CN101762746B - Cyclostationary demodulation method of complex gear transmission chain current signal based on entropy optimization - Google Patents
Cyclostationary demodulation method of complex gear transmission chain current signal based on entropy optimization Download PDFInfo
- Publication number
- CN101762746B CN101762746B CN2009102545721A CN200910254572A CN101762746B CN 101762746 B CN101762746 B CN 101762746B CN 2009102545721 A CN2009102545721 A CN 2009102545721A CN 200910254572 A CN200910254572 A CN 200910254572A CN 101762746 B CN101762746 B CN 101762746B
- Authority
- CN
- China
- Prior art keywords
- entropy
- frequency
- gear
- spectrogram
- characteristic
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to a cyclostationary demodulation method of a complex gear transmission chain current signal based on entropy optimization. The method comprises the following steps: collecting the current signal of a drive motor in a gear transmission chain, utilizing Fourier transform to obtain the cyclostationary frequency spectrogram of the current signal; calculating the entropy value of slice spectrogram of the cyclostationary frequency spectrogram under cycle frequency and drawing the curve of the entropy value; determining an entropy minimum value point in the curve of the entropy value and the characteristic slice cycle frequency of the entropy minimum value point; finding a frequency corresponding to the characteristic slice cycle frequency in the cyclostationary frequency spectrogram, determining a characteristic slice frequency spectrum; adopting frequency spectrum correction method to perform error correction to the difference of the characteristic slice cycle frequency, correcting the characteristic slice frequency spectrum to obtain an accurate frequency amplitude; and obtaining the characteristic information of fault frequency according to the accurate frequency amplitude. The demodulation method of the invention demodulates and separates the current characteristic signal of the drive motor, extracts the characteristic information of the motor fault and performs fault monitoring and diagnosis to the running state of the gear transmission system.
Description
Technical field
The invention belongs to equipment condition monitoring and diagnostic techniques field, relate to a kind of method that is used for gear drive running status malfunction monitoring and diagnosis, be specifically related to a kind of cyclostationary demodulation method of complex gear transmission chain current signal based on entropy optimization.
Background technology
Development along with maximization, serialization commercial unit, monitoring, diagnosing to gear train is had higher requirement, not only to improve the quality of diagnosis of simple gears system, and various labyrinth gear trains have also been proposed the diagnostic test demand, as the bale panoramic tables gear train of the blast-furnace distributor furnace roof gear case of the speed-changing gear box of heavy machine tool, novel high speed compressor, steel industry, steelworks and ladle tipping table reduction gear box etc.The work under bad environment of these labyrinth gear trains, and complex structure, transmission chain length, make that the installation of sensor is very difficult, even be difficult to realize (as blast-furnace distributor furnace roof gear case), cause being difficult to carry out based on the Gear Fault Diagnosis technology of method for oscillating.And existing non-contact detection method is subjected to the influence of field working conditions bigger, therefore, the diagnosis of above-mentioned labyrinth gear train has been become the blind spot that traditional diagnosis method is difficult to analyzing and processing.
As gear train, gear case is turned round certificate by the motor-driven transmission, usually adopt the stator current signal analytic approach that electrical fault is diagnosed, this method is based on the electromagnetic coupled between motor stator and rotor, the failure message of rotor and stator is reflected in the stator current, pass through the spectrum analysis of motor current signal is realized the diagnosis of motor stator, rotor and fault of eccentricity again, even can discern the state of motor rolling bearing.Therefore, the gear defects that exists in the gear train that is dragged by motor also can exert an influence to the operation of rotor, promptly cause the variation of electromagnetic coupled between electric machine rotor, thereby in stator current signal, and current signal has noncontact and the remote characteristics that detect with the gear distress message reflection.Compare with other non-metering systems with vibration, stator current signal detects to solving labyrinth gear train fault diagnosis a kind of possible means is provided.
Transmission rule according to gear train, the representation of athletic of gearbox output shaft is the vibration modulation characteristic, and pass to rotor, at stator, under the magnetic-coupled effect of rotor airgap, modulation signal shows as further modulated in stator current, effect by dynamo-electric magnetic forms repeatedly modulation like this, and under the influence of motor harmonic frequency, make gear train vibration source information show as the complex modulation phenomenon in current signal, obviously traditional Hilbert demodulation method can not effectively extract the complex frequency composition of reflection gear train state from the motor stator current signal.In order to solve the monitoring, diagnosing difficult problem of labyrinth gear train, press for the complicated demodulation method of new gear vibration.
Summary of the invention
In order to overcome above-mentioned problems of the prior art, the purpose of this invention is to provide a kind of cyclostationary demodulation method of complex gear transmission chain current signal based on entropy optimization, can from detected motor stator current signal, effectively extract the complex frequency of the labyrinth gear train generation that drags by this motor, realize monitoring, diagnosing this labyrinth gear train.
The technical solution adopted in the present invention is, cyclostationary demodulation method of complex gear transmission chain current signal based on entropy optimization, stator current signal based on drive motor in the gear transmission chain, utilize the tonality of separating of circulation autocorrelation function, cycle frequency by the entropy optimization correspondence is handled, the section demodulation of autocorrelation function circulates, obtain comprising the demodulation spectra of this cycle frequency composition and frequency multiplication thereof, and proofread and correct in the basic enterprising line frequency spectrum of this demodulation spectra, obtain the accurate diagnostic result of complicated gear system, this demodulation method carries out according to the following steps:
Step 1: adopt current sensor to gather the current signal of drive motor in the gear transmission chain, utilize Fourier transform to obtain the cyclo-stationary spectrogram of this current signal;
Step 2: under cycle frequency, the cut into slices entropy S (P) of spectrogram of calculation procedure 1 gained cyclo-stationary spectrogram, this entropy S (P) calculates by following formula:
If P is (p
1, p
2..., p
n) be a uncertain probability distribution, k is arbitrary constant, then this probability distribution P (p
1, p
2..., p
n) entropy S (P) in the scope is:
In the formula, p
iBe the probability that i information state occurs ,-lnp
iBe the quantity of information that i information state brings when occurring, S (P) is an entropy, has characterized the size of the quantity of information of i information state;
Step 3: draw the entropy curve according to the entropy that step 2 calculates;
Step 4: utilize sort method, the entropy curve that step 3 is drawn sorts, and determines the entropy minimum point in the entropy curve, then, obtains the characteristic slice cycle frequency S (P of this entropy minimum point by the entropy curve of step 3
s):
S(P
s)=min[S(P
i)]
S (P wherein
s) be the entropy minimal value, P
sBe the entropy minimum point, S (P
i) be the entropy at each frequency place;
Step 5: find the corresponding corresponding frequencies of characteristic slice cycle frequency that obtains with step 4 in the cyclo-stationary spectrogram of step 1, then the slice map of this corresponding frequencies is the characteristic slice frequency spectrum;
Step 6: adopt spectrum correction method, pass through Theoretical Calculation, the difference of the characteristic slice cycle frequency that step 4 is obtained is carried out error correction, and according to the characteristic slice cycle frequency after the correction error characteristic slice frequency spectrum in the step 5 is proofreaied and correct, and obtains accurate frequency amplitude;
Step 7:, obtain the characteristic information of failure-frequency according to the accurate frequency amplitude that step 6 obtains.
Demodulation method of the present invention has following advantage:
1. accurately hold the running status and the fault of gear train based on current of electric information, can eliminate the blind spot of gear train status monitoring and fault diagnosis.
2. adopt the associated extraction thinking of specific period signal, for the characteristic component that solves complex modulated signal provides feasible means and method, at the demodulation difficult problem of complex modulation, employing can be carried out the cyclo-stationary method of demodulation analysis to many groups amplitude modulated signal of complexity.
3. based on optimum entropy principle, can obtain each cycle frequency in the three-dimensional circulation spectrogram easily, when having solved use cyclo-stationary method current signal being handled, the signal content complicacy increases (many as frequency content, distribution range is bigger etc.), its three-dimensional spectrum complicated problems that becomes.
4. utilize spectrum correcting method that the circulation spectrum of optimizing is circulated and compose the correction at peak.
Description of drawings
Fig. 1 is the structural representation of blast furnace material distribution chute gear train assembly in the embodiment of the invention.
Fig. 2 is the driving current signal spectrogram of chute electric rotating machine in the gear train assembly of gathering shown in Figure 1.
Fig. 3 is the partial enlarged drawing of 50Hz place frequency spectrum among Fig. 2.
Fig. 4 is with the amplitude spectrum enlarged drawing after near 10 times of the refinements 50Hz.
Fig. 5 is the entropy curve map of the chute electric rotating machine current signal 11Hz place frequency band of gear train assembly shown in Figure 1.
Fig. 6 is the section spectrogram at 11.1Hz frequency place.
Fig. 7 is a 29.6Hz frequency place section spectrogram
Among Fig. 1,1. first gear, 2. second gear, 3. the 3rd gear, 4. gear ring, 5. the 4th gear, 6. the 5th gear, 7. first worm gear, 8. worm screw, 9. the 6th gear, 10. the 7th gear, 11. octadentates wheel, 12. the 9th gears, 13. the tenth gear, 14. the 11 gears, 15. the 12 gears, 16. the 13 gears, 17. turn-screw, 18. second worm gears, 19. the 14 gears, 20. the 15 gears, 21. the chute electric rotating machine, 22. chute tilting motor, 23. distributor chutes, 24. chute center lines.
Embodiment
The present invention is described in detail below in conjunction with the drawings and specific embodiments.
Demodulation method of the present invention is based on the stator current signal of drive motor in the gear transmission chain, utilize the tonality of separating of circulation autocorrelation function, cycle frequency by the entropy optimization correspondence is handled, the section demodulation of autocorrelation function circulates, obtain comprising the demodulation spectra of this cycle frequency composition and frequency multiplication thereof, and proofread and correct in the basic enterprising line frequency spectrum of this demodulation spectra, obtain the accurate diagnostic result of complicated gear system.
Stator current signal to the complex gear transmission chain drive motor, adopt the cyclo-stationary analysis to carry out demodulation, because the modulation product in the current signal is faint, the order of accuarcy of cycle frequency has determined the demodulation effect, therefore, this demodulation method is located cycle frequency rapidly and accurately based on the optimisation strategy of maximum entropy, realizes the demodulation of complex modulation current signal.
Demodulation method of the present invention, carry out according to the following steps:
Step 1: adopt current sensor to gather the current signal of drive motor in the gear transmission chain, utilize Fourier (Fourier) conversion to obtain the cyclo-stationary spectrogram of this current signal; Advantage is that the circulation autocorrelation function of signal is done Fourier (Fourier) conversion, and the spectral density function circle statistics method that obtains circulating spectrum has correlation properties, makes it have the ability that suppresses noise and undesired signal;
Step 2: under cycle frequency, the cut into slices entropy S (P) of spectrogram of calculation procedure 1 gained cyclo-stationary spectrogram, this entropy S (P) calculates by following formula:
If P is (p
1, p
2..., p
n) be a uncertain probability distribution, k is arbitrary constant, then this probability distribution P (p
1, p
2..., p
n) entropy S (P) in the scope is:
In the formula, p
iBe the probability that i information state occurs ,-lnp
iBe the quantity of information that i information state brings when occurring, S (P) is an entropy, has characterized the size of the quantity of information of i information state;
Under specifying constraint, can be from all possible compatible distribution, the probability distribution when picking out the entropy maximum, the principle of maximum entropy PME (Principle of Maximum Entropy) that Here it is proposes.Least the probability distribution of Que Dinging (equiprobability distribution) has maximum entropy, and simultaneously, the approaching more this equiprobability of probability distribution distributes, its entropy is also just big more, and conversely, probability distribution departs from equiprobability more and distributes, information entropy is just more little, and the size of information entropy has reflected the homogeneity of probability distribution.Spectral line is sparse more in spectrogram, illustrates that signal energy concentrates on the minority spectral line, like this, can utilize information entropy that energy distribution uniformity in the spectrogram is portrayed.
Step 3: utilize the entropy under each cycle frequency that step 2 calculates, and draw the entropy curve according to the entropy under this each cycle frequency;
During cyclo-stationary is analyzed, when characteristic frequency during as the section frequency, the spectral line concentration of energy on characteristic component, and when non-characteristic frequency when cutting into slices frequency, the spectral line energy approximation is in even distribution.According to principle of maximum entropy, can obtain the entropy in the pairing two-dimentional spectrogram of each cycle frequency in the three-dimensional circulation spectrogram easily, the cyclo-stationary section spectrogram entropy of characteristic frequency and crossover frequency thereof is less, otherwise entropy is bigger.
Step 4: utilize sort method, the entropy curve that step 3 is drawn sorts, and determines the entropy minimum point in the entropy curve that step 3 draws, and then, obtains the characteristic slice cycle frequency S (P of this entropy minimum point by the entropy curve of step 3
s):
S(P
s)=min[S(P
i)]
S (P wherein
s) be the entropy minimal value, P
sBe the entropy minimum point, S (P
i) be the entropy at each frequency place;
In the entropy curve of circulation spectrogram, the mixing point in the entropy minimum point difference respective signal between characteristic frequency point and the characteristic frequency.Therefore, can determine the characteristic slice cycle frequency, to improve the search efficiency of characteristic frequency according to the entropy minimum point.
Step 5: in the cyclo-stationary spectrogram of step 1, find the corresponding corresponding frequencies of characteristic slice cycle frequency that obtains with step 4, the slice map of this corresponding frequencies is the characteristic slice frequency spectrum, the entropy minimal value Frequency point that step 4 is determined is optimum cycle frequency point
The cycle frequency of non-zero can be portrayed the cyclo-stationary of signal, and cycle frequency and frequency multiplication thereof constitute the circulation spectrum.What adopt when reality is used is that mode that cyclo-stationary spectrogram (three-dimensional spectrum) combines with two dimension slicing figure is extracted the characteristic frequency composition in the signal.
Step 6: adopt spectrum correction method, pass through Theoretical Calculation, the difference of the characteristic slice cycle frequency that step 4 is obtained is carried out error correction, and according to the characteristic slice cycle frequency after the correction error characteristic slice frequency spectrum in the step 5 is proofreaied and correct, and obtains accurate frequency amplitude;
For the discrete spectrum analysis of unifrequency harmonic signal, no matter be that frequency, amplitude or phase place generally all may exist error.The error of unifrequency harmonic signal is applicable to the current signal spectrum analysis.
Step 7:, obtain the characteristic information of failure-frequency according to the accurate frequency amplitude that step 6 obtains.
Practice case
Steelworks blast furnace material distribution system utilizes material clock furnace roof technology and equipment that materials flow is controlled.And adopt the method for control weight to control materials flow, replaced the method for always continuing to use that CONTROL VOLUME and time controls materials flow of passing through,, improved furnace top pressure improving blast furnace material distribution, make full use of the chemical energy and the heat energy of coal gas, and prolong the life of the blast furnace and played vital role.
The distributor chute of this blast furnace material distribution system needs the flexible motion cloth, makes this distributor chute drive by complicated gear train assembly.Therefore,, ensure the output and the quality of product, need monitor with initial failure the gear train assembly that drives the distributor chute motion in this cloth system and diagnose in order to use material clock furnace roof technology and equipment better.Because the blast furnace field condition is very complicated, operating mode is more abominable, can't on the blast furnace rotary gear box vibration transducer be installed, and causes the monitoring of inapplicable this blast furnace gear train assembly state of traditional vibration analysis method.
The current signal analytic approach has signal and obtains convenience, signal to noise ratio (S/N ratio) height and the high characteristics of information integration degree, be applied to the status monitoring of furnace roof complicated gear case, promptly adopt the current signal analytic approach that the furnace roof gear train assembly is monitored, the current signal of the motor of gathering, then, employing is based on the cyclo-stationary analytical approach of entropy optimization, and the current signal of the electric rotating machine gathered is carried out demodulation, the characteristic frequency of extraction stove apical tooth wheel driving-chain from this current signal.
As shown in Figure 1, the gear train of above-mentioned steel-making blast furnace distributor chute gear train assembly, comprise chute electric rotating machine 21 and chute tilting motor 22, chute electric rotating machine 21 drives first gear 1, first gear 1 reaches the 9th gear 12 by second gear 2, the 3rd gear 3, the 6th gear 9, the 7th gear 10 and octadentate wheel 11 successively with power and moment of torsion, the 9th gear 12 is given distributor chute 24 by turn-screw 17 with transmission of power, drives distributor chute 24 along 25 rotations of chute center line.
The motion of following distributor chute 24 kinematic trains is analyzed and characteristic frequency is calculated.
Chute electric rotating machine 21 with power with move through fixed shaft gear train and pass to the 9th gear 12, make distributor chute 24 rotations.Thereby, chute rotational speed n
RFor:
In the formula, n
RevolveBe the rotating speed of chute electric rotating machine, negative sign is represented n
RWith n
RevolveDirection of rotation, n
12The rotating speed of representing the 9th gear 12, Z
1, Z
2, Z
9, Z
10, Z
11And Z
12The number of teeth of representing first gear 1, second gear 2, the 6th gear 9, the 7th gear 10, octadentate wheel the 11 and the 9th gear 12 respectively.
Fascinating of distributor chute 24 acted on by chute electric rotating machine 21 and chute tilting motor 22 then simultaneously.Gear ring 4 rotations in the planetary mechanism that chute electric rotating machine 21 drivings the 5th gear 6, the 4th gear 5 and gear ring 4 constitute, chute tilting motor 22 reaches the 5th gear 6 in this planetary mechanism via worm screw 8 and first worm gear 7 with power, drive 13 rotations of the tenth gear, obtain following length velocity relation:
In the formula, n
6And n
7The expression and the rotating speed of first worm gear 7 respectively, Z
7And Z
8Be respectively the number of first worm gear, 7 numbers of teeth and worm screw 8, n
InclineRotating speed for chute tilting motor 22.
In the formula, n
4The rotating speed of expression gear ring 4, Z
1, Z
2, Z
3And Z
Outside 4Represent the number of teeth of first gear 1, the number of teeth of second gear 2, the number of teeth of the 3rd gear 3 and the number of teeth of gear ring 4 external tooths respectively.
Can obtain by wheel chain structure drive connection:
In the formula, n
13Be the rotating speed of the tenth gear 13, Z
6And Z
In 4Represent the number of teeth of the 5th gear 6 and the number of teeth of gear ring 4 internal tooths respectively.
In the formula, n
14Be the rotating speed of the 11 gear 14, Z
13And Z
14Represent the number of teeth of the tenth gear 13 and the number of teeth of the 11 gear 14 respectively.
The tenth gear 13 is passed to the 15 gear 20 by the 11 gear the 14, the 12 gear the 15, the 13 gear 16, second worm gear 18 and the 14 gear 19 successively with the power that chute tilting motor 22 transmits.The 9th gear the 12, the 11 gear 14 and the 12 gear 15 are formed one group of differential gear train, according to the differential gear train transmission principle, obtain the speed of the fascinating n of distributor chute 24
TFor:
Employing detects and analyzes based on the cyclo-stationary analytical approach of the entropy optimization state to above-mentioned distributor chute kinematic train:
A. the B phase current of chute electric rotating machine 21 power supplys is sampled, sample frequency is 2500Hz, the sampled signal that collects is carried out the amplitude spectrum analysis, obtain amplitude spectrogram as shown in Figure 2, and near signal 50Hz as shown in Figure 3, in the local amplification spectrogram, can judge roughly to have some modulation products; Yet each spectrum peak is superimposed fully, has serious leakage and interference effect, therefore, need at first adopt the frequency spectrum thinning method that adjacent component in the amplitude spectrogram is separated.
B. adopt multiple modulation refinement choosing band frequency spectrum analysis method based on complex analytic band-pass filter, to 10 times of amplitude spectrogram refinements shown in Figure 2, obtain after the refinement as shown in Figure 4 near the frequency band enlarged drawing 50Hz, compare with collection of illustrative plates shown in Figure 3, signal frequency component has obtained separation to a certain degree, so each modulation product also manifested, and this has also proved in Fig. 3, because leakage and spectral line are interfered serious, the modulation product that presents is inaccurate.
C. have multiple distributing mode owing to cloth system, corresponding different tilting motor rotating speeds, and be constantly to adjust, so cause many to the unknown of gear characteristic frequency.Fig. 4 is the multiple modulation refinement choosing band frequency spectrum analysis method that adopts based on complex analytic band-pass filter, to 10 times of amplitude spectrum refinements, obtains after the refinement near the frequency band enlarged drawing 50Hz.In Fig. 4, the apparent in view modulation product that corresponding gear is right is 18.5Hz, 11Hz and 25Hz.By formula (1)---(6) calculate meshing frequency as can be known, the 25Hz modulation product is corresponding to the gyrofrequency of first gear 1 that links to each other with chute electric rotating machine 21,18.5Hz modulation product is corresponding to the meshing frequency of the 9th gear 12 and octadentate wheel 11, the meshing frequency of 11Hz modulation product correspondence the tenth gear 13 and the 11 gear 14.
D. according to above analysis, frequency band 10.8~11.3Hz (according to 3 σ theorems, when step-length equals 0.1, establishing σ=0.1) does circulation spectrogram entropy curve map around precognition frequency range 11Hz frequency.The size of information entropy has reflected the homogeneity of probability distribution, and spectral line is sparse more in spectrogram, illustrates that signal energy concentrates on the minority spectral line, like this, can utilize information entropy that energy distribution uniformity in the spectrogram is portrayed.As shown in Figure 5, being presented at the 11.1Hz place among the figure has minimum entropy, and the less energy of the relative dominant frequency of 11.1Hz is the most concentrated, illustrates that like this this frequency is a modulating frequency.
E. adopt the cyclostationary characteristic isolation technics that the B phase current signal of chute electric rotating machine 21 power supplys of collection is analyzed, circulation spectrogram when two other obvious modulation product that can obtain above sampled signal is cycle frequency, have and the similar result of 11.1Hz section spectrogram (as Fig. 6), modulation product is very obvious.
F. according to the cyclo-stationary analysis principle, also should there be corresponding peaks to exist at the crossover frequency place of three main modulating frequencies, therefore do two-dimentional cyclo-stationary analysis respectively at 29.6Hz (11.1+18.5), 7.4Hz (18.5-11.1), 36Hz (11.1+24.9), 13.8Hz (24.9-11.1), 43.4Hz (18.5+24.9) and 6.4Hz (24.9-18.5) cycle frequency place, wherein the section spectrogram at 29.6Hz frequency place as shown in Figure 7, show among the figure that gained result and theoretical analysis result fit like a glove.
Therefore, adopt and fast and effeciently to isolate modulation source, in the circulation spectrogram, can extract modulation intelligence well based on the cyclo-stationary demodulation method of entropy optimization.
Demodulation method of the present invention can carry out the characteristics of demodulation analysis to many modulation signals of complexity based on the cyclo-stationary demodulation analysis, in conjunction with the entropy optimization technology, current characteristic signal to drive motor carries out the demodulation separation, and the characteristic information of extraction electrical fault, the gear train assembly running status is carried out malfunction monitoring and diagnosis, not only solved the implementation problem of complicated gear kinematic train status monitoring, also detected and precision keeps providing a kind of new, solution route efficiently for the kinematic train operation troubles.
Claims (1)
1. based on the cyclostationary demodulation method of complex gear transmission chain current signal of entropy optimization, stator current signal based on drive motor in the gear transmission chain, utilize the tonality of separating of circulation autocorrelation function, cycle frequency by the entropy optimization correspondence is handled, the section demodulation of autocorrelation function circulates, obtain comprising the demodulation spectra of this cycle frequency composition and frequency multiplication thereof, and proofread and correct in the basic enterprising line frequency spectrum of this demodulation spectra, obtain the accurate diagnostic result of complicated gear system, it is characterized in that this demodulation method carries out according to the following steps:
Step 1: adopt current sensor to gather the current signal of drive motor in the gear transmission chain, utilize Fourier transform to obtain the cyclo-stationary spectrogram of this current signal;
Step 2: under cycle frequency, the cut into slices entropy S (P) of spectrogram of calculation procedure 1 gained cyclo-stationary spectrogram, this entropy S (P) calculates by following formula:
If P is (p
1, p
2..., p
n) be a uncertain probability distribution, k is arbitrary constant, then this probability distribution P (p
1, p
2..., p
n) entropy S (P) in the scope is:
In the formula, p
iBe the probability that i information state occurs ,-lnp
iIt is the quantity of information that i information state brings when occurring, S (P) is an entropy, characterized the size of the quantity of information of i information state, the size of information entropy has reflected the homogeneity of probability distribution, spectral line is sparse more in spectrogram, illustrate that signal energy concentrates on the minority spectral line, utilize information entropy that energy distribution uniformity in the spectrogram is portrayed;
Step 3: draw the entropy curve according to the entropy that step 2 calculates;
Step 4: utilize sort method, the entropy curve that step 3 is drawn sorts, and determines the entropy minimum point in the entropy curve, then, obtains the characteristic slice cycle frequency S (P of this entropy minimum point by the entropy curve of step 3
S):
S(P
s)=min[S(P
i)]
S (P wherein
s) be the entropy minimal value, P
sBe the entropy minimum point, S (P
i) be the entropy at each frequency place;
Step 5: find the corresponding corresponding frequencies of characteristic slice cycle frequency that obtains with step 4 in the cyclo-stationary spectrogram of step 1, then the slice map of this corresponding frequencies is the characteristic slice frequency spectrum;
Step 6: adopt spectrum correction method, pass through Theoretical Calculation, the difference of the characteristic slice cycle frequency that step 4 is obtained is carried out error correction, and according to the characteristic slice cycle frequency after the correction error characteristic slice frequency spectrum in the step 5 is proofreaied and correct, and obtains accurate frequency amplitude;
Step 7:, obtain the characteristic information of failure-frequency according to the accurate frequency amplitude that step 6 obtains.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009102545721A CN101762746B (en) | 2009-12-29 | 2009-12-29 | Cyclostationary demodulation method of complex gear transmission chain current signal based on entropy optimization |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009102545721A CN101762746B (en) | 2009-12-29 | 2009-12-29 | Cyclostationary demodulation method of complex gear transmission chain current signal based on entropy optimization |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101762746A CN101762746A (en) | 2010-06-30 |
CN101762746B true CN101762746B (en) | 2011-09-14 |
Family
ID=42494035
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2009102545721A Expired - Fee Related CN101762746B (en) | 2009-12-29 | 2009-12-29 | Cyclostationary demodulation method of complex gear transmission chain current signal based on entropy optimization |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101762746B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105223403A (en) * | 2015-08-27 | 2016-01-06 | 南京南瑞太阳能科技有限公司 | The wavelet packet extracting method of a kind of combining inverter net side resonance current information |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102607829B (en) * | 2012-02-22 | 2014-01-29 | 西安交通大学 | Quantitative recognition method for assembling loosening fault of dismountable disk drum type rotor |
CN103412057B (en) * | 2013-07-19 | 2015-11-25 | 西安交通大学 | Based on the weak oscillation damping signal parameter identification of accidental resonance and Moving Least Squares |
JP6499621B2 (en) | 2016-08-08 | 2019-04-10 | ファナック株式会社 | Control device and control system |
DE112017005650B4 (en) * | 2016-12-15 | 2023-09-07 | Mitsubishi Electric Corporation | TRANSMISSION MECHANISM ANOMALY DIAGNOSTIC DEVICE AND TRANSMISSION MECHANISM ANOMALY DIAGNOSTIC PROCEDURE |
JP7309344B2 (en) * | 2018-11-13 | 2023-07-18 | 川崎重工業株式会社 | FAILURE DIAGNOSIS DEVICE AND FAILURE DIAGNOSIS METHOD OF DRIVE MECHANISM, AND MACHINE DEVICE INCLUDING SAME FAILURE DIAGNOSIS DEVICE |
CN111896256B (en) * | 2020-03-03 | 2022-03-29 | 天津职业技术师范大学(中国职业培训指导教师进修中心) | Bearing fault diagnosis method based on deep nuclear processing |
CN111471820A (en) * | 2020-03-31 | 2020-07-31 | 鞍钢股份有限公司 | Remote monitoring method for blast furnace top distributor |
CN115060959A (en) * | 2022-05-13 | 2022-09-16 | 重庆大学 | Rail transit vehicle wheel fault monitoring method |
-
2009
- 2009-12-29 CN CN2009102545721A patent/CN101762746B/en not_active Expired - Fee Related
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105223403A (en) * | 2015-08-27 | 2016-01-06 | 南京南瑞太阳能科技有限公司 | The wavelet packet extracting method of a kind of combining inverter net side resonance current information |
Also Published As
Publication number | Publication date |
---|---|
CN101762746A (en) | 2010-06-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101762746B (en) | Cyclostationary demodulation method of complex gear transmission chain current signal based on entropy optimization | |
EP2581724B1 (en) | A method and a system for the purpose of condition monitoring of gearboxes | |
Teng et al. | Multi-fault detection and failure analysis of wind turbine gearbox using complex wavelet transform | |
CN103104509B (en) | Obtaining method of variable frequency water pump full working condition operating state | |
CN110044610B (en) | Gear fault diagnosis method | |
EP2730898B1 (en) | Bearing, housing including a set of bearings, associated method and computer program | |
CN104198186A (en) | Method and device for diagnosing gear faults based on combination of wavelet packet and spectral kurtosis | |
CN105318961A (en) | Vibration-state on-line monitoring method of high-voltage motor driving conveying belt | |
CN106124197A (en) | A kind of epicyclic gearbox sun gear partial fault detection method and system | |
CN110333071B (en) | Mechanical vibration signal processing method using narrowband cepstrum transformation | |
CN102692449A (en) | Fault diagnosis method of blast furnace top gearbox through comprehensive analysis | |
CN105548595B (en) | A kind of extraction wind turbine gearbox axis Rotating speed measring methods at different levels | |
CN103048135A (en) | Multi-fault coupling experimenter of flexible rotor rolling bearing foundation system, and fault identification method | |
CN103940607B (en) | The epicyclic gearbox signal that does not rely on time domain average separates and diagnostic method | |
CN108844733A (en) | A kind of gear condition monitoring index extracting method based on KL divergence and root-mean-square value | |
CN102175439A (en) | Method for realizing order analysis of rotating machinery | |
US20220229113A1 (en) | Power Conversion Device, Rotating Machine System, and Diagnosis Method | |
CN105841792A (en) | Gear pressure angle direction local vibration signal acquisition method based on microsensor | |
CN104111391A (en) | Three-phase instantaneous power based motorized spindle fault monitoring and diagnosing method | |
CN101452276B (en) | On-line monitoring method for running state of gear box of cold rolling and pickling drawing-correcting machine | |
CN111577553A (en) | Intelligent state monitoring system for wind generating set | |
CN104215453B (en) | Fault detection method for primary planet secondary parallel shaft gearbox | |
CN208187704U (en) | The condition monitoring and fault diagnosis acquisition system of frequency modulation gearbox | |
CN201983916U (en) | Vibration analysis test device of automobile speed changer | |
CN102661787B (en) | Transform analysis method of rotor vibration signal of variable-rotation wind generation set |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
C17 | Cessation of patent right | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20110914 Termination date: 20131229 |