CN104485113A - Multi-fault-source acoustic emission signal separation method - Google Patents
Multi-fault-source acoustic emission signal separation method Download PDFInfo
- Publication number
- CN104485113A CN104485113A CN201410808918.9A CN201410808918A CN104485113A CN 104485113 A CN104485113 A CN 104485113A CN 201410808918 A CN201410808918 A CN 201410808918A CN 104485113 A CN104485113 A CN 104485113A
- Authority
- CN
- China
- Prior art keywords
- signal
- source
- fault
- function
- noise
- 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.)
- Pending
Links
- 238000000926 separation method Methods 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 claims abstract description 21
- 230000008602 contraction Effects 0.000 claims abstract 2
- 238000000354 decomposition reaction Methods 0.000 claims description 9
- 230000002087 whitening effect Effects 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000002474 experimental method Methods 0.000 claims description 2
- 238000011946 reduction process Methods 0.000 claims description 2
- 238000001228 spectrum Methods 0.000 claims description 2
- YSIITVVESCNIPR-UHFFFAOYSA-N Troxipide Chemical compound COC1=C(OC)C(OC)=CC(C(=O)NC2CNCCC2)=C1 YSIITVVESCNIPR-UHFFFAOYSA-N 0.000 claims 1
- 238000010606 normalization Methods 0.000 claims 1
- 229960001341 troxipide Drugs 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 17
- 238000012880 independent component analysis Methods 0.000 abstract description 12
- 238000004458 analytical method Methods 0.000 abstract description 10
- 238000012545 processing Methods 0.000 description 7
- 208000037656 Respiratory Sounds Diseases 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 229910000861 Mg alloy Inorganic materials 0.000 description 1
- 206010038743 Restlessness Diseases 0.000 description 1
- 239000000956 alloy Substances 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 229940099259 vaseline Drugs 0.000 description 1
Landscapes
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a multi-fault source acoustic emission signal separation method, which mainly comprises the following steps: for the sound emission mixed signal of the noise-containing multi-fault source, firstly, denoising the noise-containing mixed signal by using a wavelet packet analysis method, reserving signals in a frequency band which accounts for 80% -85% of signal energy, then, adopting a FastICA algorithm in independent component analysis to separate each fault source in the denoised mixed signal, and finally, removing the noise in the frequency band from each separated fault signal by using a contraction function, thereby obtaining a source fault signal. The multi-fault-source signal separation method based on wavelet packet analysis and independent component analysis can separate fault signals under the conditions of strong background noise and containing multiple fault sources, the separation effect is higher than that of a method of singly adopting a FastICA algorithm, and the method is a better multi-fault-source signal separation new method.
Description
Technical field
The invention belongs to multi-channel signal processing field, specifically a kind of multiple faults source acoustic emission signal separation method based on wavelet packet analysis and independent component analysis (Independent Components Analysis, ICA).
Background technology
Easily there is the various faults such as fatigue crack and rubbing wear in the rotary parts such as lathe, aircraft device middle gear and bearing, if can not tremendous economic may be caused to lose by Timeliness coverage, even cause catastrophic failure after long-play.If acoustic emission (Acoustic Emission, AE) technology for detection just can be utilized in early days and isolate these faults, great directive significance will be had to mechanical fault diagnosis.
Mainly there is the features such as low and each source signal aliasing of signal to noise ratio (S/N ratio) in the fault-signal of early stage rotary part.Fault-signal due to rotating machinery is separate as crackle, rubbing wear etc., and therefore ICA method carries out Signal separator.Classical ICA algorithm is separated signal mainly for without situation of making an uproar, and in actual environment, observation signal is doped with various noise, and along with the raising of noise intensity, classical ICA algorithm separating effect is poorer.Much study using noise also as an independent source at present, then blind source separation method is adopted to be separated, but noise signal is except the noise that tested object itself exists, also exist from impacts such as disturbance sources in the various electronic noise of acquisition channel and transmission path, cause each channel acquisition actual to noise signal can not simply regard as from same noise source.And noise-reduction method, except the methods such as time-domain analysis, frequency-domain analysis and Wavelet Denoising Method, IC standard A model can carry out the sparse coding of uncorrelated components, can remove noise wherein by contracting function.
Therefore, wavelet packet analysis combines with ICA by the present invention, first wavelet-packet noise reduction pre-service is carried out to fault-signal, reconstruct accounts for the in-band signal of mixed signal 80%-85% energy, adopt FastICA algorithm to carry out multiple faults source blind separation to the signal in characteristic spectra again, then choose contracting function according to the pdf model of each source of trouble signal and noise reduction in frequency range is carried out to fault-signal.The present invention takes wavelet-packet noise reduction pre-service and contracting function noise reduction aftertreatment 2 noise reduction process, to improve the separating effect of feeble signal.
Summary of the invention
The present invention is in conjunction with the pre-service of wavelet packet analysis noise reduction, the aftertreatment of contracting function noise reduction and ICA multi-source blind separating method, object is to improve faint multi-source acoustic emission signal separating effect, a kind of multiple faults source signal separation method based on wavelet packet analysis and independent component analysis is provided, to improve effect of signal separation, for fault diagnosis provides information more accurately.
A kind of multiple faults source signal of the present invention separation method, comprises the following steps:
(1) to the noisy observation signal X=of the linear aliasing of the M [x that instrument gathers
1, x
2, x
m]
tcarry out WAVELET PACKET DECOMPOSITION.Every road observation signal x
iwAVELET PACKET DECOMPOSITION step as follows: carry out n-layer wavelet packet decomposition by selected wavelet packet basis functions, respectively to decompose 2
nindividual wavelet packet coefficient is reconstructed, and obtains 2 of variant frequency range
nindividual reconstruction signal, the amplitude selecting to account for mixed signal energy 80%-85% is comparatively large and frequency range continuous print reconstruction signal carries out addition obtains wavelet-packet noise reduction signal x
i'.M road signal all obtains X '=[x after noise reduction
1', x
2' ..., x
m']
t;
(2) average value processing is gone to X ' Zhong Mei road signal, make E (x '
i)=0, then carry out whitening processing and obtain signal z.M road signal in X ' all obtains Z=[z after past average whitening processing
1, z
2..., z
m]
t;
(3) make M equal the number of the Independent sources signal that will estimate, counter i=1 is set;
(4) the initial weight vector w of random generation
i, make w
i=w
i/ || w
i||;
(5)w
i←E{zg(w
i Tz)}-E{g'(w
i Tz)}w
i,w
i=w
i/||w
i||;
(6) orthonomalization process:
w
i=w
i/ || w
i||;
(7) if w
ido not restrain, then turn back to step (5);
(8) w
iconvergence, then i=i+1, if i≤M, then turns back to step (4);
(9) obtain separating mixed matrix W=[w
1, w
2..., w
m]
t, try to achieve according to Y=WZ and separate mixed signal Y=[y
1, y
2..., y
m]
t;
(10) corresponding contracting function f (y) is chosen according to the pdf model of fault-signal, after shrinking noise reduction
then the present invention is finally separated and obtains signal
Under the present invention is used for there is multiple source of trouble and the stronger situation of ground unrest simultaneously, the noise signal outside fault-signal primary band and in primary band is removed respectively by wavelet package reconstruction technology and contracting function, and in conjunction with ICA technology, multiple faults source signal is separated, be the good faint multiple faults source signal separation method of a kind of extraction effect.
Accompanying drawing explanation
A kind of multiple faults source of Fig. 1 acoustic emission signal separation method process flow diagram;
Fig. 2 source signal time-frequency figure;
Separating effect figure of the present invention under the different input signal-to-noise ratio of Fig. 3;
FastICA algorithm separating effect figure is adopted separately under the different input signal-to-noise ratio of Fig. 4.
Embodiment
Be described in detail below in conjunction with the technical scheme of accompanying drawing to invention:
Fig. 1 is a kind of multiple faults source acoustic emission signal separation method overview flow chart.The step of this method is as follows:
(1) WAVELET PACKET DECOMPOSITION is carried out to the noisy observation signal of the linear aliasing of the M of instrument collection.The WAVELET PACKET DECOMPOSITION step of every road observation signal is as follows: select sym8 wavelet packet basis functions to carry out 5 floor WAVELET PACKET DECOMPOSITION, respectively to decompose 2
5individual wavelet packet coefficient is reconstructed, and obtains 2 of variant frequency range
5individual reconstruction signal, selects the frequency range continuous print reconstruction signal of 50-180kHz to carry out addition and obtains wavelet-packet noise reduction signal x
i'.M road signal all obtains X '=[x after noise reduction
1', x
2' ..., x
m']
t;
(2) average value processing is gone to X ' Zhong Mei road signal, make E (x '
i)=0, then carry out whitening processing and obtain signal z.M road signal in X ' all obtains Z=[z after past average whitening processing
1, z
2..., z
m]
t;
(3) make M equal the number of the Independent sources signal that will estimate, M=2 in the present embodiment, counter i=1 is set;
(4) the initial weight vector w of random generation
i, make w
i=w
i/ || w
i||;
(5)w
i←E{zg(w
i Tz)}-E{g'(w
i Tz)}w
i,w
i=w
i/||w
i||;
(6) orthonomalization process:
w
i=w
i/ || w
i||;
(7) if w
ido not restrain, then turn back to step (5);
(8) w
iconvergence, then i=i+1, if i≤M, then turns back to step (4);
(9) obtain separating mixed matrix W=[w
1, w
2..., w
m]
t, try to achieve according to Y=WZ and separate mixed signal Y=[y
1, y
2..., y
m]
t;
(10) double-exponential function is adopted
come approximate crackle and tribological failure signal probability density function model, thus contracting function is
after shrinking noise reduction
then the present invention is finally separated and obtains signal
Wherein, step (1) is wavelet-packet noise reduction pre-service, and (2) ~ (9) are FastICA multi-source blind separation, and (10) are the aftertreatment of contracting function noise reduction.
Obtain simulating crack acoustic emission signal by Nielson source method of testing in the present embodiment, with 2 pieces of magnesium alloy materials rub instantaneously obtain simulation grating transmit, the time-frequency figure of two kinds of signals is as shown in Figure 2.The kurtosis of calculating simulation Signal of Cracks, friction signal is respectively 3.56 and 1.24, and thus it all has super-Gaussian.Nonlinear function choosing in FastICA algorithm.As can be seen from Figure 2, simulating crack and friction signal concentrate on 50-180kHz and 50-170kHz respectively, and therefore when using wavelet-packet noise reduction, the present embodiment adopts the frequency content in 50-180kHz to carry out wavelet package reconstruction.
All simulating signals complete collection by the PCI-2 system of PAC company of the U.S., and calibrate AE sensor model is R15 α (frequency range: 50-200kHz), 2/4/6 type prime amplifier (bandwidth 10-2000kHz).Experiment correlation parameter is as follows: sample frequency is 2MHz, and pre-amp gain is 40dB, and sampling number is 512, and threshold value is set to 30dB, adopts vaseline as couplant.Generate one group of white Gauss noise with Matlab and simulate interference noise in actual environment.
0dB is added ,-5dB ,-15dB white Gauss noise to the crackle gathered and friction signal, respectively it is separated by method of the present invention.Through repetition test, when input signal-to-noise ratio is 0dB ,-5dB ,-15dB, for its σ of Signal of Cracks
2elect 0.32 as respectively, 0.38,0.46; And its σ of friction signal
2elect 0.29,0.36,0.51 as respectively, its separating effect is better, and isolated signal as shown in Figure 3; Independent employing FastICA algorithm, to being separated it respectively, obtains the separating effect figure of Fig. 4.Two kinds of method separating effect evaluation indexes are in table 1.Adopt the separating effect of signal to noise ratio (S/N ratio) and related coefficient evaluation signal, wherein related coefficient is:
i,j=1,2
ρ in formula
11represent the related coefficient between isolated simulating crack signal source signal corresponding to it, ρ
22represent the related coefficient between isolated friction signal source signal corresponding to it.Signal to noise ratio (S/N ratio) is:
SNR in formula
1represent the signal to noise ratio (S/N ratio) of isolated simulating crack signal, SNR
2for the signal to noise ratio (S/N ratio) of isolated friction signal.
Table 1
As can be seen from the present invention's algorithm used and the separating resulting adopting separately FastICA algorithm, the inventive method to the separating effect of signals and associated noises significantly better than FastICA algorithm, and input signal-to-noise ratio its effect lower is more obvious, and FastICA algorithm declines very fast along with its separating effect of raising of noise intensity.As when input signal-to-noise ratio is 0dB, after this paper algorithm is separated, its Signal of Cracks and friction signal signal to noise ratio (S/N ratio) are respectively 9.78dB and 9.65dB, and related coefficient is respectively 0.94 and 0.92, and signal to noise ratio (S/N ratio) is respectively 3.38dB and 3.16dB in FastICA separation, related coefficient is respectively 0.55 and 0.52; When input signal-to-noise ratio is-15dB, after this paper algorithm is separated, its signal to noise ratio (S/N ratio) is for being respectively 3.28dB and 3.41dB, related coefficient is respectively 0.72 and 0.74, now signal to noise ratio (S/N ratio)-4.48dB and-4.04dB respectively after FastICA is separated, but related coefficient is all lower than 0.25, the signal obtained after separation is described and source signal have nothing to do.Can find out that FastICA algorithm is poor to feeble signal separating effect thus.
Visible by the present embodiment, can extract the multi-source fault-signal of input signal-to-noise ratio higher than-15dB.Instant invention overcomes the conventional failure method for extracting signal deficiency that separating effect is poor under multiple source of trouble and strong background noise, is that a kind of multi-source knocking noise based on wavelet packet analysis and independent component analysis transmits extracting method.
Finally it should be noted that above embodiment only in order to technical scheme of the present invention to be described, but not the restriction to usable range of the present invention.
Claims (7)
1. a multiple faults source acoustic emission signal separation method, is characterized in that, comprise the following steps:
Step one: the noisy observation signal X=of the linear aliasing of the M [x that instrument is gathered
1, x
2, x
m]
tn-layer wavelet packet decomposition is carried out, respectively to decompose 2 by selected wavelet packet basis functions
nindividual wavelet packet coefficient is reconstructed, and obtains 2 of variant frequency range
nindividual reconstruction signal, the comparatively large and frequency range continuous print reconstruction signal of the amplitude selecting to account for mixed signal energy 80%-85% carries out being added as wavelet-packet noise reduction signal;
Step 2: FastICA multi-source blind separation is carried out to described wavelet-packet noise reduction signal;
Step 3: the pdf model selecting each fault-signal, obtains the contracting function of described fault-signal, utilizes it to carry out contraction denoising to step 2 each source of trouble isolated.
2. a kind of multiple faults source acoustic emission signal separation method according to claim 1, it is characterized in that, in described step one, wavelet packet Decomposition order n is the frequency spectrum according to sensor collection signal, determines with the 80%-85% that can retain energy.
3. a kind of multiple faults source acoustic emission signal separation method according to claim 1 or 2, it is characterized in that, the defining method of the wavelet packet basis functions in described step one is the Decomposition order according to determining in claim 2, adopt different basis function to carry out denoising experiment, with optimum comprehensive in signal to noise ratio (S/N ratio) in noise reduction result and related coefficient two indices is Optimal Wavelet Packet basis function.
4. a kind of multiple faults source acoustic emission signal separation method according to claim 1, is characterized in that, in described step 2
FastICA multi-source blind separation algorithm key step is as follows:
Step 1, data carry out average, whitening pretreatment;
Step 2, initialization i=1, M are fault-signal number;
Step 3, initialization weight vector w
i, and to its normalized;
Step 4, iteration normalization: w
i← E{zg (w
i tz) }-E{g'(w
i tz) } w
i, w
i=w
i/ || w
i||;
Step 5, orthonomalization process:
If step 6 w
ido not restrain, then turn back to step 4;
Step 7, w
iconvergence, then i=i+1, if i≤M, then turns back to step 3;
Step 8, obtain separating mixed matrix W=[w
1, w
2..., w
m]
t, try to achieve according to Y=WZ and separate mixed signal Y=[y
1, y
2..., y
m]
t.
5. a kind of multiple faults source acoustic emission signal separation method according to claim 4, it is characterized in that, in described step 4, the nonlinear function of FastICA algorithm is chosen according to the Gaussian of source signal: when source signal is gaussian signal, nonlinear function selects g
1(u)=tanh (a
1u); When source signal is Super-Gaussian Signals, nonlinear function selects g
2(u)=uexp (-a
2u
2/ 2); When source signal is sub-Gaussian signals, nonlinear function selects g
3(u)=u
3, 1≤a
1≤ 2, a
2=1.
6. a kind of multiple faults source acoustic emission signal separation method according to claim 1, it is characterized in that, contracting function in described step 3 is chosen according to the pdf model of source signal: the probability density function of source signal obeys generalized L aplace distribution, and contracting function is
when the sparse degree of probability density function of source signal distributes lower than Laplace, its contracting function is
Wherein
When the probability density function of source signal
degree of rarefication when distributing higher than Laplace, its contracting function is
Wherein
Also can according to the feature of source signal, select other probability density functions that described three kinds of probability density functions are derived and contracting function, wherein, s is independent component component, is the estimated value of independent component component, is noise variance, and C is constant.
7. a kind of multiple faults source acoustic emission signal separation method according to claim 1, it is characterized in that the method is applicable to multiple faults source signal, wavelet-packet noise reduction pre-service and contracting function noise reduction aftertreatment 2 noise reduction process are carried out, and the fault-signal that the method gathers under being mainly applicable to positive fixed condition, namely the number of the source of trouble is identical with sensor acquisition channel number M.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410808918.9A CN104485113A (en) | 2014-12-23 | 2014-12-23 | Multi-fault-source acoustic emission signal separation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410808918.9A CN104485113A (en) | 2014-12-23 | 2014-12-23 | Multi-fault-source acoustic emission signal separation method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104485113A true CN104485113A (en) | 2015-04-01 |
Family
ID=52759652
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410808918.9A Pending CN104485113A (en) | 2014-12-23 | 2014-12-23 | Multi-fault-source acoustic emission signal separation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104485113A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105119850A (en) * | 2015-06-30 | 2015-12-02 | 电子科技大学 | Fast fixed point processing method used for communication signal blind separation |
CN105206283A (en) * | 2015-08-28 | 2015-12-30 | 北京航空航天大学 | amage signal blind source processing method, apparatus, and system of composite material structure |
CN106500827A (en) * | 2016-09-25 | 2017-03-15 | 郑州航空工业管理学院 | A kind of single channel method for separating vibration signal blind sources |
CN107025446A (en) * | 2017-04-12 | 2017-08-08 | 北京信息科技大学 | A kind of vibration signal combines noise-reduction method |
CN109409341A (en) * | 2018-12-10 | 2019-03-01 | 中国航发四川燃气涡轮研究院 | A kind of aero-engine noise source discrimination method near field |
CN109670536A (en) * | 2018-11-30 | 2019-04-23 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | A kind of local discharge signal clustering method in the case of multi-source electric discharge and interference superposition |
CN110398364A (en) * | 2019-07-05 | 2019-11-01 | 东南大学 | Epicyclic gearbox method for diagnosing faults based on resonance sparse decomposition and FastICA algorithm |
CN111444893A (en) * | 2020-05-06 | 2020-07-24 | 南昌航空大学 | Fault diagnosis method for main shaft device of mine hoist |
CN112116922A (en) * | 2020-09-17 | 2020-12-22 | 集美大学 | Noise blind source signal separation method, terminal equipment and storage medium |
CN117905711A (en) * | 2024-03-20 | 2024-04-19 | 江苏海拓宾未来工业科技集团有限公司 | Multi-shaft centrifugal compressor and fault diagnosis method for impeller of multi-shaft centrifugal compressor |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103433806A (en) * | 2013-08-01 | 2013-12-11 | 上海交通大学 | Self-adapting tool tiny breakage monitoring system and monitoring method |
CN103760243A (en) * | 2014-02-26 | 2014-04-30 | 长沙理工大学 | Microcrack nondestructive testing device and method |
CN103926097A (en) * | 2014-04-03 | 2014-07-16 | 北京工业大学 | Method for collecting and extracting fault feature information of low-speed and heavy-load device |
-
2014
- 2014-12-23 CN CN201410808918.9A patent/CN104485113A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103433806A (en) * | 2013-08-01 | 2013-12-11 | 上海交通大学 | Self-adapting tool tiny breakage monitoring system and monitoring method |
CN103760243A (en) * | 2014-02-26 | 2014-04-30 | 长沙理工大学 | Microcrack nondestructive testing device and method |
CN103926097A (en) * | 2014-04-03 | 2014-07-16 | 北京工业大学 | Method for collecting and extracting fault feature information of low-speed and heavy-load device |
Non-Patent Citations (3)
Title |
---|
王向红: ""混流式水轮机叶片裂纹声发射监测的若干关键技术研究"", 《中国博士学位论文全文数据库 信息科技辑》 * |
田峥译: "《统计建模的小波方法 中文版》", 31 March 2007, 高等教育出版社 * |
邹小波等: "《农产品无损检测技术与数据分析方法》", 31 January 2008, 中国轻工业出版社 * |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105119850A (en) * | 2015-06-30 | 2015-12-02 | 电子科技大学 | Fast fixed point processing method used for communication signal blind separation |
CN105206283B (en) * | 2015-08-28 | 2019-10-08 | 北京航空航天大学 | A kind of blind source processing method of the damage signal of composite structure, device and system |
CN105206283A (en) * | 2015-08-28 | 2015-12-30 | 北京航空航天大学 | amage signal blind source processing method, apparatus, and system of composite material structure |
CN106500827A (en) * | 2016-09-25 | 2017-03-15 | 郑州航空工业管理学院 | A kind of single channel method for separating vibration signal blind sources |
CN107025446A (en) * | 2017-04-12 | 2017-08-08 | 北京信息科技大学 | A kind of vibration signal combines noise-reduction method |
CN109670536A (en) * | 2018-11-30 | 2019-04-23 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | A kind of local discharge signal clustering method in the case of multi-source electric discharge and interference superposition |
CN109670536B (en) * | 2018-11-30 | 2023-05-30 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | Partial discharge signal clustering method under multi-source discharge and interference superposition condition |
CN109409341A (en) * | 2018-12-10 | 2019-03-01 | 中国航发四川燃气涡轮研究院 | A kind of aero-engine noise source discrimination method near field |
CN110398364A (en) * | 2019-07-05 | 2019-11-01 | 东南大学 | Epicyclic gearbox method for diagnosing faults based on resonance sparse decomposition and FastICA algorithm |
CN110398364B (en) * | 2019-07-05 | 2021-05-18 | 东南大学 | Planetary gearbox fault diagnosis method based on resonance sparse decomposition and FastICA algorithm |
CN111444893A (en) * | 2020-05-06 | 2020-07-24 | 南昌航空大学 | Fault diagnosis method for main shaft device of mine hoist |
CN112116922A (en) * | 2020-09-17 | 2020-12-22 | 集美大学 | Noise blind source signal separation method, terminal equipment and storage medium |
CN112116922B (en) * | 2020-09-17 | 2024-04-12 | 集美大学 | Noise blind source signal separation method, terminal equipment and storage medium |
CN117905711A (en) * | 2024-03-20 | 2024-04-19 | 江苏海拓宾未来工业科技集团有限公司 | Multi-shaft centrifugal compressor and fault diagnosis method for impeller of multi-shaft centrifugal compressor |
CN117905711B (en) * | 2024-03-20 | 2024-07-23 | 江苏海拓宾未来工业科技集团有限公司 | Multi-shaft centrifugal compressor and fault diagnosis method for impeller of multi-shaft centrifugal compressor |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104485113A (en) | Multi-fault-source acoustic emission signal separation method | |
Wang et al. | An adaptive SK technique and its application for fault detection of rolling element bearings | |
Zhang et al. | Adaptive fault feature extraction from wayside acoustic signals from train bearings | |
CN107192553B (en) | Gear-box combined failure diagnostic method based on blind source separating | |
Obuchowski et al. | Selection of informative frequency band in local damage detection in rotating machinery | |
CN102697495B (en) | Second-generation wavelet electromyographic signal noise eliminating method based on ensemble empirical mode decomposition | |
CN107991706B (en) | Coal seam hydraulic fracturing microseismic signal combined noise reduction method based on wavelet packet multiple threshold and improved empirical mode decomposition | |
Jian et al. | On the denoising method of prestack seismic data in wavelet domain | |
CN103961092B (en) | EEG Noise Cancellation based on adaptive thresholding | |
US20080262371A1 (en) | Method for Adaptive Complex Wavelet Based Filtering of Eeg Signals | |
CN103163505B (en) | Time-varying narrow-band interference suppression method based on joint approximate diagonalization of eigen-matrices (JADE) | |
CN104390781A (en) | Gear fault diagnosis method based on LMD and BP neural network | |
CN110426569B (en) | Noise reduction processing method for acoustic signals of transformer | |
CN102819043B (en) | Array signal random noise adaptive model denoising method | |
CN103190898A (en) | Cardiac magnetic signal noise adaptive filtering and eliminating design method | |
CN110244202A (en) | Based on synchronous compression wavelet transformed domain partial discharge of transformer denoising method | |
CN113642484B (en) | Magnetotelluric signal noise suppression method and system based on BP neural network | |
CN103778921A (en) | Method for eliminating nonuniform noise in speech collected by radar | |
Cao et al. | A method for extracting weak impact signal in NPP based on adaptive Morlet wavelet transform and kurtosis | |
CN112818876A (en) | Electromagnetic signal extraction and processing method based on deep convolutional neural network | |
Meng et al. | Loose parts detection method combining blind deconvolution with support vector machine | |
Jiang et al. | Differential spectral amplitude modulation and its applications in rolling bearing fault diagnosis | |
Li et al. | A sensor-dependent vibration data driven fault identification method via autonomous variational mode decomposition for transmission system of shipborne antenna | |
CN102043168B (en) | Method for carrying out simulation noise addition on digital signal | |
CN107576380A (en) | A kind of three-dimensional vibrating Modulation recognition method towards Φ OTDR techniques |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20150401 |
|
WD01 | Invention patent application deemed withdrawn after publication |