Signal processing method, device, equipment, medium and product

Abstract

The invention discloses a signal processing method, device and equipment, a medium and a product, and belongs to the technical field of signal processing. According to the invention, the original vibration signal and the multi-component signal model corresponding to the original vibration signal are obtained; according to a discretized scale basis function and a window function corresponding to an original vibration signal, firstly determining an initial sub-time-frequency representation result of each frequency component, performing denoising processing on the initial sub-time-frequency representation result based on an instantaneous frequency of the initial sub-time-frequency representation result, preliminarily removing interference of noise, and then performing denoising processing on the initial sub-time-frequency representation result based on the instantaneous frequency of the initial sub-time-frequency representation result; and then, in combination with the group delay parameter corresponding to each frequency component, further noise processing is performed on the target sub-time-frequency representation result obtained by preliminary denoising, so that the leakage of time-frequency energy is reduced, the readability of the time-frequency representation result of the original vibration signal is improved, the time-frequency cross characteristics of the original vibration signal can be accurately determined, and the accuracy of the time-frequency cross characteristics of the original vibration signal is improved. And a reliable reference is provided for fault diagnosis and state monitoring of the equipment.

Description

Technical Field

[0001] This application belongs to the field of signal processing technology, specifically relating to a signal processing method, apparatus, device, medium, and product. Background Technology

[0002] Due to their high load-bearing capacity, compact structure, and high transmission efficiency, some mechanical equipment, such as rotating machinery, is increasingly widely used in complex industrial fields such as national defense, energy, and aerospace. The reliable operation of these mechanical equipment directly determines the safety and economic benefits of industrial production; therefore, fault diagnosis and condition monitoring of these mechanical equipment are particularly important.

[0003] In practical engineering, fault diagnosis and condition monitoring of mechanical equipment mainly rely on the time-frequency crossover characteristics in the vibration signals of mechanical equipment. However, the current methods are difficult to accurately determine the time-frequency crossover characteristics of vibration signals. The crossover characteristics in time-frequency characterization suffer from severe time-frequency smearing and distortion, which seriously affects the fault diagnosis and condition monitoring of mechanical equipment. Summary of the Invention

[0004] The purpose of this application is to provide a signal processing method, apparatus, device, medium, and product that can accurately determine the time-frequency crossover characteristics of vibration signals, providing a reference for fault diagnosis and condition monitoring of mechanical equipment.

[0005] In a first aspect, embodiments of this application provide a signal processing method, including:

[0006] The original vibration signal and the corresponding multi-component signal model are obtained. The multi-component signal model includes multiple frequency components, each of which has a corresponding amplitude and frequency.

[0007] Based on the discretized scale basis function and the window function corresponding to the original vibration signal, the initial sub-time-frequency characterization result corresponding to the frequency component is determined. The window function is used to traverse the frequency component in the time domain.

[0008] Based on the instantaneous frequency of the initial sub-time-frequency characterization result corresponding to the frequency component, the initial sub-time-frequency characterization result is denoised to obtain the target sub-time-frequency characterization result corresponding to the frequency component.

[0009] Determine the group delay parameter corresponding to the frequency component based on the target sub-time-frequency characterization results;

[0010] Based on the group delay parameter, the time-frequency characterization results of the target sub-signal are denoised to obtain the time-frequency characterization results of the original vibration signal.

[0011] Based on the time-frequency characterization results, the time-frequency crossover characteristics of the original vibration signal are determined.

[0012] Secondly, embodiments of this application provide a signal processing apparatus, including:

[0013] The acquisition module is used to acquire the original vibration signal and the multi-component signal model corresponding to the original vibration signal. The multi-component signal model includes multiple frequency components, each of which has a corresponding amplitude and frequency.

[0014] The determination module is used to determine the initial sub-time-frequency characterization result corresponding to the frequency component based on the discretized scale basis function and the window function corresponding to the original vibration signal. The window function is used to traverse the frequency component in the time domain.

[0015] The processing module is used to perform noise reduction processing on the initial sub-time-frequency characterization result based on the instantaneous frequency of the initial sub-time-frequency characterization result corresponding to the frequency component, so as to obtain the target sub-time-frequency characterization result corresponding to the frequency component.

[0016] The determination module is also used to determine the group delay parameter corresponding to the frequency component based on the target sub-time-frequency characterization result;

[0017] The processing module is also used to denoise the target sub-time-frequency characterization results based on the group delay parameter to obtain the time-frequency characterization results of the original vibration signal.

[0018] The determination module is also used to determine the time-frequency crossover characteristics of the original vibration signal based on the time-frequency characterization results.

[0019] Thirdly, embodiments of this application provide an electronic device including a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the method as described in the first aspect.

[0020] Fourthly, embodiments of this application provide a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method as described in the first aspect.

[0021] Fifthly, embodiments of this application provide a computer program product stored in a storage medium, which, when executed by at least one processor, implements the steps of the method described in the first aspect.

[0022] In this embodiment, the original vibration signal and the corresponding multi-component signal model are obtained. The multi-component signal model includes multiple frequency components, each with a corresponding amplitude and frequency. Based on the discretized scaling basis function and the window function corresponding to the original vibration signal, the initial sub-time-frequency characterization result for each frequency component is first determined. Then, based on the instantaneous frequency of the initial sub-time-frequency characterization result, denoising processing is performed on the initial sub-time-frequency characterization result to initially remove noise interference. Next, combined with the group delay parameter corresponding to each frequency component, further noise processing is performed on the target sub-time-frequency characterization result obtained from the initial denoising, reducing time-frequency energy leakage and improving the readability of the time-frequency characterization result of the original vibration signal. This allows for accurate determination of the time-frequency crossover characteristics of the original vibration signal, providing a reliable reference for fault diagnosis and condition monitoring of mechanical equipment. Attached Figure Description

[0023] Figure 1 A flowchart illustrating a signal processing method provided in an embodiment of this application;

[0024] Figure 2 A flowchart illustrating another signal processing method provided in this application embodiment;

[0025] Figure 3 A flowchart illustrating another signal processing method provided in this application embodiment;

[0026] Figure 4 A schematic diagram illustrating an ideal time-frequency characterization result of an original vibration signal provided in an embodiment of this application;

[0027] Figure 5 This is a schematic diagram of the time-frequency characterization results obtained by related technologies through analysis of the original vibration signal;

[0028] Figure 6 This is a schematic diagram of the time-frequency characterization results obtained by analyzing the original vibration signal in an embodiment of this application;

[0029] Figure 7 This is a schematic diagram of the structure of a signal processing device provided in an embodiment of this application;

[0030] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0031] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0032] The terms "first," "second," etc., used in this application's specification are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, without limiting the number of objects; for example, a first object can be one or more. Furthermore, in the specification, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects have an "or" relationship.

[0033] As mentioned above, vibration signals of mechanical equipment are the basis for analyzing whether mechanical equipment has faults. For mechanical equipment used in complex industrial fields, vibration signals often have multiple nonlinear and non-stationary characteristics, and are severely interfered with by noise, making it difficult to accurately extract the time-frequency cross-characteristics of vibration signals, thus affecting the fault diagnosis and condition monitoring of mechanical equipment.

[0034] Therefore, embodiments of this application provide a signal processing method, apparatus, device, medium, and product that can accurately determine the time-frequency crossover characteristics of vibration signals, providing a reference for fault diagnosis and condition monitoring of mechanical equipment.

[0035] The signal processing methods, apparatus, devices, media, and products provided in this application will be described below with reference to the accompanying drawings and specific embodiments.

[0036] Figure 1 The flowchart illustrates a signal processing method provided in this application embodiment. This signal processing method can be applied to devices with data processing capabilities, such as laptops, tablets, desktops, industrial computers, servers, or controllers of mechanical equipment.

[0037] like Figure 1 As shown, the signal processing method may include the following steps:

[0038] S110. Obtain the original vibration signal and the multi-component signal model corresponding to the original vibration signal.

[0039] The multi-component signal model includes multiple frequency components, each with a corresponding amplitude and frequency.

[0040] S120. Based on the discretized scale basis function and the window function corresponding to the original vibration signal, determine the initial sub-time-frequency characterization result corresponding to the frequency component.

[0041] The window function is used to traverse the frequency components in the time domain.

[0042] S130. Based on the instantaneous frequency of the initial sub-time-frequency characterization result corresponding to the frequency component, perform noise reduction processing on the initial sub-time-frequency characterization result to obtain the target sub-time-frequency characterization result corresponding to the frequency component.

[0043] S140. Determine the group delay parameter corresponding to the frequency component based on the target sub-time-frequency characterization results.

[0044] S150. Based on the group delay parameter, the time-frequency characterization result of the target sub-signal is denoised to obtain the time-frequency characterization result of the original vibration signal.

[0045] S160. Based on the time-frequency characterization results, determine the time-frequency crossover characteristics of the original vibration signal.

[0046] In this embodiment, the original vibration signal and the corresponding multi-component signal model are obtained. The multi-component signal model includes multiple frequency components, each with a corresponding amplitude and frequency. Based on the discretized scaling basis function and the window function corresponding to the original vibration signal, the initial sub-time-frequency characterization result for each frequency component is first determined. Then, based on the instantaneous frequency of the initial sub-time-frequency characterization result, denoising processing is performed on the initial sub-time-frequency characterization result to initially remove noise interference. Next, combined with the group delay parameter corresponding to each frequency component, further noise processing is performed on the target sub-time-frequency characterization result obtained from the initial denoising, reducing time-frequency energy leakage and improving the readability of the time-frequency characterization result of the original vibration signal. This allows for accurate determination of the time-frequency crossover characteristics of the original vibration signal, providing a reliable reference for fault diagnosis and condition monitoring of mechanical equipment.

[0047] The above steps are explained in detail below:

[0048] In S110, the original vibration signal can be the vibration signal of the target mechanical equipment collected at a certain sampling frequency within a preset time period. The sampling frequency and the preset time period can be set according to actual needs. For example, in some embodiments, the sampling frequency can be set to 200Hz and the preset time period can be 2s.

[0049] The target mechanical equipment can be, for example, mechanical equipment applied in complex industrial fields such as defense, energy, and aerospace, or mechanical equipment applied in other complex industrial fields. For example, in some embodiments, the target mechanical equipment can be rotating machinery such as water pumps, generators, gears, etc.

[0050] This embodiment does not limit the method of acquiring the original vibration signal. For example, the original vibration signal of the target mechanical equipment can be acquired by a vibration sensor or other equipment.

[0051] The multi-component signal model is the signal model obtained corresponding to the original vibration signal. This multi-component signal model consists of multiple frequency components, and each frequency component is subjected to amplitude modulation and frequency modulation. That is, each frequency component has a corresponding amplitude and frequency.

[0052] For example, the original vibration signal can be subjected to Hilbert transform to convert the real-valued signal into a complex-valued signal, thereby obtaining the analytic signal of the original vibration signal.

[0053] For example, an analytic signal can be described by the following formula (1):

[0054]

[0055] Among them, s ′ x(t) is the analytic signal, x(t) is the original vibration signal, and x(t)∈L 2 (R), It is the imaginary unit.

[0056] Based on the analytical signal, the corresponding multi-component signal model can be obtained, as detailed in formula (2).

[0057]

[0058] Where s(t) represents the multi-component signal model, k represents the number of frequency components of the original vibration signal, and A i (t) and f i (t) represents the amplitude and instantaneous frequency of the i-th frequency component, respectively, and exp(·) represents the exponential function.

[0059] Each frequency component can be represented by the following formula (3):

[0060]

[0061] In S120, the scaling basis functions are time- and frequency-dependent basis functions, which help to obtain high-resolution and readable time-frequency representations.

[0062] For example, the scaling basis function can be expressed by the following formula (4):

[0063] Base(t) = exp(-j2πf c (c1(tt c ) 2 +c2(tt c ) 3 (4)

[0064] Where Base(t) represents the scaling basis function, t c Indicates the time center, f cRepresenting the frequency center, c1 and c2 are the core parameters that determine the matching between the scaling basis function and the signal frequency components. c1 and c2 can be expressed by the following formula (5):

[0065]

[0066] Where m0 and n0 are parameters of the nonlinearity of the two balancing tangent functions, and θ1 and θ2 are the discrete angles of the scaling basis functions, which are between (-π / 2, π / 2). For example, θ1 and θ2 can be expressed by the following formula (6):

[0067]

[0068] Where N1 and N2 are the number of discrete angles in the scaling basis function matching process, and the discretized scaling basis function can be obtained by substituting formula (5) into formula (4).

[0069] The window function here is used to traverse the frequency components in the time domain. For example, this window function can be a Gaussian window function. The width and step size of the window function can be set according to actual needs. By moving the window function in the time domain, the entire frequency component can be traversed.

[0070] The initial sub-time-frequency characterization result is the sub-time-frequency characterization result corresponding to each frequency component obtained initially, and this initial sub-time-frequency characterization result is a high-dimensional sub-time-frequency characterization result.

[0071] For example, for each frequency component, a window function can be used to traverse the frequency component, and the traversal result can be matched with the discretized scaling basis function to obtain the initial sub-time-frequency characterization result corresponding to the frequency component.

[0072] In S130, the instantaneous frequency of the initial sub-time-frequency characterization result can be obtained by solving the partial derivative of the initial sub-time-frequency characterization result with respect to time.

[0073] Based on the instantaneous frequency of the initial sub-time-frequency characterization result, noise in the initial sub-time-frequency characterization result can be initially removed to obtain the target sub-time-frequency characterization result. Compared with the initial sub-time-frequency characterization result, the target sub-time-frequency characterization result enhances the time-frequency characteristics in the signal, providing a foundation for obtaining a more readable time-frequency characterization later.

[0074] For example, a noise processing operator can be constructed based on the instantaneous frequency, and the noise in the initial sub-time-frequency characterization result can be initially removed based on the noise processing operator to obtain the target sub-time-frequency characterization result.

[0075] In S140, the group delay parameter is used to reduce leakage of time-frequency energy. For example, the group delay parameter may include at least one of the group delay corresponding to the frequency component and the group delay change rate.

[0076] For example, the group delay parameter can be obtained by solving the frequency derivative of the target sub-time-frequency characterization result.

[0077] In S150, the time-frequency characterization result is a two-dimensional time-frequency spectrum. For example, based on the group delay parameter, noise in the target sub-time-frequency result can be processed to achieve secondary noise removal, reduce time-frequency energy leakage, and improve the readability of the time-frequency characterization result.

[0078] For example, a noise processing operator can be constructed based on the group delay parameter, and the time-frequency characterization result of the target sub-sub ...

[0079] In S160, based on the time-frequency characterization results, the time-frequency crossover characteristics of the original vibration signal can be determined, providing a more accurate basis for subsequent equipment fault diagnosis.

[0080] The time-frequency crossover features here may include the time-frequency points of the crossover. Based on the time-frequency characterization results provided in the embodiments of this application, the time-frequency points of the crossover can be accurately determined.

[0081] This application embodiment improves the readability of video representation results through two noise processing steps, thereby accurately determining time-frequency crossover characteristics, providing a reference for equipment fault diagnosis and maintenance, and helping to improve the efficiency and reliability of industrial production processes.

[0082] Taking a Gaussian window function as an example, the above S120 may include the following steps:

[0083] The frequency components are traversed using a Gaussian window function to obtain the window function traversal result;

[0084] The discretized scaling basis functions are matched with the window function traversal results to obtain the matching results;

[0085] Perform a Fourier transform on the matching results to obtain the initial sub-time-frequency characterization results corresponding to the frequency components.

[0086] For example, the Gaussian window function can be expressed as the following formula (7):

[0087] g(t)=exp(-(t / σ) 2 / 2) (7)

[0088] Where g(t) represents the Gaussian window function, σ represents the window width factor, for example,

[0089] The traversal process of the Gaussian window function can be represented by the following formula (8):

[0090] Mid(t) = s i (t)g(tt c (8)

[0091] Where Mid(t) represents the result of traversing the i-th frequency component using the Gaussian window function.

[0092] The initial sub-time-frequency characterization result can be obtained by matching the discretized scale basis function with the window function traversal result and performing a Fourier transform on the matching result. For details, please refer to the following formula (9):

[0093]

[0094] Among them, Base ′ (t) represents the discretized scaling basis function, SBCT x (f c ,t c c) represents the initial sub-time-frequency characterization result.

[0095] This embodiment utilizes the discretized scaling basis function and sliding window function to analyze each frequency component of the original vibration signal through time-frequency analysis, obtaining the initial sub-time-frequency characterization result corresponding to each frequency component, which provides a foundation for accurately determining the time-frequency crossover characteristics of the original vibration signal.

[0096] Figure 2 A flowchart illustrating another signal processing method provided in this application embodiment. Figure 2 and Figure 1 The difference is that, Figure 1 S130 in the text can be further refined into Figure 2 S210-S230 in the middle.

[0097] S210. Based on the instantaneous frequency and the initial sub-time frequency characterization results, determine the ideal frequency center of the frequency component.

[0098] The instantaneous frequency of the initial sub-frequency characterization result can be expressed by the following formula (10):

[0099]

[0100] in, This represents the partial derivative with respect to the solution time of the initial sub-time-frequency characterization result, which is also the instantaneous frequency of the initial sub-time-frequency characterization result.

[0101] The actual frequency center of the frequency component is f in the above embodiment. c The ideal frequency center of the frequency component can be constructed based on the instantaneous frequency and the initial sub-time frequency characterization results, as shown in the following formula (11):

[0102]

[0103] Among them, f e (f c ,t c c) represents the ideal frequency center.

[0104] S220. Based on the ideal frequency center and the actual frequency center of the frequency component, and combined with the Dirac function, generate the first noise processing operator.

[0105] The first noise processing operator is used to remove noise from the initial sub-time-frequency characterization result and enhance the time-frequency characteristics of the signal. The first noise processing operator can be obtained based on the ideal frequency center, the actual frequency center, the Dirac function, and the following formula (12):

[0106]

[0107] Among them, HDSEO x (f c ,t c c) represents the first noise processing operator, δ(·) represents the Dirac function, specifically, when That is, f c =f e (f c ,t c c) When HDSEO x (f c ,t c c) = 1, otherwise, HDSEO x (f c ,t c c) = 0, that is, the first noise processing operator is a high-dimensional matrix composed of 0 and 1.

[0108] In some embodiments, considering the estimation error between the ideal frequency center and the actual frequency center of the signal, in order to reduce the impact of the estimation error on the first noise processing operator, for example, the following can be done: The real part is used for calculation, and the first noise processing operator is determined based on the result of the real part calculation. For details, please refer to the following formula (13):

[0109]

[0110] Where ε is a parameter of the balance error, that is, when When the real part is less than ε, it is considered that f c =f e (f c ,t c c), otherwise, consider f c ≠fe (f c ,t c ,c). Re(·) represents the real part operation.

[0111] S230. According to the first noise processing operator, the initial sub-time-frequency characterization result is denoised to obtain the target sub-time-frequency characterization result.

[0112] For example, noise reduction can be achieved using the following formula (14):

[0113] HDSECT x (f c ,t c c) = SBCT x (f c ,t c c)·HDSEO x (f c ,t c c) (14)

[0114] Among them, HDSECT x (f c ,t c c) represents the target sub-time-frequency characterization result. Compared with the initial sub-time-frequency characterization result, the target sub-time-frequency characterization result reduces a large amount of noise and enhances the time-frequency characteristics.

[0115] Based on the instantaneous frequency and the initial sub-time-frequency characterization results, the embodiments of this application construct an ideal frequency center for the frequency components. Based on the ideal frequency center and the actual frequency center, a first noise processing operator is constructed. By constructing the first noise processing operator, the interference of noise on the initial sub-time-frequency characterization results can be initially removed, and the effective components in the signal can be retained, providing a foundation for improving the accuracy of the time-frequency crossover features in the future.

[0116] Taking the group delay parameters, which include the group delay corresponding to the frequency components and the group delay change rate, as an example, S140 above may include the following steps:

[0117] The group delay corresponding to the frequency component is obtained by solving the first-order partial derivative of the frequency center of the target sub-time-frequency characterization result.

[0118] The second-order partial derivative of the frequency center is obtained by solving the time-frequency characterization result of the target sub-sub ...

[0119] For example, the group delay corresponding to the frequency component can be obtained by the following formula (15):

[0120]

[0121] in, This represents the first-order partial derivative, also known as the group delay.

[0122] For example, the group delay rate corresponding to the frequency component can be obtained by the following formula (16):

[0123]

[0124] The embodiments of this application solve the partial derivatives of the frequency center based on the target sub-time-frequency characterization results, which provides a basis for the subsequent construction of the second noise processing operator. Moreover, since the target sub-time-frequency characterization results have removed a large amount of noise compared with the initial sub-time-frequency characterization results, it helps to improve the accuracy of the second noise processing operator, thereby enabling a more accurate time-frequency characterization and time-frequency crossover feature of the original vibration signal.

[0125] Figure 3 A flowchart illustrating another signal processing method provided in this application embodiment. Figure 3 and Figure 1 The difference is that, Figure 1 S150 in the text can be further refined into Figure 3 S310-S330 in the series.

[0126] S310. Generate a second noise processing operator based on the group delay parameter and the target sub-time-frequency characterization result.

[0127] For example, a second noise processing operator can be constructed based on the group delay parameter and the target sub-time-frequency characterization result.

[0128] For example, a second noise processing operator can be constructed based on the group delay and the target sub-time-frequency characterization results.

[0129] For example, a second noise processing operator can be constructed based on the group delay change rate and the target sub-time-frequency characterization results.

[0130] For example, a second noise processing operator can be constructed based on the group delay, the group delay change rate, and the target sub-time-frequency characterization results.

[0131] Taking the group delay parameters, including the group delay corresponding to the frequency components and the group delay change rate, as an example, the second noise processing operator can correspondingly include a frequency redistribution operator and a frequency modulation redistribution operator. Based on this, in some embodiments, the above-mentioned S310 may include the following steps:

[0132] Based on the group delay change rate and the target sub-time-frequency characterization results, a frequency modulation redistribution operator is generated;

[0133] The frequency redistribution operator is generated based on the frequency redistribution operator, the time-frequency characterization results of the target sub-sub, and the group delay.

[0134] The frequency redistribution operator and the frequency modulation redistribution operator here work together to aggregate the energy of the target subtime-frequency characterization results and reduce energy leakage.

[0135] For example, the chirprate redistribution operator can be constructed using the following formula (17):

[0136]

[0137] Among them, c r (f c ,t c ,c) represents the frequency modulation redistribution operator, Im(·) represents the imaginary part operation, and c is a variable in the frequency modulation dimension.

[0138] For example, the frequency redistribution operator can be constructed using the following formula (18):

[0139]

[0140] Among them, f r (f c ,t c c) represents the frequency redistribution operator.

[0141] Based on group delay and group delay change rate, this application constructs a frequency redistribution operator and a frequency modulation redistribution operator, which can achieve secondary noise removal, reduce time-frequency energy leakage, and improve the accuracy of time-frequency characterization results.

[0142] S320. The target sub-time-frequency characterization result is denoised according to the second noise processing operator to obtain the processed target sub-time-frequency characterization result.

[0143] The processed target sub-time-frequency characterization result includes time, frequency, and frequency modulation dimensions. The magnitude of each dimension is related to time, frequency, and the number of discrete basis functions.

[0144] By performing secondary denoising, the high-dimensional target sub-time-frequency representation result can be transformed into a three-dimensional space, resulting in a sub-time-frequency representation result containing time dimension, frequency dimension, and frequency modulation dimension.

[0145] For example, the target sub-time-frequency characterization results can be denoised using the following steps:

[0146] Based on the frequency redistribution operator and the frequency modulation redistribution operator, the time-frequency energy corresponding to the target sub-time-frequency characterization result is concentrated at the target position in the plane where the frequency and frequency modulation of the frequency component are located, and the processed target sub-time-frequency characterization result is obtained.

[0147] For example, the accumulation of time-frequency energy can be achieved by using the frequency redistribution operator and the frequency modulation redistribution operator in combination with the following formula (19):

[0148]

[0149] Where, ζ γ (α,t c ,β) represents the processed time-frequency characterization result of the target sub-sub ... -5 By analyzing f c Integrating with c can eliminate HDSECT. x (f c ,t c ,c)·δ(α-f r )·δ(β-c r f in ) c And c, we get only α and t c The sub-time-frequency characterization results of β.

[0150] By using the above formula (19), noise in the target sub-time-frequency characterization result can be removed, achieving secondary noise removal and reducing time-frequency energy leakage. Formula (19) can also concentrate time-frequency energy at the target position on the plane containing the frequency component's frequency-frequency modulation. Here, the target position is the position within the plane, which is specifically related to the traversal process of the window function.

[0151] S330. Along the frequency modulation direction, the processed target sub-time-frequency characterization result is integrated to obtain the time-frequency characterization result of the original vibration signal.

[0152] Here, the frequency modulation direction is the direction corresponding to the frequency modulation dimension. For example, to obtain a two-dimensional time-frequency characterization result, the ζ obtained in the above embodiment can be... γ (α,t c Perform a β-based integral to eliminate ζ. γ (α,t c β in ,β) yields only α and t c The two-dimensional time-frequency characterization results.

[0153] For example, the time-frequency characterization result can be obtained by the following formula (20):

[0154]

[0155] Among them, FCESSCT x (α,t c ) represents the time-frequency representation result, and R represents a real number.

[0156] Based on the initial denoising, this embodiment further constructs a second noise processing operator by using a group delay parameter, and then performs further denoising processing on the time-frequency characterization results of the target sub-signal based on the second noise processing operator. This reduces the leakage of time-frequency energy, thereby improving the accuracy of the time-frequency characterization results and enabling accurate determination of the time-frequency crossover characteristics of the original vibration signal.

[0157] In some embodiments, the signal processing method may further include the following steps:

[0158] Based on the time-frequency crossover characteristics, fault diagnosis is performed on the equipment corresponding to the original vibration signal to obtain the fault diagnosis results.

[0159] For example, the time-frequency crossover features of the extracted raw vibration signal can be matched with reference time-frequency crossover features to obtain the fault diagnosis result. Alternatively, it can be combined with a fault diagnosis model to determine the fault diagnosis result. Because the extracted time-frequency crossover features are relatively accurate, fault diagnosis of the equipment can be performed accurately, improving the accuracy of the diagnostic results.

[0160] The following simulation example illustrates the solution proposed in this application.

[0161] Taking the multi-component signal model of a certain original vibration signal as an example, as described by the following formula (21):

[0162]

[0163] Where n(t) represents Gaussian white noise with a signal-to-noise ratio of 5dB, the frequency structure of the signal can be described by the following formula (22):

[0164]

[0165] The ideal time-frequency characterization results of the above-mentioned original vibration signals can be found in [reference]. Figure 4 That is, there is a clear time-frequency crossover feature at the time-frequency point (1s, 50Hz).

[0166] The analysis parameters for the simulated signal are as follows: sampling frequency fs = 200Hz, analysis duration = 2s, window function length Nw = 150, window function step size Ns = 1, number of discrete angles of the scaling basis function N1 = 70, nonlinear parameter m0 of the balanced tangent function = 10, and minimum parameter ε = 0.5. Based on the determination of the above analysis parameters, the simulated signal is analyzed according to the scheme of this application embodiment, and the time-frequency characterization is calculated to obtain... Figure 6 The results are shown. Among them, Figure 5 The time-frequency characterization results obtained by related techniques are obtained through... Figure 4 It can be seen that the time-frequency crossover features in the time-frequency representation obtained by the relevant techniques are severely damaged, and the frequency structure of the signal cannot be accurately described.

[0167] It can be seen that the readability of the time-frequency characterization results of the embodiments of this application is significantly enhanced compared with related technologies, and noise in the signal is effectively removed, which can clearly describe the time-frequency crossover characteristics of the signal.

[0168] In this embodiment, a first noise processing operator is constructed based on the instantaneous frequency of the initial sub-time-frequency characterization result to perform preliminary denoising on the initial sub-time-frequency characterization result. On this basis, a frequency redistribution operator and a frequency modulation redistribution operator are constructed based on the target sub-time-frequency characterization result to rearrange the target sub-time-frequency characterization result and project the original time-frequency characterization into a new time-frequency space. This achieves multiple removals of noise in the signal and completes the effective characterization of time-frequency crossover features, providing an accurate basis for subsequent fault diagnosis.

[0169] Based on the same inventive concept, embodiments of this application also provide a signal processing apparatus. Figure 7 This is a schematic diagram of the structure of a signal processing device provided in an embodiment of this application.

[0170] like Figure 7 As shown, the signal processing device 700 may include:

[0171] The acquisition module 701 is used to acquire the original vibration signal and the multi-component signal model corresponding to the original vibration signal. The multi-component signal model includes multiple frequency components, and each frequency component has a corresponding amplitude and frequency.

[0172] The determination module 702 is used to determine the initial sub-time-frequency characterization result corresponding to the frequency component based on the discretized scale basis function and the window function corresponding to the original vibration signal. The window function is used to traverse the frequency component in the time domain.

[0173] Processing module 703 is used to perform noise reduction processing on the initial sub-time-frequency characterization result based on the instantaneous frequency of the initial sub-time-frequency characterization result corresponding to the frequency component, so as to obtain the target sub-time-frequency characterization result corresponding to the frequency component.

[0174] The determination module 702 is also used to determine the group delay parameter corresponding to the frequency component based on the target sub-time-frequency characterization result;

[0175] The processing module 703 is also used to perform noise reduction processing on the target sub-time-frequency characterization result according to the group delay parameter to obtain the time-frequency characterization result of the original vibration signal;

[0176] The determination module 702 is also used to determine the time-frequency crossover characteristics of the original vibration signal based on the time-frequency characterization results.

[0177] In this embodiment, the original vibration signal and the corresponding multi-component signal model are obtained. The multi-component signal model includes multiple frequency components, each with a corresponding amplitude and frequency. Based on the discretized scaling basis function and the window function corresponding to the original vibration signal, the initial sub-time-frequency characterization result for each frequency component is first determined. Then, based on the instantaneous frequency of the initial sub-time-frequency characterization result, denoising processing is performed on the initial sub-time-frequency characterization result to initially remove noise interference. Next, combined with the group delay parameter corresponding to each frequency component, further noise processing is performed on the target sub-time-frequency characterization result obtained from the initial denoising, reducing time-frequency energy leakage and improving the readability of the time-frequency characterization result of the original vibration signal. This allows for accurate determination of the time-frequency crossover characteristics of the original vibration signal, providing a reliable reference for fault diagnosis and condition monitoring of mechanical equipment.

[0178] In some possible implementations of the embodiments of this application, the determining module 702 is further configured to determine the ideal frequency center of the frequency component based on the instantaneous frequency and the initial sub-time-frequency characterization result;

[0179] Processing module 703 is specifically used for:

[0180] Based on the ideal frequency center and the actual frequency center of the frequency components, and combined with the Dirac function, the first noise processing operator is generated.

[0181] Based on the first noise processing operator, the initial sub-time-frequency characterization result is denoised to obtain the target sub-time-frequency characterization result.

[0182] In some possible implementations of the embodiments of this application, the group delay parameter includes the group delay corresponding to the frequency component and the group delay change rate;

[0183] Module 702 is specifically used for:

[0184] The group delay corresponding to the frequency component is obtained by solving the first-order partial derivative of the frequency center of the target sub-time-frequency characterization result.

[0185] The second-order partial derivative of the frequency center is obtained by solving the time-frequency characterization result of the target sub-sub ...

[0186] In some possible implementations of the embodiments of this application, the processing module 703 is specifically used for:

[0187] Based on the group delay parameter and the target sub-time-frequency characterization results, a second noise processing operator is generated;

[0188] The target sub-time-frequency characterization result is denoised according to the second noise processing operator to obtain the processed target sub-time-frequency characterization result, which includes time dimension, frequency dimension and frequency modulation dimension.

[0189] Along the frequency modulation direction, the processed target sub-time-frequency characterization result is integrated to obtain the original vibration signal's time-frequency characterization result. The frequency modulation direction is the direction corresponding to the frequency modulation dimension.

[0190] In some possible implementations of the embodiments of this application, the group delay parameter includes the group delay and the group delay change rate corresponding to the frequency component, and the second noise processing operator includes a frequency redistribution operator and a frequency modulation redistribution operator.

[0191] Processing module 703 is specifically used for:

[0192] Based on the group delay change rate and the target sub-time-frequency characterization results, a frequency modulation redistribution operator is generated;

[0193] The frequency redistribution operator is generated based on the frequency redistribution operator, the time-frequency characterization results of the target sub-sub, and the group delay.

[0194] In some possible implementations of the embodiments of this application, the second noise processing operator includes a frequency redistribution operator and a frequency modulation redistribution operator;

[0195] Processing module 703 is specifically used for:

[0196] Based on the frequency redistribution operator and the frequency modulation redistribution operator, the time-frequency energy corresponding to the target sub-time-frequency characterization result is concentrated at the target position in the plane where the frequency and frequency modulation of the frequency component are located, and the processed target sub-time-frequency characterization result is obtained.

[0197] In some possible implementations of the embodiments of this application, the window function includes a Gaussian window function;

[0198] Module 702 is specifically used for:

[0199] The frequency components are traversed using a Gaussian window function to obtain the window function traversal result;

[0200] The discretized scaling basis functions are matched with the window function traversal results to obtain the matching results;

[0201] Perform a Fourier transform on the matching results to obtain the initial sub-time-frequency characterization results corresponding to the frequency components.

[0202] In some possible implementations of the embodiments of this application, the processing module 703 is further configured to perform fault diagnosis on the device corresponding to the original vibration signal based on the time-frequency crossover characteristics, and obtain the fault diagnosis result.

[0203] Based on the same inventive concept, this application also provides an electronic device, which is described below in conjunction with... Figure 8 The electronic devices provided in the embodiments of this application will be described.

[0204] like Figure 8 As shown, the electronic device 800 may include a processor 801 and a memory 802 for storing computer program instructions.

[0205] The processor 801 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0206] Memory 802 may include mass storage for data or instructions. For example, and not limitingly, memory 802 may include a hard disk drive (HDD), a floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. In one instance, memory 802 may include removable or non-removable (or fixed) media, or memory 802 may be non-volatile solid-state memory. In one instance, memory 802 may be read-only memory (ROM). In one instance, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0207] The processor 801 reads and executes computer program instructions stored in the memory 802 to achieve... Figures 1-3 The method in the illustrated embodiment achieves... Figures 1-3 The corresponding technical effects achieved by the methods in the illustrated embodiments are described briefly and will not be elaborated further here.

[0208] In one example, the electronic device 800 may also include a communication interface 803 and a bus 804. For example, Figure 8 As shown, the processor 801, memory 802, and communication interface 803 are connected through bus 804 and complete communication with each other.

[0209] The communication interface 803 is mainly used to realize communication between various modules, devices and / or equipment in the embodiments of this application.

[0210] Bus 804 includes hardware, software, or both, that couples the various components of electronic device 800 together. For example, and not as a limitation, bus 804 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 804 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0211] Furthermore, in conjunction with the signal processing methods in the above embodiments, this application embodiment can provide a readable storage medium for implementation. The readable storage medium stores program instructions; when these program instructions are executed by a processor, they implement any of the signal processing methods in the above embodiments.

[0212] Furthermore, in conjunction with the signal processing methods in the above embodiments, this application embodiment can provide a computer program product for implementation. This program product is stored in a storage medium, and when executed by at least one processor, it implements any of the signal processing methods in the above embodiments.

[0213] Although this application has been described with reference to preferred embodiments, various modifications can be made thereto and components can be replaced with equivalents without departing from the scope of this application. In particular, the technical features mentioned in the various embodiments can be combined in any manner, provided there is no structural conflict. This application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A signal processing method, characterized in that, include: The original vibration signal and the multi-component signal model corresponding to the original vibration signal are obtained. The multi-component signal model includes multiple frequency components, and each frequency component has a corresponding amplitude and frequency. Based on the discretized scaling basis function and the window function corresponding to the original vibration signal, the initial sub-time-frequency characterization result corresponding to the frequency component is determined, and the window function is used to traverse the frequency component in the time domain. Based on the instantaneous frequency of the initial sub-time-frequency characterization result corresponding to the frequency component, the initial sub-time-frequency characterization result is denoised to obtain the target sub-time-frequency characterization result corresponding to the frequency component. The group delay parameter corresponding to the frequency component is determined based on the target sub-time-frequency characterization result; Based on the group delay parameter, the time-frequency characterization result of the target sub-signal is denoised to obtain the time-frequency characterization result of the original vibration signal. Based on the time-frequency characterization results, the time-frequency crossover characteristics of the original vibration signal are determined.

2. The method according to claim 1, characterized in that, The step of denoising the initial sub-time-frequency characterization result based on the instantaneous frequency of the initial sub-time-frequency characterization result corresponding to the frequency component to obtain the target sub-time-frequency characterization result corresponding to the frequency component includes: Based on the instantaneous frequency and the initial sub-time frequency characterization results, the ideal frequency center of the frequency component is determined; Based on the ideal frequency center and the actual frequency center of the frequency component, and combined with the Dirac function, a first noise processing operator is generated; Based on the first noise processing operator, the initial sub-time-frequency characterization result is denoised to obtain the target sub-time-frequency characterization result.

3. The method according to claim 1 or 2, characterized in that, The group delay parameter includes the group delay and the rate of change of the group delay corresponding to the frequency component; Determining the group delay parameter corresponding to the frequency component based on the target sub-time-frequency characterization result includes: The group delay corresponding to the frequency component is obtained by solving the first-order partial derivative of the frequency center of the target sub-time-frequency characterization result. The second-order partial derivative of the frequency center is obtained by solving the time-frequency characterization result of the target sub-component to obtain the group delay rate of change corresponding to the frequency component.

4. The method according to claim 1 or 2, characterized in that, The step of denoising the target sub-time-frequency characterization result based on the group delay parameter to obtain the time-frequency characterization result of the original vibration signal includes: A second noise processing operator is generated based on the group delay parameter and the target sub-time-frequency characterization result; The target sub-time-frequency characterization result is denoised according to the second noise processing operator to obtain the processed target sub-time-frequency characterization result, which includes time dimension, frequency dimension and frequency modulation dimension. The processed target sub-time-frequency characterization result is integrated along the frequency modulation direction to obtain the time-frequency characterization result of the original vibration signal, wherein the frequency modulation direction is the direction corresponding to the frequency modulation dimension.

5. The method according to claim 4, characterized in that, The group delay parameter includes the group delay and group delay change rate corresponding to the frequency component, and the second noise processing operator includes a frequency redistribution operator and a frequency modulation redistribution operator; The step of generating a second noise processing operator based on the group delay parameter and the target sub-time-frequency characterization result includes: The frequency modulation redistribution operator is generated based on the group delay change rate and the target sub-time-frequency characterization result; The frequency redistribution operator is generated based on the frequency redistribution operator, the target sub-time-frequency characterization result, and the group delay.

6. The method according to claim 4, characterized in that, The second noise processing operator includes a frequency redistribution operator and a frequency modulation redistribution operator; The step of denoising the target sub-time-frequency characterization result according to the second noise processing operator to obtain the processed target sub-time-frequency characterization result includes: Based on the frequency redistribution operator and the frequency modulation redistribution operator, the time-frequency energy corresponding to the target sub-time-frequency characterization result is concentrated at the target position in the plane where the frequency and frequency modulation of the frequency component are located, to obtain the processed target sub-time-frequency characterization result.

7. The method according to claim 1 or 2, characterized in that, The window function includes the Gaussian window function; The step of determining the initial sub-time-frequency characterization result corresponding to the frequency component based on the discretized scaling basis function and the window function corresponding to the original vibration signal includes: The frequency components are traversed using the Gaussian window function to obtain the window function traversal result; The discretized scaling basis function is matched with the window function traversal result to obtain the matching result; Perform a Fourier transform on the matching result to obtain the initial sub-time-frequency characterization result corresponding to the frequency component.

8. The method according to claim 1 or 2, characterized in that, The method further includes: Based on the time-frequency crossover characteristics, fault diagnosis is performed on the equipment corresponding to the original vibration signal to obtain the fault diagnosis result.

9. A signal processing apparatus, characterized in that, include: The acquisition module is used to acquire the original vibration signal and the multi-component signal model corresponding to the original vibration signal. The multi-component signal model includes multiple frequency components, and each frequency component has a corresponding amplitude and frequency. The determination module is used to determine the initial sub-time-frequency characterization result corresponding to the frequency component based on the discretized scale basis function and the window function corresponding to the original vibration signal. The window function is used to traverse the frequency component in the time domain. The processing module is used to perform noise reduction processing on the initial sub-time-frequency characterization result based on the instantaneous frequency of the initial sub-time-frequency characterization result corresponding to the frequency component, so as to obtain the target sub-time-frequency characterization result corresponding to the frequency component. The determining module is further configured to determine the group delay parameter corresponding to the frequency component based on the target sub-time-frequency characterization result; The processing module is further configured to perform noise reduction processing on the target sub-time-frequency characterization result according to the group delay parameter, so as to obtain the time-frequency characterization result of the original vibration signal; The determining module is further configured to determine the time-frequency crossover characteristics of the original vibration signal based on the time-frequency characterization results.

10. An electronic device, characterized in that, The electronic device includes a processor and a memory, the memory storing programs or instructions that can run on the processor, the programs or instructions being executed by the processor to implement the steps of the method as described in any one of claims 1-8.

11. A readable storage medium, characterized in that, A program or instructions are stored on the readable storage medium, which, when executed by a processor, implement the steps of the method as described in any one of claims 1-8.

12. A computer program product, characterized in that, The program product is stored in a storage medium, and when executed by at least one processor, the program product implements the steps of the method as described in any one of claims 1-8.

Images