CN108801630A - The gear failure diagnosing method of single channel blind source separating - Google Patents
The gear failure diagnosing method of single channel blind source separating Download PDFInfo
- Publication number
- CN108801630A CN108801630A CN201810653028.3A CN201810653028A CN108801630A CN 108801630 A CN108801630 A CN 108801630A CN 201810653028 A CN201810653028 A CN 201810653028A CN 108801630 A CN108801630 A CN 108801630A
- Authority
- CN
- China
- Prior art keywords
- signal
- gear
- imf
- single channel
- noise reduction
- 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.)
- Granted
Links
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
The invention discloses a kind of gear failure diagnosing methods of single channel blind source separating, are related to gearbox medium gear method for diagnosing faults technical field.The method acquires the vibration signal of gear-box using single accelerator sensor；Wavelet soft-threshold noise reduction is carried out to the single channel signal of acquisition；Signal after noise reduction is subjected to CEEMD decomposition, obtains multiple IMF components and residual components；Suitable IMF components are chosen using the method being combined based on kurtosis and continuous mean-square error criteria；Using the IMF components of selection and source signal as the input signal of blind source separating, echo signal is extracted using CICA methods；Spectrum analysis is carried out to the signal of extraction, identifies the fault signature of gear.The method principle is simple, easy to implement, accurately can carry out Gear Fault Diagnosis using single channel measuring signal under very noisy.
Description
Technical field
The present invention relates to Gear Box Fault Diagnosis Technology field more particularly to a kind of gear distresses of single channel blind source separating
Diagnostic method.
Background technology
Gear is widely used in all kinds of rotating machineries, it has also become the most key one of the component of equipment, thus
Fault diagnosis important in inhibiting is carried out to it.The main method of current diagnosis gear distress is analysis of vibration signal, but due to
Fault-signal is often submerged in strong noise background, and how successfully to extract fault characteristic information becomes a step of most critical.
The vibration signal of gear is non-stationary signal, is often superimposed with very noisy, the classical filtering based on Fourier transform
It is no longer valid that method removes noise.Wavelet transformation has the characteristics that good Time-Frequency Localization and multiresolution analysis, because
And suitable for the processing of non-stationary signal and very noisy signal, have good noise reduction capability.
EMD algorithms are intuitively simple, can be a series of intrinsic mode functions by signal decomposition, these mode functions can be right
Input signal carries out the description of different scale, and has the characteristics that orthogonality, completeness and adaptivity, handles non-stationary
Signal has prodigious advantage.Although EMD has the advantages that very much, its decomposition is unstable, and there are modal overlap phenomenons, lead to certain
One IMF component includes that the signal of different scale or similar magnitude signal are present in different IMF components, this makes EMD
Decomposition has limitation very much.EEMD is the improved method of EMD, original signal is added certain white noise so that signal is not
It needs that different auxiliary white noises is repeatedly added with having continuity, this method on scale, is then disappeared by average mode
Except the influence for introducing noise, finally makes decomposable process that there is noiseproof feature, but still modal overlap phenomenon can not be completely eliminated.
CEEMD is a kind of innovatory algorithm proposed based on EMD and EEMD, aid in noise is added using positive and negative pairs of form, thus
The remaining aid in noise in reconstruction signal can be eliminated well, and the noise set number being added can be very low, calculate effect
Rate is higher, the phenomenon that further weakening modal overlap.
ICA is a kind of separation method for isolating the single source signal with independent statistics characteristic from mixed signal.By
It is seldom in the prior information for the Independent sources signal that it is needed in separation process, and have apparent separating effect, ICA is in nothing
The fields such as line communication, speech processes and mechanical fault diagnosis have a wide range of applications.In practical applications, because of source signal order
Uncertain and number is not easy the problems such as determining, limits its application in Gear Fault Feature Extraction.In recent years, exist
The CICA to grow up on the basis of ICA is improved its algorithm, and the above problem is efficiently solved, it is requiring no knowledge about source
In the case of number of signals, reference signal is generated first with prior information, and then extract interested isolated component.
Invention content
The technical problem to be solved by the present invention is to how provide one kind to utilize single pass gear-box vibration signal,
Carry out the effective method for diagnosing faults for extracting gearbox medium gear fault-signal.
In order to solve the above technical problems, the technical solution used in the present invention is：A kind of single channel blind source separating
Gear failure diagnosing method, it is characterised in that include the following steps：
Wavelet soft-threshold noise reduction is carried out to collected gear-box single channel vibration signal；
Signal after noise reduction is subjected to CEEMD decomposition, obtains several IMF components and residual components；
Suitable IMF components are chosen using the method being combined based on kurtosis and continuous mean-square error criteria；
Using the IMF components of selection and source signal as the input signal of blind source component, target is extracted using CICA methods
Signal；
Spectrum analysis is carried out to the echo signal extracted, fault signature is identified, completes the diagnosis of gear distress.
Further technical solution is that the method further includes：With the acquisition gear-box vibration of single acceleration transducer
Signal.
Further technical solution is that the method for the progress wavelet soft-threshold noise reduction includes the following steps：
The selection of wavelet basis function and Decomposition order：Wavelet basis function is selected according to the characteristics of signal, wavelet basis function
Regularity and waveform and the structure similarity degree of data can influence the effect of signal de-noising, symN wavelet basis and mechanical oscillation waveform
It is similar, therefore symN wavelet basis functions are selected to carry out wavelet soft-threshold noise reduction；Different decomposition number of plies noise reduction is different, rationally
Select Decomposition order；
Select threshold value and threshold value function：By comparing the size of wavelet coefficient and threshold value under each decomposition scale, warp
Cross the coefficient value after processing is purified；
Wavelet reconstruction：The high frequency coefficient that wavelet low frequency coefficient and each layer decompose is reconstructed, the signal after noise reduction is obtained.
Further technical solution is that the signal by after noise reduction carries out CEEMD decomposition and includes the following steps：
N groups are added into the pretreated signal of noise reduction and assist white noise, aid in noise is added in a manner of positive and negative pair
, therefore generate the IMF components of two sets of set：
Wherein：S is the pretreated signal of noise reduction；N is the white noise for meeting normal distribution；M1、M2Respectively be added just,
Bear the signal of pairs of noise；
EMD decomposition is done respectively to each signal in set, each signal obtains one group of IMF component, wherein i-th of letter
Number j-th of IMF representation in components be cij；
Decomposition result is obtained by way of multigroup component combination：
Wherein：cjIndicate that CEEMD decomposes finally obtained j-th of IMF components, n indicates the group number of auxiliary white noise.
Further technical solution is that the method being combined based on kurtosis and continuous mean-square error criteria is such as
Under：
The kurtosis for calculating each IMF, when gear-box runs well, the vibration signal amplitude distribution of acquisition can be similar to normal state
Distribution, therefore its kurtosis value is equal to 3, when gear is there are when failure, kurtosis value can become larger；
The criterion of continuous mean square error (CMSE), i.e.,:
Wherein, H is the total length of signal, and r is the IMF numbers decomposed；
Critical IMF components are determined according to following two principle：
If there are local minimums before global minimum by CMSE, take corresponding to first local minimum
Position adds 1；
If there is no local minimum, then the position corresponding to global minimum is taken to add 1, including being mostly fault message
Critical IMF components and its IMF components later, the effective IMF components of selection will be combined with kurtosis.
Further technical solution is that described is included the following steps using CICA methods extraction echo signal：
Fault-signal characteristic frequency construction reference burst signal r (t) based on gear, by echo signal y to be extracted and
The distance function of reference signal r (t) is defined as ε (y, r), for indicating the degree of closeness of echo signal and reference signal；ε(y,
R) mean square error ε (y, r)=E { (y-r) } is used to measure, shown in the mathematical model such as formula (4) and formula (5) of CICA algorithms：
Object function：
max J(y)≈ρ{E[G(y)]-E[G(v)]} (4)
Constraints：
Wherein：ρ is normal number；G () is nonlinear function；V is the gaussian variable with covariance matrix identical as y；ξ
For threshold value；Formula (5) solves it by lagrange's method of multipliers, and the best estimate of source signal can be obtained, and extracts required letter
Number.
It is using advantageous effect caused by above-mentioned technical proposal：The method is carried out using wavelet soft-threshold at noise reduction
Reason, can effectively improve signal-to-noise ratio so that blind source separating has good effect.It decomposes to obtain multichannel by CEEMD and virtually lead to
Road is chosen suitable IMF components using the method being combined based on kurtosis and continuous mean-square error criteria, itself and source signal is made
For the input signal of blind source separating, intended vibratory signal is extracted by CICA methods, identifies that fault signature, the method are a kind of
Effective gear failure diagnosing method.
Description of the drawings
The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
Fig. 1 is the flow chart of the method for the embodiment of the present invention；
Fig. 2 is the structure chart of DDS experimental benches in the embodiment of the present invention；
Fig. 3 is middle gear case kinematic scheme of the embodiment of the present invention；
Fig. 4 is failure gear position figure in the embodiment of the present invention；
Fig. 5 is the gear graph of local broken teeth in the embodiment of the present invention；
Fig. 6 a are acquired original signal time-domain diagrams；
Fig. 6 b are the time-domain diagrams after acquired original signal wavelet de-noising；
Fig. 7 a are the Signal Amplitudes after wavelet de-noising；
Fig. 7 b are the signal envelope spectrums after wavelet de-noising；
Fig. 8 is the kurtosis value of each IMF；
Fig. 9 is the CMSE of each IMF；
Figure 10 is the suitable IMF components selected；
Figure 11 is reference signal and extraction fault-signal time domain waveform；
Figure 12 is the amplitude spectrum for extracting fault-signal；
Figure 13 is the envelope spectrum for extracting fault-signal.
Specific implementation mode
With reference to the attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground describes, it is clear that described embodiment is only a part of the embodiment of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Many details are elaborated in the following description to facilitate a thorough understanding of the present invention, still the present invention can be with
Implemented different from other manner described here using other, those skilled in the art can be without prejudice to intension of the present invention
In the case of do similar popularization, therefore the present invention is not limited by following public specific embodiment.
As shown in Figure 1, the embodiment of the invention discloses a kind of gear failure diagnosing methods of single channel blind source separating, specifically
Embodiment include the following steps：
Step 1, gear-box vibration signal is acquired with single acceleration transducer, small echo is carried out to collected vibration signal
Soft threshold de-noising；
Step 2, the signal after noise reduction is subjected to CEEMD decomposition, obtains multiple IMF components and residual components；
Step 3, suitable IMF components are chosen using the method being combined based on kurtosis and continuous mean-square error criteria；
Step 4, it using the IMF components of selection and source signal as the input signal of blind source component, is extracted using CICA methods
Go out echo signal；
Step 5, spectrum analysis is carried out to the echo signal extracted, identifies fault signature, completes the diagnosis of gear distress.
In the step 1, wavelet soft-threshold noise-reduction method includes the following steps：
1-1) the selection of wavelet basis function and Decomposition order：Wavelet basis function, wavelet basis letter are selected according to the characteristics of signal
The structure similarity degree of several regularities and waveform and data can influence the effect of signal de-noising, symN wavelet basis and mechanical oscillation
Waveform is similar, therefore selects symN wavelet basis functions to carry out wavelet soft-threshold noise reduction；Different decomposition number of plies noise reduction is different,
Reasonably select Decomposition order.
1-2) selection threshold value and threshold value function：By comparing the big of wavelet coefficient under each decomposition scale and threshold value
It is small, the coefficient value after processing is purified.
1-3) wavelet reconstruction：The high frequency coefficient that wavelet low frequency coefficient and each layer decompose is reconstructed, after obtaining noise reduction
Signal.
In the step 2, included the following steps with CEEMD decomposed signals：
N groups are added into the pretreated signal of noise reduction and assist white noise, aid in noise is added in a manner of positive and negative pair
, therefore generate the IMF components of two sets of set：
Wherein：S is the pretreated signal of noise reduction；N is the white noise for meeting normal distribution；M1、M2Respectively be added just,
Bear the signal of pairs of noise；
EMD decomposition is done respectively to each signal in set, each signal obtains one group of IMF component, wherein i-th of letter
Number j-th of IMF representation in components be cij；
Decomposition result is obtained by way of multigroup component combination：
Wherein：cjIndicate that CEEMD decomposes finally obtained j-th of IMF components；N indicates the group number of auxiliary white noise.
In the step 3, the method being combined based on kurtosis and continuous mean-square error criteria is as follows：
The kurtosis for calculating each IMF, when gear-box runs well, the vibration signal amplitude distribution of acquisition can be similar to normal state
Distribution, therefore its kurtosis value is equal to 3, when gear is there are when failure, kurtosis value can become larger.
The criterion of continuous mean square error (CMSE), i.e.,:
Wherein, H is the total length of signal, and r is the IMF numbers decomposed.
Critical IMF components are determined according to following two principle：If there are local poles before global minimum by CMSE
Small value, then the position corresponding to first local minimum should be taken to add 1；If there is no local minimum, then rounding body is most
Position corresponding to small value adds 1, including fault message it is more be critical IMF components and its IMF components later.It will be with
Kurtosis is combined the effective IMF components of selection；
The CICA methods extraction echo signal used in the step 4 includes the following steps：
Fault-signal characteristic frequency construction reference burst signal r (t) based on gear, by echo signal y to be extracted and
The distance function of reference signal r (t) is defined as ε (y, r), for indicating the degree of closeness of echo signal and reference signal.ε(y,
R) mean square error ε (y, r)=E { (y-r) } can be used to measure, shown in the mathematical model such as formula (4) and formula (5) of CICA algorithms：
Object function：
max J(y)≈ρ{E[G(y)]-E[G(v)]} (4)
Constraints：
Wherein：ρ is normal number；G () is nonlinear function；V is the gaussian variable with covariance matrix identical as y；ξ
For threshold value.Formula (5) is actually a constrained optimization problem, is solved to it by lagrange's method of multipliers, source signal can be obtained
Best estimate, extract target source signal.
Embodiment 1：
In order to verify the above method validity, using SpectraQuest companies design simulate industrial power pass
Dynamic fault diagnosis comprehensive experiment table (DDS) carries out experimental analysis, sees Fig. 2.The experimental bench power drive system is by 1 grade of row
Star gear-box, 2 grades of parallel-shaft gearboxes, a bearing load and a programmable magnetic brake composition, gear-box
Kinematic scheme it is as shown in Figure 3.
Driving wheel Z6 in the output stage Z6/Z7 of fixed axis gear case is arranged in gear local fault by the method research
It sets, position as shown in Figure 4, other positions gear (containing epicyclic gearbox) and the equal fault-free of all bearings.Fixed axis gear case tooth
The single failure of wheel Z6 is set as local broken teeth failure, and the width of wherein broken teeth is about the 30% of the facewidth, as shown in Figure 5；Motor
Input speed is 40Hz (2400r/min), and the frequency that turns for being transmitted to parallel teeth roller box jackshaft is 2.537Hz, the number of teeth 36, engagement
Frequency is 91.35Hz.Utilize DASP data collecting instrument gathered datas, sample frequency 5120Hz, a length of 10s when always sampling.
The single pass vibration data of gear-box is acquired, wavelet soft-threshold noise reduction first is pre-processed to improve signal-to-noise ratio,
Fig. 6 a- Fig. 6 b are the signal after the time domain source signal and noise reduction of acquisition；Fig. 7 a-7b be pretreated amplitude spectrum and envelope spectrum,
It can be seen that obvious peak value 39.86Hz, turns frequency 40Hz, it is further seen that close on jackshaft close to motor input from amplitude spectrum
The meshing frequency 91.18Hz peak values of gear, but do not occur apparent sideband, the also influence of other frequencies, and envelope spectrum
Middle gear guilty culprit axis turns still differentiate frequently, it is seen that it can reach certain noise reduction using wavelet soft-threshold, but
It directly can not effectively tell the fault signature of gear.
CEEMD decomposition further is carried out to the single channel signal after preliminary noise reduction, obtains 10 IMF components and a remnants
Component.The kurtosis value and CMSE values of each IMF are calculated, as shown in Figure 8 and Figure 9.It can be seen that IMF3, IMF5 and IMF6 are high and steep by kurtosis figure
Angle value is bigger；Select suitable IMF for the component after IMF3 with principle described herein from CMES figures.Synthesis is examined
Consider, finally judges that IMF3, IMF5 and IMF6 belong to vibration mode component, as shown in Figure 10.The IMF components selected and source are believed
Input signal number as CICA, it is as shown in figure 11 by the obtained reference signals of CICA and Objective extraction signal.
Amplitude spectrum analysis is done to the gear distress signal extracted, as shown in figure 12, can clear view to failure gear
Meshing frequency (theoretical value 91.35Hz) and its both sides are turned the side frequency of frequency 2.53Hz modulation；Envelope spectrum analysis, such as Figure 13 are done, it can
Intermediate shaft rotation frequency (theoretical value 2.53Hz) and its frequency multiplication ingredient being clearly visible where failure gear.
Test data handling result in of the embodiment of the present invention demonstrates the validity of this method.
Claims (6)
1. a kind of gear failure diagnosing method of single channel blind source separating, it is characterised in that include the following steps：
Wavelet soft-threshold noise reduction is carried out to collected gear-box single channel vibration signal；
Signal after noise reduction is subjected to CEEMD decomposition, obtains several IMF components and residual components；
Suitable IMF components are chosen using the method being combined based on kurtosis and continuous mean-square error criteria；
Using the IMF components of selection and source signal as the input signal of blind source component, echo signal is extracted using CICA methods；
Spectrum analysis is carried out to the echo signal extracted, fault signature is identified, completes the diagnosis of gear distress.
2. the gear failure diagnosing method of single channel blind source separating as described in claim 1, it is characterised in that the method is also
Including：Gear-box vibration signal is acquired with single acceleration transducer.
3. the gear failure diagnosing method of single channel blind source separating as described in claim 1, it is characterised in that the progress
The method of wavelet soft-threshold noise reduction includes the following steps：
The selection of wavelet basis function and Decomposition order：Wavelet basis function, the canonical of wavelet basis function are selected according to the characteristics of signal
Property and waveform and the structure similarity degree of data can influence the effect of signal de-noising, symN wavelet basis and mechanical oscillation waveform phase
Seemingly, therefore symN wavelet basis functions are selected to carry out wavelet soft-threshold noise reduction；Different decomposition number of plies noise reduction is different, rationally selects
Select Decomposition order；
Select threshold value and threshold value function：By comparing the size of wavelet coefficient and threshold value under each decomposition scale, by place
Manage the coefficient value after being purified；
Wavelet reconstruction：The high frequency coefficient that wavelet low frequency coefficient and each layer decompose is reconstructed, the signal after noise reduction is obtained.
4. the gear failure diagnosing method of single channel blind source separating as described in claim 1, it is characterised in that described will drop
Signal after making an uproar carries out CEEMD decomposition and includes the following steps：
N groups are added into the pretreated signal of noise reduction and assist white noise, aid in noise is added in a manner of positive and negative pair,
Therefore the IMF components of two sets of set are generated：
Wherein：S is the pretreated signal of noise reduction；N is the white noise for meeting normal distribution；M1、M2Respectively be added it is positive and negative at
To the signal of noise；
EMD decomposition is done respectively to each signal in set, each signal obtains one group of IMF component, wherein i-th signal
J-th of IMF representation in components is cij；
Decomposition result is obtained by way of multigroup component combination：
Wherein：cjIndicate that CEEMD decomposes finally obtained j-th of IMF components, n indicates the group number of auxiliary white noise.
5. the gear failure diagnosing method of single channel blind source separating as described in claim 1, it is characterised in that it is described based on
The method that kurtosis and continuous mean-square error criteria are combined is as follows：
The kurtosis for calculating each IMF, when gear-box runs well, the vibration signal amplitude distribution of acquisition can be similar to normal state point
Cloth, therefore its kurtosis value is equal to 3, when gear is there are when failure, kurtosis value can become larger；
The criterion of continuous mean square error (CMSE), i.e.,:
Wherein, H is the total length of signal, and r is the IMF numbers decomposed；
Critical IMF components are determined according to following two principle：
If there are local minimums before global minimum by CMSE, the position corresponding to first local minimum is taken
Add 1；
If there is no local minimum, then the position corresponding to global minimum is taken to add 1, including mostly fault message being to be faced
Boundary's IMF components and its IMF components later will be combined the effective IMF components of selection with kurtosis.
6. the gear failure diagnosing method of single channel blind source separating as described in claim 1, it is characterised in that：The use
CICA methods extraction echo signal includes the following steps：
Fault-signal characteristic frequency construction reference burst signal r (t) based on gear, by echo signal y to be extracted and reference
The distance function of signal r (t) is defined as ε (y, r), for indicating the degree of closeness of echo signal and reference signal；ε (y, r) is used
Mean square error ε (y, r)=E { (y-r) } is measured, shown in the mathematical model such as formula (4) and formula (5) of CICA algorithms：
Object function：
maxJ(y)≈ρ{E[G(y)]-E[G(v)]} (4)
Constraints：
Wherein：ρ is normal number；G () is nonlinear function；V is the gaussian variable with covariance matrix identical as y；ξ is threshold
Value；Formula (5) solves it by lagrange's method of multipliers, and the best estimate of source signal can be obtained, and extracts required signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810653028.3A CN108801630B (en) | 2018-06-22 | 2018-06-22 | Gear fault diagnosis method for single-channel blind source separation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810653028.3A CN108801630B (en) | 2018-06-22 | 2018-06-22 | Gear fault diagnosis method for single-channel blind source separation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108801630A true CN108801630A (en) | 2018-11-13 |
| CN108801630B CN108801630B (en) | 2020-04-14 |
Family
ID=64084679
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810653028.3A Active CN108801630B (en) | 2018-06-22 | 2018-06-22 | Gear fault diagnosis method for single-channel blind source separation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108801630B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110146291A (en) * | 2019-05-31 | 2019-08-20 | 昆明理工大学 | A kind of Rolling Bearing Fault Character extracting method based on CEEMD and FastICA |
| CN111079710A (en) * | 2019-12-31 | 2020-04-28 | 江苏理工学院 | Multilayer noise reduction method based on improved CEEMD rolling bearing signal |
| CN111310589A (en) * | 2020-01-20 | 2020-06-19 | 河北科技大学 | Fault diagnosis method and device for mechanical system and terminal |
| CN112183259A (en) * | 2020-09-17 | 2021-01-05 | 昆明理工大学 | Rolling bearing fault diagnosis method based on CEEMD and kurtosis weighted average threshold denoising |
| CN112945546A (en) * | 2021-01-20 | 2021-06-11 | 南京航空航天大学 | Accurate diagnosis method for complex fault of gear box |
| CN113074935A (en) * | 2021-04-01 | 2021-07-06 | 西华大学 | Acoustic separation and diagnosis method for impact fault characteristics of gearbox |
| CN113884236A (en) * | 2021-08-24 | 2022-01-04 | 西安电子科技大学 | Multi-sensor fusion dynamic balance analysis method, system, equipment and medium |
| CN116756490A (en) * | 2023-06-15 | 2023-09-15 | 沈阳航空航天大学 | Rolling bearing fault early warning method based on beta distribution and EEMD-CMSE |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107192553A (en) * | 2017-06-28 | 2017-09-22 | 石家庄铁道大学 | Gear-box combined failure diagnostic method based on blind source separating |
| CN107255563A (en) * | 2017-06-28 | 2017-10-17 | 石家庄铁道大学 | Realize gear-box mixed fault signal blind source separation method |
| US20180138896A1 (en) * | 2016-11-17 | 2018-05-17 | Hwashen Electronics Co., Ltd. | Fuzzy entropy based noisy signal processing method and iterative singular spectrum analysis soft threshold de-noising method |
-
2018
- 2018-06-22 CN CN201810653028.3A patent/CN108801630B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180138896A1 (en) * | 2016-11-17 | 2018-05-17 | Hwashen Electronics Co., Ltd. | Fuzzy entropy based noisy signal processing method and iterative singular spectrum analysis soft threshold de-noising method |
| CN107192553A (en) * | 2017-06-28 | 2017-09-22 | 石家庄铁道大学 | Gear-box combined failure diagnostic method based on blind source separating |
| CN107255563A (en) * | 2017-06-28 | 2017-10-17 | 石家庄铁道大学 | Realize gear-box mixed fault signal blind source separation method |
Non-Patent Citations (2)
| Title |
|---|
| 孙伟峰等: "基于EMD的激光超声信号去噪方法", 《山东大学学报（工学版）》 * |
| 杨孟等: "基于CEEMD和小波包的降噪方法研究", 《西安邮电大学学报（自然科学版）》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110146291A (en) * | 2019-05-31 | 2019-08-20 | 昆明理工大学 | A kind of Rolling Bearing Fault Character extracting method based on CEEMD and FastICA |
| CN111079710A (en) * | 2019-12-31 | 2020-04-28 | 江苏理工学院 | Multilayer noise reduction method based on improved CEEMD rolling bearing signal |
| CN111079710B (en) * | 2019-12-31 | 2023-04-18 | 江苏理工学院 | Multilayer noise reduction method based on improved CEEMD rolling bearing signal |
| CN111310589A (en) * | 2020-01-20 | 2020-06-19 | 河北科技大学 | Fault diagnosis method and device for mechanical system and terminal |
| CN111310589B (en) * | 2020-01-20 | 2023-04-28 | 河北科技大学 | Fault diagnosis method, fault diagnosis device and terminal of mechanical system |
| CN112183259A (en) * | 2020-09-17 | 2021-01-05 | 昆明理工大学 | Rolling bearing fault diagnosis method based on CEEMD and kurtosis weighted average threshold denoising |
| CN112183259B (en) * | 2020-09-17 | 2022-11-04 | 昆明理工大学 | Rolling bearing fault diagnosis method based on CEEMD and kurtosis weighted average threshold denoising |
| CN112945546A (en) * | 2021-01-20 | 2021-06-11 | 南京航空航天大学 | Accurate diagnosis method for complex fault of gear box |
| CN113074935A (en) * | 2021-04-01 | 2021-07-06 | 西华大学 | Acoustic separation and diagnosis method for impact fault characteristics of gearbox |
| CN113884236A (en) * | 2021-08-24 | 2022-01-04 | 西安电子科技大学 | Multi-sensor fusion dynamic balance analysis method, system, equipment and medium |
| CN113884236B (en) * | 2021-08-24 | 2022-06-21 | 西安电子科技大学 | Multi-sensor fusion dynamic balance analysis method, system, equipment and medium |
| CN116756490A (en) * | 2023-06-15 | 2023-09-15 | 沈阳航空航天大学 | Rolling bearing fault early warning method based on beta distribution and EEMD-CMSE |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108801630B (en) | 2020-04-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108801630A (en) | The gear failure diagnosing method of single channel blind source separating | |
| Wang et al. | Subband averaging kurtogram with dual-tree complex wavelet packet transform for rotating machinery fault diagnosis | |
| Chen et al. | Fault diagnosis method based on integration of RSSD and wavelet transform to rolling bearing | |
| WO2022261805A1 (en) | Diesel engine gearbox fault diagnosis method | |
| Wang et al. | Complete ensemble local mean decomposition with adaptive noise and its application to fault diagnosis for rolling bearings | |
| CN102937522B (en) | Composite fault diagnosis method and system of gear case | |
| CN105115594B (en) | Gear-box vibration signal fault signature extracting method based on Wavelet Entropy and information fusion | |
| CN104714925B (en) | A kind of gear transmission noises analysis method based on Fourier Transform of Fractional Order and SVMs | |
| Wang et al. | Optimal demodulation subband selection for sun gear crack fault diagnosis in planetary gearbox | |
| CN105928702B (en) | Variable working condition box bearing method for diagnosing faults based on form PCA | |
| CN106778694A (en) | A kind of gear transmission noises analysis method based on set empirical mode decomposition and SVMs | |
| CN110598593B (en) | Planetary gearbox fault diagnosis method based on resonance sparse decomposition and FastICA algorithm | |
| CN101587017A (en) | Gear fault diagnosis method based on part mean decomposition cycle frequency spectrum | |
| CN104316163B (en) | Gear-box coupling modulation signal separating method based on interior product transformation and correlation filtering | |
| CN104215456B (en) | Plane clustering and frequency-domain compressed sensing reconstruction based mechanical fault diagnosis method | |
| CN107608936A (en) | A kind of epicyclic gearbox combined failure feature extracting method | |
| CN110398364B (en) | Planetary gearbox fault diagnosis method based on resonance sparse decomposition and FastICA algorithm | |
| CN110333071B (en) | Mechanical vibration signal processing method using narrowband cepstrum transformation | |
| CN110672327A (en) | Asynchronous motor bearing fault diagnosis method based on multilayer noise reduction technology | |
| CN104502099A (en) | Cyclic frequency extraction method for characteristic components of transient conditions of gearbox | |
| CN109540523A (en) | A kind of ship centrifugal pump Fault Diagnosis of Roller Bearings | |
| CN113390631A (en) | Fault diagnosis method for gearbox of diesel engine | |
| CN106383028A (en) | Gear case fault diagnosis method | |
| Lin et al. | Application of the EEMD method to multiple faults diagnosis of gearbox | |
| Huang et al. | Feature extraction for rolling element bearing faults using resonance sparse signal decomposition |
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 |