CN109063668A - It is a kind of to retain down-sampled impact signal envelope demodulation method based on peak value - Google Patents
It is a kind of to retain down-sampled impact signal envelope demodulation method based on peak value Download PDFInfo
- Publication number
- CN109063668A CN109063668A CN201810922977.7A CN201810922977A CN109063668A CN 109063668 A CN109063668 A CN 109063668A CN 201810922977 A CN201810922977 A CN 201810922977A CN 109063668 A CN109063668 A CN 109063668A
- Authority
- CN
- China
- Prior art keywords
- signal
- envelope
- sampled
- frequency
- obtains
- 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
- 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
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The present invention, which provides, a kind of retains down-sampled impact signal envelope demodulation method based on peak value, this method realizes the acquisition of original signal envelope waveform by wave crest extraction algorithm, extraction algorithm main processes of calculation is signal differential, is multiplied and compares, after obtaining envelope waveform, it is down-sampled to envelope waveform progress to retain down-sampled method using peak value, realize the reduction of analysis frequency, Fourier transformation finally is carried out to down-sampled data and obtains the spectrum signature of fault-signal, obtains the dependent failure information of equipment.The present invention preferably solves following problems present in current envelope demodulation method: needing poor Study first, real-time, waveform peak information loss, original signal attenuation characteristic and harmonic characteristic is lost, envelope waveform is distorted after demodulation, shock characteristic of this method suitable for rotary-type Diagnosis system of mechanical failure extracts, and is easy to integrate in real-time on-line system.
Description
Technical field
The present invention relates to a kind of mechanical fault diagnosis fields, and in particular to a kind of to retain down-sampled impact letter based on peak value
Number envelope demodulation method.
Background technique
Rotating machinery is widely used at present in various industry spots, either gear or bearing etc., operation
Quality will have a direct impact on the working performance of whole equipment.Once the failure of these rotary parts is not found in time, can lead
Cause machine breakdown, the producing line shut-down even production accidents such as casualties.Therefore, for the initial failure of related similar devices into
Row monitoring analysis has a very important significance.
In mechanical fault diagnosis field, when damage or defect is locally present in the components such as rolling bearing or gear, negative
Decaying impact can be generated by carrying in operation process, this impact signal can excite the high frequency intrinsic vibration of components.This high frequency is solid
There is vibration as carrier signal, amplitude will be by base wave modulation caused by the impact as caused by physical imperfection, to make
The vibrational waveform that outside can acquire shows complicated amplitude modulation phenomenon.And it is contained in fact in these decaying shock responses
Dependent failure information corresponding to component.
It is especially more complicated in frequency content for this kind of modulated signal, when more comprising signal component, traditional frequency domain side
Method can not be accurately positioned where equipment fault.It needs effectively to extract modulated signal therein, and envelope demodulation method can
Fault message included in signal to be separated from complicated amplitude modulated oscillations signal, and it is most widely used at present
One of diagnostic method.Envelope demodulation method has very much, and currently used technology has based on Hilbert transform (HT:Hilbert
Transform) (Song Xiaomei, the rolling bearing fault diagnosis research based on Envelope Demodulation Analysis, instrument and meter and research and application,
2012), broad sense detected envelope (Li Jun, amplitude-modulated signal envelope-demodulation method and system, the patent No.: CN104218894B) and peak value
Detection (Hou Chenggang, the impact signal demodulation method based on digital peak detection, the patent No.: CN103308310B) etc..
The cardinal principle of HT method is to filter out the low frequency component in signal first with bandpass filtering, obtain as much as possible
The high fdrequency component modulated, the phase shift for then allowing test signal to generate one 90 ° by HT constitute a parsing with original signal and believe
Number, this analytic signal constitutes envelope signal；The bearing vibration signal of existing defects, become after Hilbert transform detection
Spectrum analysis is carried out at envelope signal, then to this envelope signal, then contains the excited frequency of low frequency, i.e. defect frequency in frequency spectrum
Rate.Broad sense detected envelope method obtains amplitude-modulated signal and corresponding frequency of carrier signal first, makees four points after obtaining cycle time
One of period delay and the delay of half period, half time delayed signal is multiplied with the difference of itself and original signal, general
To signal and a quarter period time delayed signal seek root value and obtain the envelope signal of original amplitude modulated signal.Peak detection is then
It is the method that impact signal demodulation is realized using digital peak detection, this method simulates RC charge and discharge process, by with Ae-t/τIt is filtered for the Morphological Filtering Algorithm of structural element, enables initial impact peak A first0According to A0e-t/τDecay, obtains
To pad value Ai, by next collected impact peak value A1With AiCompare, if A1≤Ai, then A1 is abandoned, with AiAs peak value
After detection in waveform a little and continue to decay；It is on the contrary then stop this decaying, by A1As the new punching in waveform after peak detection
Peak value is hit, and according to A1e-t/τRestart attenuation process.
There are following one or more shortcomings and deficiencies in three kinds of above-mentioned envelope demodulation methods:
Traditional HT demodulation method real-time in envelope finding process is poor, it is difficult to which canbe used on line is needed by different
Method determines suitable Study first (filtered band and filter parameter), if filter range selection does not conform to when low signal-to-noise ratio
Suitable, demodulation effect can be very unobvious, and the waveforms amplitude after demodulation often has distortion, cannot keep under compared with harmonic analysis frequency
The originally peak value of actual signal.
Broad sense detected envelope method needs to obtain the Study first (frequency of amplitude-modulated signal and carrier signal in advance before treatment
Information), amount of delay is obtained with this, and in actual rotating machinery fault characteristic extraction procedure, it is to need through envelope solution
It transfers to obtain characteristic frequency, it is impossible to which this parameter is known in budget, thus the envelope demodulation method is in mechanical fault diagnosis field
It is not available.
Although peak detection method can keep the peak value part of original waveform, in the case where impacting unconspicuous situation
It is poor to demodulate effect, usually constructs envelope signal with the structural elements of exponential damping, the attenuation characteristic that can lose original signal is harmonious
Wave characteristic, in the model, the selection of Study first τ (time constant) can cause result directly to influence, inappropriate value meeting
Obtain full of prunes envelope signal.
In addition, when carrying out envelope spectrum analysis to signal, since characteristic frequency often focuses on low-frequency range, it usually needs reduce
Frequency is analyzed, to improve low-frequency range resolution ratio, reduce data volume and operand.It can generally be realized using interpolation resampling point
The reduction of frequency is analysed, no matter which kind of above-mentioned method is used at this time, during resampling, if corresponding to the peak value of original waveform
Moment does not have the time point of corresponding coincidence in new signal time sequence, and the amplitude information that will lead to envelope signal is lost
It loses.
Summary of the invention
These problems for existing method and insufficient, the present invention propose a kind of to retain down-sampled impact letter based on peak value
Number envelope demodulation method.This method realizes the acquisition of original signal envelope waveform by wave crest extraction algorithm, does not need any
Additional parameter overcomes the deficiency that the above several method needs to rely on Study first.After obtaining envelope waveform, protected using peak value
Down-sampled method is stayed, realizes that the reduction of analysis frequency can be to envelope signal under the premise of being sufficiently reserved original signal peak value
Low frequency part carry out specific aim analysis.The present invention has good real-time, is easy to transplant and realize on embedded device.
The envelope waveform of extraction can keep the attenuation characteristic of original signal well.
Goal of the invention to realize the present invention retains down-sampled impact signal packet based on peak value the present invention provides a kind of
Network demodulation method, the described method comprises the following steps:
With certain sample frequency F during rotating machinery rotary worksIt is sampled, obtains original vibration signal
x(i)；
Its differential signal d (i)=x (i)-x (i-1) is calculated for original vibration signal x (i), obtains two paths of signals, d1
(i)=d (1:n-1), d2(i)=d (2:n), wherein i=0,1,2 ..., n, n are step-length；
Pass through Rule of judgment d1·d2< 0&d1> 0 extracts the wave crest information of original vibration signal, obtains envelope signal X
(i)；
Fourier transformation is carried out to envelope signal, obtains envelope spectrum, signature analysis is carried out to envelope spectrum, obtains corresponding event
Hinder diagnosis.
Wherein, the sample frequency FsFor 102.4kHz, sampling number N=131068.
Before carrying out Fourier transformation to envelope signal, following steps are also executed:
According to original signal samples frequency Fs, require analysis frequency FaDrop is carried out to envelope signal X (i) with spectral line number L to adopt
Sample obtains down-sampled signal Xres(j), (1-1) n+1 < j < ln, wherein l is sampling window serial number；
Calculate t at the time of corresponding to the maximum value X (j) in each step-length nl；
In Xres(j) it is found in and all tlImmediate moment Tl, and update Xres(Tl) value be X (tl), it is updated
X afterwardsres(j)；
Judge whether the update that whole segment signal is completed, i.e. whether len (X (i)) > l*n is true, and len (X (i)) indicates letter
Number length；
If the update of whole segment signal is completed, by updated Xres(j) retain down-sampled packet as final peak value
Network signal；
If not completing the update of whole segment signal, when returning corresponding to the maximum value X (j) calculated in each step-length n
Carve tlThe step of and continue to update.
Wherein, it is F that input parameter, which includes: sample frequency,s=102.4kHz, it is desirable that analysis frequency be Fa=1kHz, spectral line
Number is L=3200, and the corresponding time span of spectral line number 3200 is 1.25s.
Wherein, step-length n can be according to the frequency range F of concernaimWith sample frequency FsIt is estimated, n=Fs/Faim。
Wherein, the sample frequency FsFor 256Hz.
Variable element peak value proposed by the invention, which retains envelope demodulation method, can well solve prior art institute
There are the problem of, its advantages specifically include that
1. realizing the acquisition of original signal envelope waveform by wave crest extraction, any additional parameter is not needed, is overcome
Conventional method needs the problem of by Study first, eliminates the dependence to parameter preset, and calculating process is very simple, tool
There is good real-time, is easy to transplant and realize on embedded device；
2. the decaying of original signal can be kept to impact well using the extracted envelope waveform of method of the invention to ring
It answers, wave character is not made any it is assumed that retaining method for resampling using peak value, realize the reduction of analysis frequency, sufficiently protecting
Under the premise of staying original signal peak value and signal characteristic, specific aim analysis is carried out to the low frequency part of envelope signal, reduces system
Operand and data store pressure.
By referring to the following drawings and to the detailed description of a specific embodiment of the invention, feature and advantage of the invention
It will become apparent.
Detailed description of the invention
Fig. 1 shows the structural schematic diagram of Diagnosis system of mechanical failure of the invention；
Fig. 2 shows the Acceleration pulse and frequency spectrum of vibration signal；
Fig. 3 shows the method flow diagram of impact signal envelope demodulation method of the invention；
Fig. 4 shows the envelope waveform extracted using wave crest and frequency spectrum；
Fig. 5, which is shown, retains the envelope waveform and frequency spectrum that down-sampled method obtains using peak value of the invention；
Fig. 6, which is shown, utilizes the obtained packet of Hilbert transform and conventional method for resampling when sample frequency is 256Hz
Network waveform and frequency spectrum；
Fig. 7, which is shown, retains the Envelop waves that down-sampled method obtains using peak value of the invention when sample frequency is 256Hz
Shape and frequency spectrum.
Specific embodiment
Embodiments of the present invention provide it is a kind of retain down-sampled impact signal envelope demodulation method based on peak value, should
Method realizes the acquisition of original signal envelope waveform by wave crest extraction algorithm, and extraction algorithm main processes of calculation is signal difference
Divide, be multiplied and compare, it is down-sampled to envelope waveform progress using the down-sampled method of peak value reservation after obtaining envelope waveform, it is real
The reduction of frequency is now analyzed, Fourier transformation finally is carried out to down-sampled data and obtains the spectrum signature of fault-signal, is set
Standby dependent failure information.In general, the present invention preferably solves following present in current envelope demodulation method ask
Topic: poor Study first, real-time, waveform peak information loss, original signal attenuation characteristic and harmonic characteristic is needed to lose, demodulate
Envelope waveform is distorted afterwards, and shock characteristic of this method suitable for rotary-type Diagnosis system of mechanical failure extracts, and is easy to real-time
It is integrated in on-line system.
A specific embodiment of the invention and algorithmic procedure are described in detail with reference to the accompanying drawing.
Fig. 1 shows the structural schematic diagram of Diagnosis system of mechanical failure of the invention.Diagnosis system of mechanical failure includes rolling
Dynamic bearing, bearing block and acceleration transducer, rolling bearing are mounted on bearing block, and acceleration transducer is installed on bearing block
On, and sampled during rolling bearing rotary work with certain frequency, obtain vibration signal x (i), sample frequency Fs=
102.4kHz sampling number N=131068.Fig. 2 shows the Acceleration pulse and frequency spectrum of vibration signal, upper figure be signal when
Domain waveform, the following figure are its corresponding frequency spectrum.From in the frequency spectrum of signal as can be seen that vibration signal in as produced by bearing fault
Frequency of impact (31.25Hz) be almost difficult to find, energy is higher instead for 4 frequencys multiplication.
Fig. 3 shows the method flow diagram of impact signal envelope demodulation method of the invention.
First against the original vibration signal x (i) of sensor acquisition, its differential signal d (i)=x (i)-x (i-1) is calculated,
Obtain two paths of signals d1(i)=d (1:n-1), d2(i)=d (2:n), wherein i=0,1,2 ..., n, n are step-length, i.e., each sampling
Length of window.Pass through Rule of judgment d1·d2< 0&d1> 0 extracts the wave crest information of x (i), obtains envelope signal X (i) (step
Rule of judgment in rapid can also add | x (i) | > Th, Th are the hard -threshold being arranged according to impact signal peak value).Corresponding packet
Network waveform and frequency spectrum are as shown in Figure 4.Apparent failure-frequency 31.25Hz can have been told from envelope spectrum and its high order is humorous
Wave, to judge the corresponding fault signature of surveyed bearing.The result shows by simple difference and can compare fortune
It calculates, to replace traditional Hilbert transform calculating process, realizes the envelope solution of rotary-type mechanical fault signals (impact signal)
It adjusts.Fourier transformation is carried out to envelope signal, obtains envelope spectrum, signature analysis is carried out to envelope spectrum, so that it may be obtained corresponding
Fault diagnosis conclusion.Simple algorithmic procedure makes it easier to the embedded platforms such as handheld instrument, monitoring device at the scene
Upper use has more preferably timeliness and timeliness.
But during actual engineer application, in order to obtain the fault signature for being modulated onto high band, sample frequency
It is relatively high, and relevant treatment is carried out to the signal of 102.4kHz sample frequency, it is right especially when original signal slowly runs
Answer failure-frequency lower, have in the case where increasing the sampling time to obtain fault signature apparent enough, to storage,
It is all greatly to challenge for operand.And actually often relatively low (the several hundred Hz even tens of our frequencies of concern
Hz), enough fault messages can not only be retained, but also the fortune of hardware can be substantially reduced by reasonable down-sampled means at this time
Calculation amount.
Down-sampled process is retained for peak value of the invention, input parameter includes: that sample frequency is Fs=102.4kHz,
It is required that analysis frequency be Fa=1kHz, spectral line number are L=3200, and the corresponding time span of spectral line number 3200 is 1.25s, original
Signal length is met the requirements.Need to handle envelope signal X (i) the down-sampled signal for sample frequency 2560Hz to this.
Specifically, firstly, according to original signal samples frequency Fs, require analysis frequency FaEnvelope is believed with spectral line number L
Number X (i) carries out down-sampled, obtains down-sampled signal Xres(j), (l-1) n+1 < j < ln, wherein l is sampling window sequence
Number.(it can use using conventional down-sampled method and close on the interpolation sides such as interpolation, bilinear interpolation or cubic convolution interpolation
Method) envelope signal is calculated, obtain down-sampled signal.
Then, t at the time of corresponding to the maximum value X (j) in each step-length n is calculatedl, step-length n can be according to the frequency of concern
Range is estimated, it is assumed that concern frequency range Faim=200Hz frequency content below, in conjunction with FsN=F is calculateds/Faim
=512.
Then, in Xres(j) it is found in and all tlImmediate moment Tl, and update Xres(Tl) value be X (tl), it obtains
To updated Xres(j)。
Judge whether the update that whole segment signal is completed, i.e. whether len (X (i)) > l*n is true, and len (X (i)) indicates letter
Number length.If the update of whole segment signal is completed, by updated Xres(j) retain down-sampled packet as final peak value
Network signal.If not completing the update of whole segment signal, at the time of return corresponding to the maximum value X (j) calculated in each step-length n
tlThe step of and continue to update.
Then down-sampled envelope signal progress Fourier transformation is retained to final peak value and obtains frequency spectrum, as shown in Figure 5.It is right
Than the difference and little of both Fig. 4,5 discoveries, Fig. 5 is to be calculated from the signal that data volume reduces 40 times, when should
When algorithm integral transplanting is into embedded system, resource consumption will be greatlyd save, obtains and more timely monitors or diagnose knot
By.
As a comparison, Fig. 6 shows using Hilbert transform and is obtained by conventional method for resampling processing vibration signal
The envelope waveform and frequency spectrum arrived.In order to increase contrast effect, the sample frequency of original signal is down to 256Hz, i.e., compared to original
The sample frequency come reduces 400 times, and handled data points decrease 400 times, the Envelop waves that two methods obtain at this time
Shape and frequency spectrum are respectively as Fig. 6 (using Hilbert transform and passing through conventional resampling) and Fig. 7 (are retained using peak value of the invention
Down-sampled method) shown in.
Compare processing result shown by two width figures, it can be seen that the time domain waveform shock characteristic in Fig. 6 is obviously than in Fig. 7
Result it is poor, the periodicity in time domain has been difficult to judge, it is also seen that the noise of failure-frequency is relatively low from frequency spectrum.And
It is in Fig. 7 the result shows that, the impact peak value of original vibration signal has obtained retaining well during down-sampled, envelope waveform
In periodic impulse it is still fairly obvious, characteristic frequency and its frequency multiplication are significant in frequency spectrum.As it can be seen that the peak value mentioned using the present invention
Retain envelope demodulation and down-sampled method, the real impact peak value of signal, comparison can be kept under low sample frequency well
Envelope waveform in Fig. 7,5 and 4 can be seen that the amplitude of impact signal almost without decaying, can want in this way reducing operation
True surge waveform and energy are obtained in the case where asking, and obtain more accurate breakdown judge.
Finally, to updated Xres(j) it carries out the envelope spectrum that Fourier transformation obtains and carries out signature analysis, can obtain
Corresponding fault diagnosis conclusion.
The processing signal type of heretofore described method is not limited to vibration signal, be also possible to it is other kinds of can
The signal (acoustical signal etc.) for reflecting unit failure, correspondingly, used sensor can be acceleration transducer, is also possible to
Vibrating sensor, displacement sensor, sound transducer etc..The data acquisition target being previously mentioned in invention can be bearing, can also
To be other as rotating machineries such as gear, axis.The solution enveloping method being related to when obtaining envelope signal is not limited to institute of the present invention
The difference wave crest extraction algorithm of proposition is also possible to other envelope demodulation algorithms such as maximum value envelope, phase sensitive detection.In addition,
Down-sampled algorithm and required input condition of the invention, is not limited to the present embodiment middle proposed interpolation method, it is possible to use its
His down-sampled method.
Described above be only it is illustrative, and it is to be understood that it is described herein arrangement and details modification and
Variation will be apparent to those skilled in the art, any under the premise of no disengaging inventive concept of the invention
Obvious replacement is within that scope of the present invention.It is therefore intended that only by scope of the appended claims rather than by
The specific detail that is presented by way of above description and explanation limits.
Claims (6)
1. a kind of retain down-sampled impact signal envelope demodulation method based on peak value, which is characterized in that the method includes with
Lower step:
With certain sample frequency F during rotating machinery rotary worksIt is sampled, obtains original vibration signal x (i)；
Its differential signal d (i)=x (i)-x (i-1) is calculated for original vibration signal x (i), obtains two paths of signals, d1(i)=d
(1:n-1), d2(i)=d (2:n), wherein i=0,1,2 ..., n, n are step-length；
Pass through Rule of judgment d1·d2< 0&d1> 0 extracts the wave crest information of original vibration signal, obtains envelope signal X (i)；
Fourier transformation is carried out to envelope signal, obtains envelope spectrum, signature analysis is carried out to envelope spectrum, corresponding failure is obtained and examines
Disconnected conclusion.
2. according to the method described in claim 1, wherein, the sample frequency FsFor 102.4kHz, sampling number N=
131068。
3. according to the method described in claim 1, also executing following steps before carrying out Fourier transformation to envelope signal:
According to original signal samples frequency Fs, require analysis frequency FaIt is down-sampled to envelope signal X (i) progress with spectral line number L,
Obtain down-sampled signal Xres(j), (1-1) n+1 < j < ln, wherein l is sampling window serial number；
Calculate t at the time of corresponding to the maximum value X (j) in each step-length nl；
In Xres(j) it is found in and all tlImmediate moment Tl, and update Xres(Tl) value be X (tl), it obtains updated
Xres(j)；
Judge whether the update that whole segment signal is completed, i.e. whether len (X (i)) > l*n is true, and len (X (i)) indicates Chief Signal Boatswain
Degree；
If the update of whole segment signal is completed, by updated Xres(j) retain down-sampled envelope letter as final peak value
Number；
If not completing the update of whole segment signal, t at the time of corresponding to the maximum value X (j) calculated in each step-length n is returnedl's
Step simultaneously continues to update.
4. according to the method described in claim 3, wherein, input parameter includes: that sample frequency is Fs=102.4kHz, it is desirable that
Analysis frequency is Fa=1kHz, spectral line number are L=3200, and the corresponding time span of spectral line number 3200 is 1.25s.
5. according to the method described in claim 3, wherein, step-length n can be according to the frequency range F of concernaimWith sample frequency FsInto
Row estimation, n=Fs/Faim。
6. according to the method described in claim 3, wherein, the sample frequency FsFor 256Hz.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810922977.7A CN109063668B (en) | 2018-08-14 | 2018-08-14 | Impact signal envelope demodulation method based on peak value retention and down-sampling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810922977.7A CN109063668B (en) | 2018-08-14 | 2018-08-14 | Impact signal envelope demodulation method based on peak value retention and down-sampling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109063668A true CN109063668A (en) | 2018-12-21 |
| CN109063668B CN109063668B (en) | 2021-09-14 |
Family
ID=64678364
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810922977.7A Active CN109063668B (en) | 2018-08-14 | 2018-08-14 | Impact signal envelope demodulation method based on peak value retention and down-sampling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109063668B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110160767A (en) * | 2019-06-14 | 2019-08-23 | 安徽智寰科技有限公司 | Impulse period automatic identification and extracting method and system based on Envelope Analysis |
| CN110569812A (en) * | 2019-09-12 | 2019-12-13 | 天津工业大学 | envelope demodulation method and envelope demodulation system for fault signal |
| CN110987435A (en) * | 2019-12-18 | 2020-04-10 | 北京化工大学 | Bearing fault diagnosis method based on improved K-SVD dictionary learning and compressed sensing |
| CN113297922A (en) * | 2021-04-30 | 2021-08-24 | 广西电网有限责任公司电力科学研究院 | High-voltage switch cabinet fault diagnosis method and device and storage medium |
| CN114754860A (en) * | 2022-04-13 | 2022-07-15 | 哈工大机器人（合肥）国际创新研究院 | Wireless vibration monitoring method, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5724002A (en) * | 1996-06-13 | 1998-03-03 | Acrodyne Industries, Inc. | Envelope detector including sample-and-hold circuit controlled by preceding carrier pulse peak(s) |
| CN103308310A (en) * | 2013-06-05 | 2013-09-18 | 西安交通大学 | Method for demodulating impact signals on basis of all-digital peak detection |
| CN105388012A (en) * | 2015-10-22 | 2016-03-09 | 兰州理工大学 | Order tracking method based on nonlinear frequency modulation wavelet transformation |
| CN107421435A (en) * | 2017-03-24 | 2017-12-01 | 浙江大学 | Homodyne orthogonal laser interference signal generation method based on digit shifter |
| CN107976304A (en) * | 2016-10-21 | 2018-05-01 | 计算***有限公司 | The mechanical disorder prediction analyzed based on the periodical information to signal |
-
2018
- 2018-08-14 CN CN201810922977.7A patent/CN109063668B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5724002A (en) * | 1996-06-13 | 1998-03-03 | Acrodyne Industries, Inc. | Envelope detector including sample-and-hold circuit controlled by preceding carrier pulse peak(s) |
| CN103308310A (en) * | 2013-06-05 | 2013-09-18 | 西安交通大学 | Method for demodulating impact signals on basis of all-digital peak detection |
| CN105388012A (en) * | 2015-10-22 | 2016-03-09 | 兰州理工大学 | Order tracking method based on nonlinear frequency modulation wavelet transformation |
| CN107976304A (en) * | 2016-10-21 | 2018-05-01 | 计算***有限公司 | The mechanical disorder prediction analyzed based on the periodical information to signal |
| CN107421435A (en) * | 2017-03-24 | 2017-12-01 | 浙江大学 | Homodyne orthogonal laser interference signal generation method based on digit shifter |
Non-Patent Citations (1)
| Title |
|---|
| 田晶 等: "基于峰值保持降采样算法的中介轴承故障声发射数据缩减技术", 《推进技术》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110160767A (en) * | 2019-06-14 | 2019-08-23 | 安徽智寰科技有限公司 | Impulse period automatic identification and extracting method and system based on Envelope Analysis |
| CN110160767B (en) * | 2019-06-14 | 2021-03-02 | 安徽智寰科技有限公司 | Impact period automatic identification and extraction method and system based on envelope analysis |
| CN110569812A (en) * | 2019-09-12 | 2019-12-13 | 天津工业大学 | envelope demodulation method and envelope demodulation system for fault signal |
| CN110569812B (en) * | 2019-09-12 | 2022-11-01 | 天津工业大学 | Envelope demodulation method and envelope demodulation system for fault signals |
| CN110987435A (en) * | 2019-12-18 | 2020-04-10 | 北京化工大学 | Bearing fault diagnosis method based on improved K-SVD dictionary learning and compressed sensing |
| CN113297922A (en) * | 2021-04-30 | 2021-08-24 | 广西电网有限责任公司电力科学研究院 | High-voltage switch cabinet fault diagnosis method and device and storage medium |
| CN113297922B (en) * | 2021-04-30 | 2023-05-05 | 广西电网有限责任公司电力科学研究院 | High-voltage switch cabinet fault diagnosis method, device and storage medium |
| CN114754860A (en) * | 2022-04-13 | 2022-07-15 | 哈工大机器人（合肥）国际创新研究院 | Wireless vibration monitoring method, electronic equipment and storage medium |
| CN114754860B (en) * | 2022-04-13 | 2024-05-14 | 哈工大机器人（合肥）国际创新研究院 | Wireless vibration monitoring method, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109063668B (en) | 2021-09-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109063668A (en) | It is a kind of to retain down-sampled impact signal envelope demodulation method based on peak value | |
| CN107356432B (en) | Fault Diagnosis of Roller Bearings based on frequency domain window experience small echo resonance and demodulation | |
| Lu et al. | Condition monitoring and fault diagnosis of motor bearings using undersampled vibration signals from a wireless sensor network | |
| Miao et al. | Improvement of kurtosis-guided-grams via Gini index for bearing fault feature identification | |
| CN107461611B (en) | The leakage detection method and leak detecting device combined is reconstructed based on small echo and EMD | |
| Sheen | A complex filter for vibration signal demodulation in bearing defect diagnosis | |
| CN109682601A (en) | The initial failure recognition methods of rolling bearing under a kind of variable speed operating condition | |
| DK2085902T3 (en) | Order tracking method and system | |
| CN103699513B (en) | A kind of stochastic resonance method adjusted based on multiple dimensioned noise | |
| Xu et al. | Periodicity-based kurtogram for random impulse resistance | |
| CN110160765A (en) | A kind of shock characteristic recognition methods and system based on sound or vibration signal | |
| CN110160767A (en) | Impulse period automatic identification and extracting method and system based on Envelope Analysis | |
| Li et al. | Cyclic statistics in rolling bearing diagnosis | |
| CN104316323A (en) | Method for confirming optimal resonance frequency band based on period target | |
| Zhang et al. | Improved local cepstrum and its applications for gearbox and rolling bearing fault detection | |
| CN105510032A (en) | Deconvolution method based on harmonic to noise ratio guidance | |
| CN106596105A (en) | Method, apparatus and system for diagnosing bearing faults | |
| CN111693283B (en) | Bearing unsteady state fault feature non-rotating speed extraction method | |
| Jiang et al. | Differential spectral amplitude modulation and its applications in rolling bearing fault diagnosis | |
| Ding et al. | Multiple instantaneous frequency ridge based integration strategy for bearing fault diagnosis under variable speed operations | |
| CN113702036A (en) | Fault frequency band identification method for vibration signal of speed reducer under variable rotating speed working condition | |
| CN108181098A (en) | A kind of high pedestal jib crane low-speed heave-load unit failure feature extracting method | |
| Henry et al. | Prism signal processing for sensor condition monitoring | |
| Li et al. | Multi-fault diagnosis of rotating machinery via iterative multivariate variational mode decomposition | |
| JP5077847B2 (en) | Reverberation time estimation apparatus and reverberation time estimation method |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220713 Address after: Room 308, building 3, intelligent science park, No. 3963 Susong Road, Hefei Economic and Technological Development Zone, Anhui Province, 230000 Patentee after: Anhui Hagong Zhanlu Technology Equipment Co.,Ltd. Address before: 230601 intelligent equipment science and Technology Park, 3963 Susong Road, Hefei Economic and Technological Development Zone, Anhui Province Patentee before: HRG INTERNATIONAL INSTITUTE FOR RESEARCH & INNOVATION |