Planetary bearing fault feature enhancement and diagnosis method and device

Abstract

The invention relates to a planet bearing fault feature enhancement and diagnosis method and device. VMD (variational mode decomposition) and MNAD (minimum noise amplitude deconvolution) are combined; firstly, a vibration signal of a shell of the planetary reduction gearbox is collected, the output rotating speed of the planetary reduction gearbox is measured, and a fault characteristic frequency interval is introduced by calculating the theoretical fault characteristic frequency under the working condition; inputting a vibration signal into the VMD, improving the VMD by taking the maximum envelope spectrum amplitude in the interval as an optimization target so that the VMD can adaptively decompose an optimal component, and taking the frequency under the maximum envelope spectrum amplitude as the actual fault characteristic frequency; inputting the optimal component and the actual fault feature frequency into the MNAD to realize further fault feature enhancement; and finally, identifying a fault type through an envelope spectrum. According to the method, the fault features can be effectively extracted from complex planetary bearing fault signals in an enhanced mode, then fault diagnosis is achieved, meanwhile, the proposed method can process the signals in a self-adaptive mode, and intelligent diagnosis can be achieved in industry.

Description

Technical Field

[0001] The present invention relates to the technical field related to fault diagnosis, and in particular to a planetary bearing fault feature enhancement and diagnosis method and device. Background Art

[0002] Planetary reducers are widely used in the field of heavy industry due to their large transmission ratios. Planetary bearings, as key transmission parts, are in a condition of heavy load impact for a long time, and it is inevitable that they will fail. Timely diagnosis of faults is of great significance for safe production and avoiding economic losses. Although many scholars have proposed good methods in the field of fixed-axis bearing fault diagnosis, fault diagnosis of planetary bearings is still a difficult problem. This is because there are multi-stage planetary transmissions inside the planetary reducer, and the structure is complex. The impact signal of the planetary bearing fault will be modulated by the gear meshing signal, the shaft signal, etc., and the sensor installation position is limited. The impact energy is attenuated in the complex transmission path. In addition, the strong environmental noise in actual work makes the fault characteristics of the collected signal extremely weak. Therefore, how to effectively reduce the noise of complex signals and enhance the extraction of fault features is the key to solving the problem of planetary bearing fault diagnosis. Summary of the invention

[0003] The purpose of the present invention is to solve at least one of the deficiencies of the prior art and to provide a method and device for enhancing and diagnosing fault characteristics of a planetary bearing.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] Specifically, a planetary bearing fault feature enhancement and diagnosis method is proposed, including the following:

[0006] Collect vibration signals on the planetary reduction gearbox housing and measure the output angular velocity of the planetary reduction gearbox;

[0007] Using adaptive VMD to process the vibration signal to obtain an optimal component, achieve preliminary noise reduction and fault feature enhancement, and determine an actual fault feature frequency;

[0008] Input the optimal component and the actual fault characteristic frequency into the MNAD algorithm to further reduce noise and enhance the fault characteristic output processed signal;

[0009] The processed signal is subjected to Fourier transform spectrum and Hilbert envelope spectrum analysis to determine the fault status and output a diagnosis result.

[0010] Further, specifically, collecting the vibration signal on the planetary reduction box housing and measuring the output angular velocity of the planetary reduction box includes:

[0011] Use an acceleration sensor to measure the vibration signal on the shell, and select a position and direction with the largest signal period impact energy for collection. The signal period impact energy E is defined as follows:

[0012]

[0013] In the formula, X e is the envelope spectrum amplitude corresponding to the vibration signal;

[0014] The encoder is used to measure the output angular velocity of the planetary reducer. The output angular velocity calculation formula is:

[0015]

[0016] Where Δt is the time interval for recording encoder readings, count is the encoder reading, and resolution is the encoder resolution.

[0017] Further, specifically, using adaptive VMD to process the vibration signal to obtain the optimal component includes:

[0018] Calculate the theoretical fault characteristic frequency f of the inner and outer rings of the planetary bearing based on the output angular velocity bpfi and f bpfo , introduce the fault characteristic frequency interval [f bpfi -1,f bpfi +1] and [f bpfo -1,f bpfo +1], the key parameters [k, α] of VMD are optimized for the vibration signal, the optimization target is the maximum envelope spectrum amplitude in the two intervals of the fault characteristic frequency interval, the optimization range of the decomposition layer number K is [3, 8], and the search step is 1; the optimization range of the penalty factor α is [100, 4000], and the search step is 100; the component corresponding to the maximum envelope spectrum amplitude is searched as the optimal component, and the frequency corresponding to the maximum envelope spectrum amplitude is the actual fault characteristic frequency.

[0019] Further, specifically, the optimal component and the actual fault characteristic frequency are input into the MNAD algorithm to further reduce noise and enhance the fault characteristic output signal, including:

[0020] The optimal component obtained by searching is input into the MNAD algorithm for filtering, and then the parameter fault characteristic frequency in the MNAD is set as the actual fault characteristic frequency to update the actual fault characteristic frequency.

[0021] Further, specifically, the processed signal is subjected to Fourier transform spectrum and Hilbert envelope spectrum analysis to analyze the fault state, including:

[0022] Observe whether the fault characteristic frequency is obvious in the Fourier transform spectrum and the Hilbert envelope spectrum, and observe whether there is no obvious noise in the envelope spectrum except the fault characteristic frequency and its multiples. If so, it is identified as a planetary bearing fault, otherwise the planetary bearing is not faulty.

[0023] Furthermore, the method further comprises:

[0024] The fault characteristic frequency in the Hilbert envelope spectrum is compared with the fault characteristic frequency interval to determine the inner race or outer race fault.

[0025] The present invention also proposes a device for enhancing and diagnosing planetary bearing fault characteristics, comprising the following:

[0026] A data acquisition module is used to collect vibration signals on the planetary reduction gearbox housing and measure the output angular velocity of the planetary reduction gearbox;

[0027] A VMD algorithm module, used for processing the vibration signal using adaptive VMD to obtain an optimal component, achieve preliminary noise reduction and fault feature enhancement, and determine an actual fault feature frequency;

[0028] MNAD algorithm module, used for inputting the optimal component and the actual fault characteristic frequency into the MNAD algorithm to further reduce noise and enhance the fault characteristic output processed signal;

[0029] The diagnosis result analysis module is used to analyze the fault status of the processed signal through Fourier transform spectrum and Hilbert envelope spectrum, and output the diagnosis result.

[0030] The beneficial effects of the present invention are:

[0031] The present invention proposes a planetary bearing fault feature enhancement and diagnosis method and device, which has at least the following four advantages:

[0032] 1. The present invention only needs to collect the vibration signal of the reducer outer shell, and provides a method for determining the collection and measurement points of the vibration signal, which is universal and simple.

[0033] 2. The present invention improves the traditional VMD algorithm so that it can adapt to different signals and effectively decompose them, effectively realize preliminary noise reduction and feature extraction, and facilitate intelligent diagnosis;

[0034] 3. The present invention combines adaptive VMD with MNAD, uses adaptive VMD to determine the actual fault characteristic frequency, provides accurate fault characteristic frequency input for the MNAD algorithm, and can give full play to the ability of MNAD to enhance fault periodic impulse signals.

[0035] 4. The present invention extracts periodic impact signals from complex signals, and its excellent fault feature extraction capability can also be applied to fixed-axis bearing fault diagnosis and even to a wider field of vibration signal research. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The above and other features of the present disclosure will become more apparent by describing in detail the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings of the present disclosure represent the same or similar elements. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure. For those skilled in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work. In the accompanying drawings:

[0037] Figure 1 Shown is a schematic diagram of a planetary bearing fault feature enhancement and diagnosis method of the present invention;

[0038] Figure 2 Shown is a schematic diagram of a test bench for simulating faults of a planetary reducer used in one embodiment of the present invention;

[0039] Figure 3 Shown is a transmission diagram of a planetary reducer and a schematic diagram of the faulty bearing pattern and position in one embodiment of the present invention;

[0040] Figure 4 The figure shows a schematic diagram of the arrangement of sensors in one embodiment of the present invention;

[0041] FIG5(a), FIG5(b) and FIG5(c) are respectively a time domain waveform, a frequency spectrum and an envelope spectrum of a collected signal in one embodiment;

[0042] Figure 6 FIG. 1 is a schematic diagram of an optimal component envelope spectrum obtained under adaptive VMD in Example 1;

[0043] FIG. 7( a ) and FIG. 7( b ) are schematic diagrams of the final signal spectrum and envelope spectrum in Example 1 after being processed by the method of the present invention;

[0044] Figure 8 Shown is a flow chart of a planetary bearing fault feature enhancement and diagnosis method of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments in this application and the features in the embodiments can be combined with each other without conflict. The same reference numerals used throughout the drawings indicate the same or similar parts.

[0046] Considering the periodicity of the fault signal of rotating parts under steady working conditions, the variational mode decomposition (VMD) method has the advantage of anti-modal aliasing compared with other decomposition methods, and can better decompose the periodic signal in the original signal. The decomposition effect of the VMD algorithm depends on the selection of two key parameters: the number of decomposition layers K and the penalty factor α, and the optimal combination of decomposition parameters is different when processing different signals. The minimum noise amplitude deconvolution (MNAD) algorithm updates and solves an optimal finite impulse response (FIR) filter, which is used to offset the attenuation of the transmission path, the interference of noise, etc., and enhance the fault impulse signal from the collected signal. The effective implementation of the MNAD algorithm requires a relatively accurate fault characteristic frequency as input.

[0047] Based on the above theoretical basis, Example 1, referring to Figure 8 The present invention proposes a planetary bearing fault feature enhancement and diagnosis method, which integrates VMD and MNAD, gives full play to the advantages of the two algorithms and makes improvements, including the following:

[0048] Step 110: collecting vibration signals on the planetary reduction box housing and measuring the output angular velocity of the planetary reduction box;

[0049] Step 120: Use adaptive VMD to process the vibration signal to obtain an optimal component, achieve preliminary noise reduction and fault feature enhancement, and determine an actual fault feature frequency;

[0050] Step 130, inputting the optimal component and the actual fault characteristic frequency into the MNAD algorithm to further reduce noise and enhance the fault characteristic output processed signal;

[0051] Step 140: Analyze the fault status of the processed signal through Fourier transform spectrum and Hilbert envelope spectrum, and output a diagnosis result.

[0052] In this embodiment 1, variational mode decomposition (VMD) is combined with minimum noise amplitude deconvolution (MNAD); first, the vibration signal of the planetary reducer housing is collected and the output speed of the planetary reducer is measured, and the fault characteristic frequency interval is introduced by calculating the theoretical fault characteristic frequency under the working condition; then the vibration signal is input into the variational mode decomposition (VMD), and the VMD is improved with the maximum envelope spectrum amplitude in the interval as the optimization target, so that VMD can adaptively decompose the optimal component, and the frequency under the maximum envelope spectrum amplitude is used as the actual fault characteristic frequency; then the optimal component and the actual fault characteristic frequency are input into the minimum noise amplitude deconvolution (MNAD) to achieve further fault feature enhancement; finally, the fault type is identified by the envelope spectrum; the scheme proposed in this embodiment can effectively enhance the extraction of fault features from complex planetary bearing fault signals, and then realize fault diagnosis, and the proposed method can adaptively process signals, which is conducive to the realization of intelligent diagnosis in industry.

[0053] As a preferred embodiment of the present invention, specifically, collecting the vibration signal on the planetary reduction box housing and measuring the output angular velocity of the planetary reduction box include:

[0054] Use an acceleration sensor to measure the vibration signal on the shell, and select a position and direction with the largest signal period impact energy for collection. The signal period impact energy E is defined as follows:

[0055]

[0056] In the formula, X e is the envelope spectrum amplitude corresponding to the vibration signal;

[0057] The encoder is used to measure the output angular velocity of the planetary reducer. The output angular velocity calculation formula is:

[0058]

[0059] Where Δt is the time interval for recording encoder readings, count is the encoder reading, and resolution is the encoder resolution.

[0060] In this preferred embodiment, the vibration signal of the planetary reducer housing and the output speed of the planetary reducer are collected, three-directional vibration signals of multiple measuring points are pre-collected, Hilbert envelope demodulation is performed on all signals to obtain demodulated signals, and then Fourier transform is used to obtain envelope spectrum; the signal periodic impact energy values ​​E of multiple measuring points and directions are calculated according to the envelope spectrum; the vibration signals at the measuring points and directions with the largest E are selected for analysis; the output angular velocity of the planetary reducer is measured using an angle encoder, and the data of the angle encoder is continuously read at the same time interval, and the time interval is recommended to be 0.5 to 1s; the angular velocity is calculated based on the data read at the time interval, and the relevant data of the planetary reducer is collected in the above manner to obtain accurate data to be analyzed.

[0061] As a preferred embodiment of the present invention, specifically, using adaptive VMD to process the vibration signal to obtain the optimal component includes:

[0062] Calculate the theoretical fault characteristic frequency f of the inner and outer rings of the planetary bearing based on the output angular velocity bpfi and f bpfo , introduce the fault characteristic frequency interval [f bpfi -1,f bpfi +1] and [f bpfo -1,f bpfo+1], the key parameters [k, α] of VMD are optimized for the vibration signal, the optimization target is the maximum envelope spectrum amplitude in the two intervals of the fault characteristic frequency interval, the optimization range of the decomposition layer number K is [3, 8], and the search step is 1; the optimization range of the penalty factor α is [100, 4000], and the search step is 100; the component corresponding to the maximum envelope spectrum amplitude is searched as the optimal component, and the frequency corresponding to the maximum envelope spectrum amplitude is the actual fault characteristic frequency.

[0063] In this preferred embodiment, according to the rotation speed calculated in step 110, the theoretical fault characteristic frequency of the planetary bearing is calculated in combination with the structural parameters of the planetary reduction box transmission parts, wherein the structural parameters of the components to be measured include the number of teeth of the first and second sun gears, the number of teeth of the first and second planetary gears, the number of teeth of the first and second ring gears, the inner diameter, outer diameter, number of rollers, and roller diameter of the planetary bearing; then, the fault characteristic frequencies f of the inner and outer rings of the first and second planetary bearings are calculated respectively. bpfi and f bpfo ; Based on this, the fault characteristic frequency interval [f bpfi -1,f bpfi +1] and [f bpfo -1,f bpfo +1]; input the signal collected in step 110 into the VMD algorithm, set the initial decomposition parameters K=3, α=100, search for the maximum envelope spectrum amplitude in the fault characteristic frequency interval of all components, and update the optimal component; then update K with the interval [3, 8] and the step size of 1, and update α with the interval [100, 4000] and the step size of 100, continue to search for the maximum envelope spectrum amplitude in the fault characteristic frequency interval of all components, and update the optimal component; after the search is completed, determine the optimal component, and determine the actual fault characteristic frequency FCF.

[0064] As a preferred embodiment of the present invention, specifically, the optimal component and the actual fault characteristic frequency are input into the MNAD algorithm to further reduce noise and enhance the fault characteristic output signal, including:

[0065] The optimal component obtained by searching is input into the MNAD algorithm for filtering, and then the parameter fault characteristic frequency in the MNAD is set as the actual fault characteristic frequency to update the actual fault characteristic frequency.

[0066] In this preferred embodiment, the optimal component in step 120 is input into MNAD for deconvolution, and the input parameters are set: the characteristic frequency to be enhanced is set to the actual fault characteristic frequency determined in step 120, the filter length is recommended to be 2^13, and the number of iterative updates is recommended to be 500; the filtered signal obtained is the signal with noise reduction and fault feature enhancement.

[0067] As a preferred embodiment of the present invention, specifically, the processed signal is subjected to Fourier transform spectrum and Hilbert envelope spectrum analysis to analyze the fault state, including:

[0068] Observe whether the fault characteristic frequency is obvious in the Fourier transform spectrum and the Hilbert envelope spectrum, and observe whether there is no obvious noise in the envelope spectrum except the fault characteristic frequency and its multiples. If so, it is identified as a planetary bearing fault, otherwise the planetary bearing is not faulty.

[0069] In this preferred embodiment, observe whether the fault characteristic frequency is obvious in the Fourier transform spectrum and the Hilbert envelope spectrum, and observe whether there is no obvious noise in the envelope spectrum except the fault characteristic frequency and its multiples. If so, it is identified as a planetary bearing failure. Specifically, the spectral line corresponding to the fault characteristic frequency on the spectrum is longer, and should include the sum and difference frequencies of the main frequency of the equipment (the main frequency is determined by a pre-selected method) and the fault characteristic frequency, and the spectral line corresponding to the fault characteristic frequency on the envelope spectrum should be the longest, and the spectral line of its multiples gradually weakens but is still prominent; except for the spectral lines of the above frequencies, the proportion of other frequency components should be extremely low (corresponding to a preset threshold, and it is judged to be extremely low if it is lower than the proportion threshold). Otherwise, there is no fault. The specific observation method can be judged by pre-training the relevant network model of machine learning, or by obtaining observation feedback from the staff.

[0070] As a preferred embodiment of the present invention, the method further comprises:

[0071] The fault characteristic frequency in the Hilbert envelope spectrum is compared with the fault characteristic frequency interval to determine the inner race or outer race fault.

[0072] Specifically, the experimental analysis in this embodiment is:

[0073] 1. Take a company's winch planetary reducer as the implementation object and build a platform such as Figure 2 As shown in the figure, the platform device includes 1-hydraulic drive device, 2-industrial computer, 3-driving winch (with a two-stage planetary reducer inside), 3-loading winch (with a two-stage planetary reducer inside); the first-stage planetary bearing outer ring fault sample is used for analysis, and the outer ring of the bearing is processed by wire cutting to simulate damage. The internal transmission diagram of the reducer and the location of the faulty bearing are shown in the figure. Figure 3 According to the method of step 110, the vibration signal is collected and the output speed is measured, and the installation position of the acceleration sensor is determined by calculation. Figure 4 As shown, the Y direction signal is selected; the angle encoder measures the winch speed (i.e., the output speed of the planetary reducer); the collected signal time domain waveform, spectrum and envelope spectrum are shown in Figure 5; the calculated output speed is 5.1r / min.

[0074] 2. According to the method of step 120, the structural parameters of the internal transmission parts of the planetary reducer are obtained as shown in Table 1 below;

[0075] Table 1

[0076]

[0077] The fault characteristic frequencies of the inner and outer rings of the first-stage planetary bearing are calculated to be f bpfi =9.6Hz and f bpfo =6.3Hz; therefore, the fault characteristic frequency intervals introduced are [5.3,7.3] and [8.6,10.6]; the collected signal is input into the VMD algorithm, and the IMF 7 under [K,α]=[7,100] is searched through optimization. When the frequency is 7.21Hz, the local envelope spectrum has the maximum amplitude, and the amplitude is 2.1, such as Figure 6 As shown, it is determined as the optimal component and the fault characteristic frequency FCF is determined to be 7.21 Hz.

[0078] 3. According to the method of step 130, the optimal component of the second step is input into MNAD, the filter length is 2^13, the number of iterative updates is 500, and the frequency domain and envelope spectrum of the filtered signal are obtained as shown in Figure 7.

[0079] 4. According to the method of step 140, it is observed that in the spectrum diagram of Figure 7(a), the fault characteristic frequency FCF component is significantly improved, and the main components in the spectrum are the sum frequency and difference frequency of the fault characteristic frequency and the gear meshing frequency, and the noise component is relatively small; in the envelope spectrum of Figure 7(b), the fault impact is obvious, and the FCF and its multiples are clear; further, by comparing the FCF with the fault characteristic frequency range, the FCF belongs to the fault characteristic frequency range of the fault outer ring, and it is determined that the planetary bearing has an outer ring fault.

[0080] Through the analysis of the examples, the specific use process of the method of the present invention can be clearly demonstrated, and the effectiveness of the present invention in fault feature extraction and diagnosis can be clearly explained.

[0081] The present invention also proposes a device for enhancing and diagnosing planetary bearing fault characteristics, comprising the following:

[0082] A data acquisition module is used to collect vibration signals on the planetary reduction gearbox housing and measure the output angular velocity of the planetary reduction gearbox;

[0083] A VMD algorithm module, used for processing the vibration signal using adaptive VMD to obtain an optimal component, achieve preliminary noise reduction and fault feature enhancement, and determine an actual fault feature frequency;

[0084] MNAD algorithm module, used for inputting the optimal component and the actual fault characteristic frequency into the MNAD algorithm to further reduce noise and enhance the fault characteristic output processed signal;

[0085] The diagnosis result analysis module is used to analyze the fault status of the processed signal through Fourier transform spectrum and Hilbert envelope spectrum, and output the diagnosis result.

[0086] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of software functional modules.

[0087] If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or system that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0088] Although the description of the present invention has been quite detailed and specifically describes several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, but should be regarded as providing a broad possible interpretation of these claims in view of the prior art by reference to the appended claims, thereby effectively covering the intended scope of the present invention. In addition, the above description of the present invention is based on the embodiments foreseeable by the inventor, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that have not yet been foreseen may still represent equivalent changes to the present invention.

[0089] The above is only a preferred embodiment of the present invention. The present invention is not limited to the above implementation. As long as the technical effect of the present invention is achieved by the same means, it should belong to the protection scope of the present invention. Within the protection scope of the present invention, its technical scheme and / or implementation method can have various modifications and changes.

Claims

1. A planetary bearing fault feature enhancement and diagnosis method, characterized in that: Includes the following: Collect vibration signals on the planetary reduction gearbox housing and measure the output angular velocity of the planetary reduction gearbox; Using adaptive VMD to process the vibration signal to obtain an optimal component, achieve preliminary noise reduction and fault feature enhancement, and determine an actual fault feature frequency; Input the optimal component and the actual fault characteristic frequency into the MNAD algorithm to further reduce noise and enhance the fault characteristic output processed signal; The processed signal is subjected to Fourier transform spectrum and Hilbert envelope spectrum analysis to determine the fault status and output a diagnosis result.

2. A planetary bearing fault feature enhancement and diagnosis method according to claim 1, characterized in that: Specifically, the vibration signal on the planetary reduction box housing is collected, and the output angular velocity of the planetary reduction box is measured, including: Use an acceleration sensor to measure the vibration signal on the shell, and select a position and direction with the largest signal period impact energy for collection. The signal period impact energy E is defined as follows: In the formula, X e is the envelope spectrum amplitude corresponding to the vibration signal; The encoder is used to measure the output angular velocity of the planetary reducer. The output angular velocity calculation formula is: Where Δt is the time interval for recording encoder readings, count is the encoder reading, and resolution is the encoder resolution.

3. A planetary bearing fault feature enhancement and diagnosis method according to claim 2, characterized in that: Specifically, the vibration signal is processed using adaptive VMD to obtain the optimal component, including: Calculate the theoretical fault characteristic frequency f of the inner and outer rings of the planetary bearing based on the output angular velocity bpfi and f bpfo , introduce the fault characteristic frequency interval [f bpfi -1,f bpfi +1] and [f bpfo -1,f bpfo +1], the key parameters [k, α] of VMD are optimized for the vibration signal, the optimization target is the maximum envelope spectrum amplitude in the two intervals of the fault characteristic frequency interval, the optimization range of the decomposition layer number K is [3, 8], and the search step is 1; the optimization range of the penalty factor α is [100, 4000], and the search step is 100; the component corresponding to the maximum envelope spectrum amplitude is searched as the optimal component, and the frequency corresponding to the maximum envelope spectrum amplitude is the actual fault characteristic frequency.

4. A planetary bearing fault feature enhancement and diagnosis method according to claim 3, characterized in that: Specifically, the optimal component and the actual fault characteristic frequency are input into the MNAD algorithm to further reduce noise and enhance the fault characteristic output signal, including: The optimal component obtained by searching is input into the MNAD algorithm for filtering, and then the parameter fault characteristic frequency in the MNAD is set as the actual fault characteristic frequency to update the actual fault characteristic frequency.

5. A planetary bearing fault feature enhancement and diagnosis method according to claim 1, characterized in that: Specifically, the processed signal is subjected to Fourier transform spectrum and Hilbert envelope spectrum analysis to analyze the fault state, including: Observe whether the fault characteristic frequency is obvious in the Fourier transform spectrum and the Hilbert envelope spectrum, and observe whether there is no obvious noise in the envelope spectrum except the fault characteristic frequency and its multiples. If so, it is identified as a planetary bearing fault, otherwise the planetary bearing is not faulty.

6. A planetary bearing fault feature enhancement and diagnosis method according to claim 5, characterized in that: The method further comprises, The fault characteristic frequency in the Hilbert envelope spectrum is compared with the fault characteristic frequency interval to determine the inner race or outer race fault.

7. A device for enhancing and diagnosing fault characteristics of planetary bearings, characterized in that: Includes the following: A data acquisition module is used to collect vibration signals on the planetary reduction gearbox housing and measure the output angular velocity of the planetary reduction gearbox; A VMD algorithm module, used for processing the vibration signal using adaptive VMD to obtain an optimal component, achieve preliminary noise reduction and fault feature enhancement, and determine an actual fault feature frequency; MNAD algorithm module, used for inputting the optimal component and the actual fault characteristic frequency into the MNAD algorithm to further reduce noise and enhance the fault characteristic output processed signal; The diagnosis result analysis module is used to analyze the fault status of the processed signal through Fourier transform spectrum and Hilbert envelope spectrum, and output the diagnosis result.

Images