CN103728123B - Rotary machine fault characteristic extraction method based on Kalman order domain-holographic spectrum principle - Google Patents
Rotary machine fault characteristic extraction method based on Kalman order domain-holographic spectrum principle Download PDFInfo
- Publication number
- CN103728123B CN103728123B CN201310471141.7A CN201310471141A CN103728123B CN 103728123 B CN103728123 B CN 103728123B CN 201310471141 A CN201310471141 A CN 201310471141A CN 103728123 B CN103728123 B CN 103728123B
- Authority
- CN
- China
- Prior art keywords
- holographic
- kalman
- vibration signal
- rank
- analysis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001228 spectrum Methods 0.000 title claims abstract description 18
- 238000000605 extraction Methods 0.000 title claims abstract description 10
- 238000004458 analytical method Methods 0.000 claims abstract description 32
- 230000001360 synchronised Effects 0.000 claims description 5
- 239000000969 carrier Substances 0.000 claims description 4
- 238000000034 method Methods 0.000 abstract description 12
- 238000010586 diagram Methods 0.000 abstract description 9
- 238000003745 diagnosis Methods 0.000 abstract description 5
- 230000004927 fusion Effects 0.000 abstract description 2
- 238000001093 holography Methods 0.000 abstract description 2
- 230000001351 cycling Effects 0.000 description 17
- 238000005070 sampling Methods 0.000 description 9
- 230000003595 spectral Effects 0.000 description 7
- 239000000284 extract Substances 0.000 description 6
- 230000000875 corresponding Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000003032 molecular docking Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000002194 synthesizing Effects 0.000 description 2
- 238000010977 unit operation Methods 0.000 description 2
- 206010011376 Crepitations Diseases 0.000 description 1
- 208000008425 Protein Deficiency Diseases 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000005039 chemical industry Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000005272 metallurgy Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 230000000630 rising Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000001052 transient Effects 0.000 description 1
Abstract
The invention discloses a kind of rotary machine fault characteristic extraction method based on Kalman rank ratio holographic spectrum principle, the ability that can the most reliably reflect set state is composed in the ability and the holography that fully utilize Kalman order ratio analysis process non-stationary signal.Kalman order ratio analysis and holographic analysis of spectrum are carried out fusion two-dimension holographic spectrum can either be utilized when the fluctuation of speed to carry out fault diagnosis, easily a stop sign can be carried out again power and couple is decomposed, holographic Waterfall plot analysis, the present invention greatly extends the operation strategies of conventional photographic spectrum, achieve and a stop sign is carried out various features parameter extraction, breach conventional situation and can only carry out the limitation of Bode diagram analysis, expand level and the scope of rotating machinery fault monitoring, diagnosing, reliably run to provide for rotating machinery health and provide powerful support for.
Description
Technical field
The invention belongs to mechanical fault diagnosis field, be specifically related to a kind of based on Kalman order domain-holographic spectrum
The rotary machine fault characteristic extraction method of principle.
Background technology
Rotating machinery is the class machinery extensively used in modern industry, at aviation, electric power, oil, chemical industry etc.
Industry is used widely, and is mostly the critical component of production equipment, once has an accident, it will cause
Economic loss greatly.Therefore, the condition monitoring and fault diagnosis of rotating machinery is extremely important.
Under stabilized (steady-state) speed, holographic spectral technology is utilized can the most reliably to reflect that the amplitude of unit, frequency, phase place are believed
Breath, can more precisely identify hidden danger present in unit operation than conventional method in actual production.Holographic
Spectral technology achieves the accurately identification of homogeneity fault by information fusion and separates, from the operation of rotor overall space
Angle is more beneficial for holding the running status that unit is overall, and the holography grown up based on holographic spectral technology is moved
Balance method effect emulation, the fine setting of counterweight amount and aspect such as counterweight scheme optimization before balance achieve breakthrough.
Correlation technique has obtained promotion and application widely in industries such as electric power, petrochemical industry, metallurgy, its effectiveness and
Reliability is checked in substantial amounts of practical engineering application and is approved.
The vibration signal playing docking process is equivalent to the dynamic response under a broadband excitation, and ratio is under stabilized (steady-state) speed
Vibration signal contains more rich information.Make full use of the vibration information of docking process, to playing docking process
Do Bode diagram, power and couple decomposition, holographic Waterfall plot etc. to analyze, it will be appreciated that unit dynamic characteristic, for excavating
Unit incipient fault provides reliable foundation with realizing unit fault Precise Diagnosis, to holding set state, prediction
Unit operation trend and prevent major accident from suffering from important effect.
But, holographic spectral technology there is also some shortcomings: Main Analysis object is the steady-state signal of fixed rotating speed,
It is difficult to the non-stationary signal of the fluctuation of speed and Up and down speeding is analyzed so that it is range of application is the most limited
System.
Summary of the invention
It is an object of the invention to overcome existing holographic spectrum diagnostic techniques can only analyze the stationary signal under stabilized (steady-state) speed
Deficiency, it is provided that a kind of rotary machine fault characteristic extraction method based on Kalman order domain-holographic spectrum principle,
Kalman order ratio analysis is merged by the method with holographic spectrum principle, has the non-stationary letter having analyzed docking process
Number ability.
For reaching above-mentioned purpose, the technical solution adopted in the present invention comprises the following steps:
1) vibration signal and the key signal of the constant duration of synchronous acquisition are obtained；
2) use key signal that vibration signal is carried out key phase processor；
3) to step 2) in vibration signal after key phase processor carry out Kalman order ratio analysis, obtain complex envelope
And carrier wave；
4) complex envelope by obtaining in step 3) tries to achieve the amplitude under each rotating speed and phase place；
5) utilize the amplitude obtained in step 4) and phase place to draw holographic Waterfall plot, and carry out fault distinguishing.
Vibration signal in described step 1) includes that the two-way to each measuring surface synchronous acquisition is orthogonal and shakes
Dynamic signal.
In described step 3), complex envelope and carrier product are reached than subscale in rank by Kalman order ratio analysis
Form.
Described vibration signal has been stop sign, after order ratio analysis, draws power and power for playing stop sign
Even exploded view and holographic Waterfall plot.:
Compared with prior art, the method have the advantages that
Rotary machine fault characteristic extraction method based on Kalman order domain-holographic spectrum principle of the present invention, be
According to order ratio analysis, there is the ability of process non-stationary signal and holographic composing can the most reliably reflect shaking of unit
The emotionally feature of condition, carries out effective integration by rank ratio method and holographic spectral theory, extends the fortune of conventional photographic spectrum
By scope, the method has been applicable to parking procedure fault feature extraction.
Accompanying drawing explanation
Fig. 1 is that the embodiment of the present invention simulates cycling start time domain beamformer；
Fig. 2 is the schematic diagram that embodiment of the present invention key drifts about mutually；
Fig. 3 is embodiment of the present invention speed diagram mutually drift-corrected to key, and wherein rising speed is 10000 revs/min；
Fig. 4 be the embodiment of the present invention to the speed diagram after key phase drift correction, wherein rise speed be 10000 turns/
Minute；
Fig. 5 is the flow chart of embodiment of the present invention Kalman order ratio analysis；
Fig. 6 is 1 order component after the embodiment of the present invention a certain passage cycling start vibration signal Kalman order ratio analysis
Time domain beamformer；
Fig. 7 is the Bode diagram of 1 order component of embodiment of the present invention rotor left cross section X passage；
Fig. 8 is the Bode diagram of 1 order component of embodiment of the present invention rotor left cross section Y passage；
Fig. 9 is that embodiment of the present invention rotor start/stop machine fault signature extracts flow chart；
Figure 10 is that the cycling start rank of embodiment of the present invention rotor unbalance fault are than holographic Waterfall plot；
Figure 11 is that the cycling start rank of embodiment of the present invention rotor misalignment fault are than holographic Waterfall plot；
Figure 12 is that the cycling start rank of embodiment of the present invention Rub-impact fault are than holographic Waterfall plot；
Figure 13 is that the cycling start rank of embodiment of the present invention rotor crack fault are than holographic Waterfall plot.
Detailed description of the invention
Below in conjunction with the accompanying drawings and embodiment the invention will be further described.
The present invention first pass through key signal calibrated after calculate each revolution of mean speed, then pass through in every turn
Linear interpolation calculates the tachometer value corresponding to each sampled point of vibration signal.Utilize the tach signal sequence calculated
The vibration signal of each vibrating channel is carried out Kalman order ratio analysis, extracts the rank on 1 to 4 rank than component,
Can obtain each component amplitude under corresponding rotating speed and phase place by the complex envelope of each order component, recycling is complete
Breath spectrum principle draws out the rank during cycling start than holographic Waterfall plot.Can be observed intuitively by holographic Waterfall plot
The situation of change of various characteristic parameters during cycling start, in order to carry out fault distinguishing.
Specifically comprise the following steps that
(1) according to the holographic spectral method requirement to data signal sampling and processing, to each measuring surface synchronous acquisition
The vibration signal of two orthogonal direction of vibration；
(2) error caused to make up key to drift about mutually, utilizes number to adopt the enumerator merit that card tachometer passage provides
Key can be corrected programming mutually, after correction, calculated the tachometer value of corresponding each sampled point by linear interpolation；
(3) identical for Shi Ge road signal initial time, and initial time is that on rotor, key phase groove senses mutually with key
Device just to moment, need to utilize key signal that vibration signal carries out key phase pretreatment；
(4) utilize the rotating speed sequence calculated in (2), appropriate bandwidth is set, by the score of Kalman rank
1 to 4 order components of cycling start process vibrations signal are extracted by analysis method；
(5) amplitude and the phase of vibration under every speed can be calculated than the complex envelope part of component by each rank
Position, draws out Bode diagram；
(6) by the vibration width under the corresponding rotating speed of each order component of orthogonal two passages in a certain cross section
Value and phase place draw out holographic Waterfall plot.
Exist as it is shown in figure 1, the cycling start vibration signal of acquired original has noise, be not Striking symmetry,
Local also can be jagged, in the presence of faulty, also can there is the component of high order, so passing through order ratio analysis
Extract each order component to filter noise and extract corresponding order, be that a kind of effective of cycling start vibration signal is analyzed
Method.
During as in figure 2 it is shown, key signal to be carried out constant duration sampling, the signal width when forwarding at key phase groove
Value will occur an obvious saltus step, imply that rotor have rotated one week.Preferably way is every time critical
Point A samples, and thus has a fixing reference point, can guarantee that and is precisely one between two key signals
Complete cycle.But owing to being constant duration sampling, it is impossible to just sample at A point, give and calculate band every time
Bigger error.
Utilize tachometer channel acquisition key signal special for DT9837B.Key signal is carried out constant duration
During sampling, when forwarding at key phase groove, signal amplitude will occur an obvious saltus step, imply that rotor have rotated
One week.As shown in Figure 1, it is desirable to just sample at A point, thus there is a fixing reference every time
Point, can guarantee that and be precisely a complete cycle between two key signals.But owing to being constant duration sampling, usually exist
B point is just sampled, and has carried bigger error to calculating.Therefore when writing capture program, can be by programming
Realize the counting from A point to B point, then calculate the time from A point to B point.Its principle is as follows: when turning
During speed sensor alignment A point, the pulse voltage that sensor produces is more than 2.5V, and enumerator starts counting up, to B
Hour counter stopping timing of sampling is carried out during point.Therefore the time of A point can be by the time of B point and counting number
Revise,Wherein fcThe highest frequency acquisition 48MHz, N for capture card represent counting number.
As it is shown on figure 3, when liter speed is fixed (7500 revs/min), not being corrected key signal will
Produce the biggest error, it is impossible to ensure order ratio analysis and the precision of holographic analysis of spectrum.Such as: as rotor 8000rpm
Time, key assign in the error of generation of drifting about mutuallyThe time of one turn is 0.0075s, therefore
Its speed error reaches 500rpm, sees Fig. 4, and after key corrects mutually, precision is greatly improved, and meets analysis demand.
Order ratio analysis method generally comprises three kinds: hardware rank ratio, calculating rank ratio and modern rank ratio.The present invention uses
Kalman rank ratio in the ratio of modern rank, it is possible to separate approximation rank ratio and the rank ratio that intersects, and pass through complex envelope very
Easily obtaining vibration amplitude and phase place, Fig. 5 is the flow chart of Kalman order ratio analysis.
As shown in Figure 6,1 order component pair that cycling start process vibrations signal extracts after Kalman order ratio analysis
Claim and the most smooth, filtered unnecessary noise.
As shown in FIG. 7 and 8, extract after Kalman order ratio analysis according to two direction of vibration signals of X, Y
The complex envelope of 1 order component, vibration amplitude and phase place under each rotating speed can be obtained very easily, draw out Byrd
Figure.
As it is shown in figure 9, after Kalman order ratio analysis, obtain horizontal and vertical in the same measuring surface of rotor
The complex envelope of vibration signal each order component in two directions, is obtained under each frequency by the complex envelope of each order
Vibration amplitude phase place, then according to respective frequencies carries out synthesizing and being arranged on a spectrogram, i.e. obtains two dimension rank
Than holographic Waterfall plot, fault signature can be summarized than holographic Waterfall plot by two dimension rank, be used for carrying out event
Barrier identifies.
Figure 10 to Figure 13 for simulate respectively on rotor imbalance, misalign, touch mill and four kinds of faults of crackle after
The rank that the cycling start vibration signal gathered is drawn after Kalman order ratio analysis are than holographic Waterfall plot.
Figure 10 is that rotor unbalance cycling start rank are than holographic Waterfall plot.Can be seen that leap is faced under imbalance fault
Before and after boundary's rotating speed, 1 rank just phase point has overturn about 180 °, and whole ellipse has also overturn 180 °, violent to electrical path length
Diminish after first becoming greatly.And after critical point, slowly, first phase point is relatively stable in oval change, oval eccentricity
The least.Additionally it is also seen that the principal vibration of rotor is 1 rank power frequency from figure, other harmonics vibrate
The least.
Figure 11 is that rotor misalignment cycling start rank are than holographic Waterfall plot.Can be seen that under misaligning fault, 1,2
Order component is bigger；1 rank oval eccentric rate is bigger；2 rank magnitude of vibrations are obvious with rotation speed change；2 rank are faced 1/2
One bigger oval appearance 2 subcritical resonance in rank occurs near boundary's rotating speed.
Figure 12 is that Rub-impact cycling start rank are than holographic Waterfall plot.Can be seen that under Rubbing faults,
During 1 oval eccentric Shuai Hen Datong District, rank, 1 rank ellipse incline direction is the most consistent；2,3 order components are the least, and 2
Rank occur the bigger oval appearance 2 subcritical resonance in rank near 1/2 critical speed；Special
Levy is that 1 rank backward whirl occur.
Figure 13 is that the cycling start rank of cracked rotor are than holographic Waterfall plot.Can be seen that, under crack fault, 1 rank are being faced
Near boundary's rotating speed, just phase point occurs 180 degree of deflections similar with imbalance, but ellipse upset in critical zone is slower；2、3
Order component is the biggest, and amplitude is inconspicuous with rotation speed change；2 rank occur one near 1/2 critical speed
Bigger ellipse, there is larger vibration on 3 rank near 1/3 critical speed, i.e. occur in that 2, the 3 subcritical resonance in rank
Phenomenon.
The ability and the holographic spectral synthesis that present invention utilizes order ratio analysis process non-stationary signal reliably reflect unit
Both are carried out merging and can solve the unstable signals such as stop sign and cannot be carried out traditional complete by the ability of state
The weakness of breath analysis of spectrum, can be broken away from the average effect of FFT, preferably retain by Kalman order ratio analysis
Transient state characteristic, the present invention greatly extends the operation strategies of conventional photographic spectrum, it is achieved that enter playing stop sign
Row feature extraction, breaches tradition stop sign and can only carry out the limitation of Bode diagram analysis, expanded whirler
The level of tool malfunction monitoring diagnosis and scope, reliably run offer for rotating machinery health and provide powerful support for.
Claims (2)
1. a rotary machine fault characteristic extraction method based on Kalman order domain-holographic spectrum principle, its feature
It is, comprises the following steps:
1) vibration signal and the key signal of the constant duration of synchronous acquisition are obtained；
Vibration signal includes the orthogonal vibration signal of two-way to each measuring surface synchronous acquisition；
2) use key signal that vibration signal is carried out key phase processor；
3) to step 2) in vibration signal after key phase processor carry out Kalman order ratio analysis, obtain complex envelope
And carrier wave；
Kalman order ratio analysis is the form that rank are reached complex envelope and carrier product than subscale；
4) by step 3) in the complex envelope that obtains try to achieve the amplitude under each rotating speed and phase place；
5) utilize step 4) in the amplitude that obtains and phase place draw holographic Waterfall plot, and carry out fault distinguishing.
A kind of rotating machinery fault based on Kalman order domain-holographic spectrum principle is special
Levy extracting method, it is characterised in that: described vibration signal has been stop sign, after order ratio analysis, for
Play stop sign and draw power and couple exploded view and holographic Waterfall plot.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310471141.7A CN103728123B (en) | 2013-10-10 | Rotary machine fault characteristic extraction method based on Kalman order domain-holographic spectrum principle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310471141.7A CN103728123B (en) | 2013-10-10 | Rotary machine fault characteristic extraction method based on Kalman order domain-holographic spectrum principle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103728123A CN103728123A (en) | 2014-04-16 |
| CN103728123B true CN103728123B (en) | 2016-11-30 |
Family
ID=
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102998110A (en) * | 2012-11-29 | 2013-03-27 | 西安交通大学 | Rotary machine fault characteristic extraction method based on order-holospectrum principle |
| CN103335785A (en) * | 2013-06-24 | 2013-10-02 | 西安交通大学 | Unsteady dynamic balance method based on order domain-holographic spectrum principle |
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102998110A (en) * | 2012-11-29 | 2013-03-27 | 西安交通大学 | Rotary machine fault characteristic extraction method based on order-holospectrum principle |
| CN103335785A (en) * | 2013-06-24 | 2013-10-02 | 西安交通大学 | Unsteady dynamic balance method based on order domain-holographic spectrum principle |
Non-Patent Citations (2)
| Title |
|---|
| 基于阶比跟踪的全息谱技术;魏玉果 等;《重庆大学学报(自然科学版)》;20070228;第30卷(第2期);第9-12、21页 * |
| 旋转机械阶比跟踪技术研究进展综述;赵晓平 等;《地震工程与工程振动》;20081231;第29卷(第6期);第213-219页 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102998110B (en) | Rotary machine fault characteristic extraction method based on order-holospectrum principle | |
| Abboud et al. | Angle⧹ time cyclostationarity for the analysis of rolling element bearing vibrations | |
| Chow et al. | Induction machine fault diagnostic analysis with wavelet technique | |
| CN103323274B (en) | Condition monitoring for rotating machinery and fault diagnosis system and method | |
| CN102967414B (en) | Method for extracting imbalanced components of micro-speed-difference double-rotor system based on frequency spectrum correction | |
| Meng et al. | General synchroextracting chirplet transform: Application to the rotor rub-impact fault diagnosis | |
| CN104034412A (en) | Rotary machine fault feature extraction method based on fractional order holographic principle | |
| CN103575523A (en) | Rotating machine fault diagnosis method based on Fast ICA-spectrum kurtosis-envelope spectrum analysis | |
| CN102087138A (en) | System and method for vibration analysis and phase analysis of vibration waveforms using dynamic statistical averaging of tachometer data to accurately calculate rotational speed | |
| CN101907089A (en) | Fault diagnosis method of compressor shafting based on three-dimensional space axle center orbit | |
| CN106759538A (en) | A kind of low strain integrity testing method based on frequency-domain analysis | |
| Zhang et al. | On bearing fault diagnosis by nonlinear system resonance | |
| CN102721462A (en) | Method for quickly computing Bode plot and Nyquist plot of rotary mechanical vehicle starting and parking processes | |
| CN106340889A (en) | Oscillation discrimination and disturbance source positioning method based on oscillation-starting analysis | |
| CN109459131A (en) | A kind of the time-frequency characteristics extracting method and device of rotating machinery multi-channel Vibration Signal | |
| CN103335785A (en) | Unsteady dynamic balance method based on order domain-holographic spectrum principle | |
| CN111307426A (en) | Rotating machinery fault feature extraction method based on FrFT-EWT principle | |
| Ding et al. | Multiple instantaneous frequency ridge based integration strategy for bearing fault diagnosis under variable speed operations | |
| CN106441840A (en) | Rotary machine pseudo-subsynchronous fault holographic diagnosis method based on model | |
| Yang et al. | Rub-impact detection of rotor systems using time-frequency techniques | |
| Song et al. | Multispectral balanced automatic fault diagnosis for rolling bearings under variable speed conditions | |
| CN103728123B (en) | Rotary machine fault characteristic extraction method based on Kalman order domain-holographic spectrum principle | |
| Zhen et al. | Phase-compensation-based dynamic time warping for fault diagnosis using the motor current signal | |
| CN103674221A (en) | Method for detecting, analyzing and screening vibration signals of motor of static-pressure liquid-floated gyro-accelerometer | |
| CN203349773U (en) | Testing device for radial revolution error and inclination angle revolution error of main shaft of precise centrifuge |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |