CN103542261B - Pipeline leakage acoustic emission signals processing method based on compressed sensing and mask signal method HHT - Google Patents

Pipeline leakage acoustic emission signals processing method based on compressed sensing and mask signal method HHT Download PDF

Info

Publication number
CN103542261B
CN103542261B CN201310459351.4A CN201310459351A CN103542261B CN 103542261 B CN103542261 B CN 103542261B CN 201310459351 A CN201310459351 A CN 201310459351A CN 103542261 B CN103542261 B CN 103542261B
Authority
CN
China
Prior art keywords
signal
acoustic emission
frequency
hht
mask
Prior art date
Application number
CN201310459351.4A
Other languages
Chinese (zh)
Other versions
CN103542261A (en
Inventor
陶然
毕贵红
王�华
司莉
魏永刚
孙云波
胡建航
原天龙
李新仕
梁波
Original Assignee
云南省特种设备安全检测研究院
昆明理工大学
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 云南省特种设备安全检测研究院, 昆明理工大学 filed Critical 云南省特种设备安全检测研究院
Priority to CN201310459351.4A priority Critical patent/CN103542261B/en
Publication of CN103542261A publication Critical patent/CN103542261A/en
Application granted granted Critical
Publication of CN103542261B publication Critical patent/CN103542261B/en

Links

Abstract

The present invention relates to a kind of pipeline leakage acoustic emission signals processing method based on compressed sensing and mask signal method HHT.Mainly can apply in the leakage acoustic emission signal detection of pipeline, boiler, main processing steps: the first step: obtain acoustic emission primary signal, utilize digital filter to filter the high-frequency noise beyond leakage acoustic emission signal frequency domain;Second step: introduce compressive sensing theory and acoustic emission signal is compressed sampling;3rd step: utilize OMP algorithm that compressed signal is carried out Accurate Reconstruction;4th step: use the EMD of mask signal method to decompose acoustic emission signal, the component of different frequency in signal is separated from high frequency to low frequency one by one;5th step: each acoustic emission signal frequency component is done Hilbert conversion, determines the beginning and ending time of acoustic emission signal.

Description

Pipeline leakage acoustic emission signals based on compressed sensing and mask signal method HHT processes Method
Technical field
The present invention relates to a kind of pipeline leakage acoustic emission signals process side based on compressed sensing and mask signal method HHT Method.
Background technology
Signals collecting mode based on tradition Nyqusit sampling thheorem, it is desirable to sample frequency is at least signal highest frequency 2 times.The frequency range of acoustic emission signal be a few KHz to hundreds of KHz, general in reality use multichannel data to adopt at a high speed Collecting system, sample frequency is 5 ~ 10 times of signal highest frequency, necessarily leads to Pang when monitoring acoustic emission signal in real time Big data so that hardware implementation cost is higher, for subsequent data transmission, store and process is exerted heavy pressures on, and this is to leakage The long term monitoring of signal is the most disadvantageous.
Acoustic emission signal is the transient signal containing noisy non-stationary, to the transition acoustic emission containing noisy non-stationary Signal decomposites acoustic emission signal, and determine the start-stop position of the generation of acoustic emission signal be sound emission signal characteristic extract and The key identified.HHT generally can be used to analyze.
HHT (Hilbert-Huang Transform, Hilbert-Huang transform) is by empirical mode decomposition EMD (Empirical Mode Decomposition, empirical mode decomposition) and Hilbert convert two parts composition, and wherein EMD is Core.EMD is a kind of effective treating method for non-stationary signal, and it is from characteristic time scale, in signal Each frequency component decomposes out according to the height of frequency one by one.But, when needing the signal decomposed to there is multiple close frequencies, Modal overlap phenomenon will be there is in traditional E MD decomposition method.The EEMD method proposed afterwards, constantly adds during signal decomposition Enter white noise, make the white noise that in whole time frequency space, uniform distribution is additional, then signal is carried out independent test, use foot All averages of enough tests, noise will be eliminated, and EEMD can solve the modal overlap problem in EMD to a certain extent, but For signal amplitude-frequency than undesirable situation, preferable Detection results still can not be reached.
Summary of the invention
The problem that the object of the invention exists for above classical signal processing method, propose a kind of by compressive sensing theory and The EMD adding mask signal is applied to the processing method in pipeline leakage acoustic emission signals detection technique.
The technical scheme is that: a kind of based on compressed sensing with the pipeline leakage acoustic emission signals of mask signal method HHT Processing method, it is characterised in that: compressive sensing theory is incorporated in acoustic emission signal detection technique, then utilizes and add mask The EMD decomposition method of signal realizes the decomposition to acoustic emission signal, and component does Hilbert conversion, determines what leakage occurred Beginning and ending time, concrete process step is as follows:
The first step: obtain acoustic emission primary signal, utilizes digital filter to filter beyond leakage acoustic emission signal frequency domain High-frequency noise;
Second step: the acoustic emission that the first step is obtained by applied compression perception theory is launched signal and is compressed sampling, with one Individual length, much smaller than the observing matrix perception acoustic emission original signal of primary signal, obtains a group observations;
3rd step: utilize OMP (Orthogonal Matching Pursuit, orthogonal matching pursuit) algorithm to second step The compressed signal obtained is reconstructed, and obtains reconstructing acoustic emission signal;
4th step: use the EMD of mask signal method to decompose acoustic emission signal, by different in the 3rd step reconstruct acoustic emission signal The component of frequency separates from high frequency to low frequency successively;
5th step: each acoustic emission signal frequency component separating the 4th step does Hilbert conversion, determines acoustic emission signal Beginning and ending time.
The described first step and second step are known calculation matrix, a length of N of acoustic emission signal X, and degree of rarefication is K, abides by Follow compressive sensing theory, X is projected to, obtain the one group of length measured value Y much smaller than N, according to measured value and calculation matrix Do inverse operation and just can obtain sparse bayesian learning with reconstruction signal X
Described second step makes signal sampling no longer be limited by Nyquist sampling thheorem, improves efficiency of data compression, reduces Data acquisition, transmission and carrying cost.
3rd step uses OMP algorithm, substantially completes the optimization of first step norm, it is achieved to compression acoustic emission signal Quick Accurate Reconstruction.
Described interpolation mask signal in the 4th step acoustic emission signal, is carried out the acoustic emission signal containing mask signal HHT (Hilbert-Huang Transform, Hilbert-Huang transform).
The mask signal added in the 4th step acoustic emission signal is, i.e. the frequency values of mask signal be adjacent two Altofrequency sum, amplitude is the amplitude of highest frequency.
Described 4th step, when acoustic emission signal comprises N number of near by frequency component, needs constantly to add in catabolic process N-1 mask signal, until last single frequency component remaining.
The acoustic emission signal component decomposited is done Hilbert conversion by described 5th step, it may be determined that its start-stop occurred Time.
A kind of pipeline leakage acoustic emission signals processing method based on compressed sensing and mask signal method HHT, i.e. compresses sense Know that theory is applied to leak in acoustic emission signal detection technique.
The invention has the advantages that: (1) makes signal sampling no longer be limited by Nyquist sampling thheorem, reduces data acquisition Collection cost, improves efficiency of data compression;(2) transmission and the storage problem of big data are solved;(3) effectively suppress in catabolic process The modal overlap phenomenon existed, makes decomposition result more accurately effectively, it is simple to the feature extraction of acoustic emission signal;(4) it is a kind of New and effective acoustic emission signal compression and processing method.
Accompanying drawing explanation
Fig. 1 is the flow chart that patent of the present invention processes acoustic emission signal.
Fig. 2 is that patent compressed sensing of the present invention is to signal sampling compression process figure.
Fig. 3 is patent OMP algorithm flow chart of the present invention.
Fig. 4 is the patent of the present invention mask signal decomposition process figure as a example by IMF1.
Fig. 5 is present invention signal reconstruction based on CS example.
Fig. 6 is that present invention EMD based on mask signal decomposes example.
Detailed description of the invention
Present invention aim at the hyperfrequency in solution acoustic emission signal detection technique and the transmission of big data and storage problem, Obtain accurate decomposition result and be characterized extraction, classify and offer is provided possible.Acoustic emission will be incorporated into by compressive sensing theory In signal detection technique, the EMD decomposition method adding mask signal is then utilized to realize the decomposition to acoustic emission signal, and to dividing Amount does Hilbert conversion, determines the beginning and ending time that leakage occurs, compressive sensing theory;And add the Martin Hilb of mask signal The signal analysis method of spy-Huang.
Concrete processing scheme is:
The first step: obtain the filtering of acoustic emission primary signal.Owing to leakage acoustic emission signal has certain frequency domain, can design Butterworth low pass filter, filters the high-frequency noise beyond leakage acoustic emission signal frequency domain.
Second step: introduce compressive sensing theory.Analyze the openness and non-correlation of signal.Acoustic emission signal is pressed Contracting sampling, by a length much smaller than the observing matrix perception original signal of primary signal, obtains a group observations;
3rd step: compressed signal is reconstructed.(Orthogonal Matching Pursuit, orthogonal coupling chases after OMP Track) algorithm is a kind of modified hydrothermal process on the basis of MP algorithm.Utilize OMP algorithm that compressed signal is carried out Accurate Reconstruction, To reconstruct acoustic emission signal;
4th step: decomposed and reconstituted acoustic emission signal.The EMD adding mask signal method is used to decompose acoustic emission signal, will In signal, the component of different frequency separates with this from high frequency to low frequency.
5th step: each acoustic emission signal frequency component is done Hilbert conversion, determines the beginning and ending time of acoustic emission signal.
Entirety is sent out by leakage Acoustic emission signal processing side based on compressive sensing theory and mask signal method empirical mode decomposition Process is as shown in Figure 1.Leakage acoustic emission signal is during actual propagation, inevitably by various external interference, so that The primary signal collected is mingled with a lot of noise.Owing to the frequency of acoustic emission signal has a range of frequency domain, for signal The too high part of medium frequency can use wave filter to be filtered.The present invention designs a Butterworth wave filter and filters former High-frequency noise in beginning signal.It is then based on compressive sensing theory, filtered signal is compressed sampling, in order to signal Transmission and storage.At the receiving terminal of signal, combine OMP (Orthogonal Matching by less observation Pursuit, orthogonal matching pursuit) the algorithm realization Accurate Reconstruction to signal.Finally reconstruction signal is carried out fast Fourier change Change, analyze the amplitude-frequency characteristic of signal, it is simple to structure mask signal.Mask signal is constantly added, after superposition in reconstruction signal Signal be EMD decompose, draw the single-frequency component after decomposition.Finally each acoustic emission signal frequency component is Hilbert Conversion, determines the beginning and ending time that leakage acoustic emission signal occurs.
Below in conjunction with the accompanying drawings, handling principle and the process of the present invention are further illustrated.
It is illustrated in figure 2 Signal Compression process based on compressed sensing.For signal X, transformed toOn territory, its Being expressed as on this territory:.This is because, the signal for compressed sensing must have openness, and the most certainly So signal is not sparse, and these signals have rarefaction representation on certain special domain,Degree of rarefication be K.With known Observing matrixGo perception, obtain one group of length much smaller than original signal measurement of length value Y, measured value Y is transmitted with During storage, just can save processing cost.Receiving terminal at signal carries out inverse transformation to measured value, it is possible to reconstruction signal.
Fig. 3 is OMP algorithm flow chart.At known observing matrix, in the case of observation Y and degree of rarefication K, can be in order to With OMP algorithm reconstruction signal, show that signal X's approaches value.During initialization, make residual error, sky indexed set and iteration are set Counting.First find out the footmark that residual error is corresponding with observing matrix row inner product maximum, after finding out footmark, update indexed set, record weight Build atom, show that one of signal is approached value, then updates residual values secondly by method of least square, until iterations is more than Being iteration ends equal to degree of rarefication, circulation searching the most always approaches value.
Fig. 4 show the mask signal method EMD decomposition step as a example by IMF1.Need before decomposition signal X is done quickly Fourier transformation, primarily determines that mask signal to be constructed according to its amplitude-frequency characteristic.Concrete catabolic process is as follows:
(1) for analyzed signal, construct mask signal.Wherein, the frequency of mask signal Value is adjacent two highest frequency sums, and amplitude is the amplitude of highest frequency.
(2) rightCarry out EMD decomposition, take its first IMF, be designated as, then haveEMD decomposition takes it First IMF is designated as
(3) average is calculated, result is as the IMF1 of signal decomposition.
When needing the signal decomposed comprises N number of appropriate frequency components, need to add N-1 mask in catabolic process Signal, constantly repeats three above step, until last single frequency component remaining.
According to above theoretical foundation, Fig. 5 is a signal reconstruction example based on CS, it can be seen that reconstruct letter Number the least with the error of primary signal, it is achieved that the Accurate Reconstruction to signal.
Fig. 6 show an EMD based on mask signal and decomposes example.It can be seen that primary signal from decomposition result Being modulated by the signal of three different frequencies and form, the signal of different frequency can substantially make a distinction, and well inhibits mode Aliasing.
The present invention is limited by Nyquist sampling thheorem when having broken classical signal collection, reduces data acquisition cost, The Accurate Reconstruction to acoustic emission signal is realized by restructing algorithm.The EMD resolution adding mask signal effectively suppresses to decompose Modal overlap phenomenon present in journey, makes decomposition result more accurately effectively.

Claims (9)

1. a pipeline leakage acoustic emission signals processing method based on compressed sensing and mask signal method HHT, it is characterised in that: Compressive sensing theory is incorporated in acoustic emission signal detection technique, then utilizes the EMD decomposition method adding mask signal real The now decomposition to acoustic emission signal, and component is done Hilbert conversion, determine the beginning and ending time that leakage occurs, specifically process step Rapid as follows:
The first step: obtain acoustic emission primary signal, utilizes digital filter to filter the high frequency beyond leakage acoustic emission signal frequency domain Noise;
Second step: applied compression perception theory is compressed sampling to acoustic emission signal X that the first step obtains, remote by a length Less than the calculation matrix perception acoustic emission primary signal of primary signal, obtain a group observations;
3rd step: utilize OMP (Orthogonal Matching Pursuit, orthogonal matching pursuit) algorithm that second step is obtained Compressed signal X ' be reconstructed, obtain reconstruct acoustic emission signal;
4th step: use the EMD of mask signal method to decompose acoustic emission signal, by different frequency in the 3rd step reconstruct acoustic emission signal Component separate successively from high frequency to low frequency;
5th step: each acoustic emission signal frequency component separating the 4th step does Hilbert conversion, determines rising of acoustic emission signal The only time.
Pipeline leakage acoustic emission signals based on compressed sensing and mask signal method HHT the most according to claim 1 processes Method, it is characterised in that: the first step and second step known calculation matrix Φ, the acoustic emission signal X length that the first step obtains For N, degree of rarefication is K, it then follows compressive sensing theory, projects on calculation matrix Φ by acoustic emission signal X that the first step obtains, To one group of length much smaller than the measured value Y of N, do inverse operation according to measured value and calculation matrix and just can reconstruct what second step obtained Compressed signal X ', obtains sparse bayesian learning
At pipeline leakage acoustic emission signals based on compressed sensing and mask signal method HHT the most according to claim 1 and 2 Reason method, it is characterised in that: described second step makes signal sampling no longer be limited by Nyquist sampling thheorem, improves data pressure Contracting efficiency, reduces data acquisition, transmission and carrying cost.
Pipeline leakage acoustic emission signals based on compressed sensing and mask signal method HHT the most according to claim 1 processes Method, it is characterised in that: the 3rd step uses OMP algorithm, substantially completes the optimization of first step norm, it is achieved to second step The compressed signal quick Accurate Reconstruction of X ' obtained.
Pipeline leakage acoustic emission signals based on compressed sensing and mask signal method HHT the most according to claim 1 processes Method, it is characterised in that: in the 4th step acoustic emission signal, add mask signal, the acoustic emission signal containing mask signal is entered Row HHT Hilbert-Huang transform.
The most according to claim 1 or 5 at pipeline leakage acoustic emission signals based on compressed sensing and mask signal method HHT Reason method, it is characterised in that: the mask signal added in the 4th step acoustic emission signal is { S1(t)……Sn-1(t) }, i.e. The frequency values of mask signal is adjacent two highest frequency sums, and amplitude is the amplitude of highest frequency.
Pipeline leakage acoustic emission signals based on compressed sensing and mask signal method HHT the most according to claim 1 processes Method, it is characterised in that: the 4th step, when acoustic emission signal comprises N number of near by frequency component, needs in catabolic process constantly Add N-1 mask signal, until last single frequency component remaining.
Pipeline leakage acoustic emission signals based on compressed sensing and mask signal method HHT the most according to claim 1 processes Method, it is characterised in that: the acoustic emission signal component decompositing the 5th step does Hilbert conversion, it may be determined that it occurs Beginning and ending time.
9. according to the pipe leakage acoustic emission based on compressed sensing and mask signal method HHT described in any one of claim 1-8 Signal processing method is applied to pipeline leakage testing.
CN201310459351.4A 2013-10-07 2013-10-07 Pipeline leakage acoustic emission signals processing method based on compressed sensing and mask signal method HHT CN103542261B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310459351.4A CN103542261B (en) 2013-10-07 2013-10-07 Pipeline leakage acoustic emission signals processing method based on compressed sensing and mask signal method HHT

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310459351.4A CN103542261B (en) 2013-10-07 2013-10-07 Pipeline leakage acoustic emission signals processing method based on compressed sensing and mask signal method HHT

Publications (2)

Publication Number Publication Date
CN103542261A CN103542261A (en) 2014-01-29
CN103542261B true CN103542261B (en) 2016-08-10

Family

ID=49965995

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310459351.4A CN103542261B (en) 2013-10-07 2013-10-07 Pipeline leakage acoustic emission signals processing method based on compressed sensing and mask signal method HHT

Country Status (1)

Country Link
CN (1) CN103542261B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104035434A (en) * 2014-06-13 2014-09-10 武汉理工大学 Air leakage monitoring system for diesel engine air valve
CN104747912B (en) * 2015-04-23 2017-04-12 重庆邮电大学 Fluid conveying pipe leakage acoustic emission time-frequency positioning method
CN109140241B (en) * 2018-08-21 2019-10-29 吉林大学 A kind of compressed sensing based pipeline leakage positioning method
CN109813417A (en) * 2019-01-18 2019-05-28 国网江苏省电力有限公司检修分公司 A kind of shunt reactor method for diagnosing faults based on improvement EMD
CN110131595B (en) * 2019-05-21 2020-09-04 北京化工大学 Method, device and system for processing pipeline slow leakage signal
CN110553156B (en) * 2019-08-30 2020-09-25 清华大学 Pipeline ultrasonic guided wave compression sensing health monitoring method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2258865C1 (en) * 2004-01-14 2005-08-20 Ефремов Александр Иванович Method of detecting location of leakage in pipelines
CN101230953A (en) * 2008-01-24 2008-07-30 深圳东方锅炉控制有限公司 Pipeline leakage detecting system and pipeline leakage detecting system with remote monitoring
CN102588745A (en) * 2012-03-05 2012-07-18 北京化工大学 Pipeline leakage positioning method
CN102588747A (en) * 2012-03-23 2012-07-18 中国人民解放军重庆通信学院 Online leakage monitoring method for pipelines on basis of burst type acoustic signal detection technology

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102235575B (en) * 2010-04-29 2013-12-25 国际商业机器公司 Data processing method and system for checking pipeline leakage

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2258865C1 (en) * 2004-01-14 2005-08-20 Ефремов Александр Иванович Method of detecting location of leakage in pipelines
CN101230953A (en) * 2008-01-24 2008-07-30 深圳东方锅炉控制有限公司 Pipeline leakage detecting system and pipeline leakage detecting system with remote monitoring
CN102588745A (en) * 2012-03-05 2012-07-18 北京化工大学 Pipeline leakage positioning method
CN102588747A (en) * 2012-03-23 2012-07-18 中国人民解放军重庆通信学院 Online leakage monitoring method for pipelines on basis of burst type acoustic signal detection technology

Also Published As

Publication number Publication date
CN103542261A (en) 2014-01-29

Similar Documents

Publication Publication Date Title
Jing-Yi et al. A new wavelet threshold function and denoising application
CN104655423B (en) A kind of Fault Diagnosis of Roller Bearings based on time-frequency domain multi-dimensional vibration Fusion Features
CN102435934B (en) Random sampling analog circuit compressed sensing measurement and signal reconstruction method
CN104523266A (en) Automatic classification method for electrocardiogram signals
Zhang et al. Time-varying singular value decomposition for periodic transient identification in bearing fault diagnosis
CN105827250B (en) A kind of power quality data compression reconfiguration method based on self-adapting dictionary study
CN103983850B (en) Harmonious Waves in Power Systems compressed signal based on compressed sensing reconstruct and detection method
CN104729853A (en) Rolling bearing performance degradation evaluation device and method
CN103902844A (en) Transformer vibration signal de-noising method based on EEMD kurtosis threshold value
CN102539150B (en) Self-adaptive failure diagnosis method of rotary mechanical component based on continuous wavelet transformation
Gao et al. Irregular seismic data reconstruction based on exponential threshold model of POCS method
CN107461611B (en) The leakage detection method and leak detecting device combined is reconstructed based on small echo and EMD
Tang et al. Sparse classification of rotating machinery faults based on compressive sensing strategy
CN108168891B (en) Method and equipment for extracting weak fault signal characteristics of rolling bearing
CN102680860B (en) Automatic fault-point locating method for traveling-wave based fault location of high-voltage electric power lines
Shao et al. Order tracking by discrete Gabor expansion
CN104473631B (en) A kind of based on non-negative blind separation Fetal ECG instantaneous heart rate recognition methods and system
CN107727399B (en) Fault Diagnosis of Roller Bearings based on dual-tree complex wavelet packet stream shape domain noise reduction
CN109781412B (en) Rolling bearing self-adaptive resonance demodulation method based on EEMD
CN103293515B (en) Ship and warship line spectrum noise source longitudinal distribution characteristic measuring method
CN104392427B (en) SAR (synthetic aperture radar) image denoising method combining empirical mode decomposition with sparse representation
CN103156599B (en) Detection method of electrocardiosignal R characteristic waves
CN105181122B (en) Mechanical oscillation signal data compression acquisition method
CN105266800B (en) One kind is based on fetus electrocardio blind separation under Low SNR
Lu et al. Two-dimensional digital filters with sparse coefficients

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20160810

Termination date: 20181007

CF01 Termination of patent right due to non-payment of annual fee