CN106096530B - A kind of Modal Parameters Identification under strong background noise environment - Google Patents
A kind of Modal Parameters Identification under strong background noise environment Download PDFInfo
- Publication number
- CN106096530B CN106096530B CN201610397420.7A CN201610397420A CN106096530B CN 106096530 B CN106096530 B CN 106096530B CN 201610397420 A CN201610397420 A CN 201610397420A CN 106096530 B CN106096530 B CN 106096530B
- Authority
- CN
- China
- Prior art keywords
- impulse response
- response signal
- level
- spectrum
- calculated
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
A kind of Modal Parameters Identification under strong background noise environment, it first passes through power hammer tap test and measures level-one impulse response signal, then preliminary noise reduction is carried out to level-one impulse response signal by spectrum-subtraction, obtain second level impulse response signal, it reuses Minimum Mean Square Error short time spectrum method and secondary noise reduction is carried out to second level impulse response signal, obtain ideal impulse response signal, finally ideal impulse response is believed using Modal Parameter Identification algorithm and carries out Modal Parameter Identification, the present invention has adaptive, calculating speed is fast, the advantages that Modal Parameter Identification is accurate under strong noise environment.
Description
Technical field
Modal Parameter Identification side the present invention relates to mould measurement field, in particular under a kind of strong background noise environment
Method.
Background technique
Modal Parameter Identification technology plays important in fields such as fault diagnosis, dynamic response analysis and Modifying models
Effect, identification parameter includes intrinsic frequency, damping ratio and the vibration shape.Common Modal Parameters Identification is divided into frequency domain and time domain
Two major classes.Frequency domain and Time domain identification method all carry out modal parameter knowledge to test obtained impulse response signal as basic data
Not, thus obtain accurately and reliably impulse response signal it is most important.Due to signal measurement, in transmission process inevitably
By different degrees of noise pollution, especially in the mould measurement of large scale structure, the components such as surrounding motor, pump, air-conditioning are produced
Raw very noisy is affected to impulse response signal, and the accuracy of Modal Parameter Identification is by extreme influence.
In view of the above-mentioned problems, mainly being made an uproar at present using modes such as statistical average, wavelet de-noising and singular value decompositions to eliminate
Interference of the sound to impulse response signal improves Modal Parameter Identification accuracy.Wherein, statistical average must assure that in measurement process
Middle parameter does not change, otherwise nonsensical using statistical average.And wavelet de-noising and singularity value decomposition require root
Different threshold parameters are set according to noise power, no survey noise reduction effect is impacted.At the same time, it is shown currently without result of study,
When the periodic component in very noisy and when intrinsic frequency is close or overlapping, existing method can accurately identify modal parameter.Cause
This, there are limitations for current Modal Parameters Identification, and there are also rooms for improvement.
Summary of the invention
In order to overcome the disadvantages of the above prior art, the purpose of the present invention is to provide under a kind of strong background noise environment
Modal Parameters Identification improves Precision of Estimating Modal Parameter, has adaptively by force, and calculating speed is fast, solid under strong noise environment
Have the advantages that frequency identification is accurate.
In order to achieve the above object, the technical scheme adopted by the invention is as follows:
A kind of Modal Parameters Identification under strong background noise environment, comprising the following steps:
Step 1 hammers tap test into shape by power, measures level-one impulse response signal y (t), the level-one impulse response letter measured
Number y (t) includes ideal impulse response signalThe time is represented with ambient noise signal n (t), t；
Step 2, using Optimal Smoothing Algorithm, the noise Estimation Algorithm of minimum statistics, or minimum controlled recursive average method
In any one obtain level-one impulse response signal y (t) ambient noise estimated value
Step 3 estimates the ambient noise of the level-one impulse response signal y (t) and level-one impulse response signal y (t) that measure
EvaluationFraming is carried out, adds Hanning window and FFT to handle, obtains the amplitude spectrum Y (p, k) and phase spectrum φ (p, k) of y (t), andAmplitude spectrumY (p, k) indicates the kth root spectral component amplitude of the pth frame signal of y (t), and φ (p, k) indicates y
(t) the kth root spectral component phase of pth frame signal,It indicatesPth frame signal kth root spectral component amplitude,
When framing, every frame length is 0.02 times of sample frequency, and 50% degree of overlapping is used between before and after frames；
Step 4 carries out at noise reduction the amplitude spectrum Y (p, k) for measuring level-one impulse response signal y (t) by spectrum-subtraction
Reason obtains preliminary noise reduction amplitude spectrumCalculation formula are as follows:
Wherein, spectrum subtracts factor alpha=9, β=0.05；
Step 5, the phase spectrum φ (p, k) that step 3 is obtained and preliminary noise reduction amplitude spectrum obtained in step 4In conjunction with obtaining second level impulse response signal y by inverse fourier transform and overlap-add1(t)；
Step 6, using Optimal Smoothing Algorithm, the noise Estimation Algorithm of minimum statistics, or minimum controlled recursive average method
In any one obtain second level impulse response signal y1(t) ambient noise estimated value
Step 7, to second level impulse response signal y1(t) and second level impulse response signal y1(t) ambient noise estimated valueFraming is carried out, adds Hanning window and FFT to handle, obtains second level impulse response signal y1(t) amplitude spectrum Y1(p, k) and phase
Compose φ1(p, k), andAmplitude spectrum 1(p, k) indicates y1(t) the kth root spectral component width of pth frame signal
Value, φ1(p, k) indicates y1(t) the kth root spectral component phase of pth frame signal,It indicatesPth frame signal
Kth root spectral component amplitude, when framing, the sample frequency that every frame length is 0.2 times uses 50% overlapping between before and after frames
Degree；
Step 8 calculates second level impulse response signal y1(t) each frame posteriori SNR estimated value, calculation formula are as follows:Wherein,It representsVariance,With Y1(p, k) is in step
It is calculated in seven；
Step 9, by Minimum Mean Square Error short time spectrum method to second level impulse response signal y1(t) amplitude spectrum Y1
(p, k) carries out secondary noise reduction process, obtains secondary noise reduction amplitude spectrumThe calculating process using endless form into
It goes, circulation, to a last frame end, successively carries out below 1.~5. step calculating, steps are as follows since first frame:
1. using classics DD algorithm, second level impulse response signal y is calculated1(t) pth frame a priori SNR estimation value, meter
Calculate formula are as follows:
Wherein, attenuation coefficient
β '=0.98；It is calculated in step 8；Initial value be set as 0 vector, later is each
5. frame walks calculated result using the；The posteriori SNR for representing the kth root spectral component of+1 frame signal of pth is estimated
Evaluation, the value are calculated in step 8；
2. calculating second level impulse response signal y1(t) noise suppression factor of pth frame spectral component, calculation formula are as follows:WhereinCalculated result is 1. walked using；
3. using prior weight evaluation method, second level impulse response signal y is calculated1(t) pth frame prior weight is estimated
Evaluation, calculation formula are as follows:
Its
In, attenuation coefficient β '=0.98；Y1(p, k) is calculated in step 7；It is calculated in step 8；
GDD2. (p, k) walks calculated result using the；It is calculated in step 8；
4. calculating second level impulse response signal y1(t) noise suppression factor of pth frame spectral component, calculation formula are as follows:WhereinCalculated result is 3. walked using；
5. by Minimum Mean Square Error short time spectrum method to Y1The pth frame spectral component of (p, k) carries out secondary noise reduction process,
Obtain secondary noise reduction amplitude spectrumCalculation formula are as follows:Wherein G (p, k) is used
4. walks calculated result；Y1(p, k) is calculated in step 7；
Step 10, by phase spectrum φ obtained in step 71(p, k) and using after Minimum Mean Square Error short time spectrum method
Obtained secondary noise reduction amplitude spectrumIn conjunction with obtaining ideal pulse and ring by inverse fourier transform and overlap-add
Induction signal
Step 11, by Modal Parameters Identification to ideal impulse response signalCarry out modal parameter knowledge
Not.
Beneficial effects of the present invention: the study found that this method can effectively eliminate the steady letter in test pulse response signal
Continuous part in number part and non-stationary signal retains the transient portion thereof in non-stationary signal, i.e., ideal impulse response letter
Number, to improve Precision of Estimating Modal Parameter.
Compared to existing method, this method has adaptively by force, and calculating speed is fast, and intrinsic frequency identifies under strong noise environment
The advantages that accurate.Even if this method is strong in test noise, cyclic component and structural natural frequencies are extremely connect in test noise
In nearly situation, structural natural frequencies can also be accurately identified.
Detailed description of the invention
Fig. 1 is the steam turbine generator body diagram for Modal Parameter Identification.
Fig. 2 is level-one impulse response signal y (t) time-domain diagram for vibrating measuring point.
Fig. 3 is level-one impulse response signal y (t) frequency domain figure for vibrating measuring point.
Fig. 4 is the time-domain diagram after the level-one impulse response signal y (t) of vibration measuring point is processed by the invention.
Fig. 5 is the frequency domain figure after the level-one impulse response signal y (t) of vibration measuring point is processed by the invention.
Specific embodiment
The present invention will be described in detail with case study on implementation with reference to the accompanying drawing.
There is generator housing mesomerism in the generating set of certain nuclear power plant, in order to solve to resonate in the process of running
Problem, demand accurately identify the structural natural frequencies of generator housing, since site environment noise is strong, and in test noise
Cyclic component and generator housing intrinsic frequency are very close, and existing method cannot accurately identify the intrinsic frequency of generator housing
Rate.This is solved the problems, such as below by the present invention.
A kind of Modal Parameters Identification under strong background noise environment, comprising the following steps:
Step 1 hammers Knock test into shape by power, measures the level-one impulse response signal y that measuring point is vibrated on generator housing
(t), sample frequency 500Hz, referring to Fig.1, Fig. 1 are shown steam turbine generator body diagram, the casing of generator 3 only according to
It is placed on installation foundation 4 by gravity, the rotor 5 of generator 3 is connected on generator's cover 2 by sliding bearing, generator 3
Casing top be connected with the hydrogen-cooled device 1 of 2 generators, vibration point position is arranged on the hydrogen-cooled device 1 of 1 generator；Fig. 2 and figure
3 be the level-one impulse response signal time domain waveform and frequency domain by vibrating measuring point on the generator housing that obtains after the step respectively
Waveform diagram, since the ambient noise around steam turbine generator is larger, the level-one impulse response signal shown in Fig. 2 is by noise
Signal contamination, the prominent frequency (49.82Hz) shown in Fig. 3 are periodic noise；
Step 2 obtains the ambient noise estimated value of level-one impulse response signal y (t) using Optimal Smoothing Algorithm
Step 3 estimates the ambient noise of the level-one impulse response signal y (t) and level-one impulse response signal y (t) that measure
EvaluationFraming is carried out, adds Hanning window and FFT to handle, obtains the amplitude spectrum Y (p, k) and phase spectrum φ (p, k) of y (t), andAmplitude spectrumWhen framing, every frame length is 0.02 times of sample frequency, and 50% weight is used between before and after frames
Folded degree；
Step 4 carries out noise reduction process by amplitude spectrum Y (p, k) of the spectrum-subtraction to level-one impulse response signal y (t), obtains
It is composed to preliminary noise reduction amplitudeCalculation formula are as follows:
Wherein, spectrum subtracts factor alpha=9, β=0.05；
Step 5, the phase spectrum φ (p, k) that step 3 is obtained and preliminary noise reduction amplitude spectrum obtained in step 4In conjunction with obtaining second level impulse response signal y by inverse fourier transform and overlap-add1(t)；
Step 6 obtains second level impulse response signal y using Optimal Smoothing Algorithm1(t) ambient noise estimated value
Step 7, to second level impulse response signal y1(t) and second level impulse response signal y1(t) ambient noise estimated valueFraming is carried out, adds Hanning window and FFT to handle, obtains second level impulse response signal y1(t) amplitude spectrum Y1(p, k) and phase
Compose φ1(p, k), andAmplitude spectrumWhen framing, every frame length is 0.2 times of sample frequency, before and after frames it
Between use 50% degree of overlapping；
Step 8 calculates second level impulse response signal y1(t) each frame posteriori SNR estimated value, calculation formula are as follows:Wherein,It representsVariance,With Y1(p, k) is in step
It is calculated in rapid seven；
Step 9, by Minimum Mean Square Error short time spectrum method to second level impulse response signal y1(t) amplitude spectrum Y1
(p, k) carries out secondary noise reduction process, the amplitude spectrum after obtaining secondary noise reductionThe calculating process using endless form into
It goes, circulation, to a last frame end, successively carries out below 1.~5. step calculating, steps are as follows since first frame:
1. using classics DD algorithm, second level impulse response signal y is calculated1(t) pth frame a priori SNR estimation value, meter
Calculate formula are as follows:
Wherein, attenuation coefficient
β '=0.98；It is calculated in step 8；Initial value be set as 0 vector, later is each
5. frame walks calculated result using the；The posteriori SNR for representing the kth root spectral component of+1 frame signal of pth is estimated
Evaluation, the value are calculated in step 8；
2. calculating second level impulse response signal y1(t) noise suppression factor of pth frame spectral component, calculation formula are as follows:WhereinCalculated result is 1. walked using；
3. using improved prior weight evaluation method, second level impulse response signal y is calculated1(t) pth frame priori letter
It makes an uproar compared estimate value, calculation formula are as follows:
Its
In, attenuation coefficient β '=0.98；Y1(p, k) is calculated in step 7；It is calculated in step 8；
GDD2. (p, k) walks calculated result using the；It is calculated in step 8；
4. calculating second level impulse response signal y1(t) noise suppression factor of pth frame spectral component, calculation formula are as follows:WhereinCalculated result is 3. walked using；
5. by Minimum Mean Square Error short time spectrum method to Y1The pth frame spectral component of (p, k) carries out secondary noise reduction process,
Obtain secondary noise reduction amplitude spectrumCalculation formula are as follows:Wherein G (p, k) is used
4. walks calculated result；Y1(p, k) is calculated in step 7；
Step 10, by phase spectrum φ obtained in step 71(p, k) and using after Minimum Mean Square Error short time spectrum method
Obtained secondary noise reduction amplitude spectrumIn conjunction with obtaining ideal pulse and ring by inverse fourier transform and overlap-add
Induction signalReferring to Fig. 4, Fig. 4 is ideal impulse response signalTime domain waveform；
Step 11, to ideal impulse response signalFourier transform is carried out, is identified from corresponding frequency domain figure
Generator housing intrinsic frequency, Fig. 4 show that the ambient noise in level-one impulse response signal y (t) has been removed；Fig. 5 is ideal
Impulse response signalFrequency domain figure, can there are the intrinsic of 48.99Hz with generator housing visible in detail from Fig. 5
Frequency, and then can determine whether that generator housing mesomerism problem is the intrinsic frequency (48.99Hz) and work due to generator housing
Make to turn to get too close to frequently (50Hz) caused.
The above content is a further detailed description of the present invention in conjunction with specific preferred embodiments, and it cannot be said that
A specific embodiment of the invention is only limitted to this, for those of ordinary skill in the art to which the present invention belongs, is not taking off
Under the premise of from present inventive concept, several simple deduction or replace can also be made, all shall be regarded as belonging to the present invention by institute
Claims of submission determine scope of patent protection.
Claims (1)
1. the Modal Parameters Identification under a kind of strong background noise environment, which comprises the following steps:
Step 1 hammers tap test into shape by power, measures level-one impulse response signal y (t), the level-one impulse response signal y measured
It (t) include ideal impulse response signalThe time is represented with ambient noise signal n (t), t；
Step 2, using in Optimal Smoothing Algorithm, the noise Estimation Algorithm of minimum statistics, or minimum controlled recursive average method
Anticipate it is a kind of obtain level-one impulse response signal y (t) ambient noise estimated value
Step 3, to the ambient noise estimated value of the level-one impulse response signal y (t) and level-one impulse response signal y (t) that measureFraming is carried out, adds Hanning window and FFT to handle, obtains the amplitude spectrum Y (p, k) and phase spectrum φ (p, k) of y (t), and
Amplitude spectrumY (p, k) indicates the kth root spectral component amplitude of the pth frame signal of y (t), and φ (p, k) indicates y (t)
The kth root spectral component phase of pth frame signal,It indicatesPth frame signal kth root spectral component amplitude, framing
When, every frame length is 0.02 times of sample frequency, and 50% degree of overlapping is used between before and after frames；
Step 4 carries out noise reduction process to the amplitude spectrum Y (p, k) for measuring level-one impulse response signal y (t) by spectrum-subtraction, obtains
It is composed to preliminary noise reduction amplitudeCalculation formula are as follows:
Wherein, spectrum subtracts factor alpha=9, β=0.05；
Step 5, the phase spectrum φ (p, k) that step 3 is obtained and preliminary noise reduction amplitude spectrum obtained in step 4
In conjunction with obtaining second level impulse response signal y by inverse fourier transform and overlap-add1(t)；
Step 6, using in Optimal Smoothing Algorithm, the noise Estimation Algorithm of minimum statistics, or minimum controlled recursive average method
Anticipate a kind of acquisition second level impulse response signal y1(t) ambient noise estimated value
Step 7, to second level impulse response signal y1(t) and second level impulse response signal y1(t) ambient noise estimated value
Framing is carried out, adds Hanning window and FFT to handle, obtains second level impulse response signal y1(t) amplitude spectrum Y1(p, k) and phase spectrum φ1
(p, k), andAmplitude spectrumY1(p, k) indicates y1(t) the kth root spectral component amplitude of pth frame signal,
φ1(p, k) indicates y1(t) the kth root spectral component phase of pth frame signal,It indicatesPth frame signal
K root spectral component amplitude, when framing, the sample frequency that every frame length is 0.2 times uses 50% degree of overlapping between before and after frames；
Step 8 calculates second level impulse response signal y1(t) each frame posteriori SNR estimated value, calculation formula are as follows:
Wherein,It representsVariance,With Y1(p, k) exists
It is calculated in step 7；
Step 9, by Minimum Mean Square Error short time spectrum method to second level impulse response signal y1(t) amplitude spectrum Y1(p,k)
Secondary noise reduction process is carried out, secondary noise reduction amplitude spectrum is obtainedThe calculating process is carried out using endless form, circulation
To a last frame end since first frame, 1.~5. step calculating is successively carried out below, steps are as follows:
1. using classics DD algorithm, second level impulse response signal y is calculated1(t) pth frame a priori SNR estimation value, calculation formula
Are as follows:Wherein, attenuation coefficient β '=
0.98；It is calculated in step 8；Initial value be set as 0 vector, later each frame makes
Calculated result is 5. walked with；Represent the posteriori SNR estimation of the kth root spectral component of+1 frame signal of pth
Value, the value are calculated in step 8；
2. calculating second level impulse response signal y1(t) noise suppression factor of pth frame spectral component, calculation formula are as follows:
WhereinCalculated result is 1. walked using；
3. using prior weight evaluation method, second level impulse response signal y is calculated1(t) pth frame a priori SNR estimation value,
Calculation formula are as follows:
Wherein, attenuation coefficient β '=0.98；Y1(p, k) is calculated in step 7；It is calculated in step 8
It arrives；GDD2. (p, k) walks calculated result using the；It is calculated in step 8；
4. calculating second level impulse response signal y1(t) noise suppression factor of pth frame spectral component, calculation formula are as follows:
WhereinCalculated result is 3. walked using；
5. by Minimum Mean Square Error short time spectrum method to Y1The pth frame spectral component of (p, k) carries out secondary noise reduction process, obtains
Secondary noise reduction amplitude spectrumCalculation formula are as follows:Wherein G (p, k) using the 4.
Walk calculated result；Y1(p, k) is calculated in step 7；
Step 10, by phase spectrum φ obtained in step 71It is obtained after (p, k) and use Minimum Mean Square Error short time spectrum method
Secondary noise reduction amplitude spectrumIn conjunction with obtaining ideal impulse response letter by inverse fourier transform and overlap-add
Number
Step 11, by Modal Parameters Identification to ideal impulse response signalCarry out Modal Parameter Identification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610397420.7A CN106096530B (en) | 2016-06-07 | 2016-06-07 | A kind of Modal Parameters Identification under strong background noise environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610397420.7A CN106096530B (en) | 2016-06-07 | 2016-06-07 | A kind of Modal Parameters Identification under strong background noise environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106096530A CN106096530A (en) | 2016-11-09 |
| CN106096530B true CN106096530B (en) | 2019-04-16 |
Family
ID=57448569
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610397420.7A Active CN106096530B (en) | 2016-06-07 | 2016-06-07 | A kind of Modal Parameters Identification under strong background noise environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106096530B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112067116B (en) * | 2020-07-13 | 2022-08-19 | 东南大学 | Method for testing and analyzing impact vibration of medium and small bridges with noise resistance |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246890A (en) * | 2013-05-15 | 2013-08-14 | 中国石油大学(华东) | Modal parameter recognizing method based on multi-input multi-output signal noise reduction |
| CN105205461A (en) * | 2015-09-18 | 2015-12-30 | 中国石油大学(华东) | Signal noise reducing method for modal parameter identification |
-
2016
- 2016-06-07 CN CN201610397420.7A patent/CN106096530B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246890A (en) * | 2013-05-15 | 2013-08-14 | 中国石油大学(华东) | Modal parameter recognizing method based on multi-input multi-output signal noise reduction |
| CN105205461A (en) * | 2015-09-18 | 2015-12-30 | 中国石油大学(华东) | Signal noise reducing method for modal parameter identification |
Non-Patent Citations (6)
| Title |
|---|
| A Multiple Random Decrement Method for Modal Parameter Identification of Stay Cables Based on Ambient Vibration Signals;Wen-Hwa Wu et al;《Advances in Structural Engineering》;20121231;第15卷(第6期);969-982页 |
| Identification of modal parameters including unmeasured forces and transient effects;B.Cauberghe et al;《Journal of Sound and Vibration》;20030814;第265卷(第3期);609-625页 |
| Improved Signal-to-Noise Ratio Estimation for Speech Enhancement;Cyril Plapous et al;《IEEE Transactions on Audio，Speech and Language Processing》;20061231;第14卷(第6期);2098-2108页 |
| Model order determination and noise removal for modal parameter estimation;Sau-Lon James Hu et al;《Mechanical Systems and Signal Processing》;20101231;第24卷(第6期);1605-1620页 |
| Speech Enhancement Using a Minimum Mean-Square Error Short-Time Spectral Amplitude Estimator;YARIV EPHRAIM et al;《IEEE Transactions on Acoustics，Speech，and Signal Processing》;19841231;第32卷(第6期);1109-1121页 |
| 一种模态参数识别的虚假模态剔除技术;王树青等;《中国海洋大学学报》;20120930;第42卷(第9期);97-101页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106096530A (en) | 2016-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108168891B (en) | Method and equipment for extracting weak fault signal characteristics of rolling bearing | |
| CN107356432B (en) | Fault Diagnosis of Roller Bearings based on frequency domain window experience small echo resonance and demodulation | |
| Zhang et al. | An optimized time varying filtering based empirical mode decomposition method with grey wolf optimizer for machinery fault diagnosis | |
| Xu et al. | A novel rolling bearing fault diagnosis method based on empirical wavelet transform and spectral trend | |
| CN101561314B (en) | Method for testing stochastic resonance-chaotic weak signal | |
| CN110987438A (en) | Method for detecting periodical vibration impact signals of hydraulic generator in variable rotating speed process | |
| Ding et al. | Sparsity-based algorithm for condition assessment of rotating machinery using internal encoder data | |
| CN107831013B (en) | Utilize the Method for Bearing Fault Diagnosis of probability principal component analysis enhancing cyclic bispectrum | |
| Sun et al. | Fault detection of rolling bearing using sparse representation-based adjacent signal difference | |
| Pang et al. | Rolling bearing fault diagnosis based on SVDP-based kurtogram and Iterative autocorrelation of Teager energy operator | |
| CN108398260B (en) | Method for quickly evaluating instantaneous angular speed of gearbox based on mixed probability method | |
| CN105180959A (en) | Anti-interference step counting method for wrist type step counting devices | |
| CN106096530B (en) | A kind of Modal Parameters Identification under strong background noise environment | |
| Sun et al. | Modal identification from non-stationary responses of high-rise buildings by variational mode decomposition and direct interpolation techniques | |
| Huang et al. | Separation method of rolling bearing compound fault characteristics based on Improved harmonic wavelet packet decomposition and fast ICA | |
| Chen et al. | Rolling bearing fault feature extraction method using adaptive maximum cyclostationarity blind deconvolution | |
| CN111523231B (en) | Subsynchronous oscillation analysis method based on EEMD and Prony method | |
| Zhang et al. | A method for extracting fault features using variable multilevel spectral segmentation framework and harmonic correlation index | |
| CN105738698B (en) | A kind of harmonic parameters algorithm for estimating based on centre frequency shift | |
| CN111353415A (en) | Detection method of harmonic component in impulse response | |
| CN110320018B (en) | Rotary machine composite fault diagnosis method based on second-order cyclostationarity | |
| CN112688325B (en) | Wind power plant subsynchronous oscillation monitoring method based on two-stage improved ITD algorithm | |
| Zhu et al. | Extraction of periodic signal without external reference by time-domain average scanning | |
| CN114924157A (en) | Parallel reactor state monitoring method and system based on 5G transmission | |
| CN113866493A (en) | Method for measuring voltage fluctuation and flicker caused by wind power |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |