Current signal denoising method, system and storage medium for fault arc detection

Abstract

The application discloses a current signal denoising method and system for fault arc detection and a storage medium, comprising: collecting the working current of the load at the current time, and analyzing to obtain the noise estimation spectrum of the working current at the current time; wavelet decomposing the noise estimation spectrum and the working current of the load at the subsequent time respectively to obtain first wavelet coefficients and second wavelet coefficients; correcting the second wavelet coefficients based on the first wavelet coefficients to obtain third wavelet coefficients; and wavelet reconstructing the third wavelet coefficients to obtain a time domain enhanced signal. The improved wavelet threshold function improves the defects of poor continuity and constant deviation of the traditional wavelet threshold function, and the current signal processed by the denoising algorithm not only suppresses the existence of noise, but also improves the arc characteristics of the signal, so that the detection performance of the fault arc detection algorithm is significantly improved.

Description

Technical Field

[0001] This invention relates to the field of electrical safety protection technology, specifically to a current signal noise reduction method, system, and storage medium for fault arc detection. Background Technology

[0002] An electric arc is a gas ionization discharge phenomenon that generally exists in power supply lines or household appliances where insulation is aged, damaged, or electrical connections are loose. When it is generated, it releases a large amount of light and heat. Electrical fires are the most common type of fire, and faulty electric arcs are one of the main causes of electrical fires.

[0003] The mainstream method for detecting fault arcs is to capture the generation of fault arcs by analyzing electrical signals. However, since the power grid itself has a certain amount of noise, and the electrical signals are inevitably contaminated by noise during the acquisition and transmission process, the accuracy and reliability of fault arc detection technology based on signal analysis are affected.

[0004] In the related technology, the fault arc detection method proposed in the patent application document with publication number CN105954628A uses DB4 wavelet for wavelet threshold denoising on the collected signal. This scheme uses a general threshold estimation algorithm for threshold estimation, and the estimation formula uses the estimation method in the speech denoising algorithm, which has two problems: (1) The target is a noisy signal with noise signal and current signal superimposed. Using the speech empirical formula for threshold estimation will result in a large error, thus affecting the subsequent denoising effect; (2) The signal is processed directly without discrimination. When fault arc, current signal and noise coexist, since fault arc and noise signal are quite similar, the fault arc will be treated as noise when performing threshold estimation, thus affecting the subsequent processing effect.

[0005] In the coefficient processing section, the scheme adopts the conventional soft thresholding method. However, soft thresholding has a constant deviation defect since its inception, which is also a consensus in the field of signal processing. The constant deviation characteristic of soft thresholding will cause the denoised signal to be distorted. In the processing of fault arc fields, since the soft thresholding function directly subtracts the threshold, it is too direct and forceful, which will destroy the time-frequency characteristics of the signal, thereby affecting the subsequent time-frequency analysis of the signal.

[0006] The patent application CN114034960A proposes a fault arc detection method based on weak arc characteristics, which uses singular value decomposition (SVD) for signal denoising. The theory of SVD aims to decompose the signal and noise into two perpendicular planes, thereby separating the signal. However, in the field of fault arcs, the fault arc itself has a wide frequency band and variable energy. This makes it difficult to find a suitable decomposition matrix to separate the noise when a fault arc is present. Therefore, SVD-based signal denoising algorithms exhibit significant deviations in applications to fault arcs. Furthermore, SVD is an empirical processing method derived from theoretical derivation, resulting in poor environmental adaptability. This means that the applicable scope of SVD is specific and easily affected by environmental interference. Singular value decomposition itself involves a large amount of matrix decomposition, leading to substantial computational overhead.

[0007] However, these common signal denoising algorithms suffer from the problem of damaging the signal arc characteristics when applied to the field of arc fault detection, thus affecting the recognition effect of arc fault detection technology. In extreme cases, they can even cause the arc fault detection algorithm to completely lose its function, which runs counter to the original purpose of signal denoising. At the same time, there is currently no evaluation index in the industry to describe the environmental universality of arc fault detection algorithms.

[0008] Therefore, it is urgent to design an improved wavelet threshold function denoising algorithm for the field of fault arc detection and an evaluation index to characterize the environmental adaptability of the detection algorithm. Summary of the Invention

[0009] The technical problem to be solved by this invention is how to significantly improve the detection capability of fault arc detection.

[0010] The present invention solves the above-mentioned technical problems through the following technical means:

[0011] In a first aspect, the present invention proposes a current signal noise reduction method for fault arc detection, the method comprising:

[0012] The current operating current of the load is collected at the current moment, and the noise estimation spectrum of the current operating current at the current moment is obtained by analysis;

[0013] Wavelet decomposition is performed on the noise estimation spectrum and the load operating current at subsequent time points to obtain the first wavelet coefficient and the second wavelet coefficient;

[0014] The second wavelet coefficients are corrected based on the first wavelet coefficients to obtain the third wavelet coefficients;

[0015] Wavelet reconstruction is performed on the third wavelet coefficients to obtain the time-domain enhanced signal.

[0016] Furthermore, the step of acquiring the current operating current of the load at the current moment and analyzing it to obtain the noise estimation spectrum of the current operating current includes:

[0017] The current operating current of the load at the current moment is collected using a current acquisition sensor;

[0018] Perform fault arc analysis on the current operating current of the load to detect whether a fault arc has occurred;

[0019] If so, then re-collect the operating current of the load;

[0020] If not, then the noise estimation spectrum of the operating current at the current moment is analyzed using an active noise estimation algorithm.

[0021] Further, the step of performing wavelet decomposition on the noise estimation spectrum and the load operating current at subsequent time points to obtain the first wavelet coefficients and the second wavelet coefficients includes:

[0022] The noise estimation spectrum and the load operating current at subsequent time points are respectively subjected to wavelet decomposition based on the dbn wavelet basis to obtain the first wavelet coefficient and the second wavelet coefficient.

[0023] Further, the step of correcting the second wavelet coefficients based on the first wavelet coefficients to obtain the third wavelet coefficients includes:

[0024] Estimate the threshold of the first wavelet coefficients;

[0025] Construct an inhibition factor matrix based on the threshold;

[0026] The second wavelet coefficients are corrected based on the suppression factor matrix to obtain the third wavelet coefficients.

[0027] Furthermore, the threshold for estimating the first wavelet coefficients is expressed by the formula:

[0028]

[0029] In the formula: λ j This represents the threshold value of the first wavelet coefficient. Let represent the i-th first wavelet coefficient of the j-th layer decomposition, and N represent the length of the wavelet coefficients of the current layer.

[0030] Furthermore, the construction of the inhibition factor matrix based on the threshold is expressed by the following formula:

[0031]

[0032] Where: H w Represents the inhibition factor matrix, λ jThe threshold value of the first wavelet coefficient is represented by w2(i,j), which represents the i-th second wavelet coefficient of the j-th layer decomposition. α and β are adjustment factors.

[0033] Furthermore, the second wavelet coefficients are corrected based on the suppression factor matrix to obtain the third wavelet coefficients, as expressed by the formula:

[0034]

[0035] In the formula: H represents the third wavelet coefficient. w Let w2(i,j) represent the suppression factor matrix, and w2(i,j) represent the i-th second wavelet coefficient of the j-th level decomposition.

[0036] Furthermore, after correcting the second wavelet coefficients based on the first wavelet coefficients to obtain the third wavelet coefficients, the method further includes:

[0037] Fault arc analysis is performed on the third wavelet signal, and the threshold margin is calculated, expressed by the following formula:

[0038] ThMar=AbNCa maxm -NCa max

[0039] In the formula: ThMar represents the threshold margin, AbNCa maxm NCa represents the m-th largest value of the fault-state arc descriptive characteristic. max This represents the maximum value of the characteristic describing a normal-state electric arc.

[0040] Secondly, the present invention also proposes a current signal noise reduction system for fault arc detection, the system comprising:

[0041] The acquisition module is used to acquire the current operating current of the load at the current moment and analyze it to obtain the noise estimation spectrum of the current operating current at the current moment;

[0042] The wavelet decomposition module is used to perform wavelet decomposition on the noise estimation spectrum and the load operating current at subsequent time points to obtain the first wavelet coefficient and the second wavelet coefficient.

[0043] The correction module is used to correct the second wavelet coefficients based on the first wavelet coefficients to obtain the third wavelet coefficients;

[0044] The wavelet reconstruction module is used to reconstruct the third wavelet coefficients to obtain a time-domain enhanced signal.

[0045] Thirdly, the present invention also proposes a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the current signal noise reduction method for fault arc detection as described above.

[0046] The advantages of this invention are:

[0047] (1) The present invention performs wavelet decomposition on the noise estimation spectrum analyzed based on the working current at the current time and the working current at subsequent times, and then uses the wavelet coefficients of the noise estimation spectrum to correct the wavelet coefficients of the working current at subsequent times, and performs wavelet reconstruction based on the corrected wavelet coefficients to obtain the denoised current signal; The improved wavelet threshold function proposed in the present invention improves the defects of the traditional wavelet threshold function, such as poor continuity and constant deviation. The current signal after the denoising algorithm not only suppresses the existence of noise, but also improves the arc characteristics of the signal, so that the detection performance of the fault arc detection algorithm is significantly improved.

[0048] (2) By introducing two adjustment factors α and β, the threshold function can be flexibly changed in specific power consumption environments, thereby improving the adaptability of the algorithm to the environment.

[0049] (3) The threshold margin is designed as an evaluation index to describe the environmental universality of the fault arc detection algorithm and the signal improvement effect of the noise reduction algorithm.

[0050] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating a current signal noise reduction method for fault arc detection proposed in this invention.

[0052] Figure 2 This is a schematic diagram of the overall process of a current signal noise reduction method for fault arc detection proposed in this invention.

[0053] Figure 3 This is the time-domain waveform of the noise estimation spectrum obtained by active noise estimation in this invention;

[0054] Figure 4 This is a schematic diagram comparing the time-domain waveforms before and after noise reduction of a normal-state resistive load in this invention. (a) is the time-domain waveform before noise reduction of a normal-state resistive load, and (b) is the time-domain waveform after noise reduction of a normal-state resistive load.

[0055] Figure 5This is a schematic diagram comparing the time-domain waveforms before and after noise reduction of the fault-state resistive load in this invention. (a) is the time-domain waveform before noise reduction of the fault-state resistive load, and (b) is the time-domain waveform after noise reduction of the fault-state resistive load.

[0056] Figure 6 This is a schematic diagram comparing the time-domain waveforms before and after noise reduction of a normal-state inductive load in this invention. (a) is the time-domain waveform before noise reduction of a normal-state inductive load, and (b) is the time-domain waveform after noise reduction of a normal-state inductive load.

[0057] Figure 7 This is a schematic diagram comparing the time-domain waveforms of the fault-state inductive load before and after noise reduction in this invention. (a) is the time-domain waveform before noise reduction of the fault-state inductive load, and (b) is the time-domain waveform after noise reduction of the fault-state inductive load.

[0058] Figure 8 This is a schematic diagram comparing the changes in fault arc characteristics before and after resistive load noise reduction in this invention. (a) is a diagram showing the changes in fault arc characteristics before resistive load noise reduction, and (b) is a diagram showing the changes in fault arc characteristics after resistive load noise reduction.

[0059] Figure 9 This is a schematic diagram comparing the changes in fault arc characteristics before and after inductive load noise reduction in this invention. (a) is a diagram showing the changes in fault arc characteristics before inductive load noise reduction, and (b) is a diagram showing the changes in fault arc characteristics after inductive load noise reduction.

[0060] Figure 10 This is a schematic diagram comparing the changes in threshold margin before and after two types of load noise reduction in this invention;

[0061] Figure 11 This is a schematic diagram of a current signal noise reduction system for fault arc detection proposed in this invention. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] like Figure 1 As shown, the first embodiment of the present invention proposes a current signal noise reduction method for fault arc detection, the method comprising the following steps:

[0064] S10. Collect the current operating current of the load at the current moment and analyze it to obtain the noise estimation spectrum of the current operating current;

[0065] S20. Perform wavelet decomposition on the noise estimation spectrum and the load operating current at subsequent time points to obtain the first wavelet coefficient and the second wavelet coefficient;

[0066] S30. Based on the first wavelet coefficients, the second wavelet coefficients are corrected to obtain the third wavelet coefficients;

[0067] S40. Perform wavelet reconstruction on the third wavelet coefficients to obtain the time-domain enhanced signal.

[0068] This embodiment performs wavelet decomposition on the noise estimation spectrum derived from the current operating current and the operating current at subsequent times. Then, it uses the first wavelet coefficients of the noise estimation spectrum to correct the second wavelet coefficients of the operating current at subsequent times, and performs wavelet reconstruction based on the corrected third wavelet coefficients to obtain the denoised current signal. The improved wavelet threshold function proposed in this invention overcomes the shortcomings of traditional wavelet threshold functions, such as poor continuity and constant deviation. The denoising algorithm reduces the noise content in the signal to be analyzed while protecting and enhancing the arc characteristics of the signal to be analyzed, thereby significantly improving the detection capability of fault arc detection technology.

[0069] In one embodiment, such as Figure 2 As shown, step S10, which involves acquiring the current operating current of the load and analyzing it to obtain the noise estimation spectrum of the current operating current, specifically includes the following steps:

[0070] S11. Use a current acquisition sensor to acquire the current operating current of the load at the current moment;

[0071] S12. Perform fault arc analysis on the current operating current of the load to detect whether a fault arc has occurred. If so, repeat step S11; otherwise, proceed to step S13.

[0072] It should be noted that the detection process for a fault arc can be as follows: Perform wavelet transform on the current signal of the load at the current moment to obtain wavelet coefficients; process the wavelet coefficients to obtain at least two fault indication features; these fault indication features must satisfy the following conditions: when a fault arc occurs, the fault indication features are not affected by non-faulty branches and can be used to determine whether a faulty branch has occurred; if at least two fault indication features meet preset judgment conditions, then a fault arc is determined to have occurred. The specific detection process can be described using the patent application document with publication number CN111707908A; detailed detection procedures will not be elaborated here.

[0073] It should be understood that those skilled in the art can also use other detection methods to detect fault arcs according to actual needs, and this embodiment does not make specific limitations.

[0074] S13. Analyze the noise estimation spectrum of the operating current at the current moment using an active noise estimation algorithm.

[0075] It should be noted that the analysis process for noise estimation spectrum can be as follows: After detecting a fault arc in the operating current, components with frequencies less than 100Hz in the operating current are filtered out to generate higher-frequency background noise; the suppression factor for each frequency of the main circuit current signal is calculated based on the frequency domain energy distribution of the load main circuit current signal and the background noise; the suppression factor is multiplied by the spectrum of the main circuit current signal to obtain the noise-reduced current spectrum. The specific detection process can be found in patent application document CN113311227A; the detailed analysis process will not be elaborated here.

[0076] It should be understood that those skilled in the art can also use other analysis methods to analyze the noise estimation spectrum of the operating current according to actual needs, and this embodiment does not make specific limitations.

[0077] In one embodiment, step S20 involves performing wavelet decomposition on the noise estimation spectrum and the load operating current at subsequent time points to obtain first wavelet coefficients and second wavelet coefficients, specifically:

[0078] The noise estimation spectrum and the load operating current at subsequent time points are respectively subjected to wavelet decomposition based on the dbn wavelet basis to obtain the first wavelet coefficient and the second wavelet coefficient.

[0079] It should be noted that this embodiment can perform 3-level db4 wavelet decomposition on the subsequent input operating current signal and noise estimation spectrum, and extract the first wavelet coefficients corresponding to the noise estimation spectrum and the second wavelet coefficients corresponding to the subsequent operating current.

[0080] It should be understood that those skilled in the art can also select the vanishing moment n of the wavelet function according to actual needs, and this embodiment does not make specific limitations.

[0081] In one embodiment, step S30, which involves correcting the second wavelet coefficients based on the first wavelet coefficients to obtain the third wavelet coefficients, specifically includes the following steps:

[0082] S31. Estimate the threshold of the first wavelet coefficients;

[0083] S32. Construct an inhibition factor matrix based on the threshold;

[0084] S33. Based on the suppression factor matrix, the second wavelet coefficients are corrected to obtain the third wavelet coefficients.

[0085] It should be noted that this embodiment constructs a suppression matrix and uses the suppression factor matrix to multiply the second wavelet coefficients to correct them, performing a proportional reduction rather than a direct subtraction. Therefore, for the time-frequency characteristics of the signal, it is only amplified or reduced, resulting in a linear change. Furthermore, this embodiment first estimates the noise of the operating current and uses the noise estimation result to set the degree of change. Therefore, the reduction is smaller for locations with stronger signals and larger for locations with stronger noise, which amplifies the time-frequency characteristics of the signal.

[0086] In one embodiment, the threshold for estimating the first wavelet coefficients is expressed by the formula:

[0087]

[0088] In the formula: λ j This represents the threshold value of the first wavelet coefficient. Let represent the i-th first wavelet coefficient of the j-th layer decomposition, and N represent the length of the wavelet coefficients of the current layer.

[0089] In one embodiment, the construction of the inhibition factor matrix based on the threshold is expressed by the following formula:

[0090]

[0091] Where: H w Represents the inhibition factor matrix, λ j The threshold value of the first wavelet coefficient is represented by w2(i,j), which represents the i-th second wavelet coefficient of the j-th layer decomposition. α and β are adjustment factors.

[0092] It should be noted that α, located in the denominator of the suppression factor, directly controls the processing intensity of the algorithm. A larger α results in more drastic signal processing. Similarly, β, acting as the order, amplifies the processing intensity as it increases. Generally, α and β can be set to 1.

[0093] This embodiment introduces two adjustment factors, α and β, which can be dynamically adjusted according to the actual application environment, so that the threshold function can be flexibly changed in specific power consumption environments, thereby improving the adaptability of the algorithm to the environment.

[0094] In one embodiment, the third wavelet coefficient is expressed by the formula:

[0095]

[0096] In the formula: H represents the third wavelet coefficient. w Let w2(i,j) represent the suppression factor matrix, and w2(i,j) represent the i-th second wavelet coefficient of the j-th level decomposition.

[0097] It should be noted that the third wavelet coefficient is actually an enhanced result of processing the second wavelet coefficient.

[0098] In one embodiment, after step S30: correcting the second wavelet coefficients based on the first wavelet coefficients to obtain the third wavelet coefficients, the method further includes the following steps:

[0099] Fault arc analysis is performed on the third wavelet signal, and the threshold margin is calculated, expressed by the following formula:

[0100] ThMar=AbNCa maxm -NCa max

[0101] In the formula: ThMar represents the threshold margin, AbNCa maxm NCa represents the m-th largest value of the fault-state arc descriptive characteristic. max This represents the maximum value of the characteristic describing a normal-state electric arc.

[0102] Specifically, the value of m can be 20 or more, and those skilled in the art can choose the value of m according to actual needs.

[0103] Currently, in the field of arc fault detection, there is no suitable evaluation metric for assessing algorithm adaptability. Furthermore, in the less-discussed field of arc fault noise reduction, there is even less of a metric for evaluating the capabilities of noise reduction algorithms. Metrics used in signal processing, such as signal-to-noise ratio (SNR) and mean square error (MSE), are clearly unsuitable for arc fault noise reduction. The goal of noise reduction in arc fault detection is to amplify arc characteristics and reduce noise interference. This embodiment designs a threshold margin as an evaluation metric to describe the environmental universality of arc fault detection algorithms and the signal improvement effect of noise reduction algorithms. The size of the threshold margin indicates the algorithm's environmental adaptability, representing how much redundancy the algorithm has to accommodate environmental interference.

[0104] The following specific example illustrates this solution:

[0105] The normal and fault-state operating currents of resistive and inductive loads under a 50Hz frequency and 220V voltage environment are collected using current transformers. An active noise estimation method is then used to obtain the noise estimation spectrum of the collected operating currents. Figure 3 As shown.

[0106] A 3-level dB4 wavelet decomposition is performed on the subsequent current signal input and the noise estimation spectrum, and the corresponding second wavelet coefficients and first wavelet coefficients are extracted. The threshold of the first wavelet coefficients corresponding to the noise estimation spectrum is estimated, and a suppression factor matrix is ​​constructed using the threshold. The second wavelet coefficients of the input subsequent current signal are then corrected. After processing, wavelet reconstruction is performed to obtain the time-domain enhanced waveform, as shown below. Figures 4-7 As shown.

[0107] Fault arc analysis of the enhanced signal can be performed using the fault arc analysis method proposed in patent application document CN111707908A, and the results are as follows: Figures 8-9 As shown, it can be clearly seen that the normal state signal characteristics decrease significantly, while the fault state signal characteristics increase to a certain extent, which further separates the normal state characteristics from the fault state characteristics, meeting the expectations of fault arc detection technology and improving the detection capability of fault arc detection technology to a certain extent.

[0108] Threshold margin analysis was performed on the operating current signals before and after noise reduction, such as... Figure 10 As shown, for resistive loads, the noise reduction algorithm proposed in this invention improves the threshold margin of the fault arc detection algorithm by 64.69%, and for inductive loads, the noise reduction algorithm proposed in this invention improves the threshold margin of the fault arc detection algorithm by 12.08%.

[0109] In addition, such as Figure 11 As shown, the second embodiment of the present invention proposes a current signal noise reduction system for fault arc detection, the system comprising:

[0110] The acquisition module 10 is used to acquire the current operating current of the load at the current moment and analyze it to obtain the noise estimation spectrum of the current operating current at the current moment;

[0111] Wavelet decomposition module 20 is used to perform wavelet decomposition on the noise estimation spectrum and the load operating current at subsequent time moments to obtain the first wavelet coefficient and the second wavelet coefficient.

[0112] Correction module 30 is used to correct the second wavelet coefficients based on the first wavelet coefficients to obtain the third wavelet coefficients;

[0113] Wavelet reconstruction module 40 is used to perform wavelet reconstruction on the third wavelet coefficients to obtain a time-domain enhanced signal.

[0114] In this embodiment, wavelet decomposition is performed on the noise estimation spectrum analyzed based on the current operating current and the operating current at subsequent times. Then, the wavelet coefficients of the operating current at subsequent times are corrected using the wavelet coefficients of the noise estimation spectrum, and wavelet reconstruction is performed based on the corrected wavelet coefficients to obtain the denoised current signal. The improved wavelet threshold function proposed in this invention overcomes the shortcomings of traditional wavelet threshold functions, such as poor continuity and constant deviation. The current signal processed by the denoising algorithm not only suppresses the presence of noise but also improves the arc characteristics of the signal, thus significantly improving the detection performance of the fault arc detection algorithm.

[0115] In one embodiment, the acquisition module 10 specifically includes:

[0116] A current acquisition unit is used to acquire the current operating current of the load at the current moment using a current acquisition sensor.

[0117] The arc detection unit is used to perform fault arc analysis on the current operating current of the load and detect whether a fault arc has occurred.

[0118] A noise estimation unit is used to analyze the noise estimation spectrum of the operating current at the current moment using an active noise estimation algorithm when the arc detection unit does not detect a fault arc.

[0119] The current acquisition unit is used to reacquire the operating current of the load when the arc detection unit detects a fault arc.

[0120] In one embodiment, the wavelet decomposition module 20 is specifically used for:

[0121] The noise estimation spectrum and the load operating current at subsequent time points are respectively subjected to wavelet decomposition based on the dbn wavelet basis to obtain the first wavelet coefficient and the second wavelet coefficient.

[0122] In one embodiment, the correction module 30 specifically includes:

[0123] A threshold estimation unit is used to estimate the threshold of the first wavelet coefficients;

[0124] Specifically, the threshold formula for the first wavelet coefficient is expressed as:

[0125]

[0126] In the formula: λ j This represents the threshold value of the first wavelet coefficient. Let represent the i-th first wavelet coefficient of the j-th layer decomposition, and N represent the length of the wavelet coefficients of the current layer.

[0127] A matrix construction unit is used to construct an inhibition factor matrix based on the threshold.

[0128] Specifically, the formula for the inhibition factor matrix is ​​expressed as follows:

[0129]

[0130] Where: H w Represents the inhibition factor matrix, λ j The threshold value of the first wavelet coefficient is represented by w2(i,j), which represents the i-th second wavelet coefficient of the j-th layer decomposition. α and β are adjustment factors.

[0131] The correction unit is used to correct the second wavelet coefficients based on the suppression factor matrix to obtain the third wavelet coefficients.

[0132] Specifically, the formula for the third wavelet coefficient is expressed as follows:

[0133]

[0134] In the formula: H represents the third wavelet coefficient. w Let w2(i,j) represent the suppression factor matrix, and w2(i,j) represent the i-th second wavelet coefficient of the j-th level decomposition.

[0135] In one embodiment, the system further includes a threshold margin calculation module, used for:

[0136] Fault arc analysis is performed on the third wavelet signal, and the threshold margin is calculated, expressed by the following formula:

[0137] ThMar=AbNCa maxm -NCa max

[0138] In the formula: ThMar represents the threshold margin, AbNCa maxm NCa represents the m-th largest value of the fault-state arc descriptive characteristic. max This represents the maximum value of the characteristic describing a normal-state electric arc.

[0139] It should be noted that other embodiments or implementation methods of the current signal noise reduction system for fault arc detection described in this invention can refer to the above-described method embodiments, and will not be repeated here.

[0140] Furthermore, the third embodiment of the present invention also proposes a computer-readable storage medium storing a computer program thereon, wherein when the computer program is executed by a processor, it implements the current signal noise reduction method for fault arc detection as described in the first embodiment above.

[0141] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0142] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0143] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0144] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for noise reduction of current signals for fault arc detection, characterized in that, The method includes: The current operating current of the load is collected at the current moment, and the noise estimation spectrum of the current operating current at the current moment is obtained by analysis; Wavelet decomposition is performed on the noise estimation spectrum and the load operating current at subsequent time points to obtain the first wavelet coefficient and the second wavelet coefficient; The third wavelet coefficient is obtained by correcting the second wavelet coefficient based on the first wavelet coefficient, including estimating a threshold for the first wavelet coefficient; constructing a suppression factor matrix based on the threshold; and correcting the second wavelet coefficient based on the suppression factor matrix to obtain the third wavelet coefficient, wherein the threshold for estimating the first wavelet coefficient is expressed by the formula: In the formula: This represents the threshold value of the first wavelet coefficient. Indicates the first j The first layer of decomposition i The first wavelet coefficient, Indicates the length of the wavelet coefficients of the current layer; The inhibition factor matrix constructed based on the threshold is expressed by the following formula: In the formula: Represents the inhibition factor matrix, This represents the threshold value of the first wavelet coefficient. Indicates the first j The first layer of decomposition i The second wavelet coefficient, , For adjustment factors; The third wavelet coefficient is obtained by correcting the second wavelet coefficient based on the suppression factor matrix, as expressed by the formula: In the formula: This represents the third wavelet coefficient. This represents the inhibition factor matrix. Indicates the first j The first layer of decomposition i The second wavelet coefficient; Wavelet reconstruction is performed on the third wavelet coefficients to obtain the time-domain enhanced signal.

2. The current signal noise reduction method for fault arc detection as described in claim 1, characterized in that, The process of acquiring the current operating current of the load and analyzing the noise estimation spectrum of the current operating current includes: The current operating current of the load at the current moment is collected using a current acquisition sensor; Perform fault arc analysis on the current operating current of the load to detect whether a fault arc has occurred; If so, then re-collect the operating current of the load; If not, then the noise estimation spectrum of the operating current at the current moment is analyzed using an active noise estimation algorithm.

3. The current signal noise reduction method for fault arc detection as described in claim 1, characterized in that, The step of performing wavelet decomposition on the noise estimation spectrum and the load operating current at subsequent time points to obtain the first wavelet coefficients and the second wavelet coefficients includes: The noise estimation spectrum and the load operating current at subsequent time points are respectively subjected to wavelet decomposition based on the dbn wavelet basis to obtain the first wavelet coefficient and the second wavelet coefficient.

4. The current signal noise reduction method for fault arc detection as described in claim 1, characterized in that, After correcting the second wavelet coefficients based on the first wavelet coefficients to obtain the third wavelet coefficients, the method further includes: Fault arc analysis is performed on the third wavelet coefficients, and the threshold margin is calculated, expressed by the formula: In the formula: Indicates threshold margin, The first characteristic representing the fault-state arc description feature m Large value, This represents the maximum value of the characteristic describing a normal-state electric arc.

5. A current signal noise reduction system for fault arc detection, characterized in that, For implementing the current signal noise reduction method for fault arc detection as described in any one of claims 1-4, the system comprises: The acquisition module is used to acquire the current operating current of the load at the current moment and analyze it to obtain the noise estimation spectrum of the current operating current at the current moment; The wavelet decomposition module is used to perform wavelet decomposition on the noise estimation spectrum and the load operating current at subsequent time points to obtain the first wavelet coefficient and the second wavelet coefficient. The correction module is used to correct the second wavelet coefficients based on the first wavelet coefficients to obtain the third wavelet coefficients; The wavelet reconstruction module is used to reconstruct the third wavelet coefficients to obtain a time-domain enhanced signal.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the current signal noise reduction method for fault arc detection as described in any one of claims 1-4.

Images