electric arc signals, and have the same effect. Attached Figure Description
[0050] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0051] Figure 1 A flowchart illustrating a method for identifying electric arc signals provided in this application embodiment;
[0052] Figure 2 A structural diagram of an arc signal identification device provided in an embodiment of this application;
[0053] Figure 3 This is a structural diagram of an arc detection device provided in another embodiment of this application;
[0054] Figure 4 This is a structural diagram of a device for identifying electric arc signals, provided in another embodiment of this application. Detailed Implementation
[0055] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.
[0056] The core of this application is to provide a method, apparatus, arc detection equipment, and medium for identifying arc signals, which can more accurately distinguish between arc signals and interference signals in AC circuits.
[0057] The vigorous development of new power systems has placed higher demands on the safe and reliable operation of power equipment. The connection of multiple devices may weaken the stability and security of the AC power grid. Therefore, under the new development background, the AC power grid still needs to address the issues of real-time equipment status perception, accurate assessment, and early warning of potential faults. Among these, arc faults are a major cause of electrical fires in distribution lines. Based on the location of the arc fault in the distribution line, it can be divided into series arc faults and parallel arc faults. Series arc faults generate relatively small currents, making traditional circuit protection devices such as circuit breakers and overcurrent protectors ineffective in detecting them. Therefore, research on series arc fault detection methods is of great significance. Existing technologies mainly use traditional arc detection equipment. Limited by the level of electronic technology development, the sampling rate of the current signal is low, and arc characteristics are examined only through low-frequency current waveforms, making them highly susceptible to environmental interference, leading to misjudgments and missed detections. Furthermore, even some detection methods that utilize model discrimination mostly rely on low-frequency features for model training. These models are constrained by computational power and are also prone to misjudgments due to environmental interference, resulting in frequent false trips and hindering widespread application. This application utilizes a high sampling rate to focus on observing the characteristic information of the high-frequency band, selects multiple suitable data feature quantities, sets static thresholds for multiple high-frequency channels respectively, and uses two strategies to perform joint discrimination simultaneously, which greatly reduces the impact of low-frequency interference in the environment on discrimination and can significantly reduce the false judgment rate.
[0058] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Figure 1 A flowchart of a method for identifying electric arc signals provided in an embodiment of this application is shown below. Figure 1 As shown, the method includes: S10: Acquiring the target signal in the AC circuit at an acquisition rate greater than or equal to 1M using an analog-to-digital converter.
[0059] In this embodiment, the target signal in the AC circuit is acquired by an analog-to-digital converter at a sampling rate greater than or equal to 1 Mbps, i.e., the high-frequency signal in the AC circuit is acquired. Specifically, a high-speed digital-to-analog converter (DAC) is used to acquire the high-frequency signal from the live wire using a transient coil or a high-frequency current transformer.
[0060] S11: Bandpass filter the target signal to obtain the target frequency channel signal.
[0061] S12: Split the target frequency channel signal into multiple single-frequency channel frequency signals.
[0062] S13: Determine whether each frequency signal is an arc signal or a non-arc signal according to a pre-set frequency feature decision strategy; wherein, the frequency feature decision strategy includes at least a frequency feature value decision strategy and a frequency feature similarity decision strategy.
[0063] After acquiring the high-frequency signal, which contains signals from multiple frequency bands, a bandpass filter is set to select the information from the appropriate frequency channels. Then, Fast Fourier Transform (FTT) digital signal processing is performed to decompose the acquired time-domain signal into the desired frequency domain signals from the time domain to the frequency domain.
[0064] In determining whether each frequency signal is an arc signal or a non-arc signal (where non-arc signals can be considered interference signals), this embodiment uses a pre-set frequency feature decision strategy. To accurately determine whether each frequency signal is an arc signal or a non-arc signal, the pre-set frequency feature decision strategy includes at least a frequency feature value decision strategy and a frequency feature similarity decision strategy; that is, at least two decision strategies are used to identify arc signals.
[0065] The method for identifying electric arc signals provided in this embodiment includes: acquiring a target signal in an AC circuit at an acquisition rate greater than or equal to 1 MHz using an analog-to-digital converter; performing bandpass filtering on the target signal to obtain a target frequency channel signal; splitting the target frequency channel signal into multiple single-frequency channel frequency signals; and determining whether each frequency signal is an electric arc signal or a non-electric arc signal according to a pre-set frequency feature decision strategy; wherein the frequency feature decision strategy includes at least a frequency feature value decision strategy and a frequency feature similarity decision strategy. Previous methods examined arc characteristics based on low-frequency current waveforms, which are easily affected by low-frequency signals in the environment, leading to errors in arc signal identification. In contrast, the method in this application uses an analog-to-digital converter to acquire the target signal in the AC circuit at a sampling rate greater than or equal to 1 MHz, i.e., acquiring a high-frequency signal. Arc signals and non-arc signals are distinguished based on the characteristics of the high-frequency signal in the AC circuit. Identifying arc signals based on the characteristics of high-frequency signals reduces the impact of low-frequency interference in the environment on the discrimination, significantly lowering the false positive rate. Furthermore, the frequency feature decision strategy used in this embodiment for arc signal identification includes at least a frequency feature value decision strategy and a frequency feature similarity decision strategy. Compared to using only one decision strategy, the method in this embodiment can improve the accuracy of identification.
[0066] To accurately identify arc signals, a preferred implementation method involves determining whether the current half-wave is an arc signal or a non-arc signal based on a pre-set frequency characteristic value decision strategy, including:
[0067] Obtain the magnitude of each feature of the current frequency channel and the threshold corresponding to each feature; wherein, the multiple features include at least the mean of the half-wave, the variance of the half-wave, and the range of the half-wave.
[0068] Compare the magnitudes of each feature quantity of the current frequency channel with the corresponding thresholds;
[0069] If the magnitude of all features in the current channel is greater than or equal to the corresponding threshold, then the decision result for the frequency feature value of the current frequency channel is passed; return to the steps of obtaining the magnitude of each feature and the threshold corresponding to each feature in the current frequency channel, until all frequency channels have been compared.
[0070] The decision result for obtaining the frequency feature value is the first number of current frequency channels that have passed;
[0071] If the first quantity meets the first preset requirement, the current half-wave is determined to be an arc signal;
[0072] Conversely, the current half-wave is determined to be a non-arc signal.
[0073] It should be noted that the frequency characteristic value decision strategy specifically refers to the threshold discrimination strategy for high-frequency characteristic values, and all current frequency channels are high-frequency channels. Specifically, the threshold discrimination strategy for high-frequency characteristic values is as follows: analysis and discrimination are performed using 10ms data segments, i.e., one single half-wave. First, P discrimination conditions are set for the selected N high-frequency channels, and different threshold values are set for different discrimination conditions according to the different frequency channels. In practice, through analysis using high sampling rates under various environments, it was found that almost all arc characteristics are high-frequency, and are most obvious in the 2-50MHz range. Therefore, this mainly focuses on setting P different representative high-frequency static threshold parameters for each frequency channel in the 2-50MHz range. These P discrimination conditions are for different feature quantities, including: half-wave maximum value, half-wave mean value, half-wave variance, half-wave coefficient of variation, half-wave range, half-wave maximum value / half-wave mean value, half-wave minimum value / half-wave mean value, half-wave range / half-wave mean value, (half-wave maximum value^2 - half-wave minimum value^2) / half-wave mean value, etc. The threshold and feature value of each channel are represented by Y. gh and T gh (where g ranges from 1 to N, representing the nth channel, and h ranges from 1 to P, representing the nth discrimination condition).
[0074] In the threshold discrimination strategy for high-frequency feature values, for a single high-frequency channel, it is considered to pass only when all P condition thresholds corresponding to it are passed. The number of high-frequency channels that have passed is recorded as the first quantity x. If the first quantity x meets the first preset requirement, the current half-wave is determined to be an arc signal; otherwise, the current half-wave is determined to be a non-arc signal. A gating ratio is set for N high-frequency channels. Here, the first preset requirement refers to the ratio of the first quantity x to N being greater than or equal to the gating ratio. A discrimination result rst(1) of a feature value is given according to the gating ratio of N high-frequency channels, and its value is 0 or 1. 0 can be considered as a non-arc signal, and 1 can be considered as an arc signal. When the number of high-frequency channels that have passed x is greater than k1, k1 represents the discrimination result of the threshold discrimination strategy for high-frequency feature values being the threshold of the passed frequency channel, and the discrimination result rst(1) of the feature value is 1, otherwise it is 0. The expression of rst(1) is as follows:
[0075]
[0076] Taking channels as the unit, assuming a total of 49 channels, each channel has P static thresholds. In a single channel, the result of this frequency channel is 1 only when all P static thresholds of the half-wave signal pass through; otherwise, it is 0. Then, when more than k1 channels out of the 49 channels are 1, the final result of the threshold discrimination strategy for high-frequency features is 1, which is the arc signal.
[0077] In this embodiment, the method determines whether the current half-wave is an arc signal or a non-arc signal based on a pre-set frequency characteristic value decision strategy. By setting multiple threshold decision conditions, the method determines that a frequency channel passes when all thresholds for each frequency channel are passed; and furthermore, if the number of passed frequency channels meets the selection ratio, the current half-wave is determined to be an arc signal. Therefore, this method uses multiple decisions to make the identification results more accurate.
[0078] The frequency feature value decision strategy has been described in the above embodiments. In practice, in addition to the frequency feature decision strategy described above, this embodiment also identifies the arc signal based on the frequency feature similarity decision strategy. A preferred embodiment is that determining whether the current half-wave is an arc signal or a non-arc signal based on a pre-set frequency feature similarity decision strategy includes:
[0079] Obtain the similarity between the current frequency channel and each of the other frequency channels;
[0080] Obtain the similarity threshold between the frequency channel and each frequency channel in all frequency channels;
[0081] The similarity of each frequency channel is compared with the corresponding similarity threshold;
[0082] If the similarity of the frequency channels is greater than or equal to the corresponding similarity threshold, the decision result for the frequency feature similarity of the current channel is passed; return to the step of obtaining the similarity between the current frequency channel and each frequency channel in all frequency channels; until all frequency channels have been compared.
[0083] The decision result based on frequency feature similarity is the second number of current frequency channels that have passed;
[0084] If the second quantity meets the second preset requirement, the current half-wave is determined to be an arc signal;
[0085] Conversely, the current half-wave is determined to be a non-arc signal.
[0086] It should be noted that the frequency feature similarity decision strategy specifically refers to the high-frequency feature similarity threshold decision strategy, and all current frequency channels are high-frequency channels. The similarity threshold and similarity score for each channel are denoted as SY. m and ST m(Where m ranges from 1 to N-1, representing the nth channel). Specifically, the high-frequency feature similarity threshold decision strategy is as follows: For a single high-frequency channel, it is considered to pass only when the similarity condition corresponding to it exceeds the threshold. The number of high-frequency channels that have passed is recorded as the second quantity z. If the second quantity z meets the second preset requirement, the current half-wave is determined to be an arc signal; otherwise, the current half-wave is determined to be a non-arc signal. A selection ratio is set for N high-frequency channels. Here, the second preset requirement refers to the ratio of the second quantity z to N being greater than or equal to the selection ratio. A feature similarity discrimination result rst(2) is given according to the selection ratio of multiple channels, and its value is 0 or 1. 0 can be considered as a non-arc signal, and 1 can be considered as an arc signal. When the number of high-frequency channels that have passed z is greater than k2, k2 represents the discrimination result of the high-frequency feature similarity threshold decision strategy as the threshold of the frequency channel that has passed. Only then is the feature similarity discrimination result rst(2) 1, otherwise it is 0. The expression of rst(2) is as follows:
[0087]
[0088] There are 49 channels in total, with 48 similarity judgments. The output for each frequency channel is 1 only if each group exceeds a static similarity threshold; otherwise, it is 0. Then, according to the gating strategy, if more than k² groups out of the 48 output 1, the final output of the high-frequency feature similarity threshold decision strategy is 1, indicating an arc signal.
[0089] In this embodiment, the method determines whether the current half-wave is an arc signal or a non-arc signal based on a pre-set frequency feature similarity judgment strategy. By setting multiple similarity thresholds, the current channel is determined to pass if the similarity between the current channel and other channels meets the corresponding similarity thresholds. Furthermore, if the number of passing frequency channels meets the selection ratio, the current half-wave is determined to be an arc signal. Therefore, this method uses multiple judgments to make the identification results more accurate.
[0090] In practice, although the current half-wave signal is determined to be an arc signal according to the above-mentioned decision strategy, in order to improve the accuracy of identification, after determining that the current half-wave is an arc signal, the method for identifying the arc signal also includes:
[0091] Based on the first quantity, the quantity of feature quantities, and the judgment result of frequency feature values, the first threshold coefficient corresponding to the frequency feature value is determined by the magnitude of each feature quantity in the current frequency channel that has passed and the corresponding threshold.
[0092] Based on the judgment result of the second quantity and frequency feature similarity, the second threshold coefficient corresponding to the frequency feature similarity is determined by the similarity of the current frequency channel passed and the corresponding similarity threshold.
[0093] The decision result for the current half-wave is determined based on the first and second overthreshold coefficients.
[0094] If the judgment meets the third preset requirement, the current half-wave is finally determined to be an arc signal.
[0095] When the discrimination result rst(1) of the eigenvalue is 1, it is also necessary to calculate the eigenvalue threshold coefficient, i.e., the first threshold coefficient w. The formula for calculating the first threshold coefficient w is as follows:
[0096]
[0097] Where i ranges from 1 to x, j ranges from 1 to P, and Y... ij and T ij These represent the characteristic value and characteristic value threshold of each high-frequency channel.
[0098] When the feature similarity discrimination result rst(2) is 1, it is also necessary to calculate the feature similarity over-threshold coefficient, i.e., the second over-threshold coefficient v. The formula for calculating the second over-threshold coefficient v is as follows:
[0099]
[0100] Where i ranges from 1 to z, ST i and SY i These represent the similarity and similarity threshold of each high-frequency channel.
[0101] After obtaining the first overthreshold coefficient w and the second overthreshold coefficient v, the decision result r of the current half-wave can be determined. The formula for calculating the decision result r of the current half-wave is as follows:
[0102] r=w / (w+v)*rst(1)+v / (w+v)*rst(2)
[0103] Here, w and v can be understood as weight coefficients, both ranging from [0, 1], and their sum is always 1. The values of rst(1) and rst(2) are 0 or 1 respectively, and the range of the decision result r is also [0, 1].
[0104] If the decision result r meets the third preset requirement, the current half-wave is ultimately determined to be an arc signal. The third preset requirement is not limited and is determined based on the actual situation. If the selected decision result r value is greater than 0.5, the half-wave is considered an arc signal; otherwise, it is a non-arc signal.
[0105] The weighting provided in this embodiment can greatly reduce the false positive rate and make the method of identifying electric arc signals more flexible.
[0106] When calculating similarity, a preferred implementation method is to obtain the similarity between the current frequency channel and each of the other frequency channels, including:
[0107] Obtain the similarity index values between the current frequency channel and each frequency channel in all frequency channels, and obtain the weight values of each similarity index; wherein, the similarity index includes at least mean similarity, variance similarity, and data point similarity;
[0108] The similarity between the current frequency channel and all other frequency channels is determined based on the values of each similarity index and the weight values of each similarity index.
[0109] When using a high-frequency feature similarity threshold decision strategy, the first step is to measure the similarity between the selected high-frequency channels, provided that the data for each high-frequency channel has been normalized. This embodiment selects at least three similarity indicators. Taking mean similarity, variance similarity, and data point similarity as an example, these three indicators are integrated into a unified similarity calculation formula based on their different weights. The similarity calculation formula is as follows:
[0110] θ(x, y)=α / (1+distance(μ(x),μ(y)))+β / (1+distance(δ(x), δ(y)))+γ / (1+distance(xy)) In the formula, the expression of distance(h, t) is as follows:
[0111]
[0112] Here, h and t represent two vectors of the same dimension; x and y refer to vectors of two different frequency channels of the same power frequency half-wave; μ(x) and μ(y) represent the mean of the two vectors; δ(x) and δ(y) represent the variance of the two vectors; distance(h, t) is the formula for calculating Euclidean distance; α, β, and γ represent the weighting coefficients of the three, all ranging from 0 to 1, and the sum of the coefficients is 1. Analysis and discrimination are performed using 10ms data segments, i.e., one single half-wave. For frequency channels from 2M to 50M, using 2M as a reference, the similarity characteristics of each channel with the 2M channel are analyzed. A similarity threshold between frequency channels is set, and the similarity calculation formula consists of three parts: mean similarity, variance similarity, and data point similarity, each with different weighting coefficients α, β, and γ.
[0113] The similarity calculation method provided in this embodiment uses multiple indicators and sets different weights for each indicator to comprehensively analyze similarity, making the calculated similarity more accurate.
[0114] In practice, to prevent spectrum leakage, a preferred implementation method is to further include, before bandpass filtering the signal in the target frequency band to obtain the signal of each frequency channel, low-pass filtering, gain adjustment, and conversion by an analog-to-digital converter on the signal in the target frequency band.
[0115] To acquire high-frequency signals, a preferred implementation method involves acquiring signals in the target frequency band of an AC circuit using an analog-to-digital converter, including:
[0116] Signals in AC circuits are acquired using an analog-to-digital converter;
[0117] The signal is filtered by an LC filter circuit to obtain the signal in the target frequency band.
[0118] LC filter circuits offer advantages such as simple structure, high reliability, and wide application range. Their main characteristics include low inductance and resistance, low DC loss, high inductance for AC, and good filtering effect. Therefore, in this embodiment, an LC filter circuit is used to filter the AC circuit to obtain the signal in the target frequency band.
[0119] In the above embodiments, the method for identifying electric arc signals has been described in detail. This application also provides embodiments of an apparatus for identifying electric arc signals and an electric arc detection device. It should be noted that this application describes the embodiments of the apparatus from two perspectives: one is based on functional modules, and the other is based on hardware.
[0120] Figure 2 A structural diagram of an arc signal identification device provided according to an embodiment of this application. This embodiment, based on functional modules, includes:
[0121] The acquisition module 10 is used to acquire target signals in the AC circuit at an acquisition rate of greater than or equal to 1M using an analog-to-digital converter.
[0122] Filtering module 11 is used to perform bandpass filtering on the target signal in order to obtain the target frequency channel signal;
[0123] The splitting module 12 is used to split the target frequency channel signal into multiple single-frequency channel frequency signals;
[0124] The determination module 13 is used to determine whether each frequency signal is an arc signal or a non-arc signal according to a pre-set frequency feature decision strategy; wherein, the frequency feature decision strategy includes at least a frequency feature value decision strategy and a frequency feature similarity decision strategy.
[0125] Since the embodiments of the apparatus and the method correspond to each other, please refer to the description of the embodiments in the method section for the embodiments of the apparatus, and they will not be repeated here. Furthermore, it has the same beneficial effects as the method for identifying electric arc signals mentioned above.
[0126] Figure 3 This is a structural diagram of an arc detection device provided in another embodiment of this application. This embodiment is based on a hardware perspective, such as... Figure 3 As shown, the arc detection equipment includes:
[0127] Memory 20 is used to store computer programs;
[0128] The processor 21 is configured to execute a computer program to implement the steps of the method for identifying arc signals as described in the above embodiments.
[0129] The arc detection device provided in this embodiment may include, but is not limited to, smartphones, tablets, laptops, or desktop computers.
[0130] The processor 21 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 21 may be implemented using at least one of the following hardware forms: Digital Signal Processor (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 21 may also include a main processor and a coprocessor. The main processor, also known as the Central Processing Unit (CPU), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 21 may integrate a Graphics Processing Unit (GPU), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 21 may also include an Artificial Intelligence (AI) processor, which is used to handle computational operations related to machine learning.
[0131] The memory 20 may include one or more computer-readable storage media, which may be non-transitory. The memory 20 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory 20 is used to store at least the following computer program 201, which, after being loaded and executed by the processor 21, is capable of implementing the relevant steps of the method for identifying arc signals disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202 and data 203, and the storage method may be temporary or permanent storage. The operating system 202 may include Windows, Unix, Linux, etc. The data 203 may include, but is not limited to, the data involved in the aforementioned method for identifying arc signals.
[0132] In some embodiments, the arc detection device may further include a display screen 22, an input / output interface 23, a communication interface 24, a power supply 25, and a communication bus 26.
[0133] Those skilled in the art will understand that Figure 3 The structure shown does not constitute a limitation on the arc detection device and may include more or fewer components than shown.
[0134] The arc detection device provided in this application includes a memory and a processor. When the processor executes the program stored in the memory, it can implement the following method: a method for identifying arc signals, with the same effect as above.
[0135] This application also provides an embodiment corresponding to a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps described in the above method embodiments.
[0136] It is understood that if the methods in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0137] The computer-readable storage medium provided in this application includes the aforementioned method for identifying electric arc signals, and has the same effect.
[0138] To enable those skilled in the art to better understand the present application, the following description is provided in conjunction with the appendix. Figure 4 The present application will be further described in detail with reference to specific embodiments. Figure 4 This is a structural diagram of a device for identifying electric arc signals, provided as another embodiment of this application. Figure 4 As shown, the device includes: an LC filter circuit 1, a hardware digital signal processing unit 2, a software signal processing unit 3, a half-wave discrimination unit 4, and a classification decision unit 5. High-frequency signals are collected from the live wire using a transient coil or a high-frequency current transformer. After low-frequency information is filtered out by the LC filter circuit 1, the signal enters the hardware digital signal processing unit 2. In the hardware digital processing unit, after the signal is sampled by a high-speed ADC, it is sent to a low-pass filter to prevent spectral leakage. After gain adjustment and conversion into a digital signal by a 400MHz ADC, it is sent to the software signal processing unit 3. In the software signal processing unit 3, the sampled signal first passes through a 2-50MHz bandpass filter to set the bandpass filter, selects the information of the corresponding frequency channel, and then performs FFT digital signal processing to decompose the information of each high-frequency signal from the time domain to the frequency domain. Using a 10ms power frequency signal, i.e., a half-wave, as the smallest discrimination unit, after FFT processing, the current half-wave is discriminated based on two high-frequency feature discrimination strategies—high-frequency feature value and high-frequency feature similarity—based on the selected 2-50MHz high-frequency signal channels, and then sent to the half-wave discrimination unit 4. In the half-wave discrimination unit 4, the arc signal is comprehensively identified based on the feature value discrimination result and the feature similarity discrimination result. Finally, the discrimination result of each single half-wave is sent to the classification decision unit 5 in real time. The set half-wave strategy classifies the arc and normal signals, and outputs "arc" or "normal" to coordinate with the circuit control of whether the trip unit operates.
[0139] Therefore, this application utilizes ultra-high frequency sampling and FFT technology to focus on information from multiple high-frequency channels while employing various feature statistical strategies for joint discrimination. This significantly reduces the impact of low-frequency interference in the environment on the discrimination, greatly lowers the false positive rate, and makes the method more flexible through weight adjustment.
[0140] The foregoing has provided a detailed description of a method, apparatus, arc detection device, and medium for identifying electric arc signals. The various embodiments in the specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.
[0141] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
Claims
1. A method for identifying electric arc signals, characterized in that, include: The target signal in the AC circuit is acquired by an analog-to-digital converter at a acquisition rate of 1 Mbps or higher. The target signal is bandpass filtered to obtain the target frequency channel signal; The target frequency channel signal is split into multiple single-frequency channel frequency signals; Using a half-wave as the smallest discrimination unit, each frequency signal is determined to be either an arc signal or a non-arc signal according to a pre-set frequency feature decision strategy; wherein, the frequency feature decision strategy includes at least a frequency feature value decision strategy and a frequency feature similarity decision strategy; The frequency feature value decision strategy includes: obtaining the magnitude of each feature quantity of the current frequency channel, wherein the feature quantity includes at least the mean, variance and range of half wave, and comparing each feature quantity with the corresponding threshold; if all feature quantities pass the threshold, the channel is determined to pass. The frequency feature similarity judgment strategy includes: calculating the similarity between the current frequency channel and all other frequency channels, the similarity being calculated based on a weighted average of mean similarity, variance similarity, and data point similarity, and comparing the similarity with a corresponding threshold; if the similarity is higher than the corresponding threshold, the frequency channel is determined to pass the similarity judgment.
2. The method for identifying electric arc signals according to claim 1, characterized in that, The smallest discriminant unit is a half-wavelength. Determining whether the current half-wave is the arc signal or the non-arc signal according to the pre-set frequency characteristic value decision strategy includes: Obtain the magnitude of each feature quantity of the current frequency channel and the threshold corresponding to each feature quantity; wherein, the multiple feature quantities include at least the mean of the half-wave, the variance of the half-wave, and the range of the half-wave; The magnitude of each of the characteristic quantities of the current frequency channel is compared with the corresponding threshold. If the magnitude of all the feature values of the current frequency channel is greater than or equal to the corresponding threshold, then the decision result of determining the frequency feature value of the current frequency channel is passed; return to the step of obtaining the magnitude of each feature value of the current frequency channel and the threshold value corresponding to each feature value, until all the frequency channels have been compared; The decision result for obtaining the frequency feature value is that the number of all frequency channels that have passed is the first number; If the first quantity meets the first preset requirement, the current half-wave is determined to be the arc signal; Conversely, the current half-wave is determined to be the non-arc signal.
3. The method for identifying electric arc signals according to claim 2, characterized in that, Determining whether the current half-wave is the arc signal or the non-arc signal based on the pre-set frequency feature similarity decision strategy includes: Obtain the similarity between the current frequency channel and each of the frequency channels in the series. Obtain the similarity threshold between the frequency channel and each frequency channel in all the frequency channels; The similarity of each frequency channel is compared with the corresponding similarity threshold; If the similarity of the frequency channel is greater than or equal to the corresponding similarity threshold, then the decision result of the frequency feature similarity of the current frequency channel is determined to be pass; return to the step of obtaining the similarity between the current frequency channel and each of the frequency channels; until all the frequency channels have been compared; The decision result for obtaining the frequency feature similarity is that the number of all frequency channels that have passed is the second number; If the second quantity meets the second preset requirement, the current half-wave is determined to be the arc signal; Conversely, the current half-wave is determined to be the non-arc signal.
4. The method for identifying electric arc signals according to claim 3, characterized in that, After determining that the current half-wave is the arc signal, the method further includes: Based on the first quantity, the quantity of the feature quantity, the judgment result of the frequency feature value, the magnitude of each feature quantity in the frequency channel that has passed, and the corresponding threshold, the first over-threshold coefficient corresponding to the frequency feature value is determined; The second threshold coefficient corresponding to the frequency feature similarity is determined based on the second quantity, the similarity of the frequency channel that has passed the frequency feature similarity judgment result, and the corresponding similarity threshold. The decision result of the current half-wave is determined based on the first overthreshold coefficient and the second overthreshold coefficient; If the judgment result meets the third preset requirement, the current half-wave is finally determined to be the arc signal.
5. The method for identifying electric arc signals according to claim 4, characterized in that, The step of obtaining the similarity between the current frequency channel and each of the frequency channels includes: The similarity index values between the current frequency channel and each of the frequency channels are obtained, as well as the weight values of each similarity index; wherein, the similarity index includes at least mean similarity, variance similarity, and data point similarity; The similarity between the current frequency channel and each of the frequency channels is determined based on the values of each similarity index and the weight values of each similarity index.
6. The method for identifying arc signals according to any one of claims 1 to 5, characterized in that, Before performing bandpass filtering on the target signal to obtain signals for each frequency channel, the method further includes: The target signal is low-pass filtered, gain adjusted, and converted using the analog-to-digital converter.
7. The method for identifying electric arc signals according to claim 6, characterized in that, The acquisition of the target signal from the AC circuit via an analog-to-digital converter includes: The signal in the AC circuit is filtered by an LC filter circuit, and the signal after being filtered by the LC filter circuit is acquired by the analog-to-digital converter in order to obtain the target signal.
8. An apparatus for identifying electric arc signals, applicable to the method for identifying electric arc signals according to any one of claims 1 to 7, characterized in that, include: The acquisition module is used to acquire target signals in AC circuits at an acquisition rate of greater than or equal to 1 Mbps using an analog-to-digital converter. The filtering module is used to perform bandpass filtering on the target signal in order to obtain the target frequency channel signal; A splitting module is used to split the target frequency channel signal into multiple single-frequency channel frequency signals; The determining module is used to determine whether each frequency signal is an arc signal or a non-arc signal according to a pre-set frequency feature decision strategy; wherein, the frequency feature decision strategy includes at least a frequency feature value decision strategy and a frequency feature similarity decision strategy.
9. An arc detection device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the method for identifying an arc signal as described in any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method for identifying arc signals as described in any one of claims 1 to 7.