CN108120598A - Square phase-couple and the bearing incipient fault detection method for improving bispectrum algorithm - Google Patents
Square phase-couple and the bearing incipient fault detection method for improving bispectrum algorithm Download PDFInfo
- Publication number
- CN108120598A CN108120598A CN201711375557.3A CN201711375557A CN108120598A CN 108120598 A CN108120598 A CN 108120598A CN 201711375557 A CN201711375557 A CN 201711375557A CN 108120598 A CN108120598 A CN 108120598A
- Authority
- CN
- China
- Prior art keywords
- bearing
- frequency
- msub
- mrow
- phase
- 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
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
Abstract
The invention discloses a kind of square phase-couples and the bearing incipient fault detection method for improving bispectrum algorithm, comprise the following steps：Obtain resampling signal sequence；The improvement bi-spectrum model counterweight sampled signal sequence detected by square phase-couple is calculated；Analysis is detected to bearing early stage minor defect characteristic component.Coupling of this method between failure behavior and system dynamic behaviour, with reference to advanced signal processing theory and Research on Methods rolling bearing fault features detection method, its target is exactly by vibration signal, by considering the nonlinear coupling effects of failure impact and recognizing the bearing fault still in early stage using improved higher order statistic analysis.This method has reliable and stable detectability to the detection of bearing early-stage weak fault, can find bearing fault in time, avoid the generation of bearing accident；It is applied widely, it can be applied to incipient fault detection and the tracking of all kinds of transmission system middle (center) bearings.
Description
Technical field
The invention belongs to condition monitoring and fault diagnosis technical field, the high-order being related in bearing system dynamics, signal processing
Statistic analysis, Nonlinear Coupling mechanism and the feature extraction problem of Weak fault detection, and in particular to a kind of secondary phase
Bearing incipient fault detection method of the position coupling with improving bispectrum algorithm.
Background technology
Rolling bearing is known as the joint of mechanical equipment, as a kind of dynamic component to concern, is widely used in various industry
In the rotating machinery in field.Bearing fault is also failure relatively common in rotating machinery, according to incompletely statistics, rotating machinery event
The 30% of barrier is as caused by bearing fault.Specifically in motor application field, almost half electrical fault is derived from bearing event
Barrier.Bearing once breaks down, it would be possible to trigger chain reaction, whole system is caused to be shut down, so as to cause huge warp
Ji loss and severe social influence.Therefore, the initial failure (Single Point of Faliure) for detecting bearing accurately and in time has important meaning
Justice.
For a long time, diversified bearing condition monitoring and method for diagnosing faults have been developed, can have been used according to it
Measurement type and classified, measured parameter generally comprises electric current, vibration, temperature and displacement signal.Commonplace
Fault Diagnosis of Roller Bearings is using the vibration signal of sensor acquisition bearing, by various signal processing methods at it
Reason realizes the extraction of fault signature, and then carries out fault identification.The most commonly used detection method is that envelope detected method is (also known as common
Shake demodulation method), the carrier wave that characterization system frequency can be extracted and the signal impact ingredient for playing modulating action, foundation
It is：Continue stand under load operating, its working surface of shaft each rotation when bearing parts abrasion, spot corrosion, deformation occur when failures
One or more impact signals can be generated, in periodical free damping, extract impact signal with regard to that can differentiate the former of the parts
Barrier and type.But this method is limited in that this method needs to determine demodulation frequency band parameters according to historical experience, and subjective
Factor can largely effect on the stability of analysis result, especially even more so when failure is in early stage.Also have using classical bispectrum
Method of estimation detects bearing fault, but can generate and false differentiate result.The most methods developed at present are for bearing event
Barrier is more effective when more apparent.
And early stage bearing fault occurs, on the one hand since load fluctuation causes the vibration signal obtained to be mostly non-stationary
Signal, conventional method of analysis are difficult to effectively analyze its fault signature；On the other hand since noisy external environment and electromagnetism are done
Disturbing etc. causes the signal obtained to contain more noise and interference components, and signal-to-noise ratio is relatively low, and conventional analysis method is difficult to send out in time
The failure symptom of existing rolling bearing.Furthermore the local damage class failure of inside and outside erosion of punctuating about accounts for bearing fault sum again
90%, vibration signal often shows typical non-linear, non-stationary characteristic, and conventional method is more difficult to carry out effective failure spy
Sign extraction.
The content of the invention
For the above-mentioned problems in the prior art, calculated the present invention provides a kind of square phase-couple with improving bispectrum
The bearing incipient fault detection method of method, coupling of this method between failure behavior and system dynamic behaviour, knot
Advanced signal processing theory and Research on Methods rolling bearing fault features detection method are closed, target is exactly to believe by vibration
Number, by considering the nonlinear coupling effects of failure impact and being recognized using improved higher order statistic analysis still in early stage
The bearing fault in stage.
The technical problems to be solved by the invention are：How tune bearing fault impact and system frequency between is explained
Similar comparativity in behavior processed between single order modulation ingredient and square phase-couple phenomenon；How by resampling technique to original
Beginning vibration signal carries out angularly resampling enhances carrying for subsequent fault signature ingredient to obtain resampling signal sequence
It takes；For resampling signal sequence, bi-spectrum estimation algorithm how is improved to eliminate false identification result.
For this purpose, present invention employs following technical schemes：
A kind of square phase-couple and the bearing incipient fault detection method for improving bispectrum algorithm, comprise the following steps：
Step 1：Obtain resampling signal sequence；
Step 2：The improvement bi-spectrum model counterweight sampled signal sequence detected by square phase-couple is calculated；
Step 3：Analysis is detected to bearing early stage minor defect characteristic component.
Further, it is believed that the interaction of bearing fault characteristics frequency and system resonance frequency is made with square phase-couple
Judge it is based on the fact that or phenomenon as a result, making this with of equal value or similar：
(1) single point defects of bearing start from the local defect on raceway or roller, when roller is by these defect faces, go out
Existing small collision simultaneously generates mechanical shock wave；These shock waves will produce the intrinsic frequency comprising the mechanical system including bearing
Raw incentive action；
(2) this incentive action is considered mechanical resonance frequency (carrier wave) by bearing fault characteristics frequency (baseband signal) institute
Modulation；
(3) the relatively other higher orders of single order side frequency in this modulation phenomenon are the result is that more significant；Inspection can such as be demodulated
Measure this single order modulation ingredient, it is meant that detect the early-stage weak fault feature of bearing；
(4) frequency is set as Fb, phase φbComponent and frequency be FC, phase φcComponent interaction (amplitude tune
AM processed), in FC+b、φc-bAnd FC-b、φc+bPlace generates two new band components, herein FbIt is equivalent to the fault signature of bearing
Frequency, FCIt is equivalent to the resonant frequency of system；Because the phase association of two band components is in FCAnd FbPhase (i.e. phase and
Phase difference), this interaction is exactly square phase-couple in fact；Two in this single order modulation result and high order equilibrium
Secondary phase coupling estimation phenomenon is equated, then the method that high order equilibrium may be employed examines bearing early-stage weak fault feature
It surveys.
Further, the detailed process of the step 1 is as follows：
(1) in terms of vibration acceleration on the timing sampling parameter synchronization acquisition bearing engine base set on signal and bearing axle
Tach signal sequence；
(2) the fluctuation situation according to rotating speed is designed based on the resampling program angularly changed；
(3) according to resampling program, resampling is carried out to bearing original vibration signal sequence；
(4) the resampling signal sequence for obtaining bearing axle turn over operation is arranged；
(5) obtain directly perceived sample frequency and sampling length of the resampling vibration signal sequence under its mean speed meaning,
Sample the parameters such as revolution.
Further, the bispectrum formula and its corresponding two-phase that the single order modulation ingredient based on resampling signal is made improvements
Dry formula is
D(f1,f2)=E { X (f2+f1)X(f2-f1)X*(f2)X*(f2)}
Wherein：Resampling signal is { x (i), i=1,2 ..., N }, and N attaches most importance to the length of sampled signal sequence, and X () is
Fourier is converted, and E () is statistical expection operator, and * represents complex conjugate.
Further, the detailed process of the step 2 is as follows：
(1) counterweight sampled signal sequence { x (i), i=1,2 ..., N } carries out trending and average is gone to pre-process, then
Calculate the Fourier conversion X () of resampling signal sequence { x (i), i=1,2 ..., N }；
(2) carrier frequency (system resonance frequency) approximate range is estimated according to X () amplitude distribution, further according to carrier frequency
(system resonance frequency) scope and bearing fault characteristics frequency determine the frequency separation that bispectrum calculates, i.e. f1、f2Interval range；
(3) resampling signal sequence { x (i), i=1,2 ..., N } is uniformly divided into K section (to be covered corresponding to resampling
The integral multiple of bearing axle turn over number), each intersegmental data is allowed to have certain overlapping；
(4) according to formula estimation bispectrum amplitude and its two-phase dry values；
(5) bispectrum amplitude and its two-phase dry values three-dimensional system of battle formations are drawn.
Further, the detailed process of the step 3 is as follows：
(1) according to the bispectrum amplitude of acquisition and its bicoherence three-dimensional system of battle formations, search bearing fault characteristics frequency is with resonating frequently
The amplitude at coordinate points that rate is formed whether there is peak value；
(2) if there is no peak value, the state for illustrating bearing is healthy, and detection process terminates；If there is peak value, after
It is continuous to be analyzed；
(3) if steadily there are amplitude, the type of bearing fault can be detected according to characteristic frequency, it is former including outer shroud
Barrier, inner ring failure or roller failure.
Compared with prior art, the beneficial effects of the invention are as follows：
(1) there is reliable and stable detectability to the detection of bearing early-stage weak fault, can find bearing fault in time,
Avoid the generation of bearing accident.
(2) it is applied widely, it can be applied to incipient fault detection and the tracking of all kinds of transmission system middle (center) bearings.
Description of the drawings
Fig. 1 is a kind of bearing incipient fault detection side of the square phase-couple provided by the present invention with improving bispectrum algorithm
The flow chart of method.
Fig. 2 is frequency relevant with bearing fault characteristics and geometric sense.
Fig. 3 is the improvement bispectrum three-dimensional width for the 6205 deep groove ball bearing inner ring early defects that the embodiment of the present invention one is provided
It is worth spectrogram.
Fig. 4 is the improvement bispectrum contour for the 6205 deep groove ball bearing inner ring early defects that the embodiment of the present invention one is provided
Scheme (10 layers have coordinate mark).
Fig. 5 is the improvement bispectrum contour for the 6205 deep groove ball bearing inner ring early defects that the embodiment of the present invention one is provided
Figure (10 layers mark without coordinate).
Fig. 6 is the improvement bispectrum for the 6205 deep groove ball bearing inner ring early defects that the embodiment of the present invention one is provided in f1=
3492.1875Hz the section spectrogram at place.
Fig. 7 is the improvement bispectrum three-dimensional width for the 6205 deep groove ball bearing outer shroud early defects that the embodiment of the present invention two is provided
It is worth spectrogram.
Fig. 8 is the improvement bispectrum contour for the 6205 deep groove ball bearing outer shroud early defects that the embodiment of the present invention two is provided
Scheme (10 layers have coordinate mark).
Fig. 9 is the improvement bispectrum contour for the 6205 deep groove ball bearing outer shroud early defects that the embodiment of the present invention two is provided
Figure (10 layers mark without coordinate).
Figure 10 is the improvement bispectrum for the 6205 deep groove ball bearing outer shroud early defects that the embodiment of the present invention two is provided in f1
Section spectrogram at=3357.4219Hz.
Figure 11 is the improvement bispectrum three-dimensional amplitude spectrum for the MB ER-10K bearing roller defects that the embodiment of the present invention three is provided
Figure.
Figure 12 is the improvement bispectrum contour map for the MB ER-10K bearing roller defects that the embodiment of the present invention three is provided
(10 layers have coordinate mark).
Figure 13 is the improvement bispectrum contour map for the MB ER-10K bearing roller defects that the embodiment of the present invention three is provided
(10 layers mark without coordinate).
Figure 14 is the improvement bispectrum for the MB ER-10K bearing roller defects that the embodiment of the present invention three is provided in f1=
2412.5Hz the section spectrogram at place.
Specific embodiment
Below in conjunction with the accompanying drawings and specific embodiment come the present invention will be described in detail, specific embodiment therein and explanation only
For explaining the present invention, but it is not as a limitation of the invention.
As shown in Figure 1, the invention discloses a kind of square phase-couples and the bearing initial failure inspection for improving bispectrum algorithm
Survey method, comprises the following steps：
Step 1：Obtain resampling signal sequence；
Step 2：The improvement bi-spectrum model counterweight sampled signal sequence detected by square phase-couple is calculated；
Step 3：Analysis is detected to bearing early stage minor defect characteristic component.
Here it is considered that the interaction of bearing fault characteristics frequency and system resonance frequency makees apparatus with square phase-couple
It is of equal value or similar as a result, it is based on the fact that or phenomenon to make this judgement：
(1) single point defects of bearing start from the local defect on raceway or roller, when roller is by these defect faces, go out
Existing small collision simultaneously generates mechanical shock wave；These shock waves will produce the intrinsic frequency comprising the mechanical system including bearing
Raw incentive action；
(2) this incentive action is considered mechanical resonance frequency (carrier wave) by bearing fault characteristics frequency (baseband signal) institute
Modulation；
(3) the relatively other higher orders of single order side frequency in this modulation phenomenon are the result is that more significant；Inspection can such as be demodulated
Measure this single order modulation ingredient, it is meant that detect the early-stage weak fault feature of bearing；
(4) frequency is set as Fb, phase φbComponent and frequency be FC, phase φcComponent interaction (amplitude tune
AM processed), in FC+b、φc-bAnd FC-b、φc+bPlace generates two new band components, herein FbIt is equivalent to the fault signature of bearing
Frequency, FCIt is equivalent to the resonant frequency of system；Because the phase association of two band components is in FCAnd FbPhase (i.e. phase and
Phase difference), this interaction is exactly square phase-couple in fact；Two in this single order modulation result and high order equilibrium
Secondary phase coupling estimation phenomenon is equated, then the method that high order equilibrium may be employed examines bearing early-stage weak fault feature
It surveys.
Single point defects usually show the contact surface of bearing in the form of spot corrosion or peeling.This kind of defect will be according to bearing table
The fault type that face is included can generate a kind of ingredient in four bearing fault characteristics frequency contents in vibration signal.This
A little characteristic frequencies are as shown in Figure 2.Bearing fault characteristics frequency calculation formula is accordingly：
Wherein：FRFor rotor (axis) frequency；FCFFor retainer failure-frequency；FIRFFor inner ring failure-frequency；FORFFor outer shroud
Failure-frequency；FBFFor roller failure-frequency；DBFor roller diameter；DPFor bearing pitch diameter；NBFor roller number；FREApply for roller
The direction of power on outer raceway；θ is roller contact angle.
The detailed process of the step 1 is as follows：
(1) in terms of vibration acceleration on the timing sampling parameter synchronization acquisition bearing engine base set on signal and bearing axle
Tach signal sequence；
(2) the fluctuation situation according to rotating speed is designed based on the resampling program angularly changed；
(3) according to resampling program, resampling is carried out to bearing original vibration signal sequence；
(4) the resampling signal sequence for obtaining bearing axle turn over operation is arranged；
(5) obtain directly perceived sample frequency and sampling length of the resampling vibration signal sequence under its mean speed meaning,
Sample the parameters such as revolution.
The single order bispectrum formula that is made improvements of modulation ingredient and its corresponding bicoherence formula based on resampling signal are
D(f1,f2)=E { X (f2+f1)X(f2-f1)X*(f2)X*(f2)}
Wherein：Resampling signal is { x (i), i=1,2 ..., N }, and N attaches most importance to the length of sampled signal sequence, and X () is
Fourier is converted, and E () is statistical expection operator, and * represents complex conjugate.
The detailed process of the step 2 is as follows：
(1) counterweight sampled signal sequence { x (i), i=1,2 ..., N } carries out trending and average is gone to pre-process, then
Calculate the Fourier conversion X () of resampling signal sequence { x (i), i=1,2 ..., N }；
(2) carrier frequency (system resonance frequency) approximate range is estimated according to X () amplitude distribution, further according to carrier frequency
(system resonance frequency) scope and bearing fault characteristics frequency determine the frequency separation that bispectrum calculates, i.e. f1、f2Interval range；
(3) resampling signal sequence { x (i), i=1,2 ..., N } is uniformly divided into K section (to be covered corresponding to resampling
The integral multiple of bearing axle turn over number), each intersegmental data is allowed to have certain overlapping；
(4) according to formula estimation bispectrum amplitude and its two-phase dry values；
(5) bispectrum amplitude and its two-phase dry values three-dimensional system of battle formations are drawn.
The detailed process of the step 3 is as follows：
(1) according to the bispectrum amplitude of acquisition and its bicoherence three-dimensional system of battle formations, search bearing fault characteristics frequency is with resonating frequently
The amplitude at coordinate points that rate is formed whether there is peak value；
(2) if there is no peak value, the state for illustrating bearing is healthy, and detection process terminates；If there is peak value, after
It is continuous to be analyzed；
(3) if steadily there are amplitude, the type of bearing fault can be detected according to characteristic frequency, it is former including outer shroud
Barrier, inner ring failure or roller failure.
Embodiment one：6205-2RS JEM SKF deep groove ball bearing inner ring incipient fault detections
Bearing structure parameter：Roller diameter DB=7.94mm；Bearing pitch diameter DP=39.04mm；Roller number NB=9.Bearing
Inner ring list defects with diameters 0.1778mm, defect depth 0.2794mm.
Bearing running environment：Motor load 2HP, rotating speed 1748r/min (29.1333Hz).
Data acquisition parameters：The sample frequency F of rotating speed and vibration acceleration meter signal on bearing blocks=12000Hz, acquisition
Length N=122136.
According to these parameters and the structural parameters of bearing, the bearing inner ring defect characteristic frequency of theoretical calculation is FIRF=
157.7628Hz, in the improvement bi-spectrum model algorithm of square phase-couple detection, it is 4096 to take fft analysis data length, frequency
Rate resolution ratio is df=Fs/ 4096=12000/4096=2.9297Hz, therefore visible bearing inner ring defect is special in actual analysis
Sign frequency is FRIRF=floor (FIRF/ df) × df=158.2031Hz.
First according to the FFT amplitude spectrums of this group of data determine the carrier frequency i.e. approximate range of resonant frequency for 3000Hz~
4000Hz, i.e. f1Section is [3000,4000].It is F according to bearing inner ring defect characteristic frequencyIRF=157.7628Hz, takes f2Area
Between for [0,400], cover twice of bearing inner ring defect characteristic frequency.Then, bispectrum is calculated by improved bispectrum algorithm
Estimation and bicoherence estimation.
The improvement bispectrum amplitude three-dimensional spectrum of acquisition is as shown in Figure 3.From reference axis f2Direction is seen, it can be clearly seen that bearing
Amplitude components at inner ring defect characteristic frequency (158.2031Hz) and twice of bearing inner ring defect characteristic frequency (316.4063Hz).
Further, Fig. 4 (having coordinate mark), the contour map that Fig. 5 (no coordinate mark) is Fig. 3 results, from reference axis f2Direction is more
It is apparent from bearing inner ring defect characteristic frequency (158.2031Hz) and twice of bearing inner ring defect characteristic frequency
Amplitude components at (316.4063Hz).From reference axis f1Also find out that the frequency interval between each component is exactly equal in bearing in direction
Ring defect characteristic frequency (158.2031Hz) is found here by calculating：3492.1875-3333.9844=158.2031
3650.3906-3492.1875=158.2031.
Fig. 6 is Fig. 3 results in f1Section spectrogram at=3492.1875Hz, bearing inner ring defect characteristic frequency
(158.2Hz) component is notable, and two-phase dry values are 0.9001, further demonstrate this conspicuousness, such 6205 deep-groove ball axis
Inner ring early defect is held just to have obtained steadily detecting.
Embodiment two：6205-2RS JEM SKF deep groove ball bearing outer shroud incipient fault detections
Bearing structure parameter：Roller diameter DB=7.94mm；Bearing pitch diameter DP=39.04mm；Roller number NB=9.Bearing
Outer shroud single point defects diameter about 0.1778mm, the deep about 0.2794mm of defect.
Bearing running environment：Motor load 2HP, rotating speed 1750r/min (29.1667Hz).
Data acquisition parameters：The sample frequency F of rotating speed and vibration acceleration meter signal on bearing blocks=12000Hz, acquisition
Length N=122136.
According to these parameters and the structural parameters of bearing, the outer race defect characteristic frequency of theoretical calculation is FORF=
104.5567Hz, in the improvement bi-spectrum model algorithm of square phase-couple detection, it is 4096 to take fft analysis data length, frequency
Rate resolution ratio is df=Fs/ 4096=12000/4096=2.9297Hz, therefore visible outer race defect is special in actual analysis
Sign frequency is FRORF=floor (FORF/ df) × df=105.4688Hz.
First according to the FFT amplitude spectrums of this group of data determine the carrier frequency i.e. approximate range of resonant frequency for 3000Hz~
4000Hz, i.e. f1Section is [3000,4000].It is F according to outer race defect characteristic frequencyIRF=104.5567Hz, takes f2Area
Between for [0,400], cover the outer race defect characteristic frequency of three times.Then, bispectrum is calculated by improved bispectrum algorithm
Estimation and bicoherence estimation.
The improvement bispectrum amplitude three-dimensional spectrum of acquisition is as shown in Figure 7.From reference axis f2Direction is seen, it can be clearly seen that bearing
Outer shroud defect characteristic frequency (105.4688Hz) and twice/three times outer race defect characteristic frequency (210.9376Hz/
316.4046Hz) at amplitude components.Further, Fig. 8 (having coordinate mark), Fig. 9 (no coordinate mark) are the contour of Fig. 7 results
Line chart, from reference axis f2Significantly more find out outer race defect characteristic frequency (105.4688Hz) and twice/three times axis in direction
Hold amplitude components at outer shroud defect characteristic frequency (210.9376Hz/316.4046Hz).From reference axis f1Also each point is found out in direction
Frequency interval between amount is exactly equal to outer race defect characteristic frequency (105.4688Hz), is found here by calculating：
3357.4219-3251.9531=105.4688 3462.8906-3357.4219=105.4687,3568.3594-
3462.8906=105.4688.
Figure 10 is Fig. 7 results in f1Section spectrogram at=3357.4219Hz, outer race defect characteristic frequency
(158.2Hz) component is notable, and two-phase dry values are 0.8406, further demonstrate this conspicuousness, such 6205 deep-groove ball axis
Outer shroud early defect is held just to have obtained steadily detecting.
Embodiment three：MB ER-10K bearing roller fault detects
Bearing structure parameter：Roller diameter DB=7.9248mm；Bearing pitch diameter DP=33.4772mm；Roller number NB=8.
Bearing running environment：Rotating speed 1807r/min (30.12Hz).
Data acquisition parameters：The sample frequency F of rotating speed and vibration acceleration meter signal on bearing blocks=25600Hz, acquisition
Length N=472000.
According to these parameters and the structural parameters of bearing, the bearing roller defect characteristic frequency of theoretical calculation is FBF=
119.9981Hz, in the improvement bi-spectrum model algorithm of square phase-couple detection, it is 4096 to take fft analysis data length, frequency
Rate resolution ratio is df=Fs/ 4096=25600/4096=6.25Hz, frequency resolution is bigger than normal, but visible axis in actual analysis
Bearing roller defect characteristic frequency is FRBF=floor (FBF/ df) × df=118.75Hz, relatively calculated value, frequency are divided
Resolution is bigger than normal to influence less testing result.
First according to the FFT amplitude spectrums of this group of data determine the carrier frequency i.e. approximate range of resonant frequency for 2000Hz~
3000Hz, i.e. f1Section is [2000,3000].It is F according to bearing roller defect characteristic frequencyBF=119.9981Hz, takes f2Area
Between for [0,400], cover the bearing roller defect characteristic frequency of three times.Then, bispectrum is calculated by improved bispectrum algorithm
Estimation and bicoherence estimation.
The improvement bispectrum amplitude three-dimensional spectrum of acquisition is as shown in figure 11.From reference axis f2Direction is seen, it can be clearly seen that axis
Bearing roller defect characteristic frequency (118.75Hz) and twice/three times bearing roller defect characteristic frequency (237.50Hz/
356.25Hz) at amplitude components.Further, Figure 12 (having coordinate mark), Figure 13 (no coordinate mark) for Figure 11 results etc.
High line chart, from reference axis f2Significantly more find out bearing roller defect characteristic frequency (118.75Hz) and twice/three times axis in direction
Amplitude components at bearing roller defect characteristic frequency (237.50Hz/356.25Hz).From reference axis f1Direction also find out each component it
Between frequency interval be exactly equal to bearing roller defect characteristic frequency (118.75Hz), here pass through calculate find：2412.5-
2293.75=118.75 2531.25-2412.5=118.75,2650-2531.25=118.75.
Figure 14 is Figure 11 results in f1Section spectrogram at=2412.5Hz, bearing roller defect characteristic frequency
(118.75Hz) component is notable, and also more significantly, such MB ER-10K bearing roller defects must for twice/three times frequency component
To steadily detecting.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
Any modification, equivalent substitution and improvement made within refreshing and spirit etc., should be included in protection scope of the present invention
Within.
Claims (6)
1. a kind of square phase-couple and the bearing incipient fault detection method for improving bispectrum algorithm, it is characterised in that：Including with
Lower step：
Step 1：Obtain resampling signal sequence；
Step 2：The improvement bi-spectrum model counterweight sampled signal sequence detected by square phase-couple is calculated；
Step 3：Analysis is detected to bearing early stage minor defect characteristic component.
2. a kind of square phase-couple according to claim 1 and the bearing incipient fault detection side for improving bispectrum algorithm
Method, it is characterised in that：Think that the interaction of bearing fault characteristics frequency and system resonance frequency is acted on square phase-couple
With of equal value or similar as a result, it is based on the fact that or phenomenon to make this judgement：
(1) single point defects of bearing start from the local defect on raceway or roller, when roller is by these defect faces, occur micro-
Small collision simultaneously generates mechanical shock wave；These shock waves, which will generate the intrinsic frequency comprising the mechanical system including bearing, to swash
The effect of encouraging；
(2) this incentive action is considered that mechanical resonance frequency (carrier wave) is adjusted by bearing fault characteristics frequency (baseband signal)
System；
(3) the relatively other higher orders of single order side frequency in this modulation phenomenon are the result is that more significant；It can such as demodulate and detect
This single order modulates ingredient, it is meant that detects the early-stage weak fault feature of bearing；
(4) frequency is set as Fb, phase φbComponent and frequency be FC, phase φcComponent interaction (amplitude modulation
AM), in FC+b、φc-bAnd FC-b、φc+bPlace generates two new band components, herein FbIt is equivalent to the fault signature frequency of bearing
Rate, FCIt is equivalent to the resonant frequency of system；Because the phase association of two band components is in FCAnd FbPhase (i.e. phase and phase
Potential difference), this interaction is exactly square phase-couple in fact；This single order modulation result with it is secondary in high order equilibrium
Phase coupling estimation phenomenon is equated, then the method that high order equilibrium may be employed examines bearing early-stage weak fault feature
It surveys.
3. a kind of square phase-couple according to claim 1 or 2 and the bearing incipient fault detection for improving bispectrum algorithm
Method, it is characterised in that：The detailed process of the step 1 is as follows：
(1) turned in terms of vibration acceleration on the timing sampling parameter synchronization acquisition bearing engine base set on signal and bearing axle
Fast signal sequence；
(2) the fluctuation situation according to rotating speed is designed based on the resampling program angularly changed；
(3) according to resampling program, resampling is carried out to bearing original vibration signal sequence；
(4) the resampling signal sequence for obtaining bearing axle turn over operation is arranged；
(5) directly perceived sample frequency and sampling length, sampling of the resampling vibration signal sequence under its mean speed meaning are obtained
The parameters such as revolution.
4. a kind of square phase-couple according to claim 3 and the bearing incipient fault detection side for improving bispectrum algorithm
Method, it is characterised in that：The bispectrum formula and its corresponding bicoherence that single order modulation ingredient based on resampling signal is made improvements
Formula is
D(f1,f2)=E { X (f2+f1)X(f2-f1)X*(f2)X*(f2)}
<mrow>
<mi>d</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mn>1</mn>
</msub>
<mo>,</mo>
<msub>
<mi>f</mi>
<mn>2</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<mo>|</mo>
<mi>D</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mn>1</mn>
</msub>
<mo>,</mo>
<msub>
<mi>f</mi>
<mn>2</mn>
</msub>
<mo>)</mo>
</mrow>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
</mrow>
<mrow>
<mi>E</mi>
<mo>{</mo>
<mo>|</mo>
<mi>X</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mn>2</mn>
</msub>
<mo>)</mo>
</mrow>
<mi>X</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mn>2</mn>
</msub>
<mo>)</mo>
</mrow>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
<mo>}</mo>
<mi>E</mi>
<mo>{</mo>
<mo>|</mo>
<mi>X</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mn>2</mn>
</msub>
<mo>+</mo>
<msub>
<mi>f</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<mi>X</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>f</mi>
<mn>2</mn>
</msub>
<mo>-</mo>
<msub>
<mi>f</mi>
<mn>1</mn>
</msub>
<mo>)</mo>
</mrow>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
<mo>}</mo>
</mrow>
</mfrac>
</mrow>
Wherein：Resampling signal is { x (i), i=1,2 ..., N }, and N attaches most importance to the length of sampled signal sequence, and X () is
Fourier is converted, and E () is statistical expection operator, and * represents complex conjugate.
5. a kind of square phase-couple according to claim 4 and the bearing incipient fault detection side for improving bispectrum algorithm
Method, it is characterised in that：The detailed process of the step 2 is as follows：
(1) counterweight sampled signal sequence { x (i), i=1,2 ..., N } carries out trending and average is gone to pre-process, and then calculates
The Fourier conversion X () of resampling signal sequence { x (i), i=1,2 ..., N }；
(2) carrier frequency (system resonance frequency) approximate range is estimated according to X () amplitude distribution, (is further according to carrier frequency
Altogether vibration frequency) scope and bearing fault characteristics frequency determine the frequency separation that bispectrum calculates, i.e. f1、f2Interval range；
(3) resampling signal sequence { x (i), i=1,2 ..., N } is uniformly divided into the K section (bearings covered corresponding to resampling
The integral multiple of axis turn over number), each intersegmental data is allowed to have certain overlapping；
(4) according to formula estimation bispectrum amplitude and its two-phase dry values；
(5) bispectrum amplitude and its two-phase dry values three-dimensional system of battle formations are drawn.
6. a kind of square phase-couple according to claim 5 and the bearing incipient fault detection side for improving bispectrum algorithm
Method, it is characterised in that：The detailed process of the step 3 is as follows：
(1) according to the bispectrum amplitude of acquisition and its bicoherence three-dimensional system of battle formations, bearing fault characteristics frequency and resonant frequency shape are searched for
Into coordinate points at amplitude whether there is peak value；
(2) if there is no peak value, the state for illustrating bearing is healthy, and detection process terminates；If there is peak value, continue into
Row analysis；
(3) if steadily there are amplitude, the type of bearing fault can be detected according to characteristic frequency, including outer shroud failure,
Inner ring failure or roller failure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711375557.3A CN108120598B (en) | 2017-12-19 | 2017-12-19 | Square phase-couple and the bearing incipient fault detection method for improving bispectrum algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711375557.3A CN108120598B (en) | 2017-12-19 | 2017-12-19 | Square phase-couple and the bearing incipient fault detection method for improving bispectrum algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108120598A true CN108120598A (en) | 2018-06-05 |
| CN108120598B CN108120598B (en) | 2019-09-13 |
Family
ID=62230273
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711375557.3A Expired - Fee Related CN108120598B (en) | 2017-12-19 | 2017-12-19 | Square phase-couple and the bearing incipient fault detection method for improving bispectrum algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108120598B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109029999A (en) * | 2018-09-19 | 2018-12-18 | 河北工业大学 | Fault Diagnosis of Roller Bearings based on enhanced modulation double-spectrum analysis |
| CN111207926A (en) * | 2019-12-27 | 2020-05-29 | 三明学院 | Fault diagnosis method based on rolling bearing, electronic device and storage medium |
| CN111259765A (en) * | 2020-01-13 | 2020-06-09 | 北京工业大学 | Order analysis method based on numerical control machine tool spindle current signal |
| CN111610364A (en) * | 2020-06-02 | 2020-09-01 | 江苏方天电力技术有限公司 | Forced oscillation mode correlation analysis method based on bispectrum |
| CN111975784A (en) * | 2020-09-03 | 2020-11-24 | 昆明理工大学 | Joint robot fault diagnosis method based on current and vibration signals |
| CN113588297A (en) * | 2021-07-28 | 2021-11-02 | 北京航天发射技术研究所 | Hydraulic steering engine fault detection method and device based on self-adaptive noise elimination |
| WO2023017226A1 (en) * | 2021-08-11 | 2023-02-16 | Safran | Title: method for detecting a bearing fault in a rotating system and monitoring system implementing this method |
| CN113588297B (en) * | 2021-07-28 | 2024-04-30 | 北京航天发射技术研究所 | Hydraulic steering engine fault detection method and device based on self-adaptive noise removal |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102798529A (en) * | 2012-08-30 | 2012-11-28 | 桂林电子科技大学 | Method and system for diagnosing bearing faults of large-size wind turbine bearing |
| CN104596766A (en) * | 2014-12-24 | 2015-05-06 | 中国船舶工业系统工程研究院 | Early fault determining method for bearing |
| CN105676169A (en) * | 2016-01-19 | 2016-06-15 | 中国电子科技集团公司第三研究所 | Orientation method and device for impulsive sound object |
| CN106772573A (en) * | 2016-11-16 | 2017-05-31 | 电子科技大学 | Seismic wavelet method for extracting signal based on maximal correlation entropy |
-
2017
- 2017-12-19 CN CN201711375557.3A patent/CN108120598B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102798529A (en) * | 2012-08-30 | 2012-11-28 | 桂林电子科技大学 | Method and system for diagnosing bearing faults of large-size wind turbine bearing |
| CN104596766A (en) * | 2014-12-24 | 2015-05-06 | 中国船舶工业系统工程研究院 | Early fault determining method for bearing |
| CN104596766B (en) * | 2014-12-24 | 2017-02-22 | 中国船舶工业系统工程研究院 | Early fault determining method and device for bearing |
| CN105676169A (en) * | 2016-01-19 | 2016-06-15 | 中国电子科技集团公司第三研究所 | Orientation method and device for impulsive sound object |
| CN106772573A (en) * | 2016-11-16 | 2017-05-31 | 电子科技大学 | Seismic wavelet method for extracting signal based on maximal correlation entropy |
Non-Patent Citations (1)
| Title |
|---|
| 周宇: "基于循环平稳信号二维平面表示的滚动轴承早期故障诊断方法研究", 《中国博士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109029999A (en) * | 2018-09-19 | 2018-12-18 | 河北工业大学 | Fault Diagnosis of Roller Bearings based on enhanced modulation double-spectrum analysis |
| WO2020057066A1 (en) * | 2018-09-19 | 2020-03-26 | 河北工业大学 | Fault diagnosis method employing enhanced modulation signal bispectrum analysis for rolling bearing |
| CN111207926A (en) * | 2019-12-27 | 2020-05-29 | 三明学院 | Fault diagnosis method based on rolling bearing, electronic device and storage medium |
| CN111207926B (en) * | 2019-12-27 | 2022-02-01 | 三明学院 | Fault diagnosis method based on rolling bearing, electronic device and storage medium |
| CN111259765A (en) * | 2020-01-13 | 2020-06-09 | 北京工业大学 | Order analysis method based on numerical control machine tool spindle current signal |
| CN111259765B (en) * | 2020-01-13 | 2024-04-16 | 北京工业大学 | Order analysis method based on numerical control machine tool spindle current signal |
| CN111610364A (en) * | 2020-06-02 | 2020-09-01 | 江苏方天电力技术有限公司 | Forced oscillation mode correlation analysis method based on bispectrum |
| CN111975784A (en) * | 2020-09-03 | 2020-11-24 | 昆明理工大学 | Joint robot fault diagnosis method based on current and vibration signals |
| CN113588297A (en) * | 2021-07-28 | 2021-11-02 | 北京航天发射技术研究所 | Hydraulic steering engine fault detection method and device based on self-adaptive noise elimination |
| CN113588297B (en) * | 2021-07-28 | 2024-04-30 | 北京航天发射技术研究所 | Hydraulic steering engine fault detection method and device based on self-adaptive noise removal |
| WO2023017226A1 (en) * | 2021-08-11 | 2023-02-16 | Safran | Title: method for detecting a bearing fault in a rotating system and monitoring system implementing this method |
| FR3126154A1 (en) * | 2021-08-11 | 2023-02-17 | Safran | Method for detecting a bearing fault of a rotating system and monitoring system implementing this method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108120598B (en) | 2019-09-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108120598B (en) | Square phase-couple and the bearing incipient fault detection method for improving bispectrum algorithm | |
| US11519821B2 (en) | Detecting faults in wind turbines | |
| Hemmati et al. | Roller bearing acoustic signature extraction by wavelet packet transform, applications in fault detection and size estimation | |
| Zhao et al. | Compound faults detection of rolling element bearing based on the generalized demodulation algorithm under time-varying rotational speed | |
| Renaudin et al. | Natural roller bearing fault detection by angular measurement of true instantaneous angular speed | |
| Singh et al. | Motor current signature analysis for bearing fault detection in mechanical systems | |
| Wang et al. | Rotating speed isolation and its application to rolling element bearing fault diagnosis under large speed variation conditions | |
| Wang et al. | An online tacholess order tracking technique based on generalized demodulation for rolling bearing fault detection | |
| Yan et al. | Multi-scale enveloping spectrogram for vibration analysis in bearing defect diagnosis | |
| CN102834701B (en) | Method and apparatus for diagnosing sliding bearing | |
| US4408294A (en) | Method for on-line detection of incipient cracks in turbine-generator rotors | |
| US7602985B2 (en) | Multi-scale enveloping spectrogram signal processing for condition monitoring and the like | |
| Yan et al. | Harmonic wavelet-based data filtering for enhanced machine defect identification | |
| EP0087813B1 (en) | Method and apparatus for monitoring cracks of a rotatable body | |
| Yan et al. | Energy-based feature extraction for defect diagnosis in rotary machines | |
| Zhao et al. | Bearing multi-fault diagnosis with iterative generalized demodulation guided by enhanced rotational frequency matching under time-varying speed conditions | |
| Saidi et al. | The use of SESK as a trend parameter for localized bearing fault diagnosis in induction machines | |
| CN111397877B (en) | Rotary machine beat vibration fault detection and diagnosis method | |
| CN110763462A (en) | Time-varying vibration signal fault diagnosis method based on synchronous compression operator | |
| JPH09113416A (en) | Method for diagnosing damage of rolling bearing | |
| Li et al. | Cyclic statistics in rolling bearing diagnosis | |
| JPH1026580A (en) | Method and device for diagnosing speed-change-type rotary mechanical equipment | |
| Yan et al. | Fault signature enhancement and skidding evaluation of rolling bearing based on estimating the phase of the impulse envelope signal | |
| Zhao et al. | Vibration health monitoring of rolling bearings under variable speed conditions by novel demodulation technique | |
| Ugwiri et al. | Vibrations measurement and current signatures for fault detection in asynchronous motor |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190913 Termination date: 20191219 |