Local Discharge Signal Denoising Method and Device Based on Singular Value Decomposition of Hankel Matrix

Abstract

The present invention discloses a local discharge signal denoising method and device based on singular value decomposition of Hankel matrix. The method is as follows: constructing a noisy signal by using a double-exponential decay oscillation function; slidingly constructing a Hankel matrix with a signal sequence of a set truncation length and performing singular value decomposition to obtain the time period when the local discharge signal in the noisy signal is emitted; slidingly constructing a Hankel matrix for the local discharge signal in the emitted time period and performing singular value decomposition, and using the proportional increase method to judge the effective order of the local discharge signal; performing signal reconstruction on the singular values of the effective order to obtain the denoised local discharge signal; setting the signals outside the time period of the denoised local discharge signal to zero to obtain a noise-free local discharge signal with the same length as the noisy signal. This method uses sliding data to obtain the time period when the local discharge signal is emitted, calculates the effective order of the local discharge signal through the proportional increase method, and realizes removing the interference of white noise signals while retaining the detailed information of the original local discharge signal.

Description

Technical Field

[0001] The invention belongs to the technical field of partial discharge signal denoising, and in particular relates to a partial discharge signal denoising method and device based on Hankel matrix singular value decomposition. Background Art

[0002] White noise is caused by thermal noise of electrical equipment and exists in large quantities at the partial discharge detection site. When conducting on-site partial discharge detection, white noise will invade the partial discharge detection device. However, the partial discharge signal PD is usually weak, and strong noise will drown out the effective PD signal. Therefore, in order to obtain a good PD signal, the white noise signal must be effectively weakened and suppressed.

[0003] In the prior art, the white noise in the partial discharge signal is mainly analyzed by wavelet analysis, but the denoising effect of this method is greatly affected by the base wavelet type, threshold selection and other methods. The singular value decomposition method is a means of suppressing partial discharge noise that has attracted the attention of researchers recently, but the application of the singular value decomposition method in partial discharge is mainly used in the suppression of narrowband interference. Since the traditional singular value decomposition method uses the entire signal sequence for singular value decomposition, the singular values of the partial discharge signal and the singular values of the white noise both show a slow downward trend, and the number is large, and it is difficult to distinguish them by singular value feature points such as singular entropy, wavelet threshold method, and differential spectrum method. If the area where the partial discharge occurs can be obtained and cut out for singular value decomposition, not only will the amount of calculation be greatly reduced, but also the concentration of the partial discharge energy can be used to reflect the waveform information of the partial discharge signal with fewer singular values, and the dividing point of the singular values of the partial discharge and the white noise can be found by searching, and then the white noise denoising of the partial discharge signal can be completed. Summary of the invention

[0004] The main purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art and provide a method and device for denoising a local discharge signal based on Hankel matrix singular value decomposition. The method is based on Hankel matrix singular value decomposition, constructs a noisy signal, uses sliding data to determine the time of the local discharge signal, uses a proportional increase method to calculate the effective order of the local discharge signal, and finally obtains a noise-free and good local discharge signal through singular value decomposition reconstruction, thereby removing the interference of the white noise signal.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] On the one hand, the present invention provides a method for denoising a partial discharge signal based on Hankel matrix singular value decomposition, characterized in that the method comprises the following steps:

[0007] S1, using double exponential decay oscillation function to construct the stained noise signal;

[0008] S2. Slide and construct a Hankel matrix for the noisy signal using a signal sequence with a set truncation length, and perform singular value decomposition to obtain the time period during which the partial discharge signal in the noisy signal is emitted;

[0009] S3. Slide and construct a Hankel matrix for the partial discharge signal in the emitted time period and perform singular value decomposition, and use the proportional increase method to determine the effective order of the partial discharge signal;

[0010] S4. Reconstruct the signal for the effective order singular values to obtain the denoised partial discharge signal;

[0011] S5. Set the signal outside the time period of the denoised partial discharge signal to zero to obtain a noise-free partial discharge signal with the same length as the noisy signal.

[0012] As a preferred technical solution, step S1 is specifically as follows:

[0013] Use a double-exponential decay oscillation function to simulate the partial discharge signal and white noise signal obeying the w~N(0,0.4 2 ) distribution respectively, and jointly form the noisy signal y(k), k = 1, 2,..., N, where N is the signal length;

[0014] The double-exponential decay oscillation function is expressed as:

[0015]

[0016] where Z4 is the signal amplitude, τ is the decay coefficient, the frequency f c is the oscillation center, t0 is the start discharge time, and m4(t) is the value corresponding to the analog signal at any time t.

[0017] As a preferred technical solution, the signal sequence with the set truncation length in step S2 is expressed as y K (s), (s = s, s + 1,..., s + K), where s is the start time of truncation and K is the truncation length;

[0018] Step S2 is specifically as follows:

[0019] S21. Let M = s;

[0020] S22. Slide and construct a Hankel matrix for the noisy signal y(k) using the signal sequence y K (M) with the set truncation length K, which is expressed as:

[0021]

[0022] where 1 < n < M, when M is even, when M is odd, m = M - n + 1;

[0023] S23. Perform singular value decomposition, and the formula is:

[0024]

[0025] where the singular value λ i satisfies λ1 ≥ λ2 ≥ … λ r > 0, r is the rank of matrix H, u i and are the eigenvectors of HH T and v i is the eigenvector of H T H, and H T is the transpose of matrix H;

[0026] S24. Record the first singular value d N = λ1, and let M = M + 1;

[0027] S25. If M < N - K, then return to step S22 to continue execution; if M ≥ N - K, then execute step S26;

[0028] S26. Plot a curve for the recorded first singular value sequence Dk = [d1, d2, …, d s+K . When the position of the peak W on the curve is consistent with the position of the peak O of the partial discharge signal, obtain the partial discharge signal y * (t) emitted by the noisy signal in the time period [t1, t2], where t1 = B + K, t2 = E + K, B is the starting point of the peak W, and E is the ending point of the peak W.

[0029] As a preferred technical solution, the specific steps of step S3 are as follows:

[0030] S31. Let the effective order of the partial discharge signal be p = 1, slide and construct a Hankel matrix for the partial discharge signal y * (t) emitted in the time period [t1, t2] and perform singular value decomposition to obtain the singular value sequence D = [λ1, λ2, …, λ r ;

[0031] S32. In the time period [t1, t2], the singular value of the partial discharge signal is greater than the singular value of the white noise signal. Take the first p singular values in the singular value sequence D and increase them L times using the proportional increase method, and keep the last r - p singular values unchanged;

[0032] S33. Use singular value decomposition to perform singular value reconstruction on the increased singular value sequence D to obtain H p , and take the first row and the last column of H p to obtain the signal For Construct a Hankel matrix and perform singular value decomposition to obtain a singular value sequence D' = [λ'1, λ'2, …, λ' r ;

[0033] S34. Calculate the ratio g(p) of the first p singular values in the singular value sequences D and D', and the formula is:

[0034]

[0035] If g(p) ≥ L, then let p = p + 1, and return to step S32 to continue execution; if g(p) < L, then take p as the effective order of the partial discharge signal.

[0036] As a preferred technical solution, the step S4 is specifically as follows:

[0037] S41. Take p singular values of the effective order in the singular value sequence D for singular value decomposition and reconstruction to obtain H * ;

[0038] S42. Take the first row and the last column of H * to obtain the denoised partial discharge signal

[0039] As a preferred technical solution, the step S5 is specifically as follows:

[0040] Set the signal outside the time period [t1, t2] of the denoised partial discharge signal to zero to obtain a noise-free partial discharge signal with the same length as the noisy signal y(k)

[0041] On the other hand, the present invention also provides a partial discharge signal denoising system based on singular value decomposition of a Hankel matrix, which is characterized in that the system includes a noisy signal construction module, a partial discharge determination module, an effective order calculation module, a denoised signal reconstruction module, and a noise-free signal acquisition module;

[0042] The noisy signal construction module constructs a noisy signal by using a double-exponential decay oscillation function;

[0043] The partial discharge determination module uses a signal sequence with a set intercept length to slide and construct a Hankel matrix for the noisy signal and perform singular value decomposition to obtain the time period when the partial discharge signal in the noisy signal is emitted;

[0044] The effective order calculation module slides and constructs a Hankel matrix for the partial discharge signal in the emission time period and performs singular value decomposition, and uses the proportional increase method to judge the effective order of the partial discharge signal;

[0045] The denoising signal reconstruction module reconstructs the signal with the effective order of singular values to obtain the denoised partial discharge signal;

[0046] The noise-free signal acquisition module zeros out the signal outside the time period of the denoised partial discharge signal to obtain a noise-free partial discharge signal with the same length as the noisy signal.

[0047] As a preferred technical solution, the signal sequence with a set truncation length in the partial discharge determination module is denoted as y K (s), (s = s, s + 1,..., s + K), where s is the start time of truncation and K is the truncation length;

[0048] The specific steps of the partial discharge determination module are as follows:

[0049] Let M = s; use the signal sequence y K (M) with a set truncation length K to slide and construct a Hankel matrix, denoted as:

[0050]

[0051] where 1 < n < M, when M is even, when M is odd, m = M - n + 1; perform singular value decomposition, and the formula is:

[0052]

[0053] where the singular value λ i satisfies λ1 ≥ λ2 ≥ … λ r > 0, r is the rank of matrix H, u i and are the eigenvectors of HH T v i is the eigenvector of H T H, and H T is the transpose of matrix H;

[0054] Record the first singular value d N = λ1, and let M = M + 1;

[0055] If M < N - K, return to slide and construct the Hankel matrix and continue to execute; if M ≥ N - K, execute the next step;

[0056] Plot a curve for the recorded first singular value sequence Dk = [d1, d2, …, d s+K , when the position of the peak W on the curve is consistent with the position of the peak O of the partial discharge signal, obtain the partial discharge signal y emitted by the noisy signal in the time period [t1, t2] *(t), where t1 = B + K, t2 = E + K, B is the starting point of the wave peak W, and E is the ending point of the wave peak W.

[0057] As a preferred technical solution, the specific steps of the effective order calculation module are as follows:

[0058] Let the effective order of the partial discharge signal be p = 1, and for the partial discharge single signal y * (t) slide to construct a Hankel matrix and perform singular value decomposition to obtain the singular value sequence D = [λ1, λ2, …, λ r ;

[0059] In the time period [t1, t2], the singular value of the partial discharge signal is greater than that of the white noise signal. Take the first p singular values in the singular value sequence D and increase them L times using the proportional increase method, and keep the last r - p singular values unchanged;

[0060] Use singular value decomposition to perform singular value reconstruction on the increased singular value sequence D to obtain H p , take the first row and the last column of H p to obtain the signal For construct a Hankel matrix and perform singular value decomposition to obtain the singular value sequence D' = [λ'1, λ'2, …, λ' r ;

[0061] Calculate the ratio g(p) of the first p singular values in the singular value sequences D and D', and the formula is:

[0062]

[0063] If g(p) ≥ L, then let p = p + 1, and return to the proportional increase method step to continue execution; if g(p) < L, then take p as the effective order of the partial discharge signal.

[0064] On the other hand, the present invention provides a computer - readable storage medium storing a program, characterized in that when the program is executed by a processor, it implements the above - mentioned partial discharge signal denoising method based on singular value decomposition of the Hankel matrix.

[0065] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0066] 1. The present invention uses a signal sequence with a set truncation length to slide - construct a Hankel matrix for the noisy signal and perform singular value decomposition, which can effectively and accurately determine the time period when the partial discharge signal is emitted according to the position of the wave peak W, avoiding the subjectivity of artificially determining the occurrence area of the partial discharge signal;

[0067] 2. The present invention constructs a Hankel matrix by sliding the partial discharge signals in the emission time period and performs singular value decomposition. By utilizing the non-correlation between the partial discharge signals and the noise signals, the proportional increase method can effectively determine the effective order of the partial discharge signals, avoiding the limitation of the traditional difference method that only takes the maximum value point and the subjectivity of the unilateral maximum value method in selection.

[0068] 3. The present invention obtains the noise-free partial discharge signals through singular value reconstruction, effectively removing the interference of white noise signals and retaining the detailed information of the original PD signals, with a small waveform distortion rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following-described drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0070] Figure 1 It is a flowchart of the partial discharge signal denoising method based on singular value decomposition of the Hankel matrix according to the embodiment of the present invention;

[0071] FIG. 2(a) is a waveform diagram of the original partial discharge signal according to the embodiment of the present invention;

[0072] FIG. 2(b) is a waveform diagram of the noisy signal according to the embodiment of the present invention;

[0073] Figure 3 It is a curve diagram composed of the singular value sequence according to the embodiment of the present invention;

[0074] Figure 4 It is a comparison diagram between the denoised partial discharge signal and the original partial discharge signal according to the embodiment of the present invention;

[0075] Figure 5 It is a waveform diagram of the noise-free partial discharge signal according to the embodiment of the present invention;

[0076] Figure 6 It is a structure diagram of the partial discharge signal denoising system based on singular value decomposition of the Hankel matrix according to the embodiment of the present invention;

[0077] Figure 7 It is a structure diagram of a computer-readable storage medium according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0078] To enable those skilled in the art to better understand the solution of this application, the following will clearly and completely describe the technical solution in the embodiments of this application in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of this application.

[0079] In this application, the mention of "embodiment" means that the specific features, structures or characteristics described in connection with the embodiment can be included in at least one embodiment of this application. The phrase appears in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described in this application can be combined with other embodiments.

[0080] As Figure 1 shown, the present invention discloses a local discharge signal denoising method based on the singular value decomposition of the Hankel matrix. In this embodiment, a local discharge signal denoising method based on the singular value decomposition of the Hankel matrix is applied to the local discharge simulation signal denoising experiment of power equipment, including the following steps:

[0081] S1. Construct a noisy signal using a double-exponential decay oscillation function;

[0082] Specifically, in step S1, a double-exponential decay oscillation function is used to simulate the local discharge signal and white noise signal obeying the w~N(0,0.4 2 ) distribution respectively. The two signals together constitute the noisy signal y(k), k = 1, 2,..., N, where N is the signal length; Fig. 2(a) is the waveform diagram of the constructed local discharge signal, and Fig. 2(b) is the waveform diagram of the constituted noisy signal.

[0083] The above double-exponential decay oscillation function is expressed as:

[0084]

[0085] where Z4 is the signal amplitude, τ is the decay coefficient, the frequency f c is the oscillation center, t0 is the start discharge time, and m4(t) is the value corresponding to the analog signal at any time t.

[0086] S2. Slide a signal sequence with a set truncation length to construct a Hankel matrix for the noisy signal and perform singular value decomposition to obtain the time period when the local discharge signal in the noisy signal is emitted;

[0087] Specifically, in step S2, the signal sequence with a set truncation length is expressed as y K(s), (s = s, s + 1,..., s + K), where s is the starting time of interception and K is the interception length;

[0088] The specific operation of step S2 is as follows:

[0089] S21. Let M = s;

[0090] S22. Use the signal sequence y K (M) of the set interception length K to slide and construct a Hankel matrix, expressed as:

[0091]

[0092] where 1 < n < M, when M is even, when M is odd, m = M - n + 1;

[0093] S23. Perform singular value decomposition, and the formula is:

[0094]

[0095] where the singular value λ i satisfies λ1 ≥ λ2 ≥ … λ r > 0, r is the rank of matrix H, u i and are the eigenvectors of HH T v i is the eigenvector of H T H, H T is the transpose of matrix H;

[0096] S24. Record the first singular value d N = λ1, and let M = M + 1;

[0097] S25. If M < N - K, return to step S22 and continue to execute; if M ≥ N - K, execute step S26;

[0098] S26. Plot a curve for the recorded first singular value sequence Dk = [d1, d2, …, d s+K . When the position of the wave peak W on the curve is consistent with the position of the partial discharge signal peak O, obtain the partial discharge signal y * (t) emitted by the noisy signal in the time period [t1, t2], where t1 = B + K, t2 = E + K, B is the starting point of the wave peak W, and E is the ending point of the wave peak W.

[0099] In this embodiment, the interception length K = 300. After sliding and constructing a Hankel matrix and performing singular value decomposition, the first singular value sequence Dk = [d1, d2, …, d Figure 3 as shown is obtained.s+K The curve graph of [], and finally determine that the time period when the partial discharge signal is emitted is [3800, 4000].

[0100] S3. Slide and construct a Hankel matrix for the partial discharge signals in the emitted time period and perform singular value decomposition, and use the proportional increase method to judge the effective order of the partial discharge signals;

[0101] Specifically, the operations in step S3 include:

[0102] S31. Let the effective order of the partial discharge signal be p = 1, and slide and construct a Hankel matrix for the partial discharge signal y * (t) in the time period [t1, t2] and perform singular value decomposition to obtain the singular value sequence D = [λ1, λ2, …, λ r ;

[0103] S32. In the time period [t1, t2], the singular value of the partial discharge signal is greater than that of the white noise signal. Take the first p singular values in the singular value sequence D and increase them by L times using the proportional increase method, and keep the last r - p singular values unchanged;

[0104] S33. Use the above singular value decomposition formula to perform singular value reconstruction on the increased singular value sequence D to obtain H p , and take the first row and the last column of H p to obtain the signal For construct a Hankel matrix and perform singular value decomposition to obtain the singular value sequence D' = [λ'1, λ'2, …, λ' r ;

[0105] S34. Calculate the ratio g(p) of the first p singular values in the singular value sequences D and D', and the formula is:

[0106]

[0107] If g(p) ≥ L, then let p = p + 1 and return to step S32 to continue execution; if g(p) < L, then take p as the effective order of the partial discharge signal.

[0108] In this embodiment, perform singular value decomposition on the partial discharge y * (t) in the time period [3800, 4000], and use the proportional increase method to perform singular value reconstruction to obtain the effective order of the partial discharge signal as p = 4.

[0109] S4. Perform signal reconstruction on the singular values of the effective order to obtain the denoised partial discharge signal;

[0110] Specifically, the operation in step S4 is:

[0111] S41. According to the above singular value decomposition formula, p effective-order singular values are taken from the singular value sequence D for singular value decomposition and reconstruction to obtain H * ;

[0112] S42. Take the first row and the last column of H * to obtain the denoised partial discharge signal

[0113] In this embodiment, Figure 4 is the denoised partial discharge signal and the contrast diagram with the original partial discharge signal y * (t). It can be seen that the waveform characteristics of the partial discharge can be well represented, with a small waveform distortion rate and a high similarity degree.

[0114] S5. Set the signals of the denoised partial discharge signal outside the time period to zero to obtain a noise-free partial discharge signal with the same length as the noisy signal. Specifically:

[0115] Set the signals of the denoised partial discharge signal outside the time period [t1, t2] to zero to obtain a noise-free partial discharge signal with the same length as the noisy signal y(k)

[0116] The waveform diagram of the noise-free partial discharge signal obtained in this embodiment is as shown in Figure 5 . By comparing with Fig. 2(a), Figure 4 , it can be seen that this method can effectively remove the interference of white noise signals, while retaining the detailed information of the original PD signal and having a small waveform distortion rate;

[0117] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously.

[0118] Based on the same idea as the partial discharge signal denoising method based on Hankel matrix singular value decomposition in the above embodiment, the present invention also provides a partial discharge signal denoising system based on Hankel matrix singular value decomposition. This system can be used to execute the above partial discharge signal denoising method based on Hankel matrix singular value decomposition. For the sake of easy explanation, in the structural schematic diagram of the partial discharge signal denoising system embodiment based on Hankel matrix singular value decomposition, only the parts related to the embodiment of the present invention are shown. Those skilled in the art can understand that the illustrated structure does not constitute a limitation on the device, and it may include more or fewer components than those shown, or combine certain components, or arrange different components.

[0119] As Figure 6 shown, another embodiment of the present invention provides a partial discharge signal denoising system based on the singular value decomposition of the Hankel matrix, including the following several modules:

[0120] The noisy signal construction module constructs a noisy signal using a double-exponential decay oscillation function;

[0121] The partial discharge determination module slides a signal sequence with a set truncation length on the noisy signal to construct a Hankel matrix and performs singular value decomposition to obtain the time period when the partial discharge signal in the noisy signal is emitted;

[0122] The effective order calculation module slides a Hankel matrix on the partial discharge signal in the emitted time period and performs singular value decomposition, and uses the proportional increase method to judge the effective order of the partial discharge signal;

[0123] The denoised signal reconstruction module reconstructs the signal from the effective order singular values to obtain the denoised partial discharge signal;

[0124] The noise-free signal acquisition module sets the signal outside the time period of the denoised partial discharge signal to zero to obtain a noise-free partial discharge signal with the same length as the noisy signal.

[0125] Specifically, the signal sequence with the set truncation length in the partial discharge determination module is denoted as y K (s), (s = s, s + 1,..., s + K), where s is the start time of truncation and K is the truncation length;

[0126] The specific steps of the partial discharge determination module are as follows:

[0127] Let M = s; use the signal sequence y K (M) with the set truncation length K to slide and construct a Hankel matrix for the noisy signal y(k), which is expressed as:

[0128]

[0129] where 1 < n < M, when M is even, when M is odd, m = M - n + 1; perform singular value decomposition, and the formula is:

[0130]

[0131] where the singular value λ i satisfies λ1 ≥ λ2 ≥... λ r > 0, r is the rank of matrix H, u i and are the eigenvectors of HH T v iIt is H T The eigenvector of H, H T is the transpose of matrix H;

[0132] Record the first singular value d N = λ1, let M = M + 1;

[0133] If M < N - K, then return to slide and construct the Hankel matrix and continue to execute; if M ≥ N - K, then execute the next step;

[0134] For the recorded first singular value sequence Dk = [d1, d2, …, d s+K , draw a curve. When the position of the peak W on the curve is consistent with the position of the peak O of the partial discharge signal, obtain the partial discharge signal y * (t) emitted by the noisy signal in the time period [t1, t2], where t1 = B + K, t2 = E + K, B is the starting point of the peak W, and E is the ending point of the peak W.

[0135] Specifically, the specific steps of the effective order calculation module are as follows:

[0136] Let the effective order of the partial discharge signal be p = 1. Slide and construct the Hankel matrix for the partial discharge signal y * (t) in the time period [t1, t2] and perform singular value decomposition to obtain the singular value sequence D = [λ1, λ2, …, λ r ;

[0137] In the time period [t1, t2], the singular value of the partial discharge signal is greater than that of the white noise signal. Take the first p singular values in the singular value sequence D and increase them L times using the proportional increase method, and keep the last r - p singular values unchanged;

[0138] Use singular value decomposition to perform singular value reconstruction on the increased singular value sequence D to obtain H p , take the first row and the last column of H p to obtain the signal For Construct the Hankel matrix and perform singular value decomposition to obtain the singular value sequence D' = [λ'1, λ'2, …, λ' r ;

[0139] Calculate the ratio g(p) of the first p singular values in the singular value sequences D and D', and the formula is:

[0140]

[0141] If g(p) ≥ L, then let p = p + 1 and return to the proportional increase method step to continue execution; if g(p) < L, then take p as the effective order of the partial discharge signal.

[0142] It should be noted that the partial discharge signal denoising system based on the singular value decomposition of the Hankel matrix of the present invention corresponds one-to-one with the partial discharge signal denoising method based on the singular value decomposition of the Hankel matrix of the present invention. The technical features and beneficial effects described in the embodiments of the above-mentioned partial discharge signal denoising method based on the singular value decomposition of the Hankel matrix are applicable to the embodiments of the partial discharge signal denoising system based on the singular value decomposition of the Hankel matrix. For specific content, please refer to the description in the method embodiments of the present invention, which will not be elaborated here. This is hereby declared.

[0143] In addition, in the implementation manner of the partial discharge signal denoising system based on the singular value decomposition of the Hankel matrix in the above embodiments, the logical division of each program module is only for illustrative purposes. In actual applications, according to needs, for example, considering the configuration requirements of the corresponding hardware or the convenience of software implementation, the above functions can be assigned to different program modules to complete, that is, the internal structure of the partial discharge signal denoising system based on the singular value decomposition of the Hankel matrix is divided into different program modules to complete all or part of the functions described above.

[0144] As Figure 7 shown, in one embodiment, a computer-readable storage medium is provided, which stores a program in a memory. When the program is executed by a processor, the partial discharge signal denoising method based on the singular value decomposition of the Hankel matrix is implemented, specifically as follows:

[0145] Construct a noisy signal using a double-exponential decay oscillation function;

[0146] Use a signal sequence with a set truncation length to slide and construct a Hankel matrix for the noisy signal and perform singular value decomposition to obtain the time period when the partial discharge signal in the noisy signal is emitted;

[0147] Slide and construct a Hankel matrix for the partial discharge signal in the emitted time period and perform singular value decomposition, and use the proportional increase method to judge the effective order of the partial discharge signal;

[0148] Reconstruct the signal for the effective order of singular values to obtain the denoised partial discharge signal;

[0149] Set the signal outside the time period of the denoised partial discharge signal to zero to obtain a noise-free partial discharge signal with the same length as the noisy signal.

[0150] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0151] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0152] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention should be equivalent replacement methods and are all included in the protection scope of the present invention.

Claims

1. A local discharge signal denoising method based on singular value decomposition of Hankel matrix, characterized in that, The method includes the following steps: S1. Construct a noisy signal using a double-exponential decay oscillation function; S2. Slide a signal sequence with a set truncation length on the noisy signal to construct a Hankel matrix and perform singular value decomposition to obtain the time period when the partial discharge signal in the noisy signal is emitted; S3. Slide a Hankel matrix on the partial discharge signal in the emitted time period and perform singular value decomposition, and use the proportional increase method to judge the effective order of the partial discharge signal. Specifically: S31. Let the effective order of the partial discharge signal be p = 1, and slide and construct a Hankel matrix for the partial discharge signal y(t) in the time period [t1, t2], and perform singular value decomposition to obtain a singular value sequence D = [λ1, λ2, …, λ * (t) and obtain a singular value sequence D = [λ1, λ2, …, λ r ; S32. In the time period [t1, t2], the singular value of the partial discharge signal is greater than that of the white noise signal. Take the first p singular values in the singular value sequence D and increase them by L times using the proportional increase method, and keep the last r - p singular values unchanged; S33. Use singular value decomposition to perform singular value reconstruction on the enlarged singular value sequence D to obtain H p , take the first row and the last column of H p to obtain the signal For construct a Hankel matrix and perform singular value decomposition to obtain the singular value sequence D’ = [λ’1, λ’2, …, λ’ r ; S34. Calculate the ratio g(p) of the first p singular values in the singular value sequences D and D', and the formula is: If g(p) ≥ L, then let p = p + 1, and return to step S32 to continue execution; if g(p) < L, then take p as the effective order of the partial discharge signal; S4. Reconstruct the signal from the singular values of the effective order to obtain the denoised partial discharge signal; S5. Set the signal outside the time period of the denoised partial discharge signal to zero to obtain a noise-free partial discharge signal with the same length as the noisy signal.

2. The partial discharge signal denoising method based on singular value decomposition of Hankel matrix according to claim 1, characterized in that The specific step S1 is as follows: The double-exponential decay oscillation function is used to simulate the partial discharge signal and the white noise signal obeying the w~N(0,0.4 2 ) distribution respectively, and they jointly constitute the noisy signal y(k), where k = 1, 2, …, N, and N is the signal length; The double-exponential decay oscillation function is expressed as: where Z4 is the signal amplification, τ is the attenuation coefficient, the frequency f c is the oscillation center, t0 is the starting discharge time, and m4(t) is the value corresponding to the analog signal at any time t.

3. The partial discharge signal denoising method based on singular value decomposition of Hankel matrix according to claim 2, wherein The signal sequence for setting the truncation length described in step S2 is denoted as y K (s),(s = s, s + 1,..., s + K), where s is the start time of truncation and K is the truncation length; The specific step S2 is as follows: S21. Let M = s; S22. Use the signal sequence y of the set truncation length K for the noise-contaminated signal y(k) K (M) Slide to construct a Hankel matrix, denoted as: where 1 < n < M, when M is an even number, when M is an odd number, m = M - n + 1; S23. Perform singular value decomposition, and the formula is: Among them, the singular value λ i satisfies λ1≥λ2≥…λ r >0, r is the rank of matrix H, u i is the eigenvector of HH T , v i is the eigenvector of H T H, and H T is the transpose of matrix H; S24. Record the first singular value d N = λ1, and let M = M + 1; S25. If M < N - K, then return to step S22 to continue execution; if M ≥ N - K, then execute step S26; S26. Plot a curve for the recorded first singular value sequence Dk = [d1, d2, …, d s+K . When the position of the peak W on the curve coincides with the position of the peak O of the partial discharge signal, the partial discharge signal y * (t) emitted by the noisy signal in the time period [t1, t2] is obtained, where t1 = B + K, t2 = E + K, B is the starting point of the peak W, and E is the ending point of the peak W.

4. The partial discharge signal denoising method based on singular value decomposition of Hankel matrix according to claim 3, characterized in that The specific step S4 is as follows: S41. Take p effective orders of singular values in the singular value sequence D for singular value decomposition and reconstruction to obtain H * ; S42. Take H * The first row and the last column of to obtain the denoised partial discharge signal 5. The partial discharge signal denoising method based on singular value decomposition of Hankel matrix according to claim 4, wherein The specific step S5 is as follows: The denoised partial discharge signal Set the signals outside the time period [t1, t2] to zero to obtain a noise-free partial discharge signal with the same length as the noisy signal y(k).

6. A partial discharge signal denoising system based on singular value decomposition of Hankel matrix, characterized in that, The system includes a noisy signal construction module, a partial discharge determination module, an effective order calculation module, a denoised signal reconstruction module, and a noise-free signal acquisition module; The noisy signal construction module constructs a noisy signal using a double-exponential decay oscillation function; The partial discharge determination module slides a signal sequence with a set truncation length on the noisy signal to construct a Hankel matrix and perform singular value decomposition to obtain the time period when the partial discharge signal in the noisy signal is emitted; The effective order calculation module slides a Hankel matrix on the partial discharge signal in the emitted time period and performs singular value decomposition, and uses the proportional increase method to judge the effective order of the partial discharge signal. Specifically: S31. Let the effective order of the partial discharge signal be p = 1, and slide the partial discharge signal y(t) in the time period [t1, t2] to construct a Hankel matrix and perform singular value decomposition to obtain the singular value sequence D = [λ1, λ2, …, λ * (t) to obtain the singular value sequence D = [λ1, λ2, …, λ r ; S32. In the time period [t1, t2], the singular value of the partial discharge signal is greater than that of the white noise signal. Take the first p singular values in the singular value sequence D and increase them by L times using the proportional increase method, and keep the last r - p singular values unchanged; S33. Use singular value decomposition to perform singular value reconstruction on the enlarged singular value sequence D to obtain H p , take the first row and the last column of H p to obtain the signal For Construct a Hankel matrix and perform singular value decomposition to obtain the singular value sequence D’ = [λ’1, λ’2, …, λ’ r ; S34. Calculate the ratio g(p) of the first p singular values in the singular value sequences D and D', and the formula is: If g(p) ≥ L, then let p = p + 1, and return to step S32 to continue execution; if g(p) < L, then take p as the effective order of the partial discharge signal; The denoised signal reconstruction module reconstructs the signal from the singular values of the effective order to obtain the denoised partial discharge signal; The noise-free signal acquisition module sets the signal outside the time period of the denoised partial discharge signal to zero to obtain a noise-free partial discharge signal with the same length as the noisy signal.

7. The partial discharge signal denoising system based on singular value decomposition of Hankel matrix according to claim 6, wherein The signal sequence with the set truncation length in the partial discharge determination module is expressed as y K (s), (s = s, s + 1,..., s + K), where s is the start time of truncation and K is the truncation length; The specific steps of the partial discharge determination module are as follows: Let M = s; the noisy signal y(k) is used to construct a Hankel matrix with a sliding window of length K for the signal sequence y K (M), denoted as: where 1 < n < M, when M is even, when M is odd, m = M - n + 1; Perform singular value decomposition, and the formula is: Among them, the singular value λ i satisfies λ1≥λ2≥…λ r > 0, r is the rank of matrix H, u i is the eigenvector of HH T , v i is the eigenvector of H T H, and H T is the transpose of matrix H; Record the first singular value d N = λ1, let M = M + 1; If M < N - K, return to slide and construct the Hankel matrix and continue to execute; if M ≥ N - K, execute the next step; For the first singular value sequence Dk = [d1, d2, …, d s+K of the record, a curve is plotted. When the position of the peak W of the curve coincides with the position of the peak O of the partial discharge signal, the partial discharge signal y * (t) emitted by the noisy signal in the time period [t1, t2] is obtained, where t1 = B + K, t2 = E + K, B is the starting point of the peak W, and E is the ending point of the peak W.

8. A computer-readable storage medium storing a program, characterized in that, When the program is executed by a processor, it implements the partial discharge signal denoising method based on singular value decomposition of the Hankel matrix according to any one of claims 1-5.

Images