Harmonic signal recovery method, device and equipment and readable storage medium

Abstract

The invention discloses a harmonic signal recovery method, a harmonic signal recovery device, harmonic signal recovery equipment and a readable storage medium, belongs to the field of missing data reconstruction, and considers that a signal gap can be accurately recovered by a signal expression of a front signal segment passing through the signal gap. The signal abrupt change point in the to-be-processed harmonic signal can be determined, and the pre-signal segment of each signal gap can be rapidly and accurately determined through the signal abrupt change point, so that the signal abrupt change point in the to-be-processed harmonic signal is firstly determined, and then the signal expression of the pre-signal segment of the signal gap is determined for any signal gap. And finally, the signal gap can be recovered according to the signal expression of the pre-signal segment of the signal gap, so that the signal gap of the to-be-processed harmonic signal can be efficiently and accurately recovered.

Description

Technical Field

[0001] The present invention relates to the field of missing data reconstruction, and in particular, to a method, device, equipment and readable storage medium for restoring harmonic signals. Background Art

[0002] With the increase of nonlinear harmonic injection sources in the power grid, harmonic signals are also doped in PQ (Power Quality) signals (such as voltage signals or current signals). The analysis of harmonic signals in PQ signals is beneficial to power grid operation management, power quality assessment, fault diagnosis, etc. In practical applications, the harmonic signals extracted from PQ signals may have gaps (that is, signal loss), and there is also a lack of a mature method for restoring harmonic signals in related technologies, resulting in poor restoration efficiency and accuracy of harmonic signals.

[0003] Therefore, how to provide a solution to the above technical problems is a problem that those skilled in the art need to solve currently. Summary of the Invention

[0004] The purpose of the present invention is to provide a method, device, equipment and readable storage medium for restoring harmonic signals. In the present invention, first, signal mutation points in the harmonic signal to be processed are determined, and then for any signal gap, the signal expression of the pre-signal segment of the signal gap is determined. Finally, the signal gap can be restored according to the signal expression of the pre-signal segment of the signal gap, so as to efficiently and accurately restore the signal gaps of the harmonic signal to be processed.

[0005] To solve the above technical problems, the present invention provides a method for restoring harmonic signals, including:

[0006] Determine the signal mutation points in the harmonic signal to be processed, where the harmonic signal to be processed is a harmonic signal with several signal gaps, and the signal mutation point is the signal sampling point where the applicable harmonic signal model changes;

[0007] For any signal gap in the harmonic signal to be processed, determine the signal expression of the pre-signal segment of the signal gap, where the pre-signal segment of the signal gap is the signal segment between the pre-breakpoint of the signal gap and the signal gap, and the pre-breakpoint of the signal gap is the signal mutation point or another signal gap that is located before the signal gap and is the closest to the signal gap;

[0008] For any signal gap in the harmonic signal to be processed, restore the signal gap according to the signal expression of the pre-signal segment of the signal gap.

[0009] On the other hand, determining the signal mutation points in the harmonic signal to be processed includes:

[0010] For any signal gap in the harmonic signal to be processed, determine whether the similarity between two signal segments before and after the signal gap meets the standard, where the signal segment is: continuous signals before or after the signal gap;

[0011] If it meets the standard, it is determined that there are no signal mutation points in the two signal segments before and after the signal gap;

[0012] If it does not meet the standard, it is determined that there is at least one signal mutation point in the signal segment before and after the signal gap;

[0013] For any signal segment determined to possibly have a signal mutation point, determine the signal mutation point in the signal segment.

[0014] On the other hand, for any signal gap in the harmonic signal to be processed, the signal expression of the pre-signal segment of the signal gap is determined to include:

[0015] For any signal gap in the harmonic signal to be processed, determine the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap;

[0016] For any signal gap in the harmonic signal to be processed, based on the data in the pre-signal segment of the signal gap, determine the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap.

[0017] On the other hand, for any signal gap in the harmonic signal to be processed, the determination of the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap based on the data in the pre-signal segment of the signal gap includes:

[0018] For any signal gap in the harmonic signal to be processed, based on the data in the pre-signal segment of the signal gap, through a signal parameter estimation method based on rotational invariance techniques, determine the parameter estimation values in the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap;

[0019] According to the total least squares method, correct the parameter estimation values to obtain parameter correction values;

[0020] Fill the parameter correction values into the expressions of the signal models of the respective harmonic sub-signals to which they belong.

[0021] On the other hand, for any signal gap in the harmonic signal to be processed, the determination of whether the similarity between two signal segments before and after the signal gap meets the standard includes:

[0022] For any signal gap in the harmonic signal to be processed, obtain the lateral shift, tilt shift, and spectral shift between two signal segments before and after the signal gap;

[0023] According to the lateral shift, tilt shift, and spectral shift, determine whether the similarity between two signal segments before and after the signal gap meets the standard.

[0024] On the other hand, determining whether the similarity between two signal segments before and after the signal gap meets the standard according to the lateral shift, tilt shift, and spectral shift includes:

[0025] Based on the lateral shift, tilt shift, and spectral shift, determine the spectral continuity parameter between two signal segments before and after the signal gap through the first relational expression;

[0026] Judge whether the frequency continuity parameter is lower than the first preset threshold;

[0027] If it is lower, it is determined that the similarity between two signal segments before and after the signal gap meets the standard;

[0028] The spectral continuity parameter includes:

[0029] ;

[0030] where SC k is the spectral continuity parameter of the signal gap with serial number k, is the lateral shift between two signal segments before and after the signal gap, is the tilt shift between two signal segments before and after the signal gap, is the spectral shift between two signal segments before and after the signal gap, x is the signal of the signal segment before the signal gap, and y is the signal of the signal segment after the signal gap.

[0031] On the other hand, for any signal segment determined to possibly have a signal mutation point, determining the signal mutation point in the signal segment includes:

[0032] For any signal segment determined to possibly have a signal mutation point, determine the initial noise subspace according to the data in the signal segment;

[0033] For any sampling signal in the signal segment, determine the subspace affinity between the sampling signal and the initial noise subspace;

[0034] Determine the change rate between the subspace affinities of any two adjacent sampling signals in the signal segment;

[0035] Take the position between a pair of sampling signals with a change rate of the subspace affinity higher than the second preset threshold as a signal mutation point.

[0036] To solve the above technical problems, the present invention also provides a harmonic signal recovery device, including:

[0037] A first determination module, configured to determine signal mutation points in a harmonic signal to be processed, where the harmonic signal to be processed is a harmonic signal with several signal gaps, and a signal mutation point is a signal sampling point at which the applicable harmonic signal model changes;

[0038] A second determination module, configured to, for any signal gap in the harmonic signal to be processed, determine the signal expression of the pre-signal segment of the signal gap, where the pre-signal segment of the signal gap is the signal segment between the pre-break point of the signal gap and the signal gap, and the pre-break point of the signal gap is the signal mutation point or another signal gap that is located before the signal gap and is the closest to the signal gap;

[0039] A recovery module, configured to, for any signal gap in the harmonic signal to be processed, recover the signal gap according to the signal expression of the pre-signal segment of the signal gap.

[0040] To solve the above technical problems, the present invention also provides a harmonic signal recovery device, including:

[0041] A memory, configured to store a computer program;

[0042] A processor, configured to implement the steps of the above-mentioned harmonic signal recovery method when executing the computer program.

[0043] To solve the above technical problems, the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned harmonic signal recovery method are implemented.

[0044] The present invention provides a harmonic signal recovery method. Considering that the signal gap can be accurately recovered through the signal expression of the pre-signal segment of the signal gap, and the pre-signal segment of each signal gap can be quickly and accurately determined through the signal mutation point, in the present invention, first, the signal mutation points in the harmonic signal to be processed are determined, then, for any signal gap, the signal expression of the pre-signal segment of the signal gap is determined, and finally, the signal gap can be recovered according to the signal expression of the pre-signal segment of the signal gap, so that the signal gaps of the harmonic signal to be processed can be recovered efficiently and accurately.

[0045] The present invention also provides a harmonic signal recovery device, device and computer-readable storage medium, which have the same beneficial effects as the above-mentioned harmonic signal recovery method. Description of the Drawings

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the attached drawings required in the related technologies and embodiments. Obviously, the attached drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other attached drawings can also be obtained based on these attached drawings.

[0047] Figure 1 It is a schematic flowchart of a method for restoring a harmonic signal provided by the present invention;

[0048] Figure 2 It is a schematic structural diagram of a harmonic signal to be processed provided by the present invention;

[0049] Figure 3 It is another schematic structural diagram of a harmonic signal to be processed provided by the present invention;

[0050] Figure 4 It is a schematic structural diagram of a device for restoring a harmonic signal provided by the present invention;

[0051] Figure 5 It is a schematic structural diagram of a device for restoring a harmonic signal provided by the present invention. Specific embodiments

[0052] The core of the present invention is to provide a method, device, equipment and readable storage medium for restoring a harmonic signal. In the present invention, the signal mutation points in the harmonic signal to be processed are first determined, and then for any signal gap, the signal expression of the pre-signal segment of the signal gap is determined. Finally, the signal gap can be restored according to the signal expression of the pre-signal segment of the signal gap, so as to efficiently and accurately restore the signal gap of the harmonic signal to be processed.

[0053] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the attached drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0054] Please refer to Figure 1 , Figure 1 It is a schematic flowchart of a method for restoring a harmonic signal provided by the present invention. The method for restoring the harmonic signal includes:

[0055] S101: Determine the signal mutation points in the harmonic signal to be processed. Here, the harmonic signal to be processed is a harmonic signal with several signal gaps, and the signal mutation point is the signal sampling point where the applicable harmonic signal model changes.

[0056] Specifically, considering the technical problems in the above background art and also considering that (1) the signal expression of the pre-signal segment passing through the signal gap can accurately restore the signal gap, and (2) the pre-signal segment of each signal gap can be quickly and accurately determined through the signal mutation point. Therefore, in the embodiments of the present invention, the signal expression of the pre-signal segment of the signal gap is intended to be used to restore the signal gap, and the signal mutation point is the prerequisite for accurately finding the "pre-signal segment of the signal gap". Therefore, in this step, the signal mutation points in the harmonic signal to be processed can be determined first, so as to use them to determine the pre-signal segment of the signal gap in the subsequent steps.

[0057] Among them, for a segment of harmonic signal to be processed, the applicable harmonic signal model may change at different positions, and the position between signals with different applicable harmonic signal models is the signal mutation point. The harmonic signal model can be composed of multiple harmonic signal sub-models, and the specific composition of the harmonic signal model depends on the signal itself corresponding to this segment of the harmonic signal model. The embodiments of the present invention do not limit this here.

[0058] S102: For any signal gap in the harmonic signal to be processed, determine the signal expression of the pre-signal segment of the signal gap. Here, the pre-signal segment of the signal gap is the signal segment between the pre-breakpoint of the signal gap and the signal gap, and the pre-breakpoint of the signal gap is the signal mutation point or another signal gap that is located before the signal gap and is the closest to the signal gap.

[0059] Specifically, since the pre-signal segment of the signal gap is defined as the signal segment between the pre-breakpoint of the signal gap (the other signal gap or signal mutation point closest to the signal gap before the signal gap) and the signal gap, after determining the signal mutation points in the signal to be processed, the expressions of the pre-signal segments of each signal gap can be determined and used as the data basis for the subsequent steps.

[0060] S103: For any signal gap in the harmonic signal to be processed, restore the signal gap according to the signal expression of the pre-signal segment of the signal gap.

[0061] Specifically, after determining the signal expressions of the pre-signal segments of each signal gap, since the signal to be restored in the signal gap is most likely a continuation of the signal in the pre-signal segment, in this step, for any signal gap in the harmonic signal to be processed, the signal gap can be restored according to the signal expression of the pre-signal segment of the signal gap, so that the restoration of the missing signal of the harmonic signal to be processed can be achieved efficiently and accurately.

[0062] The present invention provides a method for restoring a harmonic signal. Considering that the signal gap can be accurately restored through the signal expression of the pre-signal segment of the signal gap, and the pre-signal segments of each signal gap can be quickly and accurately determined through signal mutation points, in the present invention, first, the signal mutation points in the harmonic signal to be processed are determined, and then for any signal gap, the signal expression of the pre-signal segment of the signal gap is determined. Finally, the signal gap can be restored according to the signal expression of the pre-signal segment of the signal gap, so that the signal gaps of the harmonic signal to be processed can be restored efficiently and accurately.

[0063] Based on the above embodiments:

[0064] As an optional embodiment, determining the signal mutation points in the harmonic signal to be processed includes:

[0065] For any signal gap in the harmonic signal to be processed, determine whether the similarity between the two signal segments before and after the signal gap meets the standard, where the signal segment is: a continuous signal located before or after the signal gap;

[0066] If it meets the standard, it is determined that there are no signal mutation points in both signal segments before and after the signal gap;

[0067] If it does not meet the standard, it is determined that there are signal mutation points in at least one of the signal segments before and after the signal gap;

[0068] For any signal segment determined to possibly have signal mutation points, determine the signal mutation points in the signal segment.

[0069] Specifically, considering that when the harmonic signal models applicable to the two signal segments before and after the signal gap are the same, the similarity between the two signal segments before and after the signal gap is relatively high. In this case, it indicates that there are no signal mutation points in the two signal segments before and after the signal gap. Moreover, the recognition of "the similarity between the two signal segments before and after the signal gap" involves less computational effort compared to the recognition of "signal mutation points". Therefore, in the embodiments of the present invention, for any signal gap in the harmonic signal to be processed, it can be first determined whether the similarity between the two signal segments before and after the signal gap meets the standard. If it meets the standard, it is determined that there are no signal mutation points in the two signal segments before and after the signal gap. If it does not meet the standard, it can be determined that there are signal mutation points in at least one of the signal segments before and after the signal gap. This facilitates subsequent searching for signal mutation points only in the signal segments where signal mutation points are likely to exist, and there is no need to identify signal mutation points in the signal segments where signal mutation points are determined not to exist, thereby greatly reducing the computational effort and improving the recovery efficiency of the harmonic signal.

[0070] Specifically, for a better illustration of the embodiments of the present invention, please refer to Figure 2 and Figure 3 , Figure 2 which is a schematic structural diagram of a harmonic signal to be processed provided by the present invention. Figure 3 which is another schematic structural diagram of a harmonic signal to be processed provided by the present invention. In Figure 2 , the two harmonic signal models are distinguished by the expressions of "multiple horizontal lines" and "blank", that is, Figure 2 the harmonic signal to be processed in Figure 3 includes two harmonic signal models, while in Figure 3 the five harmonic signal models are distinguished by the ways of "cross, ellipse, multiple horizontal lines, triangle and pentagram", that is, Figure 2 the harmonic signal to be processed in

[0071] As an alternative embodiment, for any signal gap in the harmonic signal to be processed, the signal expression of the pre-signal segment of the signal gap is determined as follows:

[0072] For any signal gap in the harmonic signal to be processed, determine the signal models of the individual harmonic sub-signals included in the pre-signal segment of the signal gap;

[0073] For any signal gap in the harmonic signal to be processed, based on the data in the pre-signal segment of the signal gap, determine the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap.

[0074] Specifically, considering that any signal segment may be composed of the superposition of multiple harmonic sub-signals, the process of determining the signal expression of the harmonic signal model is equivalent to determining the expressions of the signal models of the respective harmonic sub-signals included in the signal segment. Therefore, in the embodiments of the present invention, for any signal gap in the harmonic signal to be processed, the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap can be determined, and then for any signal gap in the harmonic signal to be processed, based on the data in the pre-signal segment of the signal gap, determine the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap.

[0075] As an alternative embodiment, for any signal gap in the harmonic signal to be processed, determining the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap based on the data in the pre-signal segment of the signal gap includes:

[0076] For any signal gap in the harmonic signal to be processed, based on the data in the pre-signal segment of the signal gap, by using a signal parameter estimation method based on rotational invariance techniques, determine the parameter estimation values in the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap;

[0077] According to the total least squares method, correct the parameter estimation values to obtain parameter correction values;

[0078] Fill the parameter correction values into the expressions of the signal models of the respective harmonic sub-signals to which they belong.

[0079] Specifically, when determining the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap, it can be carried out based on the data in the pre-signal segment of the signal gap. The pre-signal segment of the signal gap consists of respective sampling signals. Considering that via ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques), the parameter estimation values in the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap can be efficiently estimated, and the accuracy of the parameter estimation values is generally average. Therefore, in the embodiments of the present invention, first, for any signal gap in the harmonic signal to be processed, based on the data in the pre-signal segment of the signal gap, by using the signal parameter estimation method based on rotational invariance techniques, the parameter estimation values in the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap are determined. Then, the parameter estimation values can be corrected by TLE (Total Least - Squares Estimation), so as to obtain parameter correction values with improved accuracy. Finally, the parameter correction values are filled into the expressions of the respective harmonic sub-signals to which they belong, and thus the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap are obtained efficiently and accurately.

[0080] Of course, in addition to this specific form, "determining the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap according to the data in the pre-signal segment of the signal gap" can also be implemented in many other ways, and the embodiments of the present invention do not limit this here.

[0081] Specifically, the specific operation logic for determining the expressions of the signal models of the respective harmonic sub-signals included in the pre-signal segment of the signal gap according to the data in the pre-signal segment of the signal gap can be as follows:

[0082] For signal decomposition and parameter estimation, the TLE-ESPRIT (Total Least - Squares Estimation via Signal Subspace Rotation) method is used.

[0083] The first step is to estimate the model order. For model order estimation, the logarithmic eigenvalue method is adopted. First, the eigenvalues of the sample correlation matrix are calculated, and then the highest eigenvalue is expressed as (where K is the number of the highest eigenvalues). Then, the logarithmic ratio is calculated, i = 1, 2... k; the logarithmic ratio is compared with the threshold , and this threshold Determined based on the worst-case noise level and the minimum harmonic amplitude, used to judge the model order.

[0084] After model order reduction, ESPRIT (Estimation of Signal Parameters using Rotational Invariance Techniques) is used to update the harmonic content to estimate the missing samples. This method first calculates the sample covariance matrix from the sample signal (2 base cycles after the subspace transformation point) and performs singular value decomposition. The eigenvector corresponding to the main eigenvalue d represents the signal subspace, and the remaining (M - d) eigenvectors represent the noise subspace. From the signal subspace U s , the two subspaces are calculated to form matrices U1 and U2 as follows:

[0085] ;

[0086] where M is the dimension parameter, used in related expressions such as matrix dimensions;

[0087] Then, using the least squares solution of ESPRIT, the rotation matrix is calculated as:

[0088] ;

[0089] The eigenvalues of give the diagonal invariance matrix

[0090] ;

[0091] The non-zero elements of the matrix represent the complex roots of the signal decomposition. The damping factor and frequency can be estimated from the eigenvalues as:

[0092] ;

[0093] ;

[0094] where diag is the diagonal matrix, β k is the damping factor, j represents the imaginary unit, w k is the frequency, real is the real part function, f s is the sampling frequency; imag is the imaginary part function.

[0095] The amplitude of the harmonic component can be estimated using the signal autocorrelation sequence, expressed as:

[0096] ;

[0097] The noise variance can be expressed as:

[0098] ;

[0099] where a i is the coefficient of the harmonic component, w i is the frequency of the harmonic component, is the Kronecker function, k = 0, 1, …; λ i is the i-th eigenvalue in "the eigenvalues obtained by performing eigenvalue decomposition on the covariance matrix", and the initial phase angle can be estimated by the least squares estimation method.

[0100] As an alternative embodiment, for any signal gap in the harmonic signal to be processed, determining whether the similarity between the two signal segments before and after the signal gap meets the standard includes:

[0101] For any signal gap in the harmonic signal to be processed, obtaining the lateral shift, tilt shift, and spectral shift between the two signal segments before and after the signal gap;

[0102] Based on the lateral shift, tilt shift, and spectral shift, determining whether the similarity between the two signal segments before and after the signal gap meets the standard.

[0103] Specifically, considering that the three indicators of lateral shift, tilt shift, and spectral shift can evaluate the similarity between the two signal segments before and after the signal gap from different perspectives, the embodiments of the present invention can, for any signal gap in the harmonic signal to be processed, obtain the lateral shift, tilt shift, and spectral shift between the two signal segments before and after the signal gap, and then determine whether the similarity between the two signal segments before and after the signal gap meets the standard based on the lateral shift, tilt shift, and spectral shift. Since multiple indicators reflecting different angles are integrated for similarity evaluation, the accuracy of similarity determination can be improved.

[0104] Of course, in addition to this specific method, other methods can also be used to determine whether the similarity between the two signal segments before and after the signal gap meets the standard, and the embodiments of the present invention do not limit this here.

[0105] Specifically, the lateral displacement, tilt shift, and spectral shift are introduced here:

[0106] (1) Lateral Shift (LS): Used to estimate the change in the average intensity of the signal across the data gap:

[0107] ;

[0108] where gap is the signal gap, μ gapis the lateral shift between two signal segments before and after the signal gap, L is the preset number of signal selections, X PoG (i) represents the i-th sampling signal among the total L sampling signals selected after the signal gap, X PeG and (i) represents the i-th sampling signal among the total L sampling signals selected before the signal gap.

[0109] Define a scaling factor for the lateral displacement:

[0110] ;

[0111] where, X scale is the scaling factor of the lateral displacement, , (i = 1, 2, 3…L), so the scaled lateral displacement is:

[0112] .

[0113] (2) Inclined Shift (IS): Used to estimate the standard deviation of the signal across the data gap (before and after the data gap):

[0114] ;

[0115] where, represents the inclined shift between two signal segments before and after the signal gap, , X includes X PoG or X PeG .

[0116] The scaled inclined shift is:

[0117] .

[0118] (3) Spectrum Shift (SS): Used to estimate the similarity between the spectral contents of the signals across the data gap. The spectrum shift is a new mean shift variant of the cross Teager-Kaiser energy operator, called mean-shifted cross-energy (MSCE), which can effectively eliminate the sensitivity of the operator to the initial phase angle difference between the signal pairs and the noise. MSCE calculates the difference in the mean cross-energy of the time-shifted signal pairs. First, calculate the cross-energy values of the time-shifted signals x(n) and y(n) for k samples, as shown in the following two relational expressions:

[0119] ;

[0120] ;

[0121] Among them, x(n) and y(n) represent two time-shifted signals, where n is the sample sequence number; i is the tracking variable, and its value range is i = 1, 2... k, which is used to traverse different sample displacement situations when calculating the cross-energy value; k is the number of sample displacements, which is used to define the sample range when calculating the cross-energy value; N is the upper limit in the summation operation, that is, the terms from n = 1 to n = N are summed; is the calculation result of the cross-energy value calculation formula of x(n) and y(n) signals according to k samples, which reflects the cross-energy relationship between the x signal and the y signal under different displacements i; is the calculation result of the cross-energy value calculation formula of y(n) and x(n) signals according to k samples, which reflects the cross-energy relationship between the y signal and the x signal under different displacements i;

[0122] Therefore, the average difference between the absolute time-shift cross-energy values is calculated as:

[0123] ;

[0124] Define the scaling factor of MSCE as:

[0125] ;

[0126] Among them, ;

[0127] The scaled MSCE is:

[0128] .

[0129] As an optional embodiment, determining whether the similarity of two signal segments before and after a signal gap meets the standard according to horizontal shift, tilt shift, and spectral shift includes:

[0130] Based on horizontal shift, tilt shift, and spectral shift, determine the spectral continuity parameter between two signal segments before and after the signal gap through the first relational expression;

[0131] Judge whether the frequency continuity parameter is lower than the first preset threshold;

[0132] If it is lower, it is determined that the similarity of the two signal segments before and after the signal gap meets the standard;

[0133] The spectral continuity parameter includes:

[0134] ;

[0135] Among them, SC k is the spectral continuity parameter of the signal gap with sequence number k, is the horizontal shift between two signal segments before and after the signal gap, is the tilt shift between two signal segments before and after the signal gap, is the spectral shift between two signal segments before and after the signal gap, where x is the signal of the signal segment before the signal gap and y is the signal of the signal segment after the signal gap.

[0136] Specifically, in order to better synthesize the three indicators of horizontal shift, tilt shift and spectral shift, and at the same time to more conveniently determine whether the "similarity meets the standard", the spectral continuity parameter SC is defined in the embodiments of the present invention k so as to synthesize the horizontal shift, tilt shift and spectral shift into an evaluation parameter, and then by comparing the spectral continuity parameter with a preset first preset threshold, it is possible to conveniently determine whether the "similarity meets the standard", which is beneficial to improving the efficiency and accuracy of similarity recognition.

[0137] Of course, in addition to this specific form, "judging whether the similarity of two signal segments before and after the signal gap meets the standard according to the horizontal shift, tilt shift and spectral shift" can also be other specific ways, which are not limited in the embodiments of the present invention.

[0138] As an optional embodiment, for any signal segment where a signal mutation point may exist, determining the signal mutation point in the signal segment includes:

[0139] For any signal segment where a signal mutation point may exist, determining the initial noise subspace according to the data in the signal segment;

[0140] For any sampling signal in the signal segment, determining the subspace affinity between the sampling signal and the initial noise subspace;

[0141] Determining the change rate between the subspace affinities of any two adjacent sampling signals in the signal segment;

[0142] Taking the position between a pair of sampling signals with a change rate of subspace affinity higher than the second preset threshold as a signal mutation point.

[0143] Specifically, the method in the embodiments of the present invention can be called the ROSAC (rate of subspace affinity change) method. First, the initial noise subspace can be determined according to the data in the signal segment, then the subspace affinity between the sampling signal and the initial noise subspace can be determined, and then the change rate between the subspace affinities of any two adjacent sampling signals in the signal segment can be determined. Finally, the position between a pair of sampling signals with a change rate of subspace affinity higher than the second preset threshold can be taken as a signal mutation point, and each signal mutation point in the signal segment can be determined efficiently and accurately.

[0144] Among them, for any sampling signal in the signal segment, determining the subspace affinity between the sampling signal and the initial noise subspace may include:

[0145] For any sampling signal in the signal segment, the subspace affinity between the sampling signal and the initial noise subspace is estimated by the Frobenius norm distance between the sampling signal and the initial noise subspace.

[0146] Of course, in addition to this specific method, "for any signal segment where a signal mutation point may be determined, determining the signal mutation point in the signal segment" can also be implemented by other methods, which are not limited in the embodiments of the present invention.

[0147] Specifically, a specific embodiment of determining the signal mutation point in the signal segment is:

[0148] Subspace change point detection: To track subspace changes and find change points, a method of rate of subspace affinity change (ROSAC) is proposed. This method first estimates the subspace affinity according to the Frobenius norm distance between the initial noise subspace and the sample signal Z(t). For the estimation of the initial noise subspace, the sample vector can be derived from the data before the gap; the autocorrelation matrix A is calculated from the initialized vector D as:

[0149] ;

[0150] where () H represents the conjugate transpose operation; D0(i) represents the i-th element in the initialized sample vector D0, i = 1, 2, 3... T0; T0 is a parameter when calculating the autocorrelation matrix, used for the upper limit correlation of the summation operation;

[0151] Then, the eigenvalue decomposition of A D is performed to obtain , where , are the initial signal subspace (the largest eigenvalue is r) and the noise subspace respectively.

[0152] Among them, is the initial signal subspace with a dimension of , and the largest corresponding eigenvalue is r. r is a dimension parameter related to the initial signal subspace, and M is a dimension-related parameter, used in scenarios such as subspace dimension representation; is the initial noise subspace with a dimension of , and the superscript T is the transpose operator.

[0153] The subspace affinity H between these initial noise subspaces and the sample signal Z(t) t is calculated as follows:

[0154] ;

[0155] where is the noise variance (estimated from the initial signal vector).

[0156] To effectively distinguish the subspace change points, the change rate of the affinity is calculated on the sample length N of the basic period, and the calculation formula is as follows:

[0157] ;

[0158] where N is the sample length of the basic period, which is used to calculate the change rate of the affinity; The value of the subspace affinity change rate at time t + N. If there is a spike in the ROSAC curve, it indicates that the spectral content of the sample signal at the spike has changed, that is, a signal gap has occurred.

[0159] Please refer to Figure 4 , Figure 4 which is a schematic structural diagram of a harmonic signal recovery device provided by the present invention. The harmonic signal recovery device includes:

[0160] A first determination module 41, configured to determine the signal mutation points in the harmonic signal to be processed, where the harmonic signal to be processed is a harmonic signal with several signal gaps, and the signal mutation points are the signal sampling points where the applicable harmonic signal model changes;

[0161] A second determination module 42, configured to determine the signal expression of the pre-signal segment of the signal gap for any signal gap in the harmonic signal to be processed, where the pre-signal segment of the signal gap is the signal segment between the pre-break point of the signal gap and the signal gap, and the pre-break point of the signal gap is the signal mutation point or another signal gap that is located before the signal gap and is the closest to the signal gap;

[0162] A recovery module 43, configured to recover the signal gap according to the signal expression of the pre-signal segment of the signal gap for any signal gap in the harmonic signal to be processed.

[0163] For the introduction of the harmonic signal recovery device provided by the embodiments of the present invention, please refer to the embodiments of the foregoing harmonic signal recovery method. The embodiments of the present invention will not be elaborated herein.

[0164] Please refer to Figure 5 , Figure 5 which is a schematic structural diagram of a harmonic signal recovery device provided by the present invention. The harmonic signal recovery device includes:

[0165] A memory for storing a computer program;

[0166] A processor for implementing the steps of the harmonic signal recovery method in the foregoing embodiments when executing the computer program.

[0167] For the introduction of the harmonic signal recovery device provided in the embodiments of the present invention, please refer to the embodiments of the harmonic signal recovery method described above, and the embodiments of the present invention will not be elaborated herein.

[0168] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the harmonic signal recovery method in the foregoing embodiments are implemented.

[0169] For the introduction of the computer-readable storage medium provided in the embodiments of the present invention, please refer to the embodiments of the harmonic signal recovery method described above, and the embodiments of the present invention will not be elaborated herein.

[0170] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method part. It should also be noted that in this specification, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variation thereof is intended to cover a non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the existence of additional identical elements in the process, method, article or device comprising the element.

[0171] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for restoring a harmonic signal, characterized in that: include: Determine a signal mutation point in the harmonic signal to be processed, wherein the harmonic signal to be processed is: a harmonic signal with a plurality of signal gaps, and the signal mutation point is: a signal sampling point at which an applicable harmonic signal model changes; For any signal gap in the harmonic signal to be processed, a signal expression of a preceding signal segment of the signal gap is determined, wherein the preceding signal segment of the signal gap is: a signal segment between a preceding breakpoint of the signal gap and the signal gap, and the preceding breakpoint of the signal gap is: a signal mutation point or another signal gap that is located before the signal gap and closest to the signal gap; For any signal gap in the harmonic signal to be processed, the signal gap is restored according to a signal expression of a preceding signal segment of the signal gap.

2. The method for restoring harmonic signals according to claim 1, characterized in that: Determining the signal mutation point in the harmonic signal to be processed includes: For any signal gap in the harmonic signal to be processed, determine whether the similarity between two signal segments before and after the signal gap meets the standard, wherein the signal segment is: a continuous signal before or after the signal gap; If the standard is met, it is determined that there is no signal mutation point in the two signal segments before and after the signal gap; If the standard is not met, determining that at least one signal segment before and after the signal gap has a signal mutation point; For any signal segment determined to have a possible signal mutation point, the signal mutation point in the signal segment is determined.

3. The method for restoring harmonic signals according to claim 2, characterized in that: For any signal gap in the harmonic signal to be processed, the signal expression for determining the preceding signal segment of the signal gap includes: For any signal gap in the harmonic signal to be processed, determining a signal model of each harmonic sub-signal contained in a preceding signal segment of the signal gap; For any signal gap in the harmonic signal to be processed, an expression of a signal model of each harmonic sub-signal contained in the preceding signal segment of the signal gap is determined according to data in the preceding signal segment of the signal gap.

4. The method for restoring harmonic signals according to claim 3, characterized in that: For any signal gap in the harmonic signal to be processed, the expression for determining the signal model of each harmonic sub-signal contained in the preceding signal segment of the signal gap according to the data in the preceding signal segment of the signal gap includes: For any signal gap in the harmonic signal to be processed, based on data in a preceding signal segment of the signal gap, a signal parameter estimation method based on rotational invariance technology is used to determine parameter estimation values ​​in an expression of a signal model of each harmonic sub-signal contained in the preceding signal segment of the signal gap; According to the total least squares method, the parameter estimation value is corrected to obtain a parameter correction value; The parameter correction value is filled into the expression of the signal model of each corresponding harmonic sub-signal.

5. The method for restoring harmonic signals according to claim 2, characterized in that: For any signal gap in the harmonic signal to be processed, determining whether the similarity between two signal segments before and after the signal gap meets the standard includes: For any signal gap in the harmonic signal to be processed, obtaining the lateral shift, tilt shift and spectrum shift between two signal segments before and after the signal gap; According to the lateral shift, the tilt shift and the spectrum shift, it is determined whether the similarity between the two signal segments before and after the signal gap meets the standard.

6. The method for restoring harmonic signals according to claim 5, characterized in that: Judging whether the similarity between the two signal segments before and after the signal gap meets the standard according to the lateral shift, the tilt shift and the spectrum shift includes: Based on the lateral shift, the tilt shift and the spectrum shift, a spectrum continuity parameter between two signal segments before and after the signal gap is determined by a first relational expression; Determining whether the frequency continuity parameter is lower than a first preset threshold; If it is lower than, it is determined that the similarity of the two signal segments before and after the signal gap meets the standard; The spectrum continuity parameters include: ； Among them, SC k is the spectrum continuity parameter of the signal gap with sequence number k, is the lateral shift between the two signal segments before and after the signal gap, is the tilt shift between the two signal segments before and after the signal gap, is the spectrum shift between the two signal segments before and after the signal gap, x is the signal of the signal segment before the signal gap, and y is the signal of the signal segment after the signal gap.

7. The harmonic signal recovery device according to any one of claims 2 to 6, characterized in that: For any signal segment determined to have a possible signal mutation point, determining the signal mutation point in the signal segment includes: For any signal segment determined to have a possible signal mutation point, determining an initial noise subspace according to data in the signal segment; For any sampled signal in the signal segment, determining a subspace affinity between the sampled signal and the initial noise subspace; Determining the change rate between the subspace affinities of any two adjacent sampled signals in the signal segment; The position between a pair of sampling signals where the rate of change between the subspace affinities is higher than a second preset threshold is taken as a signal mutation point.

8. A harmonic signal recovery device, characterized in that: include: The first determination module is used to determine a signal mutation point in a harmonic signal to be processed, wherein the harmonic signal to be processed is a harmonic signal with a plurality of signal gaps, and the signal mutation point is a signal sampling point at which an applicable harmonic signal model changes; A second determination module is used to determine, for any signal gap in the harmonic signal to be processed, a signal expression of a preceding signal segment of the signal gap, wherein the preceding signal segment of the signal gap is: a signal segment between a preceding breakpoint of the signal gap and the signal gap, and the preceding breakpoint of the signal gap is: a signal mutation point or another signal gap that is located before the signal gap and closest to the signal gap; The recovery module is used to recover any signal gap in the harmonic signal to be processed according to the signal expression of the preceding signal segment of the signal gap.

9. A harmonic signal recovery device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the harmonic signal recovery method according to any one of claims 1 to 7 when executing the computer program.

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

Images