CN106872171A - A kind of adaptive learning bearing calibration of Doppler's acoustic signal - Google Patents
A kind of adaptive learning bearing calibration of Doppler's acoustic signal Download PDFInfo
- Publication number
- CN106872171A CN106872171A CN201710228196.3A CN201710228196A CN106872171A CN 106872171 A CN106872171 A CN 106872171A CN 201710228196 A CN201710228196 A CN 201710228196A CN 106872171 A CN106872171 A CN 106872171A
- Authority
- CN
- China
- Prior art keywords
- signal
- frequency
- doppler
- correction
- distortion
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
Abstract
The invention discloses a kind of adaptive learning bearing calibration of Doppler's acoustic signal.The method is in correction parameter adaptive learning, based on rail side acoustic theory, by constructing shift frequency operator and tone operator, realize the pseudo- correction of one kind to distorted signal, obtaining frequency spectrum resonant belt has undistorted and high-energy aggregation transition baseline, portray in theory and reduce rail side acoustics distortion phenomenon, physically enhance self adaptation distortion correction learning mechanic；Compared with other bearing calibrations, the algorithm be capable of self adaptation from original sound signals learning to optimal correction parameter, so as to realize the complete correction to distorted signal.The algorithm has certain robustness and high efficiency, is conducive to the on-line intelligence of follow-up rail side sound system to correct and diagnosis.
Description
Technical field
The invention belongs to rail side acoustic Doppler distortion correction field, it is related to a kind of self adaptation of Doppler's acoustic signal
Bearing calibration is practised, the self adaptation to how general row distortion acoustic signal can be realized with adaptive learning to optimal distortion correction parameter
Correction learning, is particularly suited for wheel set bearing road side acoustics diagnostic system.
Background technology
The acoustics diagnosis of rail side is that a kind of acoustic signal produced with moving component is analysis object, with noncontact, many mesh
Mark, efficient monitoring, diagnosing pattern the advantages of need not shut down, have in Railway wheelset bearing diagnosis important application value with
Meaning.The acoustic signal collected by rail side measuring apparatus contains the current health and fitness information of part, but in practice, due to train
Run at a relatively high speed, the acoustic signal that moving parts is produced has relative motion with road side fixation measuring equipment so that most
The acoustic signal for observing eventually there occurs distortion (namely Doppler's distortion phenomenon), be brought to identification and the extraction of the source of trouble tired
It is difficult.Therefore, doppler correction analysis is done to acoustic signal to be had very important significance for rail side acoustics bearing failure diagnosis.
Conventional Doppler correction method is broadly divided into two kinds at present, and one kind is using instantaneous frequency crestal line (Instantaneous
Frequency Ridge Extraction) or the method such as principal component coherence analysis (such as MUSIC algorithms in short-term) obtain signal
Receiving angle sequence, by being fitted come indirect gain Doppler's distortion parameter, constructs launch time sequence；Another is to pass through
The greediness iterative algorithm such as match tracing to largely to search for (such as the Doppler's transient state mould based on Laplace small echos and spectrum correlation estimation
Type (Doppler transient model based on the Laplace wavelet and spectrum
Correlation assessment), carry out the part or all of Doppler's distortion parameter of direct access, realize the school to distorted signal
Positive analysis.The former is a kind of method based on data fitting, and robustness is subject to certain restrictions；The latter exists in terms of real-time
Than larger defect, it is unfavorable for actual diagnosis.Therefore traditional diagnostic method will be substantially reduced to train bearing fault diagnosis
Efficiency and accuracy.
The present invention proposes a kind of adaptive learning bearing calibration of Doppler's acoustic signal.The method combination rail side acoustics
Principle, constructs frequency displacement operator and tone operator physical model, and the puppet correction to Doppler's distorted signal is realized by operation operator.
The transition baseline of distorted signal is obtained based on puppet correction, not only there is its corresponding resonant belt undistorted characteristic also to have simultaneously
There is high-energy aggregation.Propose based on resonant frequency narrow band energy than maximization principle, using Global Optimization Algorithm For Analysis, it is right to realize
The model parameter adaptive learning of Doppler's distortion operator.Realize believing Doppler eventually through acoustic pressure operator and resampling technique
Number complete correction.The adaptive learning bearing calibration has the characteristic of robust adaptive, to rail side Doppler distortion acoustics letter
Number correction have very significant effect.
The content of the invention
It is an object of the invention to：A kind of adaptive learning bearing calibration of Doppler's acoustic signal is provided, for how general
Strangle the adaptively correcting study of acoustic distortion acoustic signal.
The technical solution adopted by the present invention is：A kind of adaptive learning bearing calibration of Doppler's acoustic signal, the method
Comprise the following steps：
Step (1), the analytic signal that sound measurement signal is obtained by Hilbert transform；
Step (2), based on rail side acoustic theory, give three Doppler's distortion parameters (u, M and k are respectively center of distortion
Moment, ripple ratio and resonant frequency position, these three parameters are the major parameter of distortion correction), construct under the sound-field model
The pseudo- correct operation operators of frequency displacement operator H and two kinds of tone operator G；
Step (3), foundation shift frequency operator do frequency displacement operation to the analytic signal in step (1), make frequency band to frequency
Concentrated at the k of position, realize the pseudo- correction of frequency to Doppler's distortion of signal, obtain transition baseline；
Step (4), spectrum analysis done to the transition baseline item in step (3), according to frequency band energy in resonant frequency position k
Place's aggregation, calculates at this resonant frequency narrow band energy ratio and maximum used as principle is evaluated, (setting frequency arrowband, bandwidth is designated as using it
B), optimizing, adaptive learning to optimal distortion correction parameter are carried out to pseudo- correction signal parameter based on Global Optimization Algorithm For Analysis；
Step (5), by optimal distortion correction parameter (u, M, k)opt, tone operator G is calculated, realize believing original acoustic pressure
Number the pseudo- correction of acoustic pressure, recover acoustical signal sound pressure amplitude；
Step (6), by optimal distortion correction parameter (u, M, k)opt, acoustical signal launch time sequence is rebuild, to step
Suddenly the pseudo- correction sound pressure signal in (5), using time domain resampling technique, finally realizes complete signal Doppler's distortion correction.
In the step (2)：
Construction frequency displacement operator and the pseudo- correct operation operator of two kinds of tone operator, are respectively based on rail side theory of sound propagation construction
Two physical models, its mathematical expression is as follows：
Shift frequency operator：
Tone operator：
Wherein,N is signal length, and r is vertical range (by road
Microphone and sound source direction of motion vertical range), c is SVEL, and tri- parameters of u, M and k are respectively center of distortion moment, ripple
Speed ratio and resonant frequency position.
In the step (4)：
Global Optimization Algorithm For Analysis are to produce several initial points by using a lot of point methods, are searched for using local solver
Local Extremum in basin where each, finally realizes globe optimum, and the algorithm principle is similar with genetic algorithm, can be fast
What speed was stablized searches globe optimum.
Advantages and positive effects of the present invention are：
(1) in correction parameter adaptive learning, based on rail side acoustic theory, calculated by constructing shift frequency operator and tone
Son, realizes the pseudo- correction of one kind to distorted signal, and obtaining frequency spectrum resonant belt has undistorted and high-energy aggregation mistake
Cross benchmark.Portray in theory and reduce rail side acoustics distortion phenomenon, physically enhance the study of self adaptation distortion correction
Mechanism.
(2) compared with other bearing calibrations, the algorithm can adaptively from original sound signals learning to optimal correction
Parameter, so as to realize the complete correction to distorted signal.The algorithm has certain robustness and high efficiency, is conducive to follow-up rail
On-line intelligence correction and the diagnosis of side sound system.
Brief description of the drawings
Fig. 1 is a kind of adaptive learning bearing calibration flow chart of Doppler's acoustic signal of the invention；
Fig. 2 be emulate the time domain waveform of signal, Fourier frequency amplitude spectrum (resonant frequency is diffused in 100Hz bandwidth) with
Its corresponding time-frequency distributions (wherein white horizontal line correspondence resonant frequency 1000Hz) (set Doppler parameter as [u, M, k]=
[0.250,0.0588,500])；Wherein, Fig. 2 (a) is the time domain waveform for emulating signal, and Fig. 2 (b) is the frequency spectrum for emulating signal, figure
2 (c) is the time-frequency distributions for emulating signal；
Fig. 3 is time domain waveform, Fourier frequency amplitude spectrum (resonant frequency collection based on signal after adaptive Doppler correction
In in 998Hz) corresponding time-frequency distributions (wherein white horizontal line correspondence resonant frequency 1000Hz) (Doppler's optimized parameter
[u, M, k]=[0.2415,0.0562,502])；Wherein, Fig. 3 (a) is the time domain waveform of signal after correction, and Fig. 3 (b) is correction
The frequency spectrum of signal afterwards, Fig. 3 (c) is the time-frequency distributions of signal after correction；
Fig. 4 is that respectively primary signal and the envelope power based on signal after adaptive Doppler correction composes (failure-frequency
Theoretical value 75Hz)；Wherein, Fig. 4 (a) is the envelope power spectrum of primary signal, and Fig. 4 (b) is the envelope power of signal after correction
Spectrum；
The when frequency division that Fig. 5 is the time domain waveform of practical bearing outer ring fault-signal, Fourier frequency amplitude spectrum is corresponding
Cloth (Doppler parameter [u, M, k]=[unknown, 0.0882, unknown])；Wherein, Fig. 5 (a) is the time domain ripple of outer ring signal
Shape, Fig. 5 (b) is the frequency spectrum of outer ring signal, and Fig. 5 (c) is the time-frequency distributions of outer ring signal；
Fig. 6 be based on time domain plethysmographic signal, Fourier frequency amplitude spectrum after adaptive Doppler correction it is corresponding when
Frequency division cloth (Doppler's optimized parameter [u, M, k]=[0.0756,0.0838,361])；Wherein, Fig. 6 (a) is signal after correction
Time domain waveform, Fig. 6 (b) is the frequency spectrum of signal after correction, and Fig. 6 (c) is the time-frequency distributions of signal after correction；
Fig. 7 is that respectively primary signal and the envelope power based on signal after adaptive Doppler correction composes (bearing outer ring
Failure-frequency theoretical value 138.74Hz), wherein, Fig. 7 (a) is composed for the envelope power of primary signal, and Fig. 7 (b) is letter after correction
Number envelope power spectrum.
Specific embodiment
Below in conjunction with the accompanying drawings and embodiment further illustrates the present invention.
Embodiment one：
Table 1：Doppler's distorted signal mathematical model parameter
As a example by emulating signal, emulation signal uses the unilateral decaying transient periodic signal similar to failure mechanism：
Wherein, ts, ξ, fcAnd fdRespectively the reception time of distorted signal, decay damping, resonant frequency and failure-frequency；
Analog signal is built according to the model parameter of table 1 and Morse's acoustic theory, while adding -10dB white Gaussians in primary signal
Noise obtains Doppler's distorted signal x (t), and now failure-frequency is fd=75Hz, Doppler's distortion parameter theoretical value [u, M,
K]=[0.250,0.0588,500].The corresponding time-frequency distributions of its corresponding time domain waveform, Fourier frequency amplitude spectrum are such as
Shown in Fig. 2.It is seen that signal has Doppler's distortion phenomenon from Fig. 2 (b), there is about 100Hz exhibitions in signal resonance bands
Width, disturbs the identification and analysis to resonant belt.The Doppler is distorted using adaptive learning technology proposed by the present invention is believed
Number treatment is corrected, specific operation process is as follows：
1st, its analytic signal is obtained for the primary signal of 0.5s does Hilbert transform to the duration；
2nd, general Le distortion parameter initial value [u, M, k]=[0.01,0.03,200] is randomly provided, its restriction range point is given
Wei not [0 0.2], [0.02 0.15], [100 600]；
3rd, according to the content of the invention the step of (2) and step (3), construct frequency displacement operator H, using frequency displacement operator to step (1)
In analytic signal operated, obtain transition baseline；
4th, according to the content of the invention the step of (4), the frequency spectrum of the transition baseline narrow band energy ratio at resonant frequency k is calculated
(it is 10Hz to set frequency arrowband B), the energy ratio is maximum as energy accumulating evaluation criterion, according to Global Optimization Algorithm For Analysis pair
Pseudo- correction signal parameter carries out optimizing, and output adaptive study to distortion correction parameter [u, M, k]=[0.2415,0.0562,
502]；
5th, by obtain optimal Doppler's distortion correction parameter, according to the content of the invention the step of (2) calculate tone operator
G, the pseudo- correction of pressure of being made a sound to original distortion acoustical signal.
6th, set up sound according to optimal Doppler's distortion correction parameter and receive crestal line and launch time sequence corresponding relation, profit
Time domain resampling technique is used, complete correction is done to the signal of the pseudo- sound pressure correction in step (5) and is recovered and is exported (such as Fig. 3 institutes
Show).It can be found that the doppler phenomenon of signal has obtained obvious elimination, there is this time-frequency spectrum extraordinary energy to concentrate and gather
Collection property, the resonant frequency after correction is 998Hz, basic as original resonance frequency 1000Hz.Meanwhile, the original letter of comparative analysis
Number and based on adaptive Doppler correction after signal envelope power spectrum, as shown in Figure 4.It is seen that, compared to directly to original
Beginning signal is the fault frequency f that Envelope Analysis are obtainedd1=154.8Hz, the failure for obtaining is diagnosed using this discovery method frequently
Rate fd2=75.2Hz is with theoretical value fd=75Hz fits like a glove.Also recover 2 frequencys multiplication in the envelope spectrum of signal after calibration simultaneously
Information.These experimental results reflect the present invention has a kind of effect of adaptive learning for the correction of Doppler's distorted signal
Really, it is advantageous to follow-up fault diagnosis.
Embodiment two：
Tested using the train bearing outer ring fault-signal of designed, designed.Static state is played using automobile linear motion
Fault-signal, dynamic is obtained by B＆K microphone 4944-A and NI PXI-4472/PXI-1033 cabinet in roadside
Doppler's distorted signal.Wherein train bearing model NJ (P) 3226X, design parameter is as shown in table 2.
Table 2：NJ (P) 3226X1 type train bearing parameter (units：Millimeter)
Dynamic Doppler's Collection parameter is as shown in table 3：
Table 3：Train bearing Collection parameter
It is the single failure of 0.18mm in bearing outer ring working width using wire cutting technology.In the acquisition condition of table 3
Under, bearing outer ring failure-frequency is fd=138.74Hz, the theoretical value M=0.0882 of rotating ratio.Its corresponding time domain waveform,
The corresponding time-frequency distributions of Fourier frequency amplitude spectrum are as shown in Figure 5.It is seen that signal has Doppler's distortion from Fig. 5
Phenomenon, namely signal band broadening, disturb the identification and analysis to resonant belt.Using adaptive learning skill proposed by the present invention
Art is corrected treatment to Doppler's distorted signal, and specific operation process is as follows：
1st, its analytic signal is obtained for the primary signal of 0.2s does Hilbert transform to the duration；
2nd, general Le distortion parameter initial value [u, M, k]=[0.01,0.03,200] is randomly provided, its restriction range point is given
Wei not [0 0.2], [0.02 0.15], [100 600]；
3rd, according to the content of the invention the step of (2) and step (3), construct frequency displacement operator H, using frequency displacement operator to step (1)
In analytic signal operated, obtain transition baseline；
4th, according to the content of the invention the step of (4), the frequency spectrum of the transition baseline narrow band energy ratio at resonant frequency k is calculated
(it is 10Hz to set frequency arrowband B), the energy ratio is maximum as energy accumulating evaluation criterion, according to Global Optimization Algorithm For Analysis pair
Pseudo- correction signal parameter carries out optimizing, and output adaptive study to distortion correction parameter [u, M, k]=[0.0756,0.0838,
361]；
5th, by obtain optimal Doppler's distortion correction parameter, according to the content of the invention the step of (2) calculate tone operator
G, the pseudo- correction of pressure of being made a sound to original distortion acoustical signal.
6th, set up sound according to optimal Doppler's distortion correction parameter and receive crestal line and launch time sequence corresponding relation, profit
Time domain resampling technique is used, complete correction is done to the signal of the pseudo- sound pressure correction in step (5) and is recovered and is exported (such as Fig. 6 institutes
Show).It can be found that the doppler phenomenon of signal has obtained obvious elimination, there is this time-frequency spectrum extraordinary energy to concentrate and gather
Collection property.Comparative analysis primary signal and the envelope power based on signal after adaptive Doppler correction are composed simultaneously, as shown in Figure 7.
It is seen that, compared to being directly the fault frequency f that Envelope Analysis are obtained to primary signald1=143.3Hz, using this hair
The failure-frequency f that existing method diagnosis is obtainedd2=138.7Hz is substantially with theoretical value fd=138.74Hz coincide.After calibration simultaneously
Also some frequency multiplication information are recovered in the envelope spectrum of signal.It is abnormal for Doppler that the experimental result further demonstrates the present invention
The adaptive learning calibration result of varying signal, has important value to the diagnosis of rail side acoustics.
In sum, the invention discloses a kind of adaptive learning bearing calibration of Doppler's acoustic signal, can combine
Rail side Principles of Acoustics, construct frequency displacement operator and tone operator physical model, are realized to Doppler's distorted signal by operation operator
Pseudo- correction, there is robust adaptive and high efficiency.The on-line intelligence correction of the in-orbit side acoustic monitoring system of the present invention
With in diagnosis have certain application prospect.
The foregoing description of the disclosed embodiments, enables professional and technical personnel in the field to realize or uses the present invention.
Various modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, the present invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The scope most wide for causing.
Claims (3)
1. a kind of adaptive learning bearing calibration of Doppler's acoustic signal, it is characterised in that：The method comprises the following steps：
Step (1), the analytic signal that sound measurement signal is obtained by Hilbert transform；
Step (2), based on rail side acoustic theory, give three Doppler's distortion parameters：Center of distortion moment u, ripple ratio M and altogether
Vibration frequency position k, constructs two kinds of pseudo- correct operation operators under the sound-field model：Frequency displacement operator H and tone operator G；
Step (3), foundation shift frequency operator H do frequency displacement operation to the analytic signal in step (1), make frequency band to frequency location
Concentrated at k, realize the pseudo- correction of frequency to Doppler's distortion of signal, obtain transition baseline；
Step (4), spectrum analysis done to the transition baseline item in step (3), it is poly- the k of resonant frequency position at according to frequency band energy
Collection, calculates at this resonant frequency narrow band energy ratio and it is maximum as principle is evaluated, and setting frequency arrowband, bandwidth is designated as B, base
Optimizing, last adaptive learning to distortion correction parameter are carried out to pseudo- correction signal parameter in Global Optimization Algorithm For Analysis；
Step (5), by optimal distortion correction parameter (u, M, k)opt, tone operator G is obtained, realize to original sound pressure signal
The pseudo- correction of acoustic pressure, recovers acoustical signal amplitude；
Step (6), by optimal distortion correction parameter (u, M, k)opt, acoustical signal launch time sequence is rebuild, to step
(5) the acoustic pressure puppet correction signal in, using time domain resampling technique, finally realizes complete signal Doppler's distortion correction.
2. a kind of Doppler's distortion acoustics signal calibration method based on adaptive learning according to claim 1, it is special
Levy and be：In the step (2)：Construction frequency displacement operator and the pseudo- correct operation operator of two kinds of tone operator, are respectively based on rail side
Two physical models of theory of sound propagation construction, its mathematical expression is as follows：
Shift frequency operator：
Tone operator：
Wherein,N is signal length, and r is vertical range, i.e. road side wheat
Gram wind and sound source direction of motion vertical range, c is SVEL, and tri- parameters of u, M and k are respectively center of distortion moment, velocity of wave
Than with resonant frequency position.
3. a kind of Doppler's distortion acoustics signal calibration method based on adaptive learning according to claim 1, it is special
Levy and be：In the step (4)：Global Optimization Algorithm For Analysis are to produce several initial points by using a lot of point methods, are utilized
Local Extremum in basin where local solver search each, finally realizes globe optimum, and the algorithm principle is with heredity
Algorithm is similar to, and be capable of fast and stable searches globe optimum.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710228196.3A CN106872171B (en) | 2017-04-10 | 2017-04-10 | A kind of adaptive learning bearing calibration of Doppler's acoustic signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710228196.3A CN106872171B (en) | 2017-04-10 | 2017-04-10 | A kind of adaptive learning bearing calibration of Doppler's acoustic signal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106872171A true CN106872171A (en) | 2017-06-20 |
| CN106872171B CN106872171B (en) | 2019-04-26 |
Family
ID=59161959
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710228196.3A Active CN106872171B (en) | 2017-04-10 | 2017-04-10 | A kind of adaptive learning bearing calibration of Doppler's acoustic signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106872171B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109406147A (en) * | 2018-10-29 | 2019-03-01 | 安徽大学 | A kind of train bearing rail side acoustics diagnostic method under speed change working condition |
| CN110472305A (en) * | 2019-07-26 | 2019-11-19 | 西北工业大学 | The dimensionally-optimised method of film-type acoustic metamaterial based on genetic algorithm |
| CN111770233A (en) * | 2020-06-23 | 2020-10-13 | Oppo（重庆）智能科技有限公司 | Frequency compensation method and terminal equipment |
| CN111801632A (en) * | 2018-03-05 | 2020-10-20 | 日本电气株式会社 | Diagnostic device, system, diagnostic method, and program |
| CN111814578A (en) * | 2020-06-15 | 2020-10-23 | 南京森林警察学院 | Method for extracting frequency of ultralow frequency Doppler signal |
| CN112529792A (en) * | 2020-11-20 | 2021-03-19 | 暨南大学 | Distortion correction method for distortion-free model camera |
| CN112766224A (en) * | 2021-02-01 | 2021-05-07 | 华侨大学 | Method, device, equipment and storage medium for extracting real signal from distorted signal |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1700492A1 (en) * | 1989-05-31 | 1991-12-23 | Предприятие П/Я В-2445 | Frequency meter of hydroacoustic doppler log |
| CN104502633A (en) * | 2014-12-29 | 2015-04-08 | 南京世海声学科技有限公司 | Flow field data correction method for acoustic Doppler flow velocity profiler |
| KR101580427B1 (en) * | 2015-01-29 | 2015-12-28 | 한국해양대학교 산학협력단 | Doppler Frequency Estimation and receiving method for Time-varying Underwater Acoustic Communication Channel |
| CN105277932A (en) * | 2014-07-21 | 2016-01-27 | 中国科学院声学研究所 | Doppler frequency shift correction method based on down-conversion beamforming |
| CN105424388A (en) * | 2015-11-17 | 2016-03-23 | 苏州大学 | Train wheel set bearing fault transient characteristic detection method based on parametric Doppler transient model |
| CN106226078A (en) * | 2016-07-01 | 2016-12-14 | 中国科学技术大学 | A kind of Doppler based on microphone array distorts the bearing calibration of acoustic signal |
| US9608738B2 (en) * | 2012-04-27 | 2017-03-28 | The Board Of Trustees Of The University Of Illinois | System and method for broadband doppler compensation |
-
2017
- 2017-04-10 CN CN201710228196.3A patent/CN106872171B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1700492A1 (en) * | 1989-05-31 | 1991-12-23 | Предприятие П/Я В-2445 | Frequency meter of hydroacoustic doppler log |
| US9608738B2 (en) * | 2012-04-27 | 2017-03-28 | The Board Of Trustees Of The University Of Illinois | System and method for broadband doppler compensation |
| CN105277932A (en) * | 2014-07-21 | 2016-01-27 | 中国科学院声学研究所 | Doppler frequency shift correction method based on down-conversion beamforming |
| CN104502633A (en) * | 2014-12-29 | 2015-04-08 | 南京世海声学科技有限公司 | Flow field data correction method for acoustic Doppler flow velocity profiler |
| KR101580427B1 (en) * | 2015-01-29 | 2015-12-28 | 한국해양대학교 산학협력단 | Doppler Frequency Estimation and receiving method for Time-varying Underwater Acoustic Communication Channel |
| CN105424388A (en) * | 2015-11-17 | 2016-03-23 | 苏州大学 | Train wheel set bearing fault transient characteristic detection method based on parametric Doppler transient model |
| CN106226078A (en) * | 2016-07-01 | 2016-12-14 | 中国科学技术大学 | A kind of Doppler based on microphone array distorts the bearing calibration of acoustic signal |
Non-Patent Citations (3)
| Title |
|---|
| 吴强、孔凡让、何清波、刘永斌: "基于重采样技术的声学信号多普勒畸变校正", 《信号处理》 * |
| 张海滨、陆思良、何清波、孔凡让: "多普勒畸变声学信号的伪时频分析及其校正", 《振动与冲击》 * |
| 袁仲洲、孔凡让、张海滨、胡飞、刘方、朱丕亮: "一种基于瞬时频率的多普勒畸变声信号校正方法", 《自动控制与检测》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111801632A (en) * | 2018-03-05 | 2020-10-20 | 日本电气株式会社 | Diagnostic device, system, diagnostic method, and program |
| CN109406147A (en) * | 2018-10-29 | 2019-03-01 | 安徽大学 | A kind of train bearing rail side acoustics diagnostic method under speed change working condition |
| CN109406147B (en) * | 2018-10-29 | 2020-11-13 | 安徽大学 | Train bearing rail side acoustic diagnosis method under variable speed working condition |
| CN110472305A (en) * | 2019-07-26 | 2019-11-19 | 西北工业大学 | The dimensionally-optimised method of film-type acoustic metamaterial based on genetic algorithm |
| CN110472305B (en) * | 2019-07-26 | 2022-09-13 | 西北工业大学 | Thin film type acoustic metamaterial size optimization method based on genetic algorithm |
| CN111814578A (en) * | 2020-06-15 | 2020-10-23 | 南京森林警察学院 | Method for extracting frequency of ultralow frequency Doppler signal |
| CN111770233A (en) * | 2020-06-23 | 2020-10-13 | Oppo（重庆）智能科技有限公司 | Frequency compensation method and terminal equipment |
| CN112529792A (en) * | 2020-11-20 | 2021-03-19 | 暨南大学 | Distortion correction method for distortion-free model camera |
| CN112529792B (en) * | 2020-11-20 | 2022-03-29 | 暨南大学 | Distortion correction method for distortion-free model camera |
| CN112766224A (en) * | 2021-02-01 | 2021-05-07 | 华侨大学 | Method, device, equipment and storage medium for extracting real signal from distorted signal |
| CN112766224B (en) * | 2021-02-01 | 2023-05-30 | 华侨大学 | Method, device, equipment and storage medium for extracting true signal from distorted signal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106872171B (en) | 2019-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106872171B (en) | A kind of adaptive learning bearing calibration of Doppler's acoustic signal | |
| CN105424359B (en) | A kind of gear and bearing mixed fault feature extracting method based on sparse decomposition | |
| CN104819846B (en) | Rolling bearing sound signal fault diagnosis method based on short-time Fourier transform and sparse laminated automatic encoder | |
| CN105928702B (en) | Variable working condition box bearing method for diagnosing faults based on form PCA | |
| CN108489719B (en) | A kind of combined failure of rotating machinery diagnostic method based on the unusual spectral factorization of G-P | |
| CN101426169B (en) | Time-domain tracking filter quickly detecting acoustic response parameter of sounding body and system | |
| CN104167207A (en) | Equipment sound identification method based on transformer substation patrol inspection robot | |
| CN107402131B (en) | A kind of bullet train kinematic parameter recognition methods based on rail side acoustic signal time-frequency crestal line | |
| CN106441895B (en) | A kind of train bearing rail side signal impacts component extracting method | |
| CN103018044A (en) | Bearing combined failure diagnosis method based on improved impact dictionary matching pursuit | |
| CN109813547A (en) | Rotating machinery local type method for diagnosing faults based on sparse decomposition optimization algorithm | |
| CN109214356A (en) | A kind of fan transmission system intelligent fault diagnosis method based on DCNN model | |
| CN113298134B (en) | System and method for remotely and non-contact health monitoring of fan blade based on BPNN | |
| CN108036940A (en) | A kind of Method for Bearing Fault Diagnosis | |
| Cheng et al. | Adaptive sparsest narrow-band decomposition method and its applications to rolling element bearing fault diagnosis | |
| CN106053074A (en) | Rolling bearing sound signal fault feature extraction method based on STFT and rotation inertia entropy | |
| CN116486834A (en) | Rolling sound classification method based on feature fusion and improved convolutional neural network | |
| CN108061653B (en) | Based on harmonic wave-impact Doppler modulation compound dictionary Railway wheelset bearing rail side acoustic signal | |
| CN109614887A (en) | A kind of vehicle whistle classification method based on support vector machines | |
| CN110196098A (en) | Vehicle sound quality evaluation method based on changes in heart rate | |
| CN113466616A (en) | Method and device for quickly positioning cable fault point | |
| CN103064821A (en) | Method and device for analyzing dynamic signals | |
| CN108398267B (en) | A kind of bullet train rail side kinematic parameter self-adaptive identification method | |
| CN103149047A (en) | Cooling tower acoustic diagnosis method based on nonlinear mixed model | |
| CN106593777B (en) | A kind of wind power generation system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |