CN111238808B - Compound fault diagnosis method for gearbox based on empirical mode decomposition and improved variational mode decomposition - Google Patents

Compound fault diagnosis method for gearbox based on empirical mode decomposition and improved variational mode decomposition Download PDF

Info

Publication number
CN111238808B
CN111238808B CN202010079459.0A CN202010079459A CN111238808B CN 111238808 B CN111238808 B CN 111238808B CN 202010079459 A CN202010079459 A CN 202010079459A CN 111238808 B CN111238808 B CN 111238808B
Authority
CN
China
Prior art keywords
modal
decomposition
component
mode decomposition
components
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010079459.0A
Other languages
Chinese (zh)
Other versions
CN111238808A (en
Inventor
王靖岳
李建刚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenyang Ligong University
Original Assignee
Shenyang Ligong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenyang Ligong University filed Critical Shenyang Ligong University
Priority to CN202010079459.0A priority Critical patent/CN111238808B/en
Publication of CN111238808A publication Critical patent/CN111238808A/en
Application granted granted Critical
Publication of CN111238808B publication Critical patent/CN111238808B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • G01M13/02—Gearings; Transmission mechanisms
    • G01M13/021—Gearings
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • G01M13/02—Gearings; Transmission mechanisms
    • G01M13/028—Acoustic or vibration analysis

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)

Abstract

A gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition comprises the following steps: acquiring a composite fault vibration signal of a target gearbox; self-adaptive decomposition is carried out on the vibration signal of the gearbox by using an empirical mode decomposition method; aiming at the problem of false modal components which may occur during empirical mode decomposition, an energy method is utilized to eliminate false modes; calculating to obtain modal components with larger correlation with the original signals and forming combined modal components; the method is characterized in that the mode number k and the penalty factor alpha of the variational mode decomposition are improved by combining the envelope entropy and the waveform method, and the combined mode component is decomposed; finally, envelope demodulation analysis is carried out on the finite bandwidth modal component obtained by the variational modal decomposition, and the fault characteristics of the gear are extracted. The method is applied to the analysis of the vibration signal of the gear pitting-abrasion composite fault in the actual gear box, successfully separates and extracts the fault characteristics, and obtains good effect.

Description

Compound fault diagnosis method for gearbox based on empirical mode decomposition and improved variational mode decomposition
Technical Field
The invention belongs to the technical field of composite fault diagnosis of a gearbox, and particularly relates to a composite fault diagnosis method of a gearbox based on empirical mode decomposition and improved variational mode decomposition.
Background
The gearbox is one of the most important variable speed transmission mechanisms in the rotary mechanical equipment, and the health condition of the gearbox directly influences whether other equipment can normally operate or not. The existing fault detection and diagnosis technology of the gear box is mostly based on analysis of vibration signals of the gear box, when the gear box has a compound fault, the vibration signals of the gear box usually contain multi-component amplitude modulation-frequency modulation signals, and the signals have the characteristic of instability, and a common time domain or frequency domain analysis method only analyzes the signals in a single domain, so that information loss, missed diagnosis and misdiagnosis are easily caused in the fault diagnosis process.
For multi-component complex stationary signals, Wavelet Transform (WT), Empirical Mode Decomposition (EMD), Local Mean Decomposition (LMD), etc. are commonly used at present. Although wavelet transformation can decompose a complex signal into a plurality of modal components, the selection of wavelet basis functions has no good method at present and has no self-adaptability. The EMD method is widely applied to various signal processing since the introduction, can adaptively decompose a multi-component amplitude modulation-frequency modulation signal into the sum of a plurality of Intrinsic Mode Functions (IMFs), each IMF can be approximately regarded as a single-component amplitude modulation-frequency modulation signal, and achieves the purpose of signal processing through the analysis of the related IMFs. However, the EMD method often has problems of modal aliasing, end-point effect, over-envelope and under-envelope during use, which greatly limits its application. The LMD method is improved on the basis of the theory of EMD method, and can adaptively decompose a multi-component signal into a plurality of Product Functions (PF) which are arranged from high frequency to low frequency and have physical significance, each PF component is obtained by multiplying an envelope signal and a pure frequency modulation signal, and the instantaneous amplitude and the instantaneous frequency can be obtained by the envelope signal and the pure frequency modulation signal. Compared with the EMD method, the LMD reduces the endpoint effect to a certain extent, but is greatly influenced by noise, mode aliasing inevitably occurs, and the iteration time is longer.
For the defects of the existing signal processing methods, dragomirtsky et al propose a Variation Modal Decomposition (VMD) Method in 2014, compared with the EMD and LMD methods, the VMD adopts a frequency domain non-recursive iterative solution mode, the decomposition process of the signal is transferred into a variation frame, then an augmented Lagrange function is introduced to convert the constrained state into the unconstrained state, and an optimal solution of the signal decomposition is obtained through an Alternating Direction Multiplier Method (ADMM) algorithm, so that the sum of the estimated bandwidths of each finite bandwidth modal component (BLIMF) is minimum. The VMD method overcomes the defects of EMD and other methods, is more suitable for analyzing complex multi-component non-stationary signals, and is widely applied to the fields of instantaneous frequency feature extraction and various fault diagnosis at present. However, before analyzing a signal using VMD, it is necessary to determine various parameters of the signal, such as the number k of finite bandwidth modal components of decomposition, penalty factor α, precision e and step size μ, where different parameters may have a large difference in decomposition results for the same signal, k and α have the greatest influence on the decomposition results, and the precision e and the step size μ are usually fixed. If the values of k and α are not properly selected, modal aliasing occurs in the decomposition result. At present, random selection methods are mostly adopted for selecting k and alpha values, and reasonable parameter values are obtained by continuous random combination, but the method is time-consuming and is not beneficial to processing complex multi-component signals.
Disclosure of Invention
The invention aims to solve the problem that the optimal combination of the number k of BLIMFs decomposed by the method and the parameter alpha needs to be determined when the variable mode decomposition is used for decomposing the non-stationary signals.
In order to achieve the purpose, the invention discloses a gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition, which comprises the following steps:
step 1: and collecting vibration signals of gear pitting-abrasion combined faults in the gear box.
Step 2: and carrying out empirical mode decomposition on the acquired composite fault vibration signal.
And step 3: and judging whether a false mode appears in the mode components obtained by empirical mode decomposition, if so, eliminating the false component by using an energy method, and otherwise, carrying out the next step of processing.
And 4, step 4: and calculating a correlation coefficient between the modal component obtained by empirical mode decomposition and the original signal, and selecting the modal component with a larger correlation coefficient to form a combined modal component.
And 5: and improving the variation modal decomposition by adopting an envelope entropy and waveform method, and decomposing the combined modal component by utilizing the improved variation modal decomposition.
Step 6: and carrying out envelope demodulation spectrum analysis on each BLIMF obtained by carrying out improved variation modal decomposition on the combined modal component and extracting fault features.
The gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition is characterized in that the step 3 comprises the following steps:
3.1) calculating the Total energy value E of the Compound Fault Signal of the gearbox1The magnitude of the signal energy value is expressed by a mean square value, and then the sum E of the energy values of each modal component obtained by empirical mode decomposition is calculated2。
3.2) comparison of energy values E1And E2Size of (E), if E1<E2If not, continuing the next processing.
3.3) according to the judgment of the step 3.2), if the false modal component exists, eliminating the false modal component by using an energy method, and obtaining each updated modal component.
The gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition is characterized in that the step of eliminating false modes by an energy method in the step 3.3) comprises the following steps:
3.3.1) calculating the energy value of the first-order modal component obtained by empirical mode decomposition
Figure BDA0002379738510000021
Representing, adding the first-order modal component and other components one by one and calculating the energy value
Figure BDA0002379738510000022
(i>1 and not more than the total number of modal components).
3.3.2) one by one comparison
Figure BDA0002379738510000023
And
Figure BDA0002379738510000024
size of (1), if
Figure BDA0002379738510000025
Then the ith minuteThe quantity is the true modal component, otherwise it is considered to be a spurious modal component.
3.3.3) adding all false modal components and the first-order modal components to form updated first-order modal components, and sequentially retaining the rest real modal components.
The gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition is characterized in that the step 4 comprises the following steps:
calculating the correlation coefficient of a group of new modal components obtained after eliminating the false modal components and the original signal, selecting the components with larger correlation coefficients to form combined modal components, and selecting the correlation coefficients rhoxyThe formula of (1) is as follows:
Figure BDA0002379738510000031
in the formula, muxAnd muyMean, σ, of modal components x and y, respectivelyxAnd σyIs the variance of the modal component.
The gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition is characterized in that the threshold value of the correlation coefficient in the step 4 is set to be 0.3. The correlation coefficient is much greater than 0.3.
The gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition is characterized in that the step 5 comprises the following steps:
5.1) giving the BLIMF number k of the variation modal decomposition and the initial value of a penalty factor alpha, enabling k to be 1 and alpha to be 1000, and then performing the variation modal decomposition on the combined modal component.
And 5.2) keeping the value alpha of the penalty factor unchanged at 1000, keeping k at k +1, and determining the value range of k according to a waveform method.
And 5.3) selecting a penalty factor alpha with the value range of 1000-5000, making alpha equal to alpha +1000, and performing variational modal decomposition on the combined modal function under the combination of all k and alpha.
And 5.4) carrying out Hilbert envelope demodulation on BLIMF obtained by the variable mode decomposition.
5.5) calculating the entropy value of each envelope demodulation spectrum, and selecting the minimum envelope entropy value obtained under the combination of each group of k and alpha as a undetermined value.
5.6) comparing all the envelope entropy values obtained in the step 5.5) under different combinations of k and alpha, and selecting the combination of k and alpha corresponding to the minimum envelope entropy value as the optimal parameter combination of the variational modal decomposition.
5.7) carrying out variation modal decomposition on the composition modal component according to the combination of the optimal k and alpha selected in the step 5.6).
The gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition is characterized in that the step of determining the value range of the component number k by a waveform method in the step 5.2) comprises the following steps:
keeping the initial value of the penalty factor alpha unchanged, continuously increasing the k value to obtain the central frequency of each BLIMF obtained by variational modal decomposition under different k values, and when the k value stops when aliasing occurs to the central frequency, determining that the value range of k should not be larger than k-1.
The gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition is characterized in that the step 6 comprises the following steps:
and decomposing the combined modal function by using improved variational modal decomposition, performing Hilbert envelope demodulation spectrum analysis on the obtained BLIMF, extracting the prominent frequency component in the envelope demodulation spectrum, comparing the prominent frequency component with the theoretical fault characteristic frequency of the gear, and judging the fault position and type of the gear in the gear box.
The advantages are that:
for the original VMD method, the parameters k and α need to be selected when in use, while for different fault signals, the parameter values are different. Most technicians randomly select the values of k and alpha for combination when using the VMD to diagnose faults, and the relatively optimal parameter combination is selected by continuously trying, which is time-consuming in practical application. Relatively speaking, the invention proposes that a method of combining a waveform method and an envelope entropy is adopted to improve the VMD so as to determine that the optimal parameter combination can be obtained for different fault signals, and the optimal parameter combination can be selected only according to the steps of the improved method, so that the method is more time-saving compared with the original VMD method.
Drawings
FIG. 1 is a flow chart of a gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition.
FIG. 2 is a time domain diagram of vibration signals of a normal gearbox in an embodiment.
FIG. 3 is a time domain diagram of a vibration signal when a composite fault occurs in the gearbox in the embodiment.
FIG. 4 is an IMF component obtained after EMD of a vibration signal when a gearbox has a compound fault in the embodiment.
Fig. 5 shows the energy method to eliminate the spurious mode and obtain the new IMF component.
Fig. 6 is a composite modal component composed of new IMF components having a greater correlation with the composite fault signal in the embodiment.
Fig. 7 shows the center frequencies of the BLIMF obtained by performing the metamorphic mode decomposition on the combined mode components when the penalty factor α is 1000, k is 1, and k is k +1, where (a) is the center frequency when k is 2, (b) is the center frequency when k is 3, (c) is the center frequency when k is 4, and (d) is the center frequency when k is 5.
Fig. 8 is a variation trend of the minimum envelope entropy of each group BLIMF obtained by the variation modal decomposition of the combined modal component under each k and α combination.
Fig. 9 shows BLIMF obtained by performing metamorphic modal decomposition on the combined modal component under the optimal combination of k-4 and α -1000.
Fig. 10 shows a spectrum corresponding to each BLIMF obtained by performing a metamorphic modal decomposition on the combined modal component in the optimal combination of k 4 and α 1000.
Fig. 11 shows Hilbert envelope demodulation spectra of individual BLIMF values obtained by performing a variation modal decomposition on the combined modal components in the optimal combination of k 4 and α 1000, where (a) is an envelope demodulation spectrum of BLIMF1, (b) is an envelope demodulation spectrum of BLIMF2, (c) is an envelope demodulation spectrum of BLIMF3, and (d) is an envelope demodulation spectrum of BLIMF 4.
Detailed Description
The invention will be further described with reference to the following figures and examples, but the scope of the invention is not limited thereto.
The embodiment of the invention adopts the pitting-abrasion composite fault vibration signal of the gear in the gear box as the experimental data for analysis and processing. The selected gearbox transmission is single-stage transmission, wherein, the small gears (2, one of which is a spare part gear) are driving wheels (f)r114.15Hz) and the number of teeth is z155, modulus 2, material S45C, was used to simulate wear failure. The big gears (3, two of which are spare part gears) are driven wheels (f)r210.38Hz) and the number of teeth is z2The modulus is 2 and material S45C, 75, to simulate pitting failure. The known sampling frequency is 5120Hz, the number of sampling points is 7680, the current is introduced into the electric magnetic powder brake to be 0.1A, and the rotating speed of the driving shaft is 849r/min measured by the photoelectric sensor. The time domain diagram of the vibration signal when the gearbox is normal is shown in figure 2.
The flow chart of the gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition is shown in FIG. 1, and the specific implementation steps are as follows:
step 1: vibration data of pitting-wear composite failure of the gearbox is obtained through the acceleration sensor, and a time domain diagram of the composite failure is shown in FIG. 3.
Step 2: empirical mode decomposition is carried out on the obtained gearbox pitting-abrasion composite fault vibration signal, and the obtained first 12 modal components are shown in FIG. 4.
And step 3: judging whether the modal component obtained by empirical mode decomposition contains a false mode, and the method specifically comprises the following steps:
3.1) expressing the energy of the signal by a mean square value, and calculating the energy E of the composite fault signal1137.7362, total energy of each modal component
Figure BDA0002379738510000051
3.2) judging energy according to the calculation in the step 3.1)Is E1<E2It is shown that empirical mode decomposition does not obey energy conservation, and spurious modes exist in modal components.
3.3) eliminating the false mode by using an energy method and obtaining updated modal components as shown in the following table 1:
TABLE 1 elimination of spurious modal components
Figure BDA0002379738510000052
And 4, step 4: fig. 5 shows a group of new modal components obtained after eliminating the false modal component, and the correlation magnitudes of the new modal components and the composite fault signal calculated by the correlation coefficient method are respectively: 0.7740, 0.4296, 0.2468, 0.2028, 0.0398 and 0.0004.
The relation number threshold in step 4 is set to 0.3, so the first two layers of modal components having greater correlation with the composite fault signal are selected to constitute the combined modal component, as shown in fig. 6.
And 5: the method is characterized in that the variation modal decomposition is improved by adopting an envelope entropy and waveform method, and the combined modal component is decomposed by utilizing the improved variation modal decomposition, and the specific implementation steps are as follows:
5.1) giving the BLIMF number k of the variation modal decomposition and the initial value of a penalty factor alpha, enabling k to be 1 and alpha to be 1000, and then performing the variation modal decomposition on the combined modal component.
5.2) keeping the penalty factor value α at 1000, k at k +1, the center frequency of each BLIMF obtained by metamorphic modal decomposition is shown in fig. 7. When k is 5, as shown in (d) of fig. 7, the similar central frequencies 1692Hz and 1956Hz appear in the frequency range of 1500-2000, according to the waveform method, that is, when the central frequencies of different BLIMFs are similar, the variation modal decomposition is considered to have the modal aliasing phenomenon, the increase of the k value is stopped, and the value range of k is judged to be not more than 4.
The finite bandwidth modal component u obtained by the variational modal decomposition in the step 5.2)kAnd its corresponding center frequency wkThe updates in the frequency domain are:
Figure BDA0002379738510000061
Figure BDA0002379738510000062
and repeating the iteration updating until a stopping condition is met:
Figure BDA0002379738510000063
if the inequality is true, continuing the loop iteration, otherwise, proceeding the next step.
And 5.3) selecting a penalty factor alpha with the value range of 1000-5000, setting alpha as 1000, and setting alpha as alpha +1000, and performing variation mode decomposition on the combination mode function under the combination of all k and alpha.
5.4) carrying out Hilbert envelope demodulation on BLIMF obtained by the variational modal decomposition, wherein the specific steps of the envelope demodulation are as follows:
for the bandwidth-limited modal component u firstk(t) carrying out Hilbert transform, wherein the formula is as follows:
Figure BDA0002379738510000064
the envelope signal is then found to be:
Figure BDA0002379738510000065
determination of the envelope spectrum Q from the envelope signaliComprises the following steps: qi＝|FFT(Z(t))|。
5.5) calculating the entropy value of the envelope demodulation spectrum of each BLIMF, and selecting the minimum entropy value obtained under the combination of each group of k and alpha as an undetermined value, wherein the calculation formula of the entropy is as follows:
Figure BDA0002379738510000071
in the formula, XiIs an arbitrary value of the random variable X,h (X) denotes information entropy, P (X)i) Is a probability density function of the random variable X.
The envelope entropy R of the BLIMF in step 5.5) is defined as:
Figure BDA0002379738510000072
5.6) comparing all the envelope entropy values obtained in the step 5.5) under different combinations of k and alpha, wherein the variation trend of the minimum envelope entropy value under each group of combinations of k and alpha is shown in FIG. 8, and the combination of k and alpha corresponding to the minimum envelope entropy value is the optimal parameter combination of variation modal decomposition.
5.7) according to the variation trend of the envelope entropy values in the step 5.6), selecting the combination of the optimal k and α corresponding to the minimum envelope entropy value as k being 4 and α being 1000, and then performing variation modal decomposition on the combined modal components to obtain time domain graphs of 4 BLIMFs as shown in fig. 9 and corresponding frequency spectrum graphs as shown in fig. 10.
Step 6: the envelope demodulation analysis is performed on the 4 BLIMFs obtained in 5.7), and the corresponding envelope demodulation spectrum is obtained as shown in fig. 11. As can be seen from fig. 11 (a) and 11 (d), the characteristic frequency f appears in the envelope demodulation spectra of BLIMF1 and BLIMF4r2And frequency multiplication (2 f) thereofr2，3fr2，4fr2，5fr2，7fr2) And indicating that the large gear on the driven wheel has pitting failure. As can be seen from fig. 11 (b) and fig. 11 (c), the characteristic frequency f appears in the envelope demodulation spectra of BLIMF2 and BLIMF3r1And 3fr1And the abrasion fault of the pinion on the driving wheel is explained. The effectiveness of the gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition is proved through the diagnosis of the actual gearbox pitting-abrasion composite fault. One of the prior art is the original VMD method, which requires man-made random parameter k and α combination during use, finding out the optimal parameter combination through continuous tests, and then diagnosing faults, wherein the random parameter combination actually takes time to determine specific parameter values compared with the method of combining the waveform method and the envelope entropy, which is proposed by the present invention, so the present invention is relatively less time-consumingAs is the case. The other is the prior art, such as EMD, LMD methods, etc., which have a certain adaptability and do not need to select parameters, but compared with the method proposed by the present invention, the methods cannot effectively separate the complex fault characteristic frequencies with close frequencies in the gearbox, so as to achieve the purpose of fault diagnosis. When the composite fault diagnosis of the gearbox is performed by the methods of EMD, LMD and the like, modal aliasing phenomenon can occur to the decomposed modal components, and the misdiagnosis phenomenon can occur to the diagnosis result. Compared with methods such as EMD and the like, the results prove that the method provided by the invention is more accurate in the composite fault diagnosis of the gearbox, modal aliasing phenomenon does not occur in modal components, and different fault characteristic frequencies in the composite fault vibration signal of the gearbox are separated and extracted.

Claims (4)

1. A gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition is characterized by comprising the following steps:
step 1: collecting vibration signals of gear pitting-abrasion composite faults in a gear box;
step 2: carrying out empirical mode decomposition on the acquired composite fault vibration signal;
and step 3: judging whether a false mode appears in the mode components obtained by empirical mode decomposition, if so, eliminating the false component by using an energy method, and otherwise, carrying out the next step of processing;
and 4, step 4: calculating a correlation coefficient between a modal component obtained by empirical mode decomposition and an original signal, and selecting a modal component with a larger correlation coefficient to form a combined modal component;
and 5: the variation modal decomposition is improved by adopting an envelope entropy and waveform method, and the combined modal component is decomposed by utilizing the improved variation modal decomposition;
step 6: carrying out envelope demodulation spectrum analysis on each limited bandwidth modal component obtained by improving the variational modal decomposition and extracting fault characteristics;
the step 3 comprises the following steps:
3.1) calculating the gearbox complexTotal energy value E of fault signal1The magnitude of the signal energy value is expressed by a mean square value, and then the sum E of the energy values of each modal component obtained by empirical mode decomposition is calculated2；
3.2) comparison of energy values E1And E2Size of (E), if E1<E2If not, continuing the next step of processing, wherein the empirical mode decomposition process does not obey energy conservation, and a false modal component exists in the decomposition result;
3.3) according to the judgment of the step 3.2), if the false modal component exists, eliminating the false modal component by using an energy method, and obtaining each updated modal component;
the step of eliminating the false mode by using an energy method in the step 3.3) comprises the following steps:
3.3.1) calculating the energy value of the first-order modal component obtained by empirical mode decomposition
Figure FDA0003080025580000011
Representing, adding the first-order modal component and other components one by one and calculating the energy value
Figure FDA0003080025580000012
(i>1 and not more than the total number of modal components);
3.3.2) one by one comparison
Figure FDA0003080025580000013
And
Figure FDA0003080025580000014
size of (1), if
Figure FDA0003080025580000015
The ith component is a real modal component, otherwise, the ith component is regarded as a false modal component;
3.3.3) adding all the false modal components and the first-order modal components to form updated first-order modal components, and sequentially retaining the rest real modal components;
the step 5 comprises the following steps:
5.1) giving the number k of finite bandwidth modal components of the variation modal decomposition and an initial value of a penalty factor alpha, and then performing the variation modal decomposition on the combined modal component;
5.2) keeping the initial value of the penalty factor alpha unchanged, increasing the number k of the finite bandwidth modal components, and determining the value range of k according to a waveform method;
5.3) selecting the value range of the penalty factor alpha, and performing variational modal decomposition on the combined modal function under the combination of all k and alpha;
5.4) carrying out Hilbert envelope demodulation on the finite bandwidth modal components obtained by the variational modal decomposition;
5.5) calculating the entropy value of each envelope demodulation spectrum, and selecting the minimum envelope entropy value obtained under the combination of each group of k and alpha as an undetermined value;
5.6) comparing all the envelope entropy values obtained in the step 5.5) under different combinations of k and alpha, and selecting the combination of k and alpha corresponding to the minimum envelope entropy value as the optimal parameter combination of variational modal decomposition;
5.7) carrying out variation modal decomposition on the combined modal component according to the combination of the optimal k and alpha selected in the step 5.6);
the step of determining the value range of the number k of the components by the waveform method in the step 5.2) comprises the following steps:
keeping the initial value of the penalty factor alpha unchanged, continuously increasing the k value to obtain the central frequency of each limited bandwidth modal component of the variational modal decomposition under different k values, and determining that the value range of k is not more than k-1 when the k value stops when aliasing occurs to the central frequency.
2. The gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition according to claim 1, characterized by comprising the following steps: the step 4 comprises the following steps:
calculating the correlation coefficient of a group of new modal components obtained after eliminating the false modal components and the original signal, selecting the components with larger correlation coefficients to form combined modal components, and selecting the correlation coefficients rhoxyFormula (2)The following were used:
Figure FDA0003080025580000021
in the formula, muxAnd muyMean, σ, of modal components x and y, respectivelyxAnd σyIs the variance of the modal component.
3. The gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition according to claim 1, characterized by comprising the following steps: the threshold value of the correlation coefficient in step 4 is set to 0.3.
4. The gearbox composite fault diagnosis method based on empirical mode decomposition and improved variational mode decomposition according to claim 1, characterized by comprising the following steps: the step 6 comprises the following steps:
and decomposing the combined modal function by using improved variational modal decomposition, performing Hilbert envelope demodulation spectrum analysis on the obtained finite bandwidth modal component, extracting the prominent frequency component in the envelope demodulation spectrum, comparing the prominent frequency component with the theoretical fault characteristic frequency of the gear, and judging the fault position and type of the gear in the gear box.
CN202010079459.0A 2020-02-04 2020-02-04 Compound fault diagnosis method for gearbox based on empirical mode decomposition and improved variational mode decomposition Active CN111238808B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010079459.0A CN111238808B (en) 2020-02-04 2020-02-04 Compound fault diagnosis method for gearbox based on empirical mode decomposition and improved variational mode decomposition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010079459.0A CN111238808B (en) 2020-02-04 2020-02-04 Compound fault diagnosis method for gearbox based on empirical mode decomposition and improved variational mode decomposition

Publications (2)

Publication Number Publication Date
CN111238808A CN111238808A (en) 2020-06-05
CN111238808B true CN111238808B (en) 2021-08-17

Family

ID=70863119

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010079459.0A Active CN111238808B (en) 2020-02-04 2020-02-04 Compound fault diagnosis method for gearbox based on empirical mode decomposition and improved variational mode decomposition

Country Status (1)

Country Link
CN (1) CN111238808B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112326245B (en) * 2020-10-21 2023-03-10 中国航空工业集团公司上海航空测控技术研究所 Rolling bearing fault diagnosis method based on variational Hilbert-Huang transform
CN112326236B (en) * 2020-11-02 2023-04-18 北京信息科技大学 Gear box operation state online monitoring method and system and storage medium
CN112881006B (en) * 2021-01-12 2022-09-09 北华大学 Gear fault diagnosis method
CN112837254B (en) * 2021-02-25 2024-06-11 普联技术有限公司 Image fusion method and device, terminal equipment and storage medium
CN114112013B (en) * 2021-11-04 2023-06-30 北京建筑大学 Method and device for determining safety of ancient building, electronic equipment and storage medium
CN114563181B (en) * 2022-01-10 2023-06-27 浙江工业大学之江学院 Rotary machine fault feature extraction method based on improved variation mode extraction
CN117691631B (en) * 2024-02-04 2024-04-30 西安热工研究院有限公司 Electric power frequency modulation method and system based on hybrid energy storage device
CN117713144B (en) * 2024-02-06 2024-04-30 西安热工研究院有限公司 Thermal power generating unit frequency modulation method and system based on molten salt energy storage

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101901209A (en) * 2010-02-04 2010-12-01 福州大学 Structural response analysis method based on improved EMD (Empirical Mode Decomposition) and ARMA (Autoregressive Moving-Average) model
CN101964050A (en) * 2010-09-06 2011-02-02 中国海洋大学 Method for identifying modal parameter based on model order determination and signal noise elimination
JP2011095144A (en) * 2009-10-30 2011-05-12 Asahi Kasei Electronics Co Ltd Rotation angle sensor and rotation angle calculation method
CN104699948A (en) * 2015-01-15 2015-06-10 东风朝阳朝柴动力有限公司 EMD (empirical mode decomposition) based computation method for recognizing sensitivity of signal source
CN106017926A (en) * 2016-05-13 2016-10-12 山东理工大学 Rolling bearing fault diagnosis method based on variational mode decomposition
CN106596149A (en) * 2016-12-27 2017-04-26 广州地铁集团有限公司 Method for monitoring and diagnosing flat wheel state of rail vehicle based on VMD
CN106940688A (en) * 2017-01-08 2017-07-11 广东工业大学 A kind of Mathematical Modeling modeling method of EMD chaff components identification
CN108241873A (en) * 2018-01-10 2018-07-03 中国水利水电科学研究院 A kind of intelligent failure diagnosis method towards pumping plant main equipment
CN109238717A (en) * 2018-11-09 2019-01-18 福州大学 A kind of gear-box combined failure diagnostic method based on VMD-OMEDA
CN109827776A (en) * 2019-03-15 2019-05-31 合肥工业大学 Bearing Fault Detection Method and system
CN110146289A (en) * 2019-05-28 2019-08-20 昆明理工大学 A kind of rolling bearing Weak fault feature extracting method

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104483127B (en) * 2014-10-22 2017-12-29 徐州隆安光电科技有限公司 A kind of planetary gear feature information of weak faults extracting method
CN107451557B (en) * 2017-07-29 2020-06-23 吉林化工学院 Power transmission line short-circuit fault diagnosis method based on empirical wavelet transform and local energy

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011095144A (en) * 2009-10-30 2011-05-12 Asahi Kasei Electronics Co Ltd Rotation angle sensor and rotation angle calculation method
CN101901209A (en) * 2010-02-04 2010-12-01 福州大学 Structural response analysis method based on improved EMD (Empirical Mode Decomposition) and ARMA (Autoregressive Moving-Average) model
CN101964050A (en) * 2010-09-06 2011-02-02 中国海洋大学 Method for identifying modal parameter based on model order determination and signal noise elimination
CN104699948A (en) * 2015-01-15 2015-06-10 东风朝阳朝柴动力有限公司 EMD (empirical mode decomposition) based computation method for recognizing sensitivity of signal source
CN106017926A (en) * 2016-05-13 2016-10-12 山东理工大学 Rolling bearing fault diagnosis method based on variational mode decomposition
CN106596149A (en) * 2016-12-27 2017-04-26 广州地铁集团有限公司 Method for monitoring and diagnosing flat wheel state of rail vehicle based on VMD
CN106940688A (en) * 2017-01-08 2017-07-11 广东工业大学 A kind of Mathematical Modeling modeling method of EMD chaff components identification
CN108241873A (en) * 2018-01-10 2018-07-03 中国水利水电科学研究院 A kind of intelligent failure diagnosis method towards pumping plant main equipment
CN109238717A (en) * 2018-11-09 2019-01-18 福州大学 A kind of gear-box combined failure diagnostic method based on VMD-OMEDA
CN109827776A (en) * 2019-03-15 2019-05-31 合肥工业大学 Bearing Fault Detection Method and system
CN110146289A (en) * 2019-05-28 2019-08-20 昆明理工大学 A kind of rolling bearing Weak fault feature extracting method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Composite fault diagnosis of gearbox based on empirical mode decomposition and improved variation mode decomposition;Jingyue Wang;《Journal of low frequency noise，vibration and active control》;20200226;全文 *
齿轮箱复合故障的振动特性与诊断技术研究;李建刚;《万方文库》;20200927;全文 *

Also Published As

Publication number Publication date
CN111238808A (en) 2020-06-05

Similar Documents

Publication Publication Date Title
CN111238808B (en) Compound fault diagnosis method for gearbox based on empirical mode decomposition and improved variational mode decomposition
CN107944199B (en) Gear box fault identification method based on frequency spectrum trend and variational modal decomposition
CN107505135B (en) Rolling bearing composite fault extraction method and system
CN111507221B (en) Gear signal denoising method based on VMD and maximum overlapping discrete wavelet packet transformation
CN111414893B (en) Rotor fault feature extraction method based on VMD fine composite multi-scale diffusion entropy
CN111238813B (en) Method for extracting fault features of rolling bearing under strong interference
CN107525671B (en) Method for separating and identifying compound fault characteristics of transmission chain of wind turbine generator
CN112945546B (en) Precise diagnosis method for complex faults of gearbox
CN109635306A (en) Rotary machinery fault diagnosis method based on wavelet decomposition and spectrum kurtosis
CN112098093A (en) Bearing fault feature identification method and system
CN114707537A (en) Rotary machine fault feature extraction method based on self-adaptive VMD and optimized CYCBD
Chi et al. Bearing fault diagnosis through stochastic resonance by full-wave signal construction with half-cycle delay
CN115434872A (en) Wind turbine generator gearbox composite fault diagnosis method based on AVMD and improved RSSD
CN112485028A (en) Vibration signal characteristic frequency spectrum extraction method and mechanical fault diagnosis analysis method
CN114486263A (en) Noise reduction and demodulation method for vibration signal of rolling bearing of rotary machine
Xu et al. Rolling bearing fault feature extraction via improved SSD and a singular-value energy autocorrelation coefficient spectrum
CN111323233B (en) Local mean decomposition method for low-speed rotating machine fault diagnosis
CN107490477B (en) The Fault Diagnosis of Gear Case method compared based on frequency spectrum kernel density function correlation
CN117571316A (en) Composite fault diagnosis method and system
CN112881006A (en) Gear fault diagnosis method
CN106153333B (en) A kind of envelope Analysis Method based on wavelet decomposition filtering
CN117109923A (en) Rolling bearing fault diagnosis method and system
CN115711738A (en) Method for extracting fault characteristic frequency of main bearing of aircraft engine in service environment
CN110895687B (en) Bearing fault diagnosis method based on equipartition phase empirical mode decomposition and rapid spectral kurtosis
CN112183263B (en) Improved ICEEMD and HD-based early fault signal noise reduction method for check valve

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant