Power quality signal noise reduction method and device, equipment and storage medium

Abstract

The invention provides an electric energy quality signal noise reduction method and device, equipment and a storage medium, and the method comprises the steps: carrying out the signal decomposition of an electric energy quality signal, and obtaining a plurality of candidate sub-signals; entropy discrimination is carried out on the candidate sub-signals, and an entropy value corresponding to each candidate sub-signal is obtained; obtaining a first sub-signal of which the entropy value is less than or equal to a first threshold value and a second sub-signal of which the entropy value is greater than the first threshold value and less than or equal to a second threshold value in the candidate sub-signals; performing noise reduction processing on the second sub-signal by adopting multi-resolution singular value decomposition to obtain a third sub-signal; wherein the number of layers of multi-resolution singular value decomposition is determined based on the singular value change rate; and performing signal recombination on the first sub-signal and the third sub-signal to obtain a target signal. According to the technical scheme, the noise reduction effect of the electric energy quality signal can be improved.

Description

Technical Field

[0001] The present application relates to the field of signal processing technology, and in particular to a method, device, equipment and storage medium for reducing noise of power quality signals. Background Art

[0002] The power quality signals collected from hydropower plants are a superposition of real signals and noise, with the noise often exhibiting Gaussian white noise characteristics. Traditional signal noise reduction algorithms can mistakenly filter out disturbance signals such as high-frequency, low-amplitude higher-order harmonics or transient oscillations. Therefore, achieving maximum noise reduction while preserving the original valid signal is crucial. Furthermore, for power quality signals affected by noise, it is crucial to avoid over-reduction by the noise reduction algorithm, preserving as much detail as possible from the original disturbance. Summary of the Invention

[0003] The present application aims to solve one of the technical problems in the related art at least to a certain extent.

[0004] In the first aspect, the present application proposes a method for denoising a power quality signal, the method comprising: performing signal decomposition on a power quality signal to obtain multiple candidate sub-signals; performing entropy discrimination on the candidate sub-signals to obtain an entropy value corresponding to each candidate sub-signal; obtaining a first sub-signal in the candidate sub-signals whose entropy value is less than or equal to a first threshold, and a second sub-signal whose entropy value is greater than the first threshold and less than or equal to a second threshold; wherein the first threshold is less than the second threshold; performing denoising processing on the second sub-signal using multi-resolution singular value decomposition to obtain a third sub-signal; wherein the number of layers of the multi-resolution singular value decomposition is determined based on the singular value change rate; and performing signal recombination on the first sub-signal and the third sub-signal to obtain a target signal.

[0005] In one implementation, performing entropy discrimination on the candidate sub-signals to obtain an entropy value corresponding to each candidate sub-signal includes: constructing an M-dimensional first vector and an M+1-dimensional second vector based on each candidate sub-signal; obtaining the distance between every two vectors in the first vector to obtain multiple first distances corresponding to each candidate sub-signal; obtaining the distance between every two vectors in the second vector to obtain multiple second distances corresponding to each candidate sub-signal; obtaining a first similarity ratio corresponding to each first vector based on the first distance and a preset similarity tolerance, and obtaining a second similarity ratio corresponding to each second vector based on the second distance and the similarity tolerance; obtaining a first mean of the first similarity ratio corresponding to each candidate sub-signal and a second mean of the second similarity ratio corresponding to each candidate sub-signal; and obtaining an entropy value corresponding to each candidate sub-signal based on the first mean and the second mean corresponding to each candidate sub-signal.

[0006] In one implementation, the denoising process for the second sub-signal using multi-resolution singular value decomposition to obtain the third sub-signal includes: performing multi-layer singular value decomposition on the second sub-signal to obtain an approximate signal and a detail signal corresponding to each layer of singular value decomposition; obtaining a first approximate signal, a first detail signal, a second approximate signal, and a second detail signal from the approximate signal and the detail signal; wherein the first approximate signal and the first detail signal are the approximate signal and the detail signal corresponding to the second-to-last layer of singular value decomposition, and the second approximate signal and the second detail signal are the approximate signal and the detail signal corresponding to the last layer of singular value decomposition; obtaining an approximate singular value relative change rate based on the first approximate signal and the second approximate signal, and obtaining a detail singular value relative change rate based on the first detail signal and the second detail signal; and in response to the approximate singular value relative change rate being less than or equal to a first threshold and the detail singular value relative change rate being less than or equal to a second threshold, obtaining the third sub-signal based on the other approximate signals except the second approximate signal.

[0007] In an optional implementation, the method further includes: in response to the relative change rate of the approximate singular value being greater than or equal to a first threshold, and / or the relative change rate of the detail singular value being greater than or equal to a second threshold, adding 1 to the number of layers of the singular value decomposition, and returning to execute the steps of obtaining the relative change rate of the approximate singular value based on the first approximate signal and the second approximate signal, and obtaining the relative change rate of the detail singular value based on the first detail signal and the second detail signal.

[0008] In the second aspect, the present application proposes a power quality signal noise reduction device, which includes: a signal decomposition module for performing signal decomposition on the power quality signal to obtain multiple candidate sub-signals; a first processing module for performing entropy discrimination on the candidate sub-signals to obtain an entropy value corresponding to each candidate sub-signal; a second processing module for obtaining a first sub-signal in the candidate sub-signals whose entropy value is less than or equal to a first threshold, and a second sub-signal whose entropy value is greater than the first threshold and less than or equal to a second threshold; wherein the first threshold is less than the second threshold; a signal noise reduction module for performing noise reduction processing on the second sub-signal using multi-resolution singular value decomposition to obtain a third sub-signal; wherein the number of layers of the multi-resolution singular value decomposition is determined based on the singular value change rate; a signal recombination module for performing signal recombination on the first sub-signal and the third sub-signal to obtain a target signal.

[0009] In one implementation, the first processing module can be used to: construct an M-dimensional first vector and an M+1-dimensional second vector based on each candidate sub-signal respectively; obtain the distance between each two vectors in the first vector to obtain multiple first distances corresponding to each candidate sub-signal; obtain the distance between each two vectors in the second vector to obtain multiple second distances corresponding to each candidate sub-signal; obtain a first similarity ratio corresponding to each first vector based on the first distance and a preset similarity tolerance, and obtain a second similarity ratio corresponding to each second vector based on the second distance and the similarity tolerance; obtain a first mean of the first similarity ratio corresponding to each candidate sub-signal and a second mean of the second similarity ratio corresponding to each candidate sub-signal; and obtain an entropy value corresponding to each candidate sub-signal based on the first mean and the second mean corresponding to each candidate sub-signal.

[0010] In one implementation, the signal denoising module can be used to: perform multi-layer singular value decomposition on the second sub-signal to obtain an approximate signal and a detail signal corresponding to each layer of singular value decomposition; obtain a first approximate signal, a first detail signal, a second approximate signal and a second detail signal from the approximate signal and the detail signal; wherein the first approximate signal and the first detail signal are the approximate signal and the detail signal corresponding to the second-to-last layer of singular value decomposition, and the second approximate signal and the second detail signal are the approximate signal and the detail signal corresponding to the last layer of singular value decomposition; obtain an approximate singular value relative change rate based on the first approximate signal and the second approximate signal, and obtain a detail singular value relative change rate based on the first detail signal and the second detail signal; in response to the approximate singular value relative change rate being less than or equal to a first threshold, and the detail singular value relative change rate being less than or equal to a second threshold, obtain the third sub-signal based on the other approximate signals except the second approximate signal.

[0011] In an optional implementation, the signal denoising module can also be used to: in response to the relative change rate of the approximate singular value being greater than or equal to a first threshold, and / or the relative change rate of the detail singular value being greater than or equal to a second threshold, increase the number of layers of the singular value decomposition by 1, and return to execute the steps of obtaining the relative change rate of the approximate singular value based on the first approximate signal and the second approximate signal, and obtaining the relative change rate of the detail singular value based on the first detail signal and the second detail signal.

[0012] In a third aspect, the present application provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the power quality signal denoising method according to the first aspect.

[0013] In a fourth aspect, the present application provides a computer-readable storage medium storing instructions that, when executed, cause the method according to the first aspect to be implemented.

[0014] In a fifth aspect, the present application provides a computer program product comprising a computer program that, when executed by a processor, implements the steps of the power quality signal denoising method according to the first aspect.

[0015] The power quality signal denoising method, device, equipment and storage medium provided by the present application can classify candidate sub-signals based on the entropy values of the candidate sub-signals obtained by decomposing the power quality signal, obtain pure first sub-signals and second sub-signals containing noise components, and denoise the second sub-signals to obtain third sub-signals, so as to recombine the first sub-signals and the third sub-signals to obtain target signals. The denoising effect of the power quality signal can be improved.

[0016] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0017] The above and / or additional aspects and advantages of the present application will become apparent and be readily appreciated from the following description, including the accompanying drawings, wherein:

[0018] Figure 1 is a flowchart of a power quality signal denoising method provided by an embodiment of the present application;

[0019] Figure 2 is a flowchart of another power quality signal denoising method provided by an embodiment of the present application;

[0020] Figure 3 is a flowchart of still another power quality signal denoising method provided by an embodiment of the present application;

[0021] Figure 4 is a MRSVD decomposition diagram provided by an embodiment of the present application;

[0022] Figure 5 is a denoised composite disturbance signal diagram provided by the present application;

[0023] Figure 6 This is a schematic diagram of ICEEMDAN decomposition of a noise-contaminated composite disturbance signal provided by this application;

[0024] Figure 7 It is a singular value sequence diagram provided in an embodiment of the present application;

[0025] Figure 8 This is a schematic diagram of the relative rate of change of singular values ​​provided in an embodiment of the present application;

[0026] Figure 9 This is a schematic diagram comparing signal noise reduction algorithms provided in an embodiment of the present application;

[0027] Figure 10 This is a schematic diagram comparing a noise-reduced signal and a pure signal without noise provided in an embodiment of the present application;

[0028] Figure 11 Schematic diagram of a power quality sampling signal noise reduction solution provided in an embodiment of the present application;

[0029] Figure 12 This is a structural diagram of a power quality signal noise reduction device provided in an embodiment of the present application;

[0030] Figure 13 It is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0032] The following describes a method and apparatus for reducing noise in a power quality signal according to an embodiment of the present application with reference to the accompanying drawings.

[0033] Figure 1 This is a flow chart of a method for reducing noise in power quality signals provided by an embodiment of the present application. Figure 1 As shown, the method may include but is not limited to the following steps:

[0034] Step S101: performing signal decomposition on a power quality signal to obtain a plurality of candidate sub-signals.

[0035] Exemplarily, a power quality signal of a hydropower complementary power plant is obtained, and the power quality signal is decomposed using the ICEEMDAN (Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise) method to obtain multiple candidate sub-signals.

[0036] In an optional implementation, the signal decomposition step may be as follows:

[0037] Step A1: Transform different zero-mean unit covariance Gaussian white noise signals n i (t) is added to the power quality signal x(t) to construct a new signal X i (t):

[0038] X i (t) = x(t) + e0E1(n i (t))

[0039]

[0040] where i represents the number of times noise is added (i = 1, 2, ... N), e0 represents the operation term used to reduce noise in the initial stage, ε0 represents the inverse of the expected signal-to-noise ratio between the first added noise and the original signal, std represents the standard deviation, and E1(·) represents the first IMF value of the calculated signal.

[0041] Step A2: Calculate X i The first residual signal r1(t) can be obtained by taking the average of the local means obtained by (t):

[0042] r1(t)= <M(X i (t))>

[0043] Among them, <·> represents the calculation of the overall mean, and M(·) represents the operation of generating the local mean.

[0044] Step A3: Subtract the first residual signal r1(t) from the original signal x(t) to obtain the first IMF value of the signal:

[0045] IMF1(t)=x(t)-r1(t)

[0046] Step A4: Use (r1(t)+e1E2(n i (t))) the average of the local means, calculate the second residual signal r2(t), and the second IMF value:

[0047] r2(t)= <M(r1(t)+e1E2(ni (t))).

[0048] IMF2(t) = r2(t) - r1(t)

[0049] where e1 = ε1std(x(t)) / std(E2(n i (t))).

[0050] Step A5: According to the above calculation, the decomposition state when j = 3, 4, … can be calculated:

[0051] r j (t) = <M(r j-1 (t) + e j-1 E j (n i (t))).

[0052] IMF j (t) = r j (t) - r j-1 (t)

[0053] where e j-1 = ε j-1 std(x(t)) / std(E j (n i (t))

[0054] Step S102: Entropy discrimination is performed on the candidate sub-signals to obtain an entropy value corresponding to each candidate sub-signal.

[0055] Exemplarily, SE (Sample Entropy) is used to respectively discriminate the noise content of each candidate sub-signal to obtain an entropy value corresponding to each candidate sub-signal.

[0056] It can be understood that the size of the signal SE value can reflect the strength of the randomness of the signal, and the larger the signal SE value, the higher the randomness and the greater the proportion of noise of the signal; the smaller the signal SE value, the stronger the regularity and the smaller the proportion of noise.

[0057] Step S103: A first sub-signal with an entropy value less than or equal to a first threshold value and a second sub-signal with an entropy value greater than the first threshold value and less than or equal to a second threshold value are obtained from the candidate sub-signals.

[0058] In the embodiments of the present application, the first threshold value is less than the second threshold value.

[0059] As an example, refer to Table 1, which is an example table of sub-signal category division provided by the embodiments of the present application.

[0060] Table 1 Example Table of Sub-signal Category Division

[0061]

[0062] As shown in Table 1, thresholds can be used to distinguish the decomposed sub-signals into three categories: pure signal components, useful signals with noise components, and noise signal components. The pure signal components correspond to the first sub-signal, and the useful signal with noise components corresponds to the second sub-signal.

[0063] Step S104: performing noise reduction processing on the second sub-signal using multi-resolution singular value decomposition to obtain a third sub-signal.

[0064] The number of layers of multi-resolution singular value decomposition is determined based on the rate of change of the singular values.

[0065] Exemplarily, multi-resolution singular value decomposition is used to perform noise reduction processing on the second sub-signal to further remove the noise component in the second sub-signal and obtain a third sub-signal corresponding to the second sub-signal.

[0066] Step S105: performing signal recombination based on the first sub-signal and the third sub-signal to obtain a target signal.

[0067] By implementing the embodiments of the present application, candidate sub-signals obtained by power quality signal decomposition can be classified based on their corresponding entropy values, obtaining a pure first sub-signal and a second sub-signal containing noise components. The second sub-signal is then subjected to noise reduction to obtain a third sub-signal, and the first and third sub-signals are then recombined to obtain a target signal. This can improve the noise reduction effect of the power quality signal.

[0068] In one implementation, a vector can be constructed based on the candidate sub-signals to obtain the corresponding entropy value based on the vector corresponding to the sub-signal. As an example, see Figure 2 , Figure 2 FIG. 1 is a flow chart of another method for reducing noise of power quality signals provided in an embodiment of the present application. Figure 2 As shown, the method may include but is not limited to the following steps:

[0069] Step S201: performing signal decomposition on the power quality signal to obtain multiple candidate sub-signals.

[0070] In the embodiment of the present application, step S201 can be implemented by any of the methods in the embodiments of the present application. The embodiments of the present application do not limit this and will not be described in detail.

[0071] Step S202: constructing an M-dimensional first vector and an M+1-dimensional second vector based on each candidate sub-signal.

[0072] For example, for a candidate sub-signal of length N, the M-dimensional first vector Xi It can be expressed as follows:

[0073] X i =[x(i)x(i+1)…x(i+m-1)]

[0074] Wherein, i=1, 2, ..., N-m+1, and a total of (N-m+1) first vectors are constructed.

[0075] Step S203: Obtain the distance between every two vectors in the first vector to obtain a plurality of first distances corresponding to each candidate sub-signal.

[0076] For example, for any first vector, any two vectors X i With X j Distance between It can be expressed as follows:

[0077]

[0078] Step S204: Obtain the distance between every two vectors in the second vector to obtain multiple second distances corresponding to each candidate sub-signal.

[0079] Exemplarily, the method for acquiring the second distance may be the same as the method for acquiring the first distance.

[0080] Step S205: obtaining a first similarity ratio corresponding to each first vector based on the first distance and the preset similarity tolerance, and obtaining a second similarity ratio corresponding to each second vector based on the second distance and the similarity tolerance.

[0081] In an optional implementation, the similarity tolerance r may be determined according to the signal standard deviation.

[0082] Exemplarily, the similarity tolerance may be 0.1 to 0.25 times the signal standard deviation.

[0083] Step S206: Obtain a first mean value of the first similarity ratio corresponding to each candidate sub-signal and a second mean value of the second similarity ratio corresponding to each candidate sub-signal.

[0084] For example, the first mean value can be expressed as follows:

[0085]

[0086] Step S207: Based on the first mean and the second mean, obtain the entropy value corresponding to each candidate sub-signal.

[0087] For example, the sample entropy corresponding to the candidate sub-signal can be expressed as:

[0088]

[0089] When N is a finite number, the above formula can be expressed as:

[0090]

[0091] Step S208: Obtain a first sub-signal with an entropy value less than or equal to a first threshold value, and a second sub-signal with an entropy value greater than the first threshold value and less than or equal to a second threshold value, from the candidate sub-signals.

[0092] In embodiments of the present application, step S208 can be implemented by any of the embodiments of the present application, and the embodiments of the present application do not limit this and will not be repeated.

[0093] Step S209: Perform noise reduction processing on the second sub-signal using multi-resolution singular value decomposition to obtain a third sub-signal.

[0094] In embodiments of the present application, step S209 can be implemented by any of the embodiments of the present application, and the embodiments of the present application do not limit this and will not be repeated.

[0095] Step S2010: Perform signal recombination on the first sub-signal and the third sub-signal to obtain a target signal.

[0096] In embodiments of the present application, step S2010 can be implemented by any of the embodiments of the present application, and the embodiments of the present application do not limit this and will not be repeated.

[0097] By implementing the embodiments of the present application, the adaptive noise reduction algorithm based on multi-resolution singular value decomposition can be used to perform noise reduction processing on the sub-signal containing noise components. The noise reduction effect on the power quality signal can be improved.

[0098] In one implementation, the number of layers of multi-resolution singular value decomposition can be determined according to the singular value change rate in the multi-resolution singular value decomposition process. As an example, please refer to Figure 3 , Figure 3 is another flowchart of a power quality signal noise reduction method provided by an embodiment of the present application. As shown in Figure 3 , the method can include but is not limited to the following steps:

[0099] Step S301: Perform signal decomposition on the power quality signal to obtain a plurality of candidate sub-signals.

[0100] In embodiments of the present application, step S301 can be implemented by any of the embodiments of the present application, and the embodiments of the present application do not limit this and will not be repeated.

[0101] Step S302: performing entropy discrimination on the candidate sub-signals to obtain an entropy value corresponding to each candidate sub-signal.

[0102] In the embodiment of the present application, step S302 can be implemented by any of the methods in the embodiments of the present application. The embodiments of the present application do not limit this and will not be described in detail.

[0103] Step S303: obtaining a first sub-signal having an entropy value less than or equal to a first threshold and a second sub-signal having an entropy value greater than the first threshold and less than or equal to a second threshold from among the candidate sub-signals.

[0104] In the embodiment of the present application, step S303 can be implemented by any of the methods in the embodiments of the present application. The embodiments of the present application do not limit this and will not be described in detail.

[0105] Step S304: performing multi-layer singular value decomposition on the second sub-signal to obtain an approximate signal and a detail signal corresponding to each layer of singular value decomposition.

[0106] For example, the steps of multi-layer singular value decomposition can be expressed as follows:

[0107] Step B1: For the second sub-signal x=(x1, x2, x3…x N ), and reconstruct it to obtain the trajectory matrix A.

[0108]

[0109] Step B2: Perform SVD decomposition on A to obtain singular values ​​σ1 and σ2, with σ1>σ2, and its diagonal matrix D = [diag(σ1,σ2),0].

[0110] Step B3: Assuming σ2 = 0, then D = [diag(σ1), 0], and a new trajectory matrix A0 is generated. From A0, an approximate component and a detail component D0 can be obtained.

[0111] Step B4: retain the detail component D0 and bring the approximate component A0 into step B1, and perform continuous iterations to perform multi-layer singular value decomposition.

[0112] Step S305: obtaining a first approximate signal, a first detail signal, a second approximate signal, and a second detail signal from the approximate signal and the detail signal.

[0113] The first approximate signal and the first detail signal are the approximate signal and the detail signal corresponding to the penultimate layer of singular value decomposition, and the second approximate signal and the second detail signal are the approximate signal and the detail signal corresponding to the last layer of singular value decomposition.

[0114] As an example, see Figure 4 ,Figure 4 This is a schematic diagram of MRSVD decomposition provided in an embodiment of the present application. Figure 4 Where Aj is the approximate signal obtained by decomposition at the jth layer, and Dj is the detail signal obtained by decomposition at the jth layer. A first approximate signal and a first detail signal obtained by decomposition at the j-1th layer, and a second approximate signal and a second detail signal obtained by decomposition at the jth layer are obtained.

[0115] Step S306: obtaining relative change rates of approximate singular values ​​based on the first approximate signal and the second approximate signal, and obtaining relative change rates of detail singular values ​​based on the first detail signal and the second detail signal.

[0116] In an optional implementation, the calculation formula for the relative rate of change of the approximate singular value of the approximate signal obtained by the J-th layer decomposition can be expressed as follows:

[0117]

[0118] Among them, σ x,J is the approximate singular value obtained by decomposition at the Jth layer, σ x,J+1 Approximate singular values ​​obtained from the J+1th layer decomposition.

[0119] In an optional implementation, the calculation formula for the relative change rate of the detail singular value of the detail signal obtained by the J-th layer decomposition can be expressed as follows:

[0120]

[0121] Among them, ξ x,J is the detail singular value obtained by decomposition at the Jth layer, ξ x,J+1 The detail singular values ​​obtained by decomposition at the J+1th layer.

[0122] Step S307: in response to the relative change rate of the approximate singular value being less than or equal to the first threshold and the relative change rate of the detail singular value being less than or equal to the second threshold, obtaining a third sub-signal based on other approximate signals except the second approximate signal.

[0123] As the number of decomposition layers increases, MOCσ J Will form an approximate singular value relative change rate sequence B1=[MOCσ1,MOCσ2,MOCσ3,MOCσ4....MOCσ k ];MOCξ J Will form the detail singular value relative change rate sequence B2=[MOCξ1,MOCξ2,MOCξ3,MOCξ4....MOCξ k ], where k = 1, 2, 3, 4…J. If the above formula satisfies the following constraints:

[0124] min Js.t.MOCσ J≤ε

[0125] min Js.t.MOCξ J ≤ε

[0126] Here, ε is the determined threshold. A value below the threshold indicates that the approximate singular value and the detail singular value almost no longer decrease, and the noise energy is almost 0. The first decomposition scale below the threshold is the optimal decomposition scale.

[0127] Exemplarily, in response to the relative change rate of the approximate singular value being less than or equal to the first threshold, and the relative change rate of the detail singular value being less than or equal to the second threshold, the approximate signals obtained by the singular value decomposition of layers other than the last layer are combined to obtain a third sub-signal.

[0128] In other embodiments, in response to the relative change rate of the approximate singular value being greater than or equal to a first threshold, and / or the relative change rate of the detail singular value being greater than or equal to a second threshold, the number of layers of singular value decomposition is increased by 1, and the steps of obtaining the relative change rate of the approximate singular value based on the first approximate signal and the second approximate signal, and obtaining the relative change rate of the detail singular value based on the first detail signal and the second detail signal are returned to be executed.

[0129] Exemplarily, in response to the relative rate of change of the approximate singular value being greater than or equal to the first threshold, another layer of singular value decomposition is performed on the second sub-signal, and the process returns to step S305 .

[0130] Exemplarily, in response to the detail singular value relative change rate being greater than or equal to the second threshold, another layer of singular value decomposition is performed on the second sub-signal, and the process returns to step S305.

[0131] Exemplarily, in response to the relative change rate of the approximate singular value being greater than or equal to the first threshold and the relative change rate of the detail singular value being greater than or equal to the second threshold, another layer of singular value decomposition is performed on the second sub-signal, and the process returns to step S305.

[0132] Step S308: performing signal recombination on the first sub-signal and the third sub-signal to obtain a target signal.

[0133] In the embodiment of the present application, step S308 can be implemented by any of the methods in the embodiments of the present application. The embodiments of the present application do not limit this and will not be described in detail.

[0134] By implementing the embodiments of the present application, candidate sub-signals obtained by power quality signal decomposition can be classified based on their corresponding entropy values, obtaining a pure first sub-signal and a second sub-signal containing noise components. The second sub-signal is then subjected to noise reduction to obtain a third sub-signal, and the first and third sub-signals are then recombined to obtain a target signal. This can improve the noise reduction effect of the power quality signal.

[0135] The following describes the power quality signal noise reduction method provided by this application in detail with reference to specific embodiments:

[0136] Taking the composite disturbance composed of common voltage flicker and 3rd, 5th, 7th, and 9th harmonics in the power grid as an example, the power quality signal can be expressed as:

[0137] u(t)=[1+αsin(βwt)]sin(wt)+0.38sin(3wt)+0.3sin(5wt)+0.25sin(7wt)+0.2sin(9wt)

[0138] Where α = 0.45, β = 0.2, w = 2πf0, f0 = 50Hz, the sampling frequency is 6400Hz, and the continuous acquisition is 0.5s. Figure 5 , Figure 5 This is a schematic diagram of a noise-stained composite disturbance signal provided by this application. Figure 5 As shown in Figure 2, after adding 20dB Gaussian white noise to the original signal, the signal is severely distorted and the original signal cannot be distinguished. Then, ICEEMDAN is used to process and decompose u(t). Figure 6 , Figure 6 This is a schematic diagram of ICEEMDAN decomposition of a noise-contaminated composite disturbance signal provided by this application. Figure 6 As shown in Figure 2, the noise-composite disturbance signal is decomposed into 10 modes, and the sample entropy of the 10 modes is shown in Table 2.

[0139] Table 2 Sub-signal sample entropy value table

[0140]

[0141] from Figure 6 As can be seen, the modal signals further down the spectrum change more gradually, becoming smoother and less noisy. The modal signals further up the spectrum change more dramatically, with more "burrs" present. IMF1 is almost entirely noise. Table 2 shows that the sample entropy values ​​decrease gradually, with IMF1 being the largest and IMF10 being the smallest. This indicates that the noise content also decreases gradually. IMF1 has the highest noise content, followed by IMF2, which is significantly higher than the other modal components. The sample entropies of IMF3-IMF10 are all less than 0.1, so they are considered pure signal components and retained intact.

[0142] The changing pattern of sample entropy values ​​in Table 2 is Figure 6 This corresponds to the information observed in , which further illustrates the rationality of using sample entropy to determine the modal noise content.

[0143] As shown in Table 2, the sample entropy of IMF1 is greater than 2. The set threshold can classify IMF1 as an invalid noise component and discard it. The sample entropy of IMF2 is between 1 and 2, which is considered to be a useful signal containing noise components. The adaptive multi-resolution singular value decomposition proposed in this application is used to reduce noise. Figure 7 and Figure 8 , Figure 7 is a singular value sequence diagram provided in an embodiment of the present application, Figure 8 This is a schematic diagram of the relative change rate of singular values ​​provided in the embodiment of the present application. Figure 7 and Figure 8 As shown in , let ε = 0.04, and the optimal decomposition scale of the multi-resolution singular value decomposition determined thereby is 16. Figure 8 It can be observed that the two singular values ​​change rapidly at the beginning, indicating that the energy of the noise decreases rapidly at the beginning, and then gradually slows down with the increase of J and gradually approaches a straight line, indicating that the noise removal tends to be stable, which is consistent with the multi-resolution singular value denoising principle analyzed above.

[0144] The signal after IMF2 denoising is superimposed with IMF3-IMF10 to obtain the final denoised signal. To verify the performance of the denoising algorithm proposed in this application, it was compared with wavelet soft / hard threshold denoising, wavelet packet denoising, and ICEEMDAN denoising. For wavelet packet denoising, this application conducted multiple comparative experiments on the combination of basis functions and decomposition levels, and found that the "sym6" wavelet basis was the best for two-layer wavelet packet decomposition. Similarly, for wavelet soft / hard threshold denoising, the "sym6" wavelet basis was selected for two-layer decomposition, resulting in the best denoising effect. See [1] for more information. Figure 9 and Figure 10 , Figure 9 is a schematic diagram comparing a signal noise reduction algorithm provided in an embodiment of the present application. Figure 10 This is a schematic diagram comparing a noise-reduced signal and a pure signal without noise provided in an embodiment of the present application.

[0145] The performance of an algorithm model needs to be measured using universal quantitative metrics. This application uses the signal-to-noise ratio (SNR) and root mean square error (RMSE). A higher SNR indicates better noise reduction, while a lower RMSE indicates higher noise reduction performance. Denoising algorithms always strive for both a higher SNR and a lower RMSE.

[0146]

[0147] In the above formula, s(i) is the original signal, For the denoised signal, the SNR and RMSE of various denoising algorithms are calculated as shown in Table 3.

[0148] Table 3 Comparison examples of different noise reduction algorithms

[0149]

[0150] from Figure 10 It can be observed that the dual denoising algorithm of ICEEMDAN and adaptive MRSVD proposed in this application has a good denoising effect and almost overlaps with the original pure signal. Figure 9 As can be seen from the comparison, compared with other denoising algorithms, the algorithm proposed in this application is closer to the original signal after denoising. The indicators in the above table also demonstrate the superiority of the denoising algorithm proposed in this application. Compared with other denoising algorithms, it has the highest signal-to-noise ratio and the lowest root mean square error. Compared with the single ICEEMDAN denoising algorithm, the adaptive MRSVD improves the signal-to-noise ratio and reduces the root mean square error.

[0151] Random Gaussian white noise with different noise intensities is introduced into u(t). Compared with the comparative denoising algorithm mentioned above, the SNR and RMSE results after denoising are shown in Table 3.

[0152] Table 4 Comparison of noise reduction algorithms in different signal-to-noise ratio environments

[0153]

[0154] As can be seen from the table above, the denoising algorithm proposed in the technical solution of this application has the highest SNR and the lowest RMSE in various noise environments, and has good performance under various noise intensities. Compared with ICEEMDAN denoising, the denoising performance is improved, which demonstrates the effectiveness of multiple denoising. Compared with the wavelet series denoising algorithm, ICEEMDAN does not require the selection of basis functions and decomposition levels, and can complete the signal decomposition based on the characteristics of the signal itself. In addition, the algorithm proposed in this application can adaptively determine multi-resolution singular values ​​without knowing prior information about the noise.

[0155] Please attend Figure 11 , Figure 11 Schematic diagram of a power quality sampling signal noise reduction solution provided by an embodiment of the present application. Figure 11 As shown in the figure, the original sampled signal is first adaptively decomposed into several sub-signals using ICEEMDAN based on its own characteristics. The sample entropy value of each sub-signal is calculated, and the sub-signals are divided into three categories based on the sample entropy threshold mentioned above: pure signal components, useful signal components containing noise signals, and noise-invalid signal components. The first category of signals is retained, the third category of signals is discarded, and the second category of signals is further denoised using the adaptive MRSVD algorithm mentioned above. The MRSVD-denoised signal and the first category of signals are recombined.

[0156] See Figure 12 , Figure 12 This is a schematic diagram of the structure of a power quality signal noise reduction device provided in an embodiment of the present application. Figure 12 As shown, the device 1200 includes: a signal decomposition module 1201, which is used to decompose the power quality signal to obtain multiple candidate sub-signals; a first processing module 1202, which is used to perform entropy discrimination on the candidate sub-signals to obtain an entropy value corresponding to each candidate sub-signal; a second processing module 1203, which is used to obtain a first sub-signal whose entropy value is less than or equal to a first threshold, and a second sub-signal whose entropy value is greater than the first threshold and less than or equal to a second threshold in the candidate sub-signals; wherein the first threshold is less than the second threshold; a signal denoising module 1204, which is used to perform denoising processing on the second sub-signal using multi-resolution singular value decomposition to obtain a third sub-signal; wherein the number of layers of the multi-resolution singular value decomposition is determined based on the singular value change rate; a signal recombination module 1205, which is used to perform signal recombination on the first sub-signal and the third sub-signal to obtain a target signal.

[0157] In one implementation, the first processing module 1202 can be used to: construct an M-dimensional first vector and an M+1-dimensional second vector based on each candidate sub-signal respectively; obtain the distance between each two vectors in the first vector to obtain multiple first distances corresponding to each candidate sub-signal; obtain the distance between each two vectors in the second vector to obtain multiple second distances corresponding to each candidate sub-signal; obtain a first similarity ratio corresponding to each first vector based on the first distance and a preset similarity tolerance, and obtain a second similarity ratio corresponding to each second vector based on the second distance and the similarity tolerance; obtain a first mean of the first similarity ratio corresponding to each candidate sub-signal and a second mean of the second similarity ratio corresponding to each candidate sub-signal; and obtain an entropy value corresponding to each candidate sub-signal based on the first mean and the second mean corresponding to each candidate sub-signal.

[0158] In an optional implementation, the entropy value is calculated as follows:

[0159]

[0160] Among them, SE(m,r,N) is the entropy value and r is the similarity tolerance.

[0161] In one implementation, the signal denoising module 1204 can be used to: perform multi-layer singular value decomposition on the second sub-signal to obtain an approximate signal and a detail signal corresponding to each layer of singular value decomposition; obtain a first approximate signal, a first detail signal, a second approximate signal, and a second detail signal from the approximate signal and the detail signal; wherein the first approximate signal and the first detail signal are the approximate signal and the detail signal corresponding to the second-to-last layer of singular value decomposition, and the second approximate signal and the second detail signal are the approximate signal and the detail signal corresponding to the last layer of singular value decomposition; obtain an approximate singular value relative change rate based on the first approximate signal and the second approximate signal, and obtain a detail singular value relative change rate based on the first detail signal and the second detail signal; in response to the approximate singular value relative change rate being less than or equal to a first threshold, and the detail singular value relative change rate being less than or equal to a second threshold, obtain a third sub-signal based on other approximate signals except the second approximate signal.

[0162] In an optional implementation, the signal denoising module 1204 can also be used to: in response to the relative change rate of the approximate singular value being greater than or equal to a first threshold, and / or the relative change rate of the detail singular value being greater than or equal to a second threshold, increase the number of layers of the singular value decomposition by 1, and return to execute the steps of obtaining the relative change rate of the approximate singular value based on the first approximate signal and the second approximate signal, and obtaining the relative change rate of the detail singular value based on the first detail signal and the second detail signal.

[0163] The apparatus of the embodiments of the present application can classify candidate sub-signals based on their corresponding entropy values ​​obtained by decomposing a power quality signal, obtaining a pure first sub-signal and a second sub-signal containing noise components. The second sub-signal is then subjected to noise reduction to obtain a third sub-signal, and the first and third sub-signals are recombined to obtain a target signal. This can improve the noise reduction effect of the power quality signal.

[0164] It should be noted that the above explanation of the embodiment of the power quality signal noise reduction method is also applicable to the power quality signal noise reduction device of this embodiment, and will not be repeated here.

[0165] In order to implement the above embodiment, the present application also proposes an electronic device. Figure 13 , Figure 13 Schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Figure 13 As shown, the electronic device 1300 includes: a processor 1301, and a memory 1302 communicatively connected to the processor 1301; the memory 1302 stores computer-executable instructions; the processor 1301 executes the computer-executable instructions stored in the memory to implement the method provided in the aforementioned embodiment.

[0166] In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0167] In order to implement the above embodiments, the present application also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0168] In the description of this application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in this application is merely a way to describe the association relationship of associated objects, indicating that three relationships may exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0169] In the descriptions of the foregoing embodiments, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are mutually inconsistent.

[0170] 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 the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0171] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0172] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the 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 (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program 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 the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing it in another suitable manner if necessary, and then storing it in a computer memory.

[0173] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0174] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0175] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0176] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A method for reducing noise of power quality signals, characterized in that: include: Decompose the power quality signal to obtain multiple candidate sub-signals; Performing entropy discrimination on the candidate sub-signals to obtain an entropy value corresponding to each candidate sub-signal; Acquire a first sub-signal, from the candidate sub-signals, whose entropy value is less than or equal to a first threshold, and a second sub-signal, whose entropy value is greater than the first threshold and less than or equal to a second threshold; wherein the first threshold is less than the second threshold; Performing noise reduction processing on the second sub-signal using multi-resolution singular value decomposition to obtain a third sub-signal; wherein the number of layers of the multi-resolution singular value decomposition is determined based on the rate of change of the singular values; Signal recombination is performed on the first sub-signal and the third sub-signal to obtain a target signal.

2. The method according to claim 1, characterized in that The performing entropy discrimination on the candidate sub-signals to obtain an entropy value corresponding to each candidate sub-signal includes: constructing an M-dimensional first vector and an M+1-dimensional second vector based on each of the candidate sub-signals; Obtaining the distance between every two vectors in the first vectors to obtain a plurality of first distances corresponding to each candidate sub-signal; Obtaining the distance between every two vectors in the second vectors to obtain multiple second distances corresponding to each candidate sub-signal; Obtaining a first similarity ratio corresponding to each first vector based on the first distance and a preset similarity tolerance, and obtaining a second similarity ratio corresponding to each second vector based on the second distance and the similarity tolerance; Obtaining a first mean of the first similarity ratio corresponding to each candidate sub-signal and a second mean of the second similarity ratio corresponding to each candidate sub-signal; Based on the first mean value and the second mean value corresponding to each candidate sub-signal, an entropy value corresponding to each candidate sub-signal is obtained.

3. The method according to claim 1, characterized in that The step of performing noise reduction processing on the second sub-signal by using multi-resolution singular value decomposition to obtain a third sub-signal includes: Performing multi-layer singular value decomposition on the second sub-signal to obtain an approximate signal and a detail signal corresponding to each layer of singular value decomposition; Obtaining a first approximate signal, a first detail signal, a second approximate signal, and a second detail signal from the approximate signal and the detail signal; wherein the first approximate signal and the first detail signal are the approximate signal and the detail signal corresponding to the second-to-last layer of singular value decomposition, and the second approximate signal and the second detail signal are the approximate signal and the detail signal corresponding to the last layer of singular value decomposition; Obtaining an approximate singular value relative change rate based on the first approximate signal and the second approximate signal, and obtaining a detail singular value relative change rate based on the first detail signal and the second detail signal; In response to the relative change rate of the approximate singular value being less than or equal to a first threshold and the relative change rate of the detail singular value being less than or equal to a second threshold, the third sub-signal is acquired based on the other approximate signals except the second approximate signal.

4. The method according to claim 3, characterized in that The method further comprises: In response to the relative change rate of the approximate singular value being greater than or equal to a first threshold, and / or the relative change rate of the detail singular value being greater than or equal to a second threshold, the number of layers of the singular value decomposition is increased by 1, and the steps of obtaining the relative change rate of the approximate singular value based on the first approximate signal and the second approximate signal, and obtaining the relative change rate of the detail singular value based on the first detail signal and the second detail signal are returned to be executed.

5. A power quality signal noise reduction device, characterized in that: include: A signal decomposition module is used to decompose the power quality signal to obtain multiple candidate sub-signals; A first processing module is configured to perform entropy discrimination on the candidate sub-signals to obtain an entropy value corresponding to each candidate sub-signal; a second processing module, configured to obtain, from the candidate sub-signals, a first sub-signal whose entropy value is less than or equal to a first threshold, and a second sub-signal whose entropy value is greater than the first threshold and less than or equal to a second threshold; wherein the first threshold is less than the second threshold; a signal denoising module, configured to perform denoising on the second sub-signal using multi-resolution singular value decomposition to obtain a third sub-signal; wherein the number of layers of the multi-resolution singular value decomposition is determined based on a singular value change rate; The signal recombination module is configured to recombine the first sub-signal and the third sub-signal to obtain a target signal.

6. An electronic device, characterized in that: include: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 4 when executed by a processor.

8. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 4 when being executed by a processor.

Images