A Rolling Bearing Compound Fault Diagnosis Method for Aerospace Electromechanical Equipment

Abstract

The present invention discloses a method for diagnosing compound faults of rolling bearings in aerospace electromechanical equipment, comprising the following steps: Step 1, collect vibration data of the rolling bearings in the aerospace electromechanical equipment in a fault state to obtain a vibration signal f v (t); Step 2, process the vibration signal f v (t) by using multi-synchronous compression transform to obtain a time-frequency diagram #imgabs0# Step 3, perform envelope demodulation on the energy at each instantaneous frequency in the time-frequency diagram #imgabs1# to obtain an energy aggregation spectrum S η ; Step 4, calculate 4 relative factor indexes of the energy aggregation spectrum S η ; Step 5, screen out 4 optimal demodulation frequency bands from the 4 relative factor indexes; Step 6, reconstruct the 4 optimal demodulation frequency bands to obtain a time-domain signal f v '(t); Step 7, calculate the corresponding kurtosis value according to f v '(t), and screen out the optimal demodulation frequency bands corresponding to the kurtosis value greater than 3.5; Step 8, perform envelope demodulation analysis on the energy at each optimal demodulation frequency band to obtain a time-frequency energy aggregation spectrum, and then perform compound fault diagnosis on the rolling bearings in the aerospace electromechanical equipment according to the distribution of the spectral lines.

Description

Technical Field

[0001] The present invention relates to the field of fault diagnosis, and particularly to a method for diagnosing compound faults of rolling bearings in aerospace electromechanical equipment. Background Art

[0002] A rolling bearing generally consists of four parts: an inner ring, an outer ring, rolling elements, and a cage. It is a precision mechanical component used to support a rotating shaft and the parts on the shaft, and to maintain the normal working position and rotational accuracy of the shaft. It is widely used in the rotating mechanisms of various aerospace electromechanical equipment.

[0003] As the "joint" in the transmission system of the rotating mechanism of aerospace electromechanical equipment, the rolling bearing plays a crucial and important role in the reliability and safety of the operation of aerospace electromechanical equipment. Due to the complexity of the working environment of aerospace electromechanical equipment, most of the rolling bearings therein work under harsh conditions such as alternating loads, fatigue wear, and high-temperature impact for a long time, and are one of the components with a relatively high frequency of faults in aerospace equipment.

[0004] At present, many scholars have proposed technologies for fault diagnosis of rolling bearings. Among them, demodulation analysis methods such as wavelet analysis, spectral kurtosis, empirical mode decomposition, blind source separation, and variational mode decomposition have been widely used by researchers. The patents and literature related to the above methods are as follows: (1) Chinese Patent CN202010735503.9 discloses a method for diagnosing rolling bearing faults based on optimized variational mode decomposition. The invention optimizes the parameters of variational mode decomposition through an improved bat algorithm, and obtains feature vectors according to the optimized parameters to improve the recognition rate of fault states; (2) Chinese Patent CN202010936754.3 discloses a method for diagnosing rolling bearing faults based on empirical mode decomposition and kernel correlation. The invention comprehensively uses the empirical mode decomposition method, kernel correlation, and envelope analysis method to extract signals, so as to obtain more accurate fault diagnosis; (3) Yu Zhifeng et al. proposed a method for jointly extracting sensitive fault features by variational mode decomposition and continuous wavelet transform to obtain fault features with higher recognition accuracy.

[0005] However, due to factors such as the particularity of the operating environment of the rotating mechanism of aerospace electromechanical equipment, the complexity of the internal structure, and the variability of operating conditions, the rolling bearings therein mostly present multiple or concurrent compound faults, and their dynamic signals mostly exhibit characteristics such as variable rotational speed under obvious large fluctuations, confusion of multiple vibration modes, and weak feature modulation, which bring greater difficulties to the accurate diagnosis of rolling bearing faults in aerospace electromechanical equipment.

[0006] However, the above method can only better identify a single bearing fault with obvious symptoms in a compound fault, and it is difficult to take into account the synchronous demodulation, comprehensive extraction, and scientific and correct diagnosis of the fault characteristics of multiple bearings in aerospace electromechanical equipment. Therefore, the present invention combines the multi-synchronous compression transform and the time-frequency energy aggregation spectrum method to identify the fault characteristics of rolling bearings in aerospace electromechanical equipment. Patents and literatures similar to the basic theory of the present invention are as follows: (1) Chinese Patent CN201910687690.5 discloses a synchronous compression transform order ratio analysis method, which comprehensively uses synchronous compression transform, high-order polynomial fitting, and Hilbert envelope demodulation methods to analyze signals to determine whether there is a fault in the rolling bearing and its fault type; (2) Chinese Patent CN201611079765.4 discloses a variable-speed rolling bearing fault identification method based on order envelope time-frequency energy spectrum, which comprehensively uses short-time Fourier transform, fast kurtosis spectrum, Hilbert transform and other technologies to obtain the order envelope time-frequency energy spectrum to improve the fault monitoring efficiency of rolling bearings; (3) Yu Gang et al. proposed the application of time reassignment multi-synchronous transform in bearing fault diagnosis; (4) Han Tao et al. proposed an intelligent compound fault diagnosis method for rolling bearings based on multi-wavelet transform and convolutional neural network.

[0007] Although the above patents and literatures have optimized the traditional rolling bearing fault diagnosis methods, there are still deficiencies. For example, although the methods proposed in Chinese Patent CN201910687690.5, Chinese Patent CN201611079765.4 and Yu Gang et al. are similar to the basic theory in the present invention, their fault diagnosis and analysis mainly focus on single faults, and the effectiveness of identifying the compound fault characteristics of rolling bearings in aerospace electromechanical equipment is weak. At present, most of the literatures on rolling bearing compound fault diagnosis are based on intelligent classification algorithms, such as the method proposed by Han Tao et al. However, due to the limitation of computing efficiency, it is difficult to achieve the real-time performance of rolling bearing compound fault diagnosis in aerospace electromechanical equipment. Summary of the Invention

[0008] The present invention is made to solve the above problems, and the purpose is to provide a method for diagnosing compound faults of rolling bearings in aerospace electromechanical equipment.

[0009] The present invention provides a method for diagnosing compound faults of rolling bearings in aerospace electromechanical equipment, which has the following characteristics and includes the following steps: Step 1, collect the vibration data of the rolling bearing in the aerospace electromechanical equipment in the fault state to obtain the vibration signal f v (t); Step 2, process the vibration signal f v (t) by using the multi-synchronous compression transform to obtain a time-frequency diagram that can reflect the fault state characteristics of the rolling bearing in the aerospace electromechanical equipment Step 3, for the time-frequency diagram Envelope demodulation is performed on the energy at each instantaneous frequency to obtain the energy aggregation spectrum S η ; Step 4, calculate the four relative factor indexes of the energy aggregation spectrum S η ; Step 5, screen out four optimal demodulation frequency bands from the four relative factor indexes according to the principle of maximizing the indexes; Step 6, reconstruct the four optimal demodulation frequency bands according to the reconstruction formula to obtain the time-domain signal f v '(t); Step 7, calculate the corresponding kurtosis value according to the time-domain signal f v '(t), and screen out the optimal demodulation frequency band corresponding to the kurtosis value greater than 3.5, which is the optimal demodulation frequency band with faults in the vibration signal of the rolling bearing of the aerospace electromechanical equipment; Step 8, perform envelope demodulation analysis on the energy at each optimal demodulation frequency band to obtain the time-frequency energy aggregation spectrum, and then perform composite fault diagnosis on the rolling bearing in the aerospace electromechanical equipment according to the distribution of the spectral lines.

[0010] In the method for diagnosing composite faults of rolling bearings of aerospace electromechanical equipment provided by the present invention, it may also have the following characteristics: The specific steps of the multi-synchronous compression transform are as follows:

[0011] Step 2-1, perform short-time Fourier transform processing on the vibration signal f v (t) of the rolling bearing in the aerospace electromechanical equipment, and the expression of the short-time Fourier transform is:

[0012]

[0013] In the formula, t is the time variable of the vibration signal of the rolling bearing of the aerospace electromechanical equipment, η is the frequency variable of the vibration signal of the rolling bearing of the aerospace electromechanical equipment, and g(t) ∈ L 2 (R) is a real symmetric window function,

[0014] Let g η (τ) = g(τ - t)·e i2πτ , and obtain a new short-time Fourier transform expression:

[0015]

[0016] In the formula is the Fourier transform of f v (τ), is the Fourier transform of g(τ), () * is the complex conjugate symbol;

[0017] Step 2-2, introduce the phase shift operator e iηt in the newly obtained short-time Fourier transform, and let to obtain the improved Fourier transform, and its expression is as follows:

[0018]

[0019] Step 2-3: Calculate the instantaneous frequency estimation operator of time-frequency rearrangement by taking the partial derivative of the short-time Fourier transform with respect to time. After taking the partial derivative, the formula is as follows:

[0020]

[0021] In the formula, represents taking the partial derivative with respect to the time-shift variable. If the signal f v (t), the instantaneous frequency can be expressed as:

[0022]

[0023] Step 2-4: Use synchrosqueezing to concentrate the divergent energy at the estimated instantaneous frequency. The synchrosqueezing expression is:

[0024]

[0025] In the formula, η ∈ R,

[0026] Step 2-5: Continue to perform a compression operation on the time-frequency diagram obtained by the synchrosqueezing transform. The multiple synchrosqueezing can be expressed as:

[0027]

[0028]

[0029]

[0030] In the formula, K is the number of iterations, and K > 2.

[0031] In the composite fault diagnosis method for rolling bearings of aerospace electromechanical equipment provided by the present invention, it may also have the following characteristics: Among them, in step 2, the formula for the multiple synchrosqueezing transform is:

[0032]

[0033] In the composite fault diagnosis method for rolling bearings of aerospace electromechanical equipment provided by the present invention, it may also have the following characteristics: Among them, in step 4, according to the four relative factor expressions of the fault characteristic frequency of aerospace electromechanical equipment in the energy-aggregation spectrum, calculate the 4 relative factor indexes of the aggregation spectrum. The formulas for the 4 fault characteristic frequencies of the rolling bearings of aerospace electromechanical equipment are as follows:

[0034]

[0035]

[0036]

[0037]

[0038] wherein, f r is the rotational frequency of the rotating shaft where the rolling bearing of the aerospace electromechanical equipment is located, D p is the pitch diameter of the rolling bearing of the aerospace electromechanical equipment, d is the diameter of the rolling elements in the rolling bearing of the aerospace electromechanical equipment, θ is the contact angle, n is the number of rolling elements, f c , f o , f i , f el respectively represent the fault characteristic frequencies of the cage, outer ring, inner ring, and rolling elements of the rolling bearing of the aerospace electromechanical equipment. The four relative factor expressions of the fault characteristic frequencies of the rolling bearing of the aerospace electromechanical equipment in the energy aggregation spectrum are as follows:

[0039]

[0040]

[0041]

[0042]

[0043] wherein, Δ represents the neighborhood near the fault characteristic frequency of the rolling bearing of the aerospace electromechanical equipment, S η represents the time-frequency diagram in which the energy aggregation spectrum with the instantaneous frequency of η, S η (f k ) represents the spectral line in the energy aggregation spectrum, respectively represent the ratios of the fault characteristic frequencies of the cage, outer ring, inner ring, and rolling elements in the rolling bearing of the aerospace electromechanical equipment to the maximum periodic spectral peak in the energy aggregation spectrum.

[0044] In the composite fault diagnosis method for the rolling bearing of the aerospace electromechanical equipment provided by the present invention, it may further have the following characteristics: wherein, in step 6, the reconstruction formula is as follows:

[0045]

[0046] Functions and effects of the invention

[0047] According to a composite fault diagnosis method for the rolling bearing of the aerospace electromechanical equipment involved in the present invention, since the specific diagnosis process is as follows: Step 1, collect the vibration data of the rolling bearing in the aerospace electromechanical equipment in the fault state to obtain the vibration signal f v (t); Step 2, perform multiple synchrosqueezing transform on the vibration signal f v(t) to obtain a time-frequency diagram that can reflect the fault state characteristics of rolling bearings in aerospace electromechanical equipment Step 3: For the time-frequency diagram Perform envelope demodulation on the energy at each instantaneous frequency to obtain the energy aggregation spectrum S η ; Step 4: Calculate 4 relative factor indexes of the energy aggregation spectrum S η ; Step 5: Screen out 4 optimal demodulation frequency bands from the 4 relative factor indexes according to the index maximization principle; Step 6: Reconstruct the 4 optimal demodulation frequency bands according to the reconstruction formula to obtain the time-domain signal f v '(t); Step 7: Calculate the corresponding kurtosis value according to the time-domain signal f v '(t), and screen out the optimal demodulation frequency bands corresponding to the kurtosis value greater than 3.5, which are the optimal demodulation frequency bands with faults in the vibration signals of the rolling bearings of aerospace electromechanical equipment; Step 8: Perform envelope demodulation analysis on the energy at each optimal demodulation frequency band to obtain the time-frequency energy aggregation spectrum, and then perform the composite fault diagnosis of the rolling bearings in aerospace electromechanical equipment according to the distribution of the spectral lines.

[0048] Therefore, the composite fault diagnosis method for rolling bearings of aerospace electromechanical equipment proposed by the present invention can be used to solve some difficulties existing in the current fault diagnosis methods for rolling bearings of aerospace electromechanical equipment. For example: (1) There is more research on severe fault diagnosis and less on weak fault diagnosis; (2) There is more research on single fault diagnosis methods and less on composite faults, and among the studied composite faults, most are the composite faults of the outer ring and the inner ring, and less involve rolling elements; (3) In composite fault diagnosis methods, there are fewer effective feature extraction methods.

[0049] At the same time, the present invention is directed at a series of fault monitoring and diagnosis problems existing in the current rolling bearings in aerospace electromechanical equipment due to the particularity of the working environment, aiming to provide a method that can solve the accuracy problem of the optimal demodulation frequency band, and realize the comprehensive extraction, synchronization and accurate output of strong and weak multi-fault characteristic frequency bands, providing a favorable basis for the extraction and identification of weak and composite bearing fault characteristics in the complex dynamic signals of rotating mechanisms in aerospace electromechanical equipment. Description of the Drawings

[0050] Figure 1 is a schematic flow diagram of the composite fault diagnosis method for rolling bearings of aerospace electromechanical equipment in the embodiment of the present invention;

[0051] Figure 2 is the time-domain waveform and frequency spectrum diagram of the composite fault vibration signal of the rolling bearings of aerospace electromechanical equipment collected in the embodiment of the present invention;

[0052] Figure 3 is the result diagram of analyzing the composite fault vibration signal of the rolling bearings of aerospace electromechanical equipment collected in the embodiment by using the method in the present invention;

[0053] Figure 4 It is a result graph obtained by analyzing the vibration signal of the rolling bearing compound fault of aerospace electromechanical equipment collected in the embodiment by using the variational mode decomposition method;

[0054] Figure 5 It is a result graph obtained by analyzing the vibration signal of the rolling bearing compound fault of aerospace electromechanical equipment collected in the embodiment by using the fast spectral kurtosis method. Specific implementation manners

[0055] In order to make the technical means, creative features, achieved purposes and effects of the present invention easy to understand, the following embodiments will specifically describe a method for diagnosing the compound fault of the rolling bearing of aerospace electromechanical equipment in conjunction with the accompanying drawings.

[0056] In this embodiment, a method for diagnosing the compound fault of the rolling bearing of aerospace electromechanical equipment is provided.

[0057] Figure 1 It is a schematic flow chart of a method for diagnosing the compound fault of the rolling bearing of aerospace electromechanical equipment in this embodiment.

[0058] As Figure 1 shown, a method for diagnosing the compound fault of the rolling bearing of aerospace electromechanical equipment involved in this embodiment includes the following steps:

[0059] Step S1, perform vibration data acquisition on a certain aerospace electromechanical equipment: acquire the vibration data of the rolling bearing in the equipment when it is in a fault state, and obtain the vibration signal f v (t).

[0060] Figure 2 It is the time-domain waveform and frequency spectrum diagram of the vibration signal f v (t) of the rolling bearing compound fault of aerospace electromechanical equipment collected in this embodiment.

[0061] As Figure 2 shown, it can be seen that combining the time-frequency domain cannot provide a reliable basis for bearing fault diagnosis. Therefore, the time-frequency energy aggregation spectrum method is used to analyze the signal. The method for diagnosing the compound fault of the rolling bearing based on the time-frequency energy aggregation spectrum specifically starts from step S2.

[0062] Step S2, perform multiple synchrosqueezing transform on the vibration signal f v (t) of the rolling bearing in the aerospace electromechanical equipment to obtain a time-frequency diagram that can better reflect the fault state characteristics of the rolling bearing in the aerospace electromechanical equipment The formula for the multiple synchrosqueezing transform is as follows:

[0063]

[0064] The specific implementation steps of step S2 are as follows:

[0065] Step S2-1, perform short-time Fourier transform processing on the vibration signal f v (t) of the rolling bearing in the aerospace electromechanical equipment. The short-time Fourier transform has the following expression:

[0066]

[0067] In the formula, t is the time variable of the vibration signal of the rolling bearing of the aerospace electromechanical equipment, η is its frequency variable, and g(t) ∈ L 2 (R) is a real symmetric window function. Let g η (τ) = g(τ - t) · e i2πτ , and obtain a new short-time Fourier transform expression:

[0068]

[0069] Among them, is the Fourier transform of f v (τ), is the Fourier transform of g(τ), () * is the complex conjugate symbol.

[0070] Step S2-2, introduce the phase shift operator e iηt in the newly obtained short-time Fourier transform, and let to obtain the improved Fourier transform. Its expression is as follows:

[0071]

[0072] Step S2-3, use the short-time Fourier transform to take the partial derivative with respect to time to calculate the instantaneous frequency estimation operator of time-frequency rearrangement. The formula after taking the partial derivative is as follows:

[0073]

[0074] In the formula, represents taking the partial derivative with respect to the time shift variable. Among them, if the instantaneous frequency of the signal f v (t) can be expressed as:

[0075]

[0076] Step S2-4, use synchrosqueezing to concentrate the divergent energy at the estimated instantaneous frequency. The synchrosqueezing expression is:

[0077]

[0078] Among them, η ∈ R,

[0079] Step S2-5, continue to perform a compression operation on the time-frequency diagram obtained by the synchronous compression transform. The multi-synchronous compression can be expressed as:

[0080]

[0081]

[0082]

[0083] where K is the number of iterations and K > 2.

[0084] The multi-synchronous compression can also be expressed as:

[0085]

[0086] The multi-synchronous compression transform further improves the instantaneous frequency estimation accuracy by performing multi-compression on the result of the synchronous compression transform. Therefore, a time-frequency diagram with more concentrated time-frequency energy and more accurately reflecting the fault state characteristics of rolling bearings in aerospace electromechanical equipment can be obtained.

[0087] Step S3, perform envelope demodulation on the energy at each instantaneous frequency in the time-frequency diagram of the vibration signal f v (t) of the rolling bearing in the aerospace electromechanical equipment after multi-synchronous compression transform processing to obtain the energy concentration spectrum S η .

[0088] Envelope demodulation is a commonly used method in fault diagnosis. This method can effectively identify certain impact vibrations and thus find the source of the impact vibration. Therefore, envelope demodulation can effectively separate and extract the fault signals in the composite fault vibration signal f v (t) of the rolling bearing of a certain aerospace electromechanical equipment collected.

[0089] Step S4, calculate four relative factor indexes of the concentration spectrum according to the four relative factor expressions of the fault characteristic frequency of the aerospace electromechanical equipment in the energy concentration spectrum. Among them, the formulas for the four fault characteristic frequencies of the rolling bearing of the aerospace electromechanical equipment are as follows:

[0090]

[0091]

[0092]

[0093]

[0094] where f ris the rotational frequency of the rotating shaft where the rolling bearing of the aerospace electromechanical equipment is located, D p is the pitch diameter of the rolling bearing of the aerospace electromechanical equipment, d is the diameter of the rolling elements in the rolling bearing of the aerospace electromechanical equipment, θ is the contact angle, n is the number of rolling elements, f c 、f o 、f i 、f el respectively represent the fault characteristic frequencies of the cage, outer ring, inner ring, and rolling elements in the rolling bearing of the aerospace electromechanical equipment.

[0095] The expressions of the four relative factors of the fault characteristic frequencies of the rolling bearing of the aerospace electromechanical equipment in the energy aggregation spectrum are as follows:

[0096]

[0097]

[0098]

[0099]

[0100] Among them, Δ represents the neighborhood near the fault characteristic frequency of the rolling bearing of the aerospace electromechanical equipment, S η represents the time-frequency diagram in which the energy aggregation spectrum with the instantaneous frequency of η, S η (f k ) represents the spectral line in the energy aggregation spectrum, respectively represent the ratios of the fault characteristic frequencies of the cage, outer ring, inner ring, and rolling elements in the rolling bearing of the aerospace electromechanical equipment to the maximum period spectral peak in the energy aggregation spectrum;

[0101] Step S5, select 4 optimal demodulation frequency bands by the principle of maximizing the index.

[0102] Step S6, reconstruct the 4 optimal demodulation frequency bands according to the reconstruction formula to obtain the time-domain signal f v '(t). The reconstruction formula is as follows:

[0103]

[0104] Step S7, calculate the time-domain signal f v '(t), and select the optimal demodulation frequency band corresponding to the kurtosis value greater than 3.5, which is the optimal demodulation frequency band with faults in the vibration signal of the rolling bearing of the aerospace electromechanical equipment.

[0105] Step S8, perform envelope demodulation analysis on the energy at each optimal demodulation frequency band to obtain the time-frequency energy aggregation spectrum, and then conduct composite fault diagnosis on the rolling bearing in the aerospace electromechanical equipment according to the distribution of the spectral lines.

[0106] Since the energy is periodic in the resonance frequency band, this energy periodicity is the fault characteristic frequency of the rolling bearing compound fault of aerospace electromechanical equipment. Therefore, only by performing envelope demodulation on the energy in each resonance frequency band can the characteristic frequencies of each fault be obtained. And the energy in the optimal demodulation frequency band is the highest, and its periodicity is more obvious. Therefore, compared with other methods, directly performing envelope demodulation on the energy at the optimal demodulation frequency band is simpler and more effective.

[0107] Figure 3 It is the result diagram of analyzing the vibration signal of the rolling bearing compound fault of aerospace electromechanical equipment collected in this embodiment by using the method in the present invention.

[0108] As Figure 3 shown, Figure 3 (a) shows the result of fault one. From the place pointed by the arrow in the figure, the fault characteristic frequency and its multiple frequencies of the outer ring of the rolling bearing of aerospace electromechanical equipment can be clearly seen. And Figure 3 (b) shows the result of fault two, in which the highest amplitude is the fault characteristic frequency of the rolling element of the rolling bearing of aerospace electromechanical equipment. This result shows that the method in this embodiment can effectively extract the compound fault characteristics of the outer ring and rolling elements of the rolling bearing of aerospace electromechanical equipment at the same time.

[0109] Figure 4 It is the result diagram of analyzing the vibration signal of the rolling bearing compound fault of aerospace electromechanical equipment collected in this embodiment by using the variational mode decomposition method.

[0110] From Figure 4 it can be seen the four components C1 - C4 of the variational mode decomposition (left figure) and the corresponding envelope spectrum (right figure). From the envelope spectra of components C3 and C4, the obvious fault characteristic frequency of the outer ring of the rolling bearing of aerospace electromechanical equipment (marked with a circle) can be seen, but there is no weak fault characteristic frequency of the rolling element of the rolling bearing of aerospace electromechanical equipment in all four components.

[0111] Figure 5 It is the result diagram of analyzing the vibration signal of the rolling bearing compound fault of aerospace electromechanical equipment collected in this embodiment by using the fast spectral kurtosis method. Among them, Figure 5 (a) is the time domain diagram before filtering; (b) is the time domain diagram after filtering by the fast spectral kurtosis; (c) is the envelope spectrum of the signal filtered by the fast spectral kurtosis.

[0112] As Figure 5 shown, similar to using the variational mode decomposition method, the fast spectral kurtosis can only extract the fault characteristic frequency of the outer ring of the rolling bearing of aerospace electromechanical equipment (marked with a circle in Figure 5 (c)), and cannot extract the fault characteristic frequency of its rolling element.

[0113] Comparing Figure 3 , Figure 4 andFigure 5 , it can be obtained that the compound fault diagnosis method for rolling bearings of aerospace electromechanical equipment in this embodiment can extract the strong and weak compound fault characteristics of rolling bearings of aerospace electromechanical equipment more effectively than other methods, providing a reliable basis for the bearing fault diagnosis in aerospace electromechanical equipment.

[0114] Functions and effects of the embodiment

[0115] According to a compound fault diagnosis method for rolling bearings of aerospace electromechanical equipment involved in this embodiment, the specific diagnosis process is as follows: Step 1, collect the vibration data of the rolling bearings in the aerospace electromechanical equipment in the fault state to obtain the vibration signal f v (t); Step 2, use the multi-synchronous compression transform to process the vibration signal f v (t) to obtain a time-frequency diagram that can reflect the fault state characteristics of the rolling bearings in the aerospace electromechanical equipment Step 3, perform envelope demodulation on the energy at each instantaneous frequency in the time-frequency diagram to obtain the energy aggregation spectrum S η ; Step 4, calculate 4 relative factor indicators of the energy aggregation spectrum S η ; Step 5, screen out 4 optimal demodulation frequency bands through the index maximization principle for the 4 relative factor indicators; Step 6, reconstruct the 4 optimal demodulation frequency bands according to the reconstruction formula to obtain the time-domain signal f v '(t); Step 7, calculate the corresponding kurtosis value according to the time-domain signal f v '(t), and screen out the optimal demodulation frequency bands corresponding to the kurtosis value greater than 3.5, which are the optimal demodulation frequency bands with faults in the vibration signal of the rolling bearings of the aerospace electromechanical equipment; Step 8, perform envelope demodulation analysis on the energy at each optimal demodulation frequency band to obtain the time-frequency energy aggregation spectrum, and then perform the compound fault diagnosis of the rolling bearings in the aerospace electromechanical equipment according to the distribution of the spectral lines.

[0116] The compound fault diagnosis method for rolling bearings of aerospace electromechanical equipment proposed in this embodiment can be used to solve some difficulties existing in the current fault diagnosis methods for rolling bearings of aerospace electromechanical equipment. For example: (1) There is more research on severe fault diagnosis and less on weak fault diagnosis; (2) There is more research on single fault diagnosis methods and less on compound faults, and among the compound faults studied, most are the compound faults of the outer ring and the inner ring, and less involve rolling elements; (3) In the compound fault diagnosis methods, there are fewer effective feature extraction methods.

[0117] Meanwhile, this embodiment is directed at a series of fault monitoring and diagnosis problems existing in the rolling bearings of current aerospace electromechanical equipment due to the particularity of the working environment, aiming to provide a solution to the problem of the accuracy of the optimal demodulation frequency band, and realizing the comprehensive extraction, synchronization and accurate output of the strong and weak multi-fault characteristic frequency bands, providing a favorable basis for the extraction and identification of weak and composite bearing fault characteristics in the complex dynamic signals of the rotating mechanisms in aerospace electromechanical equipment.

[0118] The above embodiments are preferred cases of the present invention and are not used to limit the protection scope of the present invention.

Claims

1. A composite fault diagnosis method for rolling bearings of aerospace electromechanical equipment, characterized in that, Including the following steps: Step 1, collect the vibration data of the rolling bearing in the aerospace electromechanical equipment under the fault state to obtain the vibration signal f v (t); Step 2, perform a multi-synchronous compression transform on the vibration signal f v (t) to obtain a time-frequency diagram that can reflect the fault state characteristics of rolling bearings in aerospace electromechanical equipment Step 3, perform envelope demodulation on the energy at each instantaneous frequency in the time-frequency diagram to obtain an energy concentration spectrum S η ; Step 4, calculate the energy aggregation spectrum S η of four relative factor indexes; Step 5: Select 4 optimal demodulation frequency bands from the 4 relative factor indicators according to the principle of maximizing the index. Step 6, reconstruct the 4 optimal demodulation frequency bands according to the reconstruction formula to obtain the time-domain signal f v '(t); Step 7, calculate the corresponding kurtosis value according to the time-domain signal f v '(t), and screen out the optimal demodulation frequency band corresponding to the kurtosis value greater than 3.5, which is the optimal demodulation frequency band where there is a fault in the vibration signal of the rolling bearing of the aerospace electromechanical equipment; Step 8: Perform envelope demodulation analysis on the energy at each optimal demodulation frequency band to obtain the time-frequency energy aggregation spectrum, and then perform composite fault diagnosis on the rolling bearing in the aerospace electromechanical equipment according to the distribution of the spectral lines.

2. The method for diagnosing the composite fault of the rolling bearing of the aerospace electromechanical equipment according to claim 1, wherein: The specific steps of the multiple synchrosqueezing transform are as follows: Step 2-1, perform short-time Fourier transform processing on the vibration signal f v (t) of the rolling bearing in the aerospace electromechanical equipment collected, where the expression of the short-time Fourier transform is: where \(t\) is the time variable of the vibration signal of the rolling bearing of the aerospace electromechanical equipment, \(\eta\) is the frequency variable of the vibration signal of the rolling bearing of the aerospace electromechanical equipment, and \(g(t)\in L\) 2 (R) is a real symmetric window function Let \(g\) η (\(\tau\)) = \(g(\tau - t)\cdot e\) i2πτ , we obtain a new short-time Fourier transform expression: where is the Fourier transform of f v (τ), is the Fourier transform of g(τ), and () * is the complex conjugate symbol; Step 2-2, introduce the phase shift operator \(e\) in the newly obtained short-time Fourier transform iηt , and let to obtain the improved Fourier transform, and its expression is as follows: Step 2-3: Use the short-time Fourier transform to calculate the instantaneous frequency estimation operator of time-frequency rearrangement by taking the partial derivative with respect to time. The formula after taking the partial derivative is as follows: wherein, denotes the partial derivative with respect to the time-shifted variable, wherein if the signal f v (t) the instantaneous frequency can be expressed as: Step 2-4: Use synchrosqueezing to concentrate the divergent energy at the estimated instantaneous frequency. The synchrosqueezing expression is: where η ∈ R, Step 2-5: Continue to perform the compression operation on the time-frequency diagram obtained by the synchrosqueezing transform. The multiple synchrosqueezing can be expressed as: In the formula, K is the number of iterations, and K > 2.

3. The method for diagnosing the composite fault of the rolling bearing of the aerospace electromechanical equipment according to claim 1, wherein: Among them, In step 2, the formula of the multiple synchrosqueezing transform is:

4. The method for diagnosing the composite fault of the rolling bearing of the aerospace electromechanical equipment according to claim 1, wherein: Among them, In step 4, calculate the 4 relative factor indicators of the aggregation spectrum according to the four relative factor expressions of the fault characteristic frequency of the aerospace electromechanical equipment in the energy aggregation spectrum. The formulas of the 4 fault characteristic frequencies of the rolling bearing of the aerospace electromechanical equipment are as follows: where f r is the rotational frequency of the rotating shaft where the rolling bearing of the aerospace electro-mechanical equipment is located, D p is the pitch diameter of the rolling bearing of the aerospace electro-mechanical equipment, d is the diameter of the rolling elements in the rolling bearing of the aerospace electro-mechanical equipment, θ is the contact angle, n is the number of rolling elements, f c , f o , f i , f el represent the fault characteristic frequencies of the cage, outer ring, inner ring, and rolling elements of the rolling bearing of the aerospace electro-mechanical equipment, respectively. The four relative factor expressions of the fault characteristic frequency of the rolling bearing of the aerospace electromechanical equipment in the energy aggregation spectrum are as follows: where Δ represents the neighborhood near the fault characteristic frequency of the rolling bearing of aerospace electromechanical equipment, and S η represents the energy concentration spectrum with the instantaneous frequency of η in the time-frequency diagram, and S in the η (f k ) represents the spectral line in the energy concentration spectrum, which respectively represent the ratios of the fault characteristic frequencies of the cage, outer ring, inner ring, and rolling elements in the rolling bearing of aerospace electromechanical equipment to the peak of the maximum period spectrum in the energy concentration spectrum.

5. The method for diagnosing the composite fault of the rolling bearing of the aerospace electromechanical equipment according to claim 1, wherein: Among them, In step 6, the reconstruction formula is as follows:

Images