CN107345858A - Method for rapidly extracting train bearing rail edge signal impact components - Google Patents
Method for rapidly extracting train bearing rail edge signal impact components Download PDFInfo
- Publication number
- CN107345858A CN107345858A CN201710609918.XA CN201710609918A CN107345858A CN 107345858 A CN107345858 A CN 107345858A CN 201710609918 A CN201710609918 A CN 201710609918A CN 107345858 A CN107345858 A CN 107345858A
- Authority
- CN
- China
- Prior art keywords
- mrow
- rail side
- msub
- signal
- side signal
- 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
- 238000000034 method Methods 0.000 title claims abstract description 33
- 230000001052 transient effect Effects 0.000 claims abstract description 28
- 238000000605 extraction Methods 0.000 claims abstract description 19
- 230000000737 periodic effect Effects 0.000 claims abstract description 7
- 239000000203 mixture Substances 0.000 claims description 33
- 238000012545 processing Methods 0.000 claims description 14
- 238000002592 echocardiography Methods 0.000 claims description 13
- 238000013507 mapping Methods 0.000 claims description 6
- 238000012952 Resampling Methods 0.000 claims description 3
- 238000000926 separation method Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 2
- 230000010355 oscillation Effects 0.000 abstract description 3
- 230000005236 sound signal Effects 0.000 abstract 2
- 238000013016 damping Methods 0.000 abstract 1
- 230000009977 dual effect Effects 0.000 abstract 1
- 238000001514 detection method Methods 0.000 description 7
- 238000003745 diagnosis Methods 0.000 description 6
- 239000000470 constituent Substances 0.000 description 5
- 239000000284 extract Substances 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 238000001228 spectrum Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000003862 health status Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 208000037656 Respiratory Sounds Diseases 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 125000002015 acyclic group Chemical group 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- 230000003116 impacting effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000004080 punching Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M17/00—Testing of vehicles
- G01M17/08—Railway vehicles
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention relates to a method for rapidly extracting an impact component of a train bearing rail side signal, which comprises the steps of roughly estimating a period parameter of a rail side sound signal by using a maximum correlation coefficient method through equal period transient Doppler wavelets, then accurately searching in a parameter range near the estimated period value, identifying an oscillation damping ratio and a system resonance center frequency, and rapidly extracting the impact component. Due to the dual effects of Doppler modulation and roller relative sliding, the impact component is aperiodic, and the traditional periodic transient wavelet cannot be matched and identified correctly. And the non-periodic traversal parameter search has large calculation amount and low practicability because of many parameters to be searched and large search range. The method has the advantages of high impact component extraction precision and small calculated amount, and can be used for extracting the fault characteristics of the sound signals on the edge of the train bearing rail and accurately judging the fault severity.
Description
Technical field
It is specifically a kind of to train bearing rail side signal the present invention relates to Condition Detection and fault diagnosis field
Impact composition and carry out accurate and quick extracting method.
Background technology
As the critical component of train, train bearing connects in working condition at a high speed, heavily loaded, each parts surface for a long time
Touch stress repetition effect, easily cause fatigue, crackle, impression thus fracture, it is deadlocked, abrasion the problems such as, can bring vibration increase,
The problems such as noise and rotary resistance, clamping stagnation occurs when serious so that whole axle system failure, or even cause considerable safety thing
Therefore.The voice signal sent during bearing operation has contained abundant health status information, and it is to realize event that analyzing and processing is carried out to it
Hinder the effective way of diagnosis, and there is the advantages of non-contact measurement compared with vibration signal, thus be widely used in certain
The monitoring and diagnosis of a little special objects.Online measuring technique can carry out health detection to it, realize fault pre-alarming, effectively prevent
The generation of catastrophe failure.
Rail side acoustic detection technology is one of effective means of Railway wheelset bearing health detection and fault diagnosis, has generation
Table is the TADS systems in the U.S. and the RailBAM systems of Australia, passes through the microphone installed in train rail both sides
Array acquisition train at a high speed by when the voice signal that sends of wheel set bearing, being extracted from the voice signal collected to reflect
The validity feature of bearing operation conditions, realizes health detection and fault diagnosis.But due to train fortune relative with the high speed of microphone
Dynamic, the voice signal of microphone collection has Doppler's distortion so that signal produces the modulation on time domain and frequency domain.Parts of bearings
During generation local defect, impact twocomponent signal is produced under the periodic excitation of defect, extraction impact composition can use from signal
In the judgement of fault type and fault degree.It is real using carrying out matching extraction with the similar small echo of impact signal morphosis
One of existing approach.But due to being influenceed by Doppler modulation, impact composition generates Doppler's distortion, and traditional small echo is
Designed for the signal gathered under quiescent conditions, the difference on waveform configuration with rail side signal impact composition be present, to punching
The effective extraction for hitting composition brings new challenge.In addition, the relative position of its roller is not absolute fixation during bearing operation
, relative slip be present, the impact composition and non-exhibiting strict periodicity of its voice signal during train uniform motion.In a word, rail
Aperiodicity is presented in the impact composition of side signal, and traditional period transient state small echo correctly can not be matched and identified, and aperiodicity
Although transient state small echo correctly can match and identify impact composition, matching and identification are time-consuming longer, influence recognition efficiency and right
The real-time of train bearing health status monitoring, thus need to propose new solution method.
There has been no pertinent literature report at present.
The content of the invention
The technology of the present invention solves the problems, such as：For rail side signal Doppler distortion feature and the deficiencies in the prior art, there is provided
A kind of train bearing rail side signal impacts composition rapid extracting method, realizes the accurate of train bearing rail side acoustical signal impact composition
And rapid extraction, provide new technical scheme for the diagnosis of train bearing rail side acoustics.
The technology of the present invention solution：A kind of train bearing rail side signal impacts composition rapid extracting method, including step
It is as follows：
(1) using microphone collection train by when rail side signal caused by bearing, build single transient state Doppler small echo；
(2) using single transient state Doppler small echo of structure, based on maximum correlation coefficient criterion, rail side signal is matched
Identification, it is reconstructed so as to extract impact composition.
In the step (1), the step of building single transient state Doppler small echo, is as follows：
(11) sounding amplitude sequence s is calculatede(n), it is assumed that the sample frequency of rail side signal is fs, expressed according to wavelet function
Into sample frequency fsUnder the discrete sounding amplitude sequence s consistent with rail side signal lengthe(n), wherein N is the rail collected
The length of side signal, n=0,1 ..., (N-1)；
(12) time series that quiets down tr(n) calculate, pass through sequential mapping function：
The phonation time sequence t as corresponding to the sounding amplitude sequence of step (11)e(n) the time series t that quiets down is calculatedr
(n)；X in formula0For the initial time train position and the lateral separation of harvester position of signal, Vs(t) it is train speed, r is
The fore-and-aft distance of harvester and track, c are the velocity of sound；
(13) time delay sequence td(n) calculate, time delay sequence td(n) to finally give time series, its value td
(n)=te(n)+R (0)/c, wherein R (0) represent sound source in starting point and the distance of microphone；
(14) amplitude that quiets down sequence sr(n) calculate, pass through amplitude mapping function：
By step (11) sounding amplitude sequence se(n) the amplitude sequence s that quiets down is calculatedr(n), wherein R represents sound source and Mike
The distance of wind, θ are sound source to the angle between the vectorial and sound source direction of motion vector of microphone, n=0,1 ..., (N-1)；
(15) interpolation fitting samples, the time series t that quiets down obtained with step (2)r(n) it is x variables, is obtained with step (4)
To the amplitude sequence s that quiets downr(n) it is y variables, with the time delay sequence t in step (3)d(n) it is interpolation x variables, performs three
Secondary spline interpolation resampling processing, obtains the i.e. single transient state Doppler small echo of the Laplace small echos after Doppler modulation.
Wavelet function expression formula in the step (11) is Morlet small echos, harmonic wavelet, Laplace small echos, unilateral
Morlet small echos, unilateral harmonic wavelet or unilateral Laplace small echos.
R (0) value in the step (13) is calculated by geometrical relationship, specific as follows：
The process that single transient state Doppler small echo using structure extracts impact composition from the signal of rail side is as follows：
(21) the rail side signal for assuming collection is x (t)；
(22) according to operating mode, geometric parameter and bearing parameter set single transient state Doppler small echo dampingratioζ, time parameter τ,
Frequency of oscillation f model parameter scope, i.e. parameter set γ={ ζ, τ, f }；
(23) value in τ is taken successively, and the τ values each determined are generated by periodic wavelet Ψ (t) with ζ, fγForm reconstruction signal y
(t)γ, y (t)γDoppler modulation processing generation ydop(t)γ；
(24) x (t) and ydop(t)γIn each reconstruction signal seek correlation coefficient function ρ, determine ρmaxCorresponding ζk, τk, fk；
(25) τ is taken outk, addition domain of walker [- N:N], determine τmScope [τk-N:τk+N]；
(26) τ is taken successivelymValue, with ζk, fkGenerate Ψ (t)γ, Ψ (t)γDoppler modulation processing generation Ψdop(t)γ；
(27) Ψ is obtained respectivelydop(t)γImpacted with x (t) first impact composition to n-th into partial correlation coefficient most
Big value, it is determined that corresponding τn, by ζk, fkWith the τ respectively obtainednGenerate Ψ (t)n, it is superimposed and obtains y (t)n；
(28)y(t)nDoppler modulation processing generation ydop(t)n, obtain reconstruction signal ydop(t)nDiagnosed.
The present invention compared with prior art the advantages of be：A kind of train bearing rail side signal impact constituents extraction of the present invention
Method extracts impact composition using single transient state Doppler small echo from the voice signal of train fault bearing rail side, available for train axle
Support rail side Acoustic detection；Due to Doppler modulation and the relative double action slided of roller, impact composition is presented aperiodicity, passed
System period transient state small echo correctly can not be matched and identified.And acyclic traversal parameter search is based on, due to needing what is searched for
Parameter is more, and the span of parameter is big, computationally intensive, the poor real low to the efficiency of rail side signal of parameter identification.For upper
Problem is stated, the present invention proposes single transient state Doppler small echo developing algorithm based on interpolation algorithm, how general produces Traditional Wavelet
Strangle modulation, and be based on maximum correlation coefficient criterion, first by the cycle Doppler's reconstruction signal to aperiodicity rail side signal
Impact interval estimated roughly, then add domain of walker on the impact interval of estimation, form new interval, most
After be again based on maximum correlation coefficient criterion, the model parameter and impact for identifying each impact composition are spaced.So as to realize to row
The stripping one by one of car faulty bearings rail side voice signal impact composition, is carried available for train bearing rail side voice signal fault signature
The accurate judgement with fault severity level is taken, a kind of new technical scheme is provided for train bearing rail side Acoustic detection.
Brief description of the drawings
Fig. 1 is that single transient state Doppler small echo in the present invention builds flow chart；
Fig. 2 is impact constituents extraction and reconstruct flow chart in the present invention；
Fig. 3 is that the rail side that the embodiment of the present invention 1 uses emulates signal (a) and its frequency spectrum (b)；
Fig. 4 is the result for the small echo of cycle Doppler modulation such as using to emulation signal impact constituents extraction, and (a) is
Signal is emulated, (b) is reconstruction signal；
Fig. 5 is that emulation signal is impacted into using non-grade cycle Doppler modulation list transient state small echo proposed by the present invention
Divide the result of extraction, (a) is emulation signal, and (b) is reconstruction signal；
Fig. 6 is the interior ring signal (a) and its frequency spectrum (b) that the embodiment of the present invention 2 uses；
Fig. 7 is the result for the internal ring signal of small echo of cycle Doppler modulation such as using impact constituents extraction, and (a) is
Interior ring signal, (b) are reconstruction signal；
Fig. 8 is to be impacted into using the non-grade cycle internal ring signal of Doppler modulation list transient state small echo proposed by the present invention
Divide the result of extraction, (a) is interior ring signal, and (b) is reconstruction signal.
Embodiment
The technical solution adopted by the present invention is：Using microphone gather train by when rail side signal, structure caused by bearing
Single transient state Doppler small echo is built, match cognization is carried out to the signal based on maximum correlation coefficient criterion, extraction impact composition is carried out
Reconstruct.The core content of the present invention includes two aspects：One is the structure of single transient state Doppler small echo, the second is utilizing structure
Single transient state Doppler small echo extracts impact composition from the signal of rail side.
First, the specific steps of single transient state Doppler small echo structure are as shown in figure 1, as follows：
(1) sounding amplitude sequence sr(n) calculate.Assuming that the sample frequency of rail side signal is fs, according to wavelet function expression formula
(such as Morlet small echos, harmonic wavelet, Laplace small echos, unilateral Morlet small echos, unilateral harmonic wavelet, unilateral Laplace
Small echo) generate the discrete sounding amplitude sequence s consistent with rail side signal length under the sample frequencye(n), n=0,
1 ..., (N-1), wherein N be the length of the rail side signal collected.
(2) time series that quiets down tr(n) calculate.Pass through sequential mapping function：
The phonation time sequence t as corresponding to the sounding amplitude sequence of step (1)e(n), n=0,1 ..., (N-1) is calculated
The time series that quiets down tr(n), n=0,1 ..., (N-1)；X in formula0Initial time train position and harvester position for signal
The lateral separation put, Vs(t) it is train speed, r is the fore-and-aft distance of harvester and track, and c is the velocity of sound.
(3) time delay sequence td(n) calculate.Time delay sequence td(n) to finally give time series, its value is equal to
te(n)+R (0)/c, wherein R (0) represent that in starting point and the distance of microphone, it can be calculated by geometrical relationship for sound source
It is worth and is：
(4) amplitude that quiets down sequence sr(n) calculate.Pass through amplitude mapping function：
By sounding amplitude sequence se(n), n=0,1 ..., (N-1) calculates the amplitude sequence s that quiets downr(n), n=0,
1 ..., (N-1) wherein R represents the distance of sound source and microphone, θ be sound source to the vector of microphone and the sound source direction of motion to
Angle between amount.
(5) interpolation fitting samples.The time series t that quiets down obtained with step (2)r(n) it is x variables, step (4) obtains
The amplitude that quiets down sequence sr(n) it is y variables, with the time delay sequence t in step (3)d(n) it is interpolation x variables, performs sample three times
The resampling of bar interpolation is handled, and obtains the small echo after Doppler modulation.
2nd, the specific steps for impacting constituents extraction are as shown in Figure 2：
(1) the rail side signal for assuming collection is x (t)；
(2) according to operating mode, geometric parameter and bearing parameter set single transient state Doppler small echo dampingratioζ, time parameter τ,
Frequency of oscillation f model parameter scope, i.e. parameter set γ={ ζ, τ, f }；
(3) value in τ is taken successively, and the τ values each determined are generated by periodic wavelet Ψ (t) with ζ, fγForm reconstruction signal y
(t)γ, y (t)γDoppler modulation processing generation ydop(t)γ；
(4) x (t) and ydop(t)γIn each reconstruction signal seek correlation coefficient function ρ, determine ρmaxCorresponding ζk, τk, fk；
(5) τ is taken outk, addition domain of walker [- N:N], determine τmScope [τk-N:τk+N]；
(6) τ is taken successivelymValue, with ζk, fkGenerate Ψ (t)γ, Ψ (t)γDoppler modulation processing generation Ψdop(t)γ；
(7) Ψ is obtained respectivelydop(t)γImpacted with x (t) first impact composition to n-th into partial correlation coefficient most
Big value, it is determined that corresponding τn, by ζk, fkWith the τ respectively obtainednGenerate Ψ (t)n, it is superimposed and obtains y (t)n；
(8)y(t)nDoppler modulation processing generation ydop(t)n, obtain reconstruction signal ydop(t)nDiagnosed.
Below by the validity of two specific embodiment analysis checking the inventive method, embodiment 1 is simulation analysis, real
The rail side stimulus analysis that example 2 is true train inner ring faulty bearings is applied, is illustrated separately below.
Embodiment 1：
Emulation rail side signal x (t) time domain waveform and its frequency spectrum is as shown in figure 3, sample frequency 20KHZ.Utilize proposition
Algorithm peels off impact composition one by one from signal, concretely comprises the following steps：
(1) five kinds of wavelet models (Morlet small echos, harmonic wavelet, Laplace small echos, unilateral Morlet small echos, lists are set
Side harmonic wavelet, unilateral Laplace small echos) parameter area and step-length, parameter set is generated, τ is taken successively according to proposed by the present invention
In value, the τ values each determined and ζ, f are generated by periodic wavelet Ψ (t)γForm reconstruction signal y (t)γ, y (t)γDoppler adjusts
System processing generation ydop(t)γ；
(2) emulation signal x (t) the Doppler's reconstruction signals obtained with step 1 are done into correlation analysis, takes coefficient correlation maximum
τ corresponding to valuek；
(3) τ is taken outk, addition domain of walker [- N:N], determine τmScope [τk-N:τk+ N], τ is taken successivelymValue, with ζk,
fkGenerate Ψ (t)γ, Ψ (t)γDoppler modulation processing generation Ψdop(t)γ；
(4) Ψ is obtained respectivelydop(t)γImpacted with x (t) first impact composition to n-th into partial correlation coefficient most
Big value, it is determined that corresponding τn, by ζk, fkWith the τ respectively obtainednGenerate Ψ (t)n, it is superimposed and obtains y (t)n；
(5)y(t)nDoppler modulation processing generation ydop(t)n, obtain reconstruction signal ydop(t)nDiagnosed.
As shown in figure 5, (a) is emulation signal, (b) is reconstruction signal for the reconstruction signal that is obtained by above step, can be with
Find out that waveform has obtained good matching.(a) in Fig. 4 is to emulate signal, and (b) in Fig. 4 is that single transient state in the cycles such as use is small
Ripple is handled obtained result.To the reconstruction signal obtained using two methods, the coefficient correlation with emulation signal is done respectively,
Method that the present invention is handled and etc. coefficient correlation corresponding to the method for period treatment be respectively 0.4883,0.8905.From Fig. 4,5
It can be seen that structural aberration caused by Doppler modulation, traditional small echo can not realize accurate matching and extraction, this hair
The method of bright proposition accurately can not only be matched and extracted, and can significantly shorten matching and extraction time.
Embodiment 2：
Handled using the Railway wheelset bearing inner race Single Point of Faliure rail side voice signal of reality.Signal waveform such as Fig. 6
In (a) shown in, sample frequency 50KHz.Handled using algorithm proposed by the present invention.Fig. 7 is the cycle methods pair such as uses
The result of inner ring signal transacting, (a) are interior ring signal, and (b) is inner ring reconstruction signal.Fig. 8 is to wait cycle methods to handle using non-
Result, (a) is interior ring signal, and (b) is inner ring reconstruction signal.The signal reconstructed using two methods is related to interior ring signal
Coefficient is respectively 0.3956,0.4744.Seen by the result for contrasting two methods, it can be seen that method proposed by the present invention
Interior ring signal is set to have obtained preferable matching and reconstruct, impact composition has obtained accurate extraction.
Table 1
| Fault type | Simulated fault | Inner ring failure |
| Conventional method | 190 (s/ seconds) | 5605 (s/ seconds) |
| Inventive method | 17 (s/ seconds) | 325 (s/ seconds) |
Table 1 is the reduced time form that the computer for the use of processor being i5-6200u runs two methods；
Illustrate that the time that parameter identifies can be greatly shortened relative to traditional method using method proposed by the present invention.
Above example is provided just for the sake of the description purpose of the present invention, and is not intended to limit the scope of the present invention.This
The scope of invention is defined by the following claims.The various equivalent substitutions that do not depart from spirit and principles of the present invention and make and repair
Change, all should cover within the scope of the present invention.
Claims (5)
1. a kind of train bearing rail side signal impacts composition rapid extracting method, it is characterised in that step is as follows：
(1) using microphone collection train by when rail side signal caused by bearing, build single transient state Doppler small echo；
(2) using single transient state Doppler small echo of structure, based on maximum correlation coefficient criterion, matching knowledge is carried out to rail side signal
Not, extraction impact composition is reconstructed, so as to complete train bearing rail side signal impact composition rapid extraction.
A kind of 2. train bearing rail side signal impact composition rapid extracting method according to claim 1, it is characterised in that：
In the step (1), the step of building single transient state Doppler small echo, is as follows：
(11) sounding amplitude sequence s is calculatede(n), it is assumed that the sample frequency of rail side signal is fs, this is expressed as according to wavelet function
Sample frequency fsUnder the discrete sounding amplitude sequence s consistent with rail side signal lengthe(n), wherein N is the rail side letter collected
Number length, n=0,1 ..., (N-1)；
(12) time series that quiets down tr(n) calculate, pass through sequential mapping function：
<mrow>
<msub>
<mi>t</mi>
<mi>r</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mi>t</mi>
<mi>e</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
<mo>+</mo>
<mfrac>
<msqrt>
<mrow>
<msup>
<mrow>
<mo>&lsqb;</mo>
<msub>
<mi>X</mi>
<mn>0</mn>
</msub>
<mo>-</mo>
<msubsup>
<mo>&Integral;</mo>
<mn>0</mn>
<mi>t</mi>
</msubsup>
<msub>
<mi>V</mi>
<mi>s</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mi>d</mi>
<mi>t</mi>
<mo>&rsqb;</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<mi>r</mi>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
<mi>c</mi>
</mfrac>
</mrow>
The phonation time sequence t as corresponding to the sounding amplitude sequence of step (11)e(n) the time series t that quiets down is calculatedr(n)；Formula
Middle X0For the initial time train position and the lateral separation of harvester position of signal, Vs(t) it is train speed, r is collection dress
The fore-and-aft distance with track is put, c is the velocity of sound；
(13) time delay sequence td(n) calculate, time delay sequence td(n) to finally give time series, its value td(n)=te
(n)+R (0)/c, wherein R (0) represent sound source in starting point and the distance of microphone；
(14) amplitude that quiets down sequence sr(n) calculate, pass through amplitude mapping function：
<mrow>
<msub>
<mi>S</mi>
<mi>r</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mi>r</mi>
<mrow>
<mi>R</mi>
<msup>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>-</mo>
<msub>
<mi>V</mi>
<mi>s</mi>
</msub>
<mi>c</mi>
<mi>o</mi>
<mi>s</mi>
<mi>&theta;</mi>
<mo>/</mo>
<mi>c</mi>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</mfrac>
<mo>&times;</mo>
<msub>
<mi>S</mi>
<mi>e</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>)</mo>
</mrow>
</mrow>
By step (11) sounding amplitude sequence se(n) the amplitude sequence s that quiets down is calculatedr(n), wherein R represents sound source and microphone
Distance, θ are sound source to the angle between the vectorial and sound source direction of motion vector of microphone, n=0,1 ..., (N-1)；
(15) interpolation fitting samples, the time series t that quiets down obtained with step (2)r(n) it is x variables, the receipts obtained with step (4)
Acoustic amplitude sequence sr(n) it is y variables, with the time delay sequence t in step (3)d(n) it is interpolation x variables, performs cubic spline
Interpolation resampling is handled, and obtains the i.e. single transient state Doppler small echo of the Laplace small echos after Doppler modulation.
A kind of 3. train bearing rail side signal impact composition rapid extracting method according to claim 1, it is characterised in that：
Using single transient state Doppler small echo of structure in the step (2), based on maximum correlation coefficient criterion, to the progress of rail side signal
With identification, the process that extraction impact composition is reconstructed is as follows：
(21) the rail side signal for assuming collection is x (t)；
(22) single transient state Doppler small echo dampingratioζ, time parameter τ, vibration are set according to operating mode, geometric parameter and bearing parameter
Frequency f model parameter scope, i.e. parameter set γ={ ζ, τ, f }；
(23) take the value in τ successively, the τ values each determined and ζ, f generate by etc. periodic wavelet Ψ (t)γForm reconstruction signal y
(t)γ, y (t)γDoppler modulation processing generation ydop(t)γ；
(24) x (t) and ydop(t)γIn each reconstruction signal seek correlation coefficient function ρ, determine ρmaxCorresponding ζk, τk, fk；
(25) τ is taken outk, addition domain of walker [- N:N], determine τmScope [τk-N:τk+N]；
(26) τ is taken successivelymValue, with ζk, fkGenerate Ψ (t)γ, Ψ (t)γDoppler modulation processing generation Ψdop(t)γ；
(27) Ψ is obtained respectivelydop(t)γMaximum of the composition to n-th of impact into partial correlation coefficient is impacted with first of x (t)
Value, it is determined that corresponding τn, by ζk, fkWith the τ respectively obtainednGenerate Ψ (t)n, it is superimposed and obtains y (t)n；
(28) to y (t)nDoppler modulation processing generation ydop(t)n, obtain reconstruction signal ydop(t)n。
4. train bearing rail side signal according to claim 3 impacts component extracting method, it is characterised in that：It is described
(23), in (24), the cycle methods such as use to estimate impact interval roughly.
A kind of 5. train bearing rail side signal impact composition rapid extracting method according to claim 3, it is characterised in that：
In (25), on the basis of obtained impact interval is estimated roughly, domain of walker is added, carries out precise search.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710609918.XA CN107345858B (en) | 2017-07-25 | 2017-07-25 | Method for rapidly extracting train bearing rail edge signal impact components |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710609918.XA CN107345858B (en) | 2017-07-25 | 2017-07-25 | Method for rapidly extracting train bearing rail edge signal impact components |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107345858A true CN107345858A (en) | 2017-11-14 |
| CN107345858B CN107345858B (en) | 2018-08-14 |
Family
ID=60257900
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710609918.XA Active CN107345858B (en) | 2017-07-25 | 2017-07-25 | Method for rapidly extracting train bearing rail edge signal impact components |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107345858B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109374291A (en) * | 2018-11-12 | 2019-02-22 | 华南师范大学 | A kind of method of rapidly extracting impulse response signal characteristic parameter |
| CN109784305A (en) * | 2019-01-29 | 2019-05-21 | 石家庄铁道大学 | Based on the matched Lapalce wavelet basis rarefaction representation dictionary construction method of wave forms impact |
| CN112284584A (en) * | 2019-07-12 | 2021-01-29 | 斯凯孚公司 | Method for estimating bearing load using strain parameters that account for contact angle variations |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105241666A (en) * | 2015-09-21 | 2016-01-13 | 华南理工大学 | Rolling bearing fault feature extraction method based on signal sparse representation theory |
| CN105424388A (en) * | 2015-11-17 | 2016-03-23 | 苏州大学 | Train wheel set bearing fault transient characteristic detection method based on parametric Doppler transient model |
| CN205262744U (en) * | 2015-11-17 | 2016-05-25 | 苏州大学 | Train wheel pair bearing trouble transient state characteristic detection device based on parameterization doppler transient state model |
| CN106441895A (en) * | 2016-10-09 | 2017-02-22 | 安徽大学 | Train bearing rail edge signal impact component extraction method |
-
2017
- 2017-07-25 CN CN201710609918.XA patent/CN107345858B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105241666A (en) * | 2015-09-21 | 2016-01-13 | 华南理工大学 | Rolling bearing fault feature extraction method based on signal sparse representation theory |
| CN105424388A (en) * | 2015-11-17 | 2016-03-23 | 苏州大学 | Train wheel set bearing fault transient characteristic detection method based on parametric Doppler transient model |
| CN205262744U (en) * | 2015-11-17 | 2016-05-25 | 苏州大学 | Train wheel pair bearing trouble transient state characteristic detection device based on parameterization doppler transient state model |
| CN106441895A (en) * | 2016-10-09 | 2017-02-22 | 安徽大学 | Train bearing rail edge signal impact component extraction method |
Non-Patent Citations (1)
| Title |
|---|
| FANG LIU等: "Doppler effect reduction scheme via acceleration-based Dopplerlet transform and resampling method for the wayside acoustic defective bearing detector system", 《JOURNAL OF MECHANICAL ENGINEERING SCIENCE》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109374291A (en) * | 2018-11-12 | 2019-02-22 | 华南师范大学 | A kind of method of rapidly extracting impulse response signal characteristic parameter |
| CN109784305A (en) * | 2019-01-29 | 2019-05-21 | 石家庄铁道大学 | Based on the matched Lapalce wavelet basis rarefaction representation dictionary construction method of wave forms impact |
| CN112284584A (en) * | 2019-07-12 | 2021-01-29 | 斯凯孚公司 | Method for estimating bearing load using strain parameters that account for contact angle variations |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107345858B (en) | 2018-08-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106441895B (en) | Train bearing rail edge signal impact component extraction method | |
| CN108426713B (en) | Rolling bearing weak fault diagnosis method based on wavelet transformation and deep learning | |
| Li et al. | Fault detection method for railway wheel flat using an adaptive multiscale morphological filter | |
| CN103335617B (en) | A kind of railway track geometric deformation detection method based on vibration signal | |
| CN105424388B (en) | A kind of Railway wheelset bearing fault transient feature detection method based on parametrization Doppler's transient model | |
| CN110135492B (en) | Equipment fault diagnosis and abnormality detection method and system based on multiple Gaussian models | |
| Zhou et al. | A new method to classify railway vehicle axle fatigue crack AE signal | |
| Li et al. | Railway wheel flat detection based on improved empirical mode decomposition | |
| Zhang et al. | Deep convolutional neural network probability imaging for plate structural health monitoring using guided waves | |
| CN109323754B (en) | Train wheel polygon fault diagnosis and detection method | |
| CN103018044B (en) | Bearing combined failure diagnosis method based on improved impact dictionary matching pursuit | |
| CN107345858A (en) | Method for rapidly extracting train bearing rail edge signal impact components | |
| CN104833534A (en) | Train running fault diagnosis device based on multi-source information fusion, and method | |
| CN108444696A (en) | A kind of gearbox fault analysis method | |
| CN105424366A (en) | Bearing fault diagnosis method based on EEMD adaptive denoising | |
| CN106226407A (en) | A kind of online preprocess method of ultrasound echo signal based on singular spectrum analysis | |
| CN103776903B (en) | A kind of wind electricity blade delamination detection method and detection system | |
| CN101587007A (en) | Output-only wavelet analytical method for recognizing flexible bridge structure kinetic parameter | |
| CN110243605A (en) | Multi-source time-frequency crestal line extracting method | |
| CN105862935A (en) | Damage recognition method used for retaining wall structural system | |
| CN103808801A (en) | Real-time detection method for high-speed rail injury based on vibration and audio composite signals | |
| Liu et al. | Damage-sensitive and domain-invariant feature extraction for vehicle-vibration-based bridge health monitoring | |
| CN110990978A (en) | Bolt state monitoring method and device | |
| CN112461934A (en) | Aero-engine blade crack source positioning method based on acoustic emission | |
| CN114263621A (en) | Test method and system for diagnosing and simulating cavitation fault of centrifugal pump |
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 |