CN108645920A - A kind of direct wave suppressing method of the rail flaw ultrasonic detection based on denoising and alignment - Google Patents

A kind of direct wave suppressing method of the rail flaw ultrasonic detection based on denoising and alignment Download PDF

Info

Publication number
CN108645920A
CN108645920A CN201810314454.4A CN201810314454A CN108645920A CN 108645920 A CN108645920 A CN 108645920A CN 201810314454 A CN201810314454 A CN 201810314454A CN 108645920 A CN108645920 A CN 108645920A
Authority
CN
China
Prior art keywords
waveform
direct wave
interception
wave
length
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
Application number
CN201810314454.4A
Other languages
Chinese (zh)
Other versions
CN108645920B (en
Inventor
韦岗
资灵薇
曹燕
杨萃
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China University of Technology SCUT
Original Assignee
South China University of Technology SCUT
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 South China University of Technology SCUT filed Critical South China University of Technology SCUT
Priority to CN201810314454.4A priority Critical patent/CN108645920B/en
Publication of CN108645920A publication Critical patent/CN108645920A/en
Application granted granted Critical
Publication of CN108645920B publication Critical patent/CN108645920B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N29/00Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
    • G01N29/44Processing the detected response signal, e.g. electronic circuits specially adapted therefor
    • G01N29/48Processing the detected response signal, e.g. electronic circuits specially adapted therefor by amplitude comparison
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N29/00Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
    • G01N29/44Processing the detected response signal, e.g. electronic circuits specially adapted therefor
    • G01N29/4454Signal recognition, e.g. specific values or portions, signal events, signatures
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N29/00Investigating or analysing materials by the use of ultrasonic, sonic or infrasonic waves; Visualisation of the interior of objects by transmitting ultrasonic or sonic waves through the object
    • G01N29/44Processing the detected response signal, e.g. electronic circuits specially adapted therefor
    • G01N29/4472Mathematical theories or simulation
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2291/00Indexing codes associated with group G01N29/00
    • G01N2291/02Indexing codes associated with the analysed material
    • G01N2291/023Solids
    • G01N2291/0234Metals, e.g. steel
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2291/00Indexing codes associated with group G01N29/00
    • G01N2291/02Indexing codes associated with the analysed material
    • G01N2291/028Material parameters
    • G01N2291/0289Internal structure, e.g. defects, grain size, texture
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2291/00Indexing codes associated with group G01N29/00
    • G01N2291/26Scanned objects
    • G01N2291/262Linear objects
    • G01N2291/2623Rails; Railroads

Abstract

The present invention provides a kind of direct wave suppressing method of the rail flaw ultrasonic detection based on denoising and alignment.This method directly utilizes the reception waveform of single measurement point, is first blocked to receiving waveform progress waveform, part of the interception comprising complete direct wave receives waveform.Based on this maximum reality of the wave energy that goes directly in waveform is received, the direct wave part in interception waveform is extracted using singular value decomposition denoising;It is aligned interception waveform and direct wave again, generates extension direct wave, finally subtracts extension direct wave waveform from reception waveform, obtains the reception signal of removal direct wave.Calculation amount of the present invention is small and can handle in real time, and the direct wave that can be effectively realized in ultrasonic inspection inhibits, and improves the signal-to-noise ratio of back wave.

Description

A kind of direct wave suppressing method of the rail flaw ultrasonic detection based on denoising and alignment
Technical field
The present invention relates to ultrasonic inspection field, more particularly to a kind of rail flaw ultrasonic detection based on denoising and alignment goes directly Wave suppressing method.
Background technology
In recent years, railway transportation occupies very important status in China's transportation system.Either growing object Stream demand or increasing personnel circulation, the importance of railway transportation more enhance.Carrying of the rail as railway transportation Component directly influences the safety of railway transportation.But rail can be caused to rub and lose for a long time by being open to traffic, in addition nature calamity Evil and atrocious weather are easy that rail is made to generate deformation or defect fracture etc..Therefore regularly rail examination is to ensure railway fortune Defeated safe important link.Currently, main steel rail flaw detection technology has Ultrasonic Flaw Defect and electromagnetic induction.Rail ultrasound is visited Wound is to emit ultrasonic wave in rail side, and ultrasonic wave reflects when encountering medium, using the amplitude size of back wave, phase, The information such as time delay can determine that the information such as the presence or absence of defect in rail, position.
Rail flaw ultrasonic detection in the application there is also some critical problems have it is to be solved, in connecing for rail ultrasonic inspection The direct-coupling part for including transmitted waveform in waveform, i.e. direct wave are received, compared with the reflection configuration that detection target generates, is gone directly Wave arrival time is earliest, and energy is maximum and its energy may much be better than back wave so that back wave is very faint, usually floods Not in direct wave and noise, direct wave will be to waveform formation severe jamming.Therefore, the reliable of rail flaw ultrasonic detection is improved Property, inhibit direct wave to be of great importance rail flaw ultrasonic detection.
The direct wave minimizing technology of major part scholar research is the progress " supple-settlement " handled by digital waveform at present, such as Wavelet transformation, artificial neural network, adaptive-filtering etc..Wavelet transformation and artificial neural network need multiple DATA REASONING, ginseng Number requires height, and due to being influenced by systematic jitters and environmental complexity etc., cannot handle in real time.Adaptive-filtering calculates essence Degree is difficult to ensure that noiseproof feature is poor, can offset part back wave simultaneously in the case that direct wave and back wave degree of overlapping are higher, Reduce signal-to-noise ratio.
It is to receive waveform to whole section to carry out waveform processing mostly in the above method.But the whole section of waveform that ultrasonic inspection receives It can be divided into:Pure noise section;Direct wave, back wave and noise aliased portion, abbreviation aliased portion;Noise-containing back wave Part.It can be seen that only part back wave, by direct-wave jamming, if receiving, Wave data amount is big, and handling whole segment data can cause to calculate Amount rises.Therefore in processing waveform advance row waveform interception, part of the interception comprising complete direct wave receives waveform and reprocesses number According to calculation amount can be effectively reduced.
Compared with direct wave waveform, the energy for the back wave that waveform obtains after being reflected through target becomes very faint, directly Much it is better than back wave up to wave energy.It is far longer than the original of back wave and noise energy based on the wave energy that goes directly in reception waveform Back wave is also regarded noise by reason, and receiving the noise in waveform by removal extracts direct wave.It can be considered to use denoising Mode obtains direct wave, proposes that a kind of direct wave of rail flaw ultrasonic detection based on denoising and alignment inhibits based on this this patent Method first uses the extraction of singular value decomposition (SVD) denoising method to receive the direct wave waveform in waveform, then do after time adjustment with original Waveform subtract each other to get to eliminate direct wave reception waveform.
For extracting direct wave with singular value decomposition denoising, there is researcher to propose the strong phase using each measurement point direct wave Guan Xing, acquire multiple measurement points reception waveform be organized into matrix after, carry out singular value decomposition and obtain direct wave, then by direct wave Inhibition direct wave is offseted by sef-adapting filter with the reception waveform of each measurement point respectively.If in this each measurement point of method Direct wave arrival time has mistake, the direct wave that singular value decomposition obtains that will have very big error.Singular value decomposition is obtained straight It is offseted respectively by sef-adapting filter inhibition with the reception waveform of each measuring point amount up to wave, essence, which is dynamic estimation, makes the two most The time delay of the direct wave waveform of small mean square error minimum, and filter calculating method is complicated and computationally intensive.Based on this, carry herein A kind of waveform received using single measurement point is gone out, the direct wave of the rail flaw ultrasonic detection based on waveform denoising and alignment inhibits Method.
Invention content
Caused reflection wave extraction difficulty increase is covered by direct wave waveform to solve back wave in ultrasonic inspection The problem of, the present invention provides a kind of direct wave suppressing method of the rail flaw ultrasonic detection based on denoising and alignment, and this method calculates It measures small and can handle in real time, the direct wave that can be effectively realized in ultrasonic inspection inhibits, and improves the signal-to-noise ratio of back wave.
The present invention is realized using following technical scheme.
A kind of direct wave suppressing method of the rail flaw ultrasonic detection based on denoising and alignment, directly connecing using single measurement point Waveform is received, is first blocked to receiving waveform progress waveform, part of the interception comprising complete direct wave receives waveform.Based on receiving wave This maximum reality of the wave energy that goes directly in shape, the direct wave part in interception waveform is extracted using singular value decomposition denoising.Again Alignment interception waveform and direct wave generate extension direct wave, finally subtract extension direct wave waveform from reception waveform, are removed The reception signal of direct wave.
Based on the above technical solution, the reception waveform includes direct wave, back wave and noise, wherein direct wave Energy is most the decades of times of back wave and noise.Receiving waveform can be divided into:Pure noise section;Direct wave, back wave and noise are mixed Folded part, abbreviation aliased portion;Noise-containing back wave part.
Based on the above technical solution, it is to intercept the part received wave for including complete direct wave that the waveform, which blocks, Shape.To ensure that the waveform of interception includes complete direct wave, intercept method is as follows:Starting point is intercepted to shift to an earlier date in the direct wave point of arrival, The one section waveform longer than transmitted waveform length is intercepted backward since intercepting starting point as interception waveform;Assuming that transmitted waveform length For N, interception waveform length is M, and interception starting point is t, and direct wave reaches estimation time point as T, can use interception starting point t=0.8T, Intercept waveform length M=1.5N.
Based on the above technical solution, the singular value decomposition denoising is that pending waveform is reorganized into a square Battle array carries out SVD to the matrix and converts to obtain singular value matrix, and larger singular value corresponds to direct wave waveform, smaller singular value pair Answer back wave and noise.Therefore by smaller singular value zero setting, going directly in waveform is received finally by SVD inverse transformations are extractable Wave waveform removes noise and back wave.
Based on the above technical solution, the direct wave suppression of a kind of rail flaw ultrasonic detection based on denoising and alignment Method specific implementation step processed is as follows:
(1) it to the reception waveform of single measurement point, being pre-processed, part of the interception comprising complete direct wave receives waveform, The length of M.
Intercept method is as follows:Starting point is intercepted in advance in the direct wave point of arrival, is intercepted backward since intercepting starting point than transmitting One section of waveform of waveform length length is as interception waveform.Assuming that it is L to receive waveform length, transmitted waveform length is N, intercepts waveform Length:M(M<L), interception starting point is t, and direct wave reaches estimation time point as T, can use interception starting point t=0.8T, can take M= 1.5N。
(2) noise and back wave are removed by singular value decomposition denoising to interception waveform, extracts direct wave waveform, length For M.
Assuming that interception waveform is r (n), wherein n=0,1 ..., M-1 in (1), it is Hankel squares that r (n), which is reorganized, Battle array,
Wherein q+1 is the line number of above-mentioned matrix, and q is bigger, and direct wave waveform extracting precision is higher, but calculation amount increases. Then above-mentioned matrix is converted into SVD, obtains its singular value matrix, then will be less than the singular value zero setting of all singular value mean values. The direct wave waveform in interception waveform is obtained finally by SVD inverse transformations.
(3) the time unifying processing for intercepting waveform and direct wave, generates the extension direct wave just as sample length with received wave Waveform.
Assuming that it is L to receive waveform length, the direct wave length that denoising obtains is M, and the direct wave point of arrival is x after alignment.Expand It opens up in direct wave waveform, it is 0 that the data on time point [1, x-1] and [x+M+1, L], which are mended, and [x, x+M] data are direct wave waveform Data (length M), extension direct wave length are L.
(4) it receives waveform and subtracts extension direct wave waveform, obtain the reception signal of removal direct wave.
Compared with prior art, it is an advantage of the invention that:
(1) the direct wave suppressing method of a kind of rail flaw ultrasonic detection based on denoising and alignment of the invention, is not change Become on the basis of target echo, extracts direct wave with the mode of denoising, inhibit the interference of direct wave in echo, can handle in real time And improve the accuracy of rail flaw ultrasonic detection.
(2) direct wave directly is extracted using SVD denoisings using after the data configuration matrix of single measurement point, comparison is surveyed more using The data for measuring point, computation amount and avoid error caused by time of occurrence mistake in multipoint data.
(3) part of the interception comprising complete direct wave receives waveform and carries out processing data, can effectively reduce calculation amount.
Description of the drawings
Fig. 1 is that direct wave generates schematic diagram in the embodiment of the present invention.
Fig. 2 is that ultrasonic inspection receives waveform diagram in the embodiment of the present invention.
Fig. 3 is a kind of direct wave suppressing method implementation steps of the rail flaw ultrasonic detection based on denoising and alignment of the present invention Figure.
Fig. 4 a are the reception waveform diagram of single measurement point.
Fig. 4 b are to receive waveform to single measurement point to carry out the interception waveform diagram after waveform blocks.
Fig. 4 c are the direct wave waveform extracted using singular value decomposition denoising in interception waveform.
Fig. 4 d are the extension direct wave waveform after time unifying.
Fig. 4 e are the result figure for receiving waveform and subtracting extension direct wave waveform.
Specific implementation mode
The present embodiment combination attached drawing is described further the specific implementation mode of the present invention, but the implementation of the present invention is unlimited In this.
As shown in Figure 1, being that direct wave of the present invention generates schematic diagram.In rail detection, transmitted wave passes through the ground of rail Or defect reflection back wave is received probe and receives, but receiving transducer is in addition to receiving back wave, also direct wave and noise, Wherein direct wave is the signal that transmitting probe is directly coupled to receiving transducer.Direct wave arrival time is earliest, energy it is maximum and Its energy may much be better than back wave so that back wave is very faint, is usually submerged in direct wave and noise, and direct wave is just It can be to waveform formation severe jamming.
As shown in Fig. 2, being that ultrasonic inspection of the present invention receives waveform diagram.Receiving waveform can be divided into:Pure noise section;Directly Up to wave, back wave and noise aliased portion, abbreviation aliased portion;Noise-containing back wave part.It is interception wave in dotted line frame Shape schematic diagram.
As shown in figure 3, being implementation steps figure of the present invention, it is as follows:
Step 1:Waveform blocks.The energy feature that direct wave reaches front and back received wave has extremely big difference, therefore can To judge the direct wave point of arrival using this characteristic.Direct wave reach before received wave amplitude very little, therefore its energy and Smaller, after direct wave reaches, amplitude increases, and energy and also becomes larger therewith.Therefore, (straight in window when two near direct wave Up to wave arrival time it is latter when window with it is previous when window) energy ratio can be very big, as long as finding the ceiling capacity received on waveform Ratio point, that is, direct wave the point of arrival.Energy ratio calculation formula is as follows:
Wherein t0Window starting point, t when being previous1When being previous window terminal and when the latter window starting point, t2Window when being latter Terminal, when window width τ=t1-t0=t2-t1, when window width can be arranged according to sampling precision, window moving step length be 0.5 τ.For It guarantees data integrity, it is that direct wave reaches estimation time point that the present invention, which takes ceiling capacity ratio point, and interception starting point is in advance in straight Estimation time point is reached up to wave, intercepts the one section waveform longer than transmitted waveform length backward since intercepting starting point as interception wave Shape.Assuming that it is L to receive waveform length, transmitted waveform length is N, and interception starting point is t, and direct wave arrival estimation time point is T, is cut Take waveform length:M(M<L), M=1.5N, t=T-2 τ can be taken.
Step 2:SVD denoisings.Based on the maximum characteristic of wave energy of going directly in waveform is received, back wave is also considered as noise, The method denoising for ignoring component of degree n n using singular value decomposition extraction principal component, extracts direct wave.To step 1 intercept waveform again Hankel matrix R are configured to, it is maximum due to receiving the wave energy that goes directly in waveform, it is far longer than reflection and involves noise etc., therefore Direct wave information accounts for major part in matrix R.Assuming that the waveform that step 1 intercepts is r (n), wherein n=0,1 ..., M-1, by r (n) It is matrix R to reorganize:
Wherein q+1 is the line number of above-mentioned matrix, and q is bigger, and direct wave waveform extracting precision is higher, but calculation amount increases. Then SVD transformation is carried out to matrix R, obtains the singular value of matrix R.The singular value of matrix R, which can represent, receives direct wave in waveform Situation is concentrated with the energy of indirect wave, and singular value meets δ1≥δ2≥…≥δr>=0, larger singular value (is more than singular value Mean value) the prodigious direct wave of energy in corresponding waveform, in smaller singular value (being less than singular value mean value) corresponding waveform energy compared with Small back wave and noise.By less than the singular value zero setting of all singular value mean values, then SVD inverse transformations are carried out, then reconstructs and cut Take the direct wave waveform that energy is larger in waveform, length M.
Step 3:Time unifying is extended with waveform.The direct wave and practical direct wave that step 2 is rebuild may have in time delay Little difference, starting point is not quite identical, to ensure that precision need to adjust starting point.Therefore by a small range by the direct wave wave of reconstruction Shape lag or in advance a fixed step size, accurate starting point is found by searching algorithm:The reconstruction direct wave that will be late by or shift to an earlier date every time Waveform and the reception waveform of interception carry out absolute difference operation point by point, and obtain the mean value of absolute difference sequence.In range After the completion of interior search, by the lag of gained absolute difference serial mean minimum or transfer point is considered as starting point time alignment in advance.
Assuming that ranging from [- k, the k] of delay, interception waveform is r (n), and the direct wave waveform that step 2 is rebuild is d (n), long Degree is M, and calculation expression is as follows:
Wherein m ∈ [- k, k], d (n) is front and back in calculating process mends k zero.It is starting point to acquire the corresponding m of above formula minimum value Time alignment, will direct wave waveform to be delayed after m be alignment as a result, i.e. direct wave accurate starting point.
Assuming that direct wave arrival time point is x, x=t+m in receiving waveform after alignment (t is the interception starting point of step 1). It extends in direct wave waveform, it is 0 that the data on time point [1, x-1] and [x+M+1, L], which are mended, and [x, x+M] data are direct wave wave Graphic data (length M), extension direct wave length are L.
Step 4:It receives waveform and subtracts extension direct wave waveform, obtain the reception signal of removal direct wave.
It is that each step of the present invention corresponds to oscillogram as shown in Fig. 4 a~Fig. 4 e.Fig. 4 a are the reception waveform of single measurement point Schematic diagram.Fig. 4 b are to receive waveform to single measurement point to carry out the interception waveform diagram after waveform blocks.Fig. 4 c are using strange Different value decomposes denoising and extracts the direct wave waveform intercepted in waveform.Fig. 4 d are the extension direct wave waveform after time unifying. Fig. 4 e are the result for receiving waveform and subtracting extension direct wave waveform.
The above embodiment is a preferred embodiment of the present invention, but embodiments of the present invention are not by above-described embodiment Limitation, it is other it is any without departing from the spirit and principles of the present invention made by changes, modifications, substitutions, combinations, simplifications, Equivalent substitute mode is should be, protection scope of the present invention is included in.

Claims (5)

1. a kind of direct wave suppressing method of the rail flaw ultrasonic detection based on denoising and alignment, it is characterised in that directly surveyed using single The reception waveform of point is measured, is first blocked to receiving waveform progress waveform, part of the interception comprising complete direct wave receives waveform;It receives To waveform in the wave energy that goes directly it is maximum, extract the direct wave part in interception waveform using singular value decomposition denoising;It is right again It is neat to take waveform and direct wave, extension direct wave is generated, finally extension direct wave waveform is subtracted from reception waveform, it is straight to obtain removal Up to the reception signal of wave.
2. the direct wave suppressing method of the rail flaw ultrasonic detection according to claim 1 based on denoising and alignment, feature exist It is that part reception waveform of the interception comprising complete direct wave carries out data processing again to be blocked in the waveform;To ensure the wave of interception Shape includes complete direct wave, and intercept method is as follows:Intercept starting point in advance in the direct wave point of arrival, since intercepting starting point backward The interception one section waveform longer than transmitted waveform length is as interception waveform;Assuming that transmitted waveform length is N, interception waveform length is M, interception starting point are t, and direct wave reaches estimation time point as T, can use interception starting point t=0.8T, interception waveform length M= 1.5N。
3. the direct wave suppressing method of the rail flaw ultrasonic detection according to claim 1 based on denoising and alignment, feature exist It is that single measurement point is directly received to waveform to carry out after waveform blocks, being reorganized into a matrix in the singular value decomposition denoising, Singular value decomposition is carried out to the matrix to convert to obtain singular value matrix, can extract received wave finally by singular value decomposition inverse transformation Direct wave waveform in shape removes noise and back wave.
4. the direct wave suppressing method of the rail flaw ultrasonic detection according to claim 1 based on denoising and alignment, feature exist Specific implementation step is as follows:
(1) it to the reception waveform of single measurement point, is pre-processed, part of the interception comprising complete direct wave receives waveform, length Degree is M;
(2) noise and back wave are removed by singular value decomposition denoising to interception waveform, extracts direct wave waveform, the length of M;
Assuming that interception waveform is r (n), wherein n=0,1 ..., M-1 in (1), interception waveform length is M, and r (n) is reorganized For Hankel matrixes,
Wherein q+1 is the line number of above-mentioned matrix, and q is bigger, and direct wave waveform extracting precision is higher, but calculation amount increases;Then Singular value decomposition is carried out to above-mentioned matrix, obtains its singular value matrix, then will be less than the singular value zero setting of all singular value mean values; The direct wave waveform in interception waveform is obtained finally by SVD inverse transformations;
(3) the time unifying processing for intercepting waveform and direct wave, generates the extension direct wave waveform just as sample length with received wave;
Assuming that it is L to receive waveform length, the direct wave length that denoising obtains is M, and the direct wave point of arrival is x after alignment;Extension is straight Up in wave waveform, it is 0 that the data on time point [1, x-1] and [x+M+1, L], which are mended, and [x, x+M] data are through waveform data, Extension direct wave length is L;
(4) it receives waveform and subtracts extension direct wave waveform, obtain the reception signal of removal direct wave.
5. the direct wave suppressing method of the rail flaw ultrasonic detection according to claim 4 based on denoising and alignment, feature exist Intercept method is as follows:Step (1) interception specifically includes:Starting point is intercepted in advance in the direct wave point of arrival, is opened from interception starting point Begin to intercept the one section waveform longer than transmitted waveform length backward as interception waveform;Assuming that it is L, transmitted waveform to receive waveform length Length is N, and interception waveform length is M;M<L, interception starting point are t, and direct wave reaches estimation time point as T, can use interception starting point t =0.8T can take M=1.5N.
CN201810314454.4A 2018-04-09 2018-04-09 Denoising and alignment-based direct wave suppression method for ultrasonic flaw detection of steel rail Expired - Fee Related CN108645920B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810314454.4A CN108645920B (en) 2018-04-09 2018-04-09 Denoising and alignment-based direct wave suppression method for ultrasonic flaw detection of steel rail

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810314454.4A CN108645920B (en) 2018-04-09 2018-04-09 Denoising and alignment-based direct wave suppression method for ultrasonic flaw detection of steel rail

Publications (2)

Publication Number Publication Date
CN108645920A true CN108645920A (en) 2018-10-12
CN108645920B CN108645920B (en) 2020-12-22

Family

ID=63745687

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810314454.4A Expired - Fee Related CN108645920B (en) 2018-04-09 2018-04-09 Denoising and alignment-based direct wave suppression method for ultrasonic flaw detection of steel rail

Country Status (1)

Country Link
CN (1) CN108645920B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110018429A (en) * 2019-03-29 2019-07-16 中国科学院电子学研究所 A kind of method and system for eliminating magnetic detection Platform Vibration interference magnetic field
CN112491999A (en) * 2020-11-18 2021-03-12 成都佳华物链云科技有限公司 Data reporting method and device

Citations (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0188335A2 (en) * 1985-01-15 1986-07-23 The Babcock & Wilcox Company Ultrasonic testing of materials
CN101071175A (en) * 2006-05-11 2007-11-14 中国石油集团东方地球物理勘探有限责任公司 Zero hypocentral distance vertical seismic section compressional-shear wave data depth field corridor stacked section processing method
US20090048789A1 (en) * 2007-04-13 2009-02-19 University Of South Carolina Optimized Embedded Ultrasonics Structural Radar System With Piezoelectric Wafer Active Sensor Phased Arrays For In-Situ Wide-Area Damage Detection
US20090122889A1 (en) * 2007-11-09 2009-05-14 Samsung Electronics Co., Ltd. Method and apparatus for decomposing channel in closed-loop multiple input multiple output communication system
CN102193107A (en) * 2010-03-05 2011-09-21 西安石油大学 Method for separating and denoising seismic wave field
CN102230915A (en) * 2011-04-07 2011-11-02 华南理工大学 Device and method for flaw detection of small crawler-type rails based on three-way ultrasonic diffraction
CN102608219A (en) * 2012-03-21 2012-07-25 华南理工大学 Device for expanding ultrasonic detection region and increasing detection precision and method
US20130094729A1 (en) * 2011-10-12 2013-04-18 University Of Virginia Patent Foundation Singular value filter for imaging or detection
CN104135767A (en) * 2014-08-19 2014-11-05 电子科技大学 Subsection mutual correlation method for measuring arrival time difference of signal direct waves
EP2872884A1 (en) * 2012-07-12 2015-05-20 Commissariat à l'Énergie Atomique et aux Énergies Alternatives Method for processing signals acquired by echo sounding, and corresponding computer program and echo sounding device
JP5742513B2 (en) * 2011-06-29 2015-07-01 Jfeスチール株式会社 Ultrasonic flaw detection method and ultrasonic flaw detection apparatus
CN105607125A (en) * 2016-01-15 2016-05-25 吉林大学 Seismic data noise suppression method based on block matching algorithm and singular value decompression
US20160157828A1 (en) * 2014-06-05 2016-06-09 Chikayoshi Sumi Beamforming method, measurement and imaging instruments, and communication instruments
CN105929375A (en) * 2016-07-11 2016-09-07 长沙太电子科技有限公司 Method for suppressing through-wall radar coupling signals based on minimum mean square error criterion
WO2016142692A1 (en) * 2015-03-06 2016-09-15 Micromass Uk Limited Spectrometric analysis
CN106094033A (en) * 2016-06-05 2016-11-09 吉林大学 The orientation seismic beam forming method of singular value decomposition
US9618597B2 (en) * 2013-01-04 2017-04-11 Siemens Aktiengesellschaft Method and magnetic resonance apparatus for automated analysis of the raw data of a spectrum
CN106864477A (en) * 2017-03-15 2017-06-20 华南理工大学 A kind of rail flaw ultrasonic detection intelligent detecting method based on time-frequency do-nothing function
CN107024535A (en) * 2016-03-30 2017-08-08 北京工业大学 A kind of multiple index depth detection method of the vertical defect based on surface wave
CN107704831A (en) * 2017-10-11 2018-02-16 山东科技大学 A kind of gas density data noise reduction based on singular value decomposition median method
CN107884190A (en) * 2017-11-01 2018-04-06 北京信息科技大学 The Method for Bearing Fault Diagnosis decomposed based on variation mode decomposition and wavelet singular

Patent Citations (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0188335A2 (en) * 1985-01-15 1986-07-23 The Babcock & Wilcox Company Ultrasonic testing of materials
CN101071175A (en) * 2006-05-11 2007-11-14 中国石油集团东方地球物理勘探有限责任公司 Zero hypocentral distance vertical seismic section compressional-shear wave data depth field corridor stacked section processing method
US20090048789A1 (en) * 2007-04-13 2009-02-19 University Of South Carolina Optimized Embedded Ultrasonics Structural Radar System With Piezoelectric Wafer Active Sensor Phased Arrays For In-Situ Wide-Area Damage Detection
US20090122889A1 (en) * 2007-11-09 2009-05-14 Samsung Electronics Co., Ltd. Method and apparatus for decomposing channel in closed-loop multiple input multiple output communication system
CN102193107A (en) * 2010-03-05 2011-09-21 西安石油大学 Method for separating and denoising seismic wave field
CN102230915A (en) * 2011-04-07 2011-11-02 华南理工大学 Device and method for flaw detection of small crawler-type rails based on three-way ultrasonic diffraction
JP5742513B2 (en) * 2011-06-29 2015-07-01 Jfeスチール株式会社 Ultrasonic flaw detection method and ultrasonic flaw detection apparatus
US20130094729A1 (en) * 2011-10-12 2013-04-18 University Of Virginia Patent Foundation Singular value filter for imaging or detection
CN102608219A (en) * 2012-03-21 2012-07-25 华南理工大学 Device for expanding ultrasonic detection region and increasing detection precision and method
EP2872884A1 (en) * 2012-07-12 2015-05-20 Commissariat à l'Énergie Atomique et aux Énergies Alternatives Method for processing signals acquired by echo sounding, and corresponding computer program and echo sounding device
US9618597B2 (en) * 2013-01-04 2017-04-11 Siemens Aktiengesellschaft Method and magnetic resonance apparatus for automated analysis of the raw data of a spectrum
US20160157828A1 (en) * 2014-06-05 2016-06-09 Chikayoshi Sumi Beamforming method, measurement and imaging instruments, and communication instruments
CN104135767A (en) * 2014-08-19 2014-11-05 电子科技大学 Subsection mutual correlation method for measuring arrival time difference of signal direct waves
WO2016142692A1 (en) * 2015-03-06 2016-09-15 Micromass Uk Limited Spectrometric analysis
CN105607125A (en) * 2016-01-15 2016-05-25 吉林大学 Seismic data noise suppression method based on block matching algorithm and singular value decompression
CN107024535A (en) * 2016-03-30 2017-08-08 北京工业大学 A kind of multiple index depth detection method of the vertical defect based on surface wave
CN106094033A (en) * 2016-06-05 2016-11-09 吉林大学 The orientation seismic beam forming method of singular value decomposition
CN105929375A (en) * 2016-07-11 2016-09-07 长沙太电子科技有限公司 Method for suppressing through-wall radar coupling signals based on minimum mean square error criterion
CN106864477A (en) * 2017-03-15 2017-06-20 华南理工大学 A kind of rail flaw ultrasonic detection intelligent detecting method based on time-frequency do-nothing function
CN107704831A (en) * 2017-10-11 2018-02-16 山东科技大学 A kind of gas density data noise reduction based on singular value decomposition median method
CN107884190A (en) * 2017-11-01 2018-04-06 北京信息科技大学 The Method for Bearing Fault Diagnosis decomposed based on variation mode decomposition and wavelet singular

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
CAI LIU.ET: "Random noise de-noising and direct wave eliminating based on SVD method for ground penetrating radar signals", 《JOURNAL OF APPLIED GEOPHYSICS》 *
XUEZHI ZHAO.ET: "Similarity of signal processing effect between Hankel matrix-based SVD and wavelet transform and its mechanism analysis", 《MECHANICAL SYSTEMS AND SIGNAL PROCESSING》 *
李文杰等: "SVD滤波法在直达波和折射波衰减处理中的应用", 《石油勘探与开发》 *
沈鸿雁等: "奇异值分解地震纵、横波波场分离与去噪方法", 《石油地球物理勘探》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110018429A (en) * 2019-03-29 2019-07-16 中国科学院电子学研究所 A kind of method and system for eliminating magnetic detection Platform Vibration interference magnetic field
CN110018429B (en) * 2019-03-29 2021-01-15 中国科学院电子学研究所 Method and system for eliminating magnetic field interference caused by vibration of magnetic detection platform
CN112491999A (en) * 2020-11-18 2021-03-12 成都佳华物链云科技有限公司 Data reporting method and device
CN112491999B (en) * 2020-11-18 2022-10-11 成都佳华物链云科技有限公司 Data reporting method and device

Also Published As

Publication number Publication date
CN108645920B (en) 2020-12-22

Similar Documents

Publication Publication Date Title
CN106772365B (en) A kind of multipath based on Bayes&#39;s compressed sensing utilizes through-wall radar imaging method
CN112162281B (en) Multi-channel SAR-GMTI image domain two-step processing method
CN111142105A (en) ISAR imaging method for complex moving target
CN102914773B (en) Multi-pass circumference SAR three-dimensional imaging method
CN104950305A (en) Real beam scanning radar angle super-resolution imaging method based on sparse constraint
CN104076360B (en) The sparse target imaging method of two-dimensional SAR based on compressed sensing
CN110988834B (en) Pulse arrival time measuring method based on self-adaptive threshold value double thresholds
CN103616687A (en) Polynomial fitting ISAR envelope alignment method based on piecewise linear estimation
CN103235295A (en) Method for estimating small-scene radar target range images on basis of compression Kalman filtering
CN109655827B (en) Hidden target positioning method for single building corner
CN108645920A (en) A kind of direct wave suppressing method of the rail flaw ultrasonic detection based on denoising and alignment
CN101464514B (en) Calibration method and calibration processor for step frequency radar system
CN102495393A (en) Compressive sensing radar imaging algorithm based on subspace tracking
CN105929397B (en) Displaced phase center antenna imaging method based on regularization
CN113589287B (en) Synthetic aperture radar sparse imaging method and device, electronic equipment and storage medium
CN115032638A (en) Bistatic SAR (synthetic aperture radar) phase synchronization precision improving method based on compressed sensing
CN113447915B (en) Ultra-wideband tomography method suitable for complex multipath environment
CN102830394B (en) Weak target detection method based on multispectral accumulation
CN110554377A (en) Single-channel SAR two-dimensional flow field inversion method and system based on Doppler center offset
NL2026449B1 (en) Abnormal grain condition detection method based on radio tomographic imaging
CN103245949A (en) SAR azimuth ambiguity suppression method based on improved ideal filter
CN115453490B (en) Coherent accumulation method, device and equipment based on radar signals and storage medium
CN115951349A (en) Inverse synthetic aperture radar imaging method and system based on co-prime two-channel down-sampling
CN114325700A (en) Satellite-borne multi-channel SAR moving target imaging method
CN113030943A (en) Multi-target tracking algorithm for collecting azimuth range profile based on monopulse radar signals

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

Granted publication date: 20201222

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