CN109871824A - The multi-modal separation method of supersonic guide-wave and its system based on management loading - Google Patents
The multi-modal separation method of supersonic guide-wave and its system based on management loading Download PDFInfo
- Publication number
- CN109871824A CN109871824A CN201910183672.3A CN201910183672A CN109871824A CN 109871824 A CN109871824 A CN 109871824A CN 201910183672 A CN201910183672 A CN 201910183672A CN 109871824 A CN109871824 A CN 109871824A
- Authority
- CN
- China
- Prior art keywords
- signal
- mode
- dictionary
- wave
- supersonic guide
- 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
Landscapes
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a kind of multi-modal separation method of supersonic guide-wave based on management loading and its systems, it the described method comprises the following steps: being applied to measured structure using signal s (t) as excitation, obtain the supersonic guide-wave response signal r (t) of measured structure, it includes P supersonic guide-wave mode that the supersonic guide-wave response signal, which has altogether, according to supersonic guide-wave mode obtained, the son constructed respectively under covering different propagation distance by mode propagates dictionary Di, and be combined to obtain multi-modal composite propagation dictionary D=[D1, D2..., DP]；Supersonic guide-wave response signal r (t) is subjected to rarefaction representation at multi-modal composite propagation dictionary D, establishes sparse representation model, is solved using management loading algorithm with drag: r=Dw+n, and by its corresponding sub- weight coefficient W1, W2... WPIt is separated one by one by mode；The multi-modal composite propagation dictionary D is pressed into mode decomposition, the single mode obtained under each mode propagates dictionary D1, D2..., DP, each single mode is propagated into dictionary and corresponding sub- weight coefficient wiIt is multiplied to obtain the single mode signal component of i-th of mode.
Description
Technical field
The invention belongs to signal processing technology field, especially a kind of supersonic guide-wave multimode based on management loading
State separation method and its system.
Background technique
For supersonic guide-wave compares conventional ultrasound, farther distance can be propagated, and surface and internal exergy dissipation to structure
Wound is all sensitive, has become the important tool of monitoring structural health conditions and non-destructive testing.With the monitoring technology of the passive types such as sound emission
Unlike, the structural health monitoring technology based on supersonic guide-wave is a kind of active monitoring technology, is not required to dependency structure hair
Characteristic information out, but pass through Active spurring and acquire response, so that the health status of evaluation structure, is monitoring structural health conditions neck
The important and hot research direction in domain.
Multi-modal is the inherent characteristic of supersonic guide-wave.Within the scope of any excitation frequency band, in structure exist two kinds or with
On guided wave modal.And with the increase of frequency, the guided wave modal number in structure can be gradually increased.The multi-modal phenomenon of guided wave
For monitoring structural health conditions, favorably also there is disadvantage.Its benefit is that different modalities are to various forms of damages or weld defect control degree
Difference, it is multi-modal to help to cover the monitoring to more type injuries, promote the Overall Acquisition to structural damage information；Its disadvantage exists
In the increase of mode number, the mutual aliasing of different modalities twocomponent signal improves the difficulty of signal processing.Therefore guided wave modal point
It is particularly significant for the monitoring structural health conditions based on guided wave from method.With the continuous deepening of research, to supersonic guide-wave mould
The accuracy of state separation also proposed increasingly higher demands.
Existing supersonic guide-wave modal separation method mainly has Time-frequency Analysis and matching pursuit algorithm.Time of occurrence is earlier
It is Time-frequency Analysis, however such methods, because being restricted by sea cucumber Burger uncertainty principle, time domain and frequency domain precision cannot be same
Shi Tigao because different modalities can not separate well on time frequency plane, thus limits the precision and application of such method
Range.And matching pursuit algorithm needs artificially given degree of rarefication parameter, but the parameter actually difficulty has priori, and matching pursuit algorithm is anti-
Interference performance is poor.In practical applications, algorithm above will cause the defects of separation is incomplete or separation accuracy is not high, limit base
In the development and application of the structural health monitoring technology of supersonic guide-wave.
Disclosed above- mentioned information are used only for enhancing the understanding to background of the present invention in the background section, it is thus possible to
Information comprising not constituting the prior art known to a person of ordinary skill in the art in home.
Summary of the invention
It is an object of the invention to propose a kind of super based on management loading for above-mentioned existing methods deficiency
The multi-modal separation method of guided Waves improves the precision of modal separation under conditions of without artificial determining parameter, avoids because of separation
Inaccuracy and cause wrong report to structural damage and fail to report.
The purpose of the present invention is being achieved by the following technical programs, a kind of ultrasound based on management loading is led
The multi-modal separation method of wave the following steps are included:
In first step, it is applied to measured structure using signal s (t) as excitation, the supersonic guide-wave for obtaining measured structure is rung
Induction signal r (t), it includes P supersonic guide-wave mode that the supersonic guide-wave response signal, which has altogether, wherein symbol t indicates that the time becomes
Amount；
In second step, according to supersonic guide-wave mode obtained, constructed respectively by mode under covering different propagation distance
Son propagate dictionary Di, and be combined to obtain multi-modal composite propagation dictionary D=[D1, D2..., DP]；
In third step, supersonic guide-wave response signal r (t) is subjected to rarefaction representation at multi-modal composite propagation dictionary D,
Establish sparse representation model:
min||w||0S.t.r=Dw=[D1, D2..., DP][w1, w2... wP]T,
In formula, r is by the discrete rear column vector r=[r (t formed of signal r (t)1), r (t2) ..., r (tN)]T, N is signal
Points, subscript T indicate the transposition operation of vector or matrix, w=[w1, w2... wP]T, w is column vector r in multi-modal composite propagation
Weight vectors under dictionary D, wherein wiCorresponding to sub- dictionary DiWeight coefficient, s.t. expression be constrained in；
In four steps, solved using management loading algorithm with drag: r=Dw+n,
In formula, n is system noise, obtains weight vectors w by management loading algorithm, and by its corresponding sub- power
Weight coefficient w1, w2... wPIt is separated one by one by mode；
In 5th step, the multi-modal composite propagation dictionary D is pressed into mode decomposition, the single mode obtained under each mode passes
Broadcast dictionary D1, D2..., DP, each single mode is propagated into dictionary D1, D2..., DPWith corresponding sub- weight coefficient wiIt is multiplied to obtain
The single mode signal component of i mode stores obtained single mode signal component.
In the method, in second step, the multi-modal composite propagation dictionary D passes through to measurement method, finite element
Simulation method or analytic method obtain.
In the method, the measurement method is the position by changing excitation and receiving sensor, in the tested knot of original
Excitation motivates identical signal with former in structure and same environment, measures its response and as matrix column vector, in order
It forms a matrix and obtains composite propagation dictionary.
In the method, the finite element simulation method is to propagate finite element by establishing the supersonic guide-wave of measured structure
Model, after being applied to identical original excitation signal, emulation obtain response signal under different propagation distance and as
Matrix column vector forms a matrix in order and obtains composite propagation dictionary.
In the method, the analytic method is to obtain different propagation distance x by following guided waves propagation model1,
x2..., xMUnder response signal,
Dm=[gm(x1), gm(x2) ..., gm(xM)]T
In formula, DmM-th of mode propagation dictionary of table, gm(xj) expression propagation distance be xjM-th of mode under single mode
Response signal gm(xj, t) and the column vector that is formed after discrete, gm(xj)=[gm(xj, t1), gm(xj, t2) ..., gm(xj, tN)]T,
Wherein gm(xj, t1), gm(xj, t2) ..., gm(xj, tN) respectively indicate signal gm(xj, t) the 1st, the 2nd, n-th discrete point, letter
Number gm(xj, t) and it is determined by following formula,In formula, S (ω) is former sharp
Encourage the frequency domain form of signal s (t), km(ω) is the frequency-wavenumber curve of m-th of mode, and ω is angular frequency, xMFor maximum propagation
Distance.
In the method, maximum propagation distance xMMiddle acquisition duration T and the guided wave maximum propagation including original signal
Speed vmaxProduct, i.e. xM=Tvmax。
In the method, supersonic guide-wave includes Lamb wave.
According to another aspect of the present invention, the separation system for implementing the method includes,
Measured structure is provided to propagate supersonic guide-wave, measured structure be equipped with excitation supersonic guide-wave driver and
The collector for the supersonic guide-wave propagated through measured structure is acquired,
Signal exciting unit is configured to issue supersonic guide-wave,
Signal amplification unit is configured to amplify amplifying unit one end connection signal exciting unit of the waveform signal, separately
One end connects the driver in measured structure,
Signal acquisition unit is configured to signal acquisition unit one end connection measured structure of acquisition supersonic guide-wave response signal
In collector, other end connection signal processing unit,
Signal processing unit obtains the single mode signal of each mode based on the supersonic guide-wave response signal.
In the separation system, signal processing unit includes,
Multi-modal composite propagation dictionary creation unit is configured to based on supersonic guide-wave mode obtained, by mode point
The son that Gou Jian do not cover under different propagation distance propagates dictionary DiAnd it is combined to obtain multi-modal composite propagation dictionary；
Sparse representation model generation unit, be configured to based on response signal r (t) carried out at dictionary D rarefaction representation with
Establish sparse representation model；
Signal processing unit is configured with management loading algorithm solving model r=Dw+n, obtain weight to
W is measured, and by its corresponding sub- weight coefficient W1, W2... WPIt is separated one by one by mode；
Computing unit is configured to multi-modal composite propagation dictionary D described in mode decomposition, obtains the single mode under each mode
Propagate dictionary D1, D2..., DP, each single mode is propagated into dictionary DiWith corresponding sub- weight coefficient wiIt is multiplied to obtain m-th of mode
Single mode signal component.
In the separation system, driver and collector are PZT piezoelectric ceramic piece, and the signal processing unit is number
Signal processor, application-specific integrated circuit ASIC or on-site programmable gate array FPGA, the signal processing unit include memory,
The memory includes that one or more read only memory ROMs, random access memory ram, flash memory or electronics are erasable
Except programmable read only memory EEPROM.
Beneficial effect
Compared with the conventional method, present invention has the advantage that the present invention can be realized the multi-modal supersonic guide-wave of aliasing
Modal separation, because constructing management loading model, parameter in model is automatically from the acquistion of signal middle school to not needing
It is artificial to adjust ginseng.And the present invention by original signal the rarefaction representation under excessively complete dictionary, directly separated and extracted in the time domain
Each mode is carried out without converting into time-frequency domain, to avoid the problem that precision is not high in Time-frequency Analysis.
Detailed description of the invention
By reading the detailed description in hereafter preferred embodiment, various other advantages and benefits of the present invention
It will become apparent to those of ordinary skill in the art.Figure of description only for the purpose of illustrating preferred embodiments,
And it is not to be construed as limiting the invention.It should be evident that drawings discussed below is only some embodiments of the present invention,
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings
Other attached drawings.And throughout the drawings, identical component is presented with like reference characters.
In the accompanying drawings:
Fig. 1 is the multi-modal separation method of the supersonic guide-wave based on management loading according to an embodiment of the invention
Pumping signal time domain waveform；
Fig. 2 is the multi-modal separation method of the supersonic guide-wave based on management loading according to an embodiment of the invention
Pumping signal frequency-domain waveform figure；
Fig. 3 is the multi-modal separation method of the supersonic guide-wave based on management loading according to an embodiment of the invention
Flow chart；
Fig. 4 is the multi-modal separation method of the supersonic guide-wave based on management loading according to an embodiment of the invention
Guided wave response signal r (t) time domain waveform；
Fig. 5 is the multi-modal separation method of the supersonic guide-wave based on management loading according to an embodiment of the invention
Weight coefficient schematic diagram；
Fig. 6 is the multi-modal separation method of the supersonic guide-wave based on management loading according to an embodiment of the invention
The weight map of an obtained mode ingredient；
Fig. 7 is the multi-modal separation method of the supersonic guide-wave based on management loading according to an embodiment of the invention
The weight map of another obtained mode ingredient；
Fig. 8 is the multi-modal separation method of the supersonic guide-wave based on management loading according to an embodiment of the invention
The time domain waveform of the Lamb wave ingredient of an obtained mode ingredient；
Fig. 9 is the multi-modal separation method of the supersonic guide-wave based on management loading according to an embodiment of the invention
The time domain waveform of the Lamb wave ingredient of another obtained mode ingredient；
Figure 10 is multi-modal point of the supersonic guide-wave based on management loading of implementation according to an embodiment of the invention
Separation system structural schematic diagram from method.
Below in conjunction with drawings and examples, the present invention will be further explained.
Specific embodiment
1 to Figure 10 specific embodiment that the present invention will be described in more detail below with reference to accompanying drawings.Although being shown in attached drawing
Specific embodiments of the present invention, it being understood, however, that may be realized in various forms the reality of the invention without that should be illustrated here
Example is applied to be limited.On the contrary, providing these embodiments is to be able to thoroughly understand the present invention, and can will be of the invention
Range is fully disclosed to those skilled in the art.
It should be noted that having used some vocabulary in the specification and claims to censure specific components.Ability
Field technique personnel it would be appreciated that, technical staff may call the same component with different nouns.This specification and right
It is required that not in such a way that the difference of noun is as component is distinguished, but with the difference of component functionally as differentiation
Criterion."comprising" or " comprising " as mentioned throughout the specification and claims are an open language, therefore should be solved
It is interpreted into " including but not limited to ".Specification subsequent descriptions are to implement better embodiment of the invention, so the description be with
For the purpose of the rule of specification, the range that is not intended to limit the invention.Protection scope of the present invention is when the appended right of view
It is required that subject to institute's defender.
In order to facilitate understanding of embodiments of the present invention, further by taking specific embodiment as an example below in conjunction with attached drawing to be solved
Explanation is released, and each attached drawing does not constitute the restriction to the embodiment of the present invention.
In order to better understand, Fig. 3 is the multi-modal separation method work of a supersonic guide-wave based on management loading
Make flow chart, as shown in figure 3, a kind of multi-modal separation method of supersonic guide-wave based on management loading includes following step
It is rapid:
In first step S1, it is applied to measured structure using signal s (t) as excitation, obtains the supersonic guide-wave of measured structure
Response signal r (t), it includes P supersonic guide-wave mode that the supersonic guide-wave response signal, which has altogether, wherein symbol t indicates that the time becomes
Amount；
In second step S2, according to supersonic guide-wave mode obtained, covering different propagation distance is constructed respectively by mode
Under son propagate dictionary Di, and be combined to obtain multi-modal composite propagation dictionary D=[D1, D2..., DP]；
In third step S3, supersonic guide-wave response signal r (t) is subjected to sparse table at multi-modal composite propagation dictionary D
Show, establish sparse representation model:
min||w||0S.t.r=Dw=[D1, D2..., DP][w1, w2... wP]T,
In formula, r is by the discrete rear column vector r=[r (t formed of signal r (t)1), r (t2) ..., r (tN)]T, N is signal
Points, subscript T indicate the transposition operation of vector or matrix, w=[w1, w2... wP]T, w is column vector r in multi-modal composite propagation
Weight vectors under dictionary D, wherein wiCorresponding to sub- dictionary DiWeight coefficient, s.t. expression be constrained in；
In four steps S4, solved using management loading algorithm with drag: r=Dw+n,
In formula, n is system noise, obtains weight vectors w by management loading algorithm, and by its corresponding sub- power
Weight coefficient w1, w2... wPIt is separated one by one by mode；
In 5th step S5, the multi-modal composite propagation dictionary D is pressed into mode decomposition, obtains the single mode under each mode
Propagate dictionary D1, D2..., DP, each single mode is propagated into dictionary D1, D2..., DPWith corresponding sub- weight coefficient wiIt is multiplied to obtain
The single mode signal component of i-th of mode stores obtained single mode signal component.
In one embodiment of the method, in second step S2, the multi-modal composite propagation dictionary D passes through
Measurement method, finite element simulation method or analytic method are obtained.
In one embodiment of the method, the measurement method is the position by changing excitation and receiving sensor
It sets, excitation motivates identical signal with former in former measured structure and same environment, measures its response and as matrix
Column vector forms a matrix in order and obtains composite propagation dictionary.
In one embodiment of the method, the finite element simulation method is the ultrasound by establishing measured structure
Guided waves propagation finite element model, after being applied to identical original excitation signal, emulation obtains the response under different propagation distance
Signal and as matrix column vector, forms a matrix in order and obtains composite propagation dictionary.
In one embodiment of the method, the analytic method is to obtain difference by following guided waves propagation model
Propagation distance x1, x2..., xMUnder response signal,
Dm=[gm(x1), gm(x2) ..., gm(xM)]T
In formula, gm(xj) expression propagation distance be xjM-th of mode under single modal response signal gm(xj, t) and through discrete
The column vector formed afterwards, gm(xj)=[gm(xj, t1), gm(xj, t2) ..., gm(xj, tN)]T, wherein gm(xj, t1), gm(xj,
t2) ..., gm(xj, tN) respectively indicate signal gm(xj, t) the 1st, the 2nd, n-th discrete point, signal gm(xj, t) and true by following formula
It is fixed,In formula, S (ω) is the frequency domain of original pumping signal s (t)
Form, km(ω) is the frequency-wavenumber curve of m-th of mode, and ω is angular frequency, xMFor maximum propagation distance.
In one embodiment of the method, maximum propagation distance xMMiddle acquisition duration T including original signal with
Guided wave maximum propagation speed vmxProduct, i.e. xM=Tvmax。
In one embodiment of the method, supersonic guide-wave includes Lamb wave.
For a further understanding of the present invention, with reference to the accompanying drawing and specific embodiment the invention will be further described, answer
Should be, it is emphasized that following the description be only exemplary, and application of the invention does not limit to following examples.
In one embodiment, the measured structure of the embodiment of the present invention is preferably the aluminium alloy plate of model 6061, thickness
For 2mm, material mechanical parameters are as follows: density p=2690kg/m3, elastic modulus E=70GPa, Poisson ratioσ=0.33.Structure
It is middle to use piezoelectric ceramic piece PZT as energy converter, it is respectively used to the driver and collector of supersonic guide-wave.
Pumping signal used by the present embodiment is 5 periods of the Hanning window modulation that a centre frequency is 200kHz
Sinusoidal impulse signal, corresponding time domain waveform and spectrogram difference are as depicted in figs. 1 and 2.
Referring to Fig. 3, the present embodiment provides a kind of multi-modal separation method of the supersonic guide-wave based on management loading,
Process the following steps are included:
In first step S1, using signal s (t) as excitation, guided wave response r (t) of measured structure, the response signal are obtained
It altogether include P guided wave modal.Wherein symbol t indicates time variable.Referring to Fig. 4, the guided wave response letter obtained for the present embodiment
The time domain waveform of number r (t) includes altogether P=2 guided wave modal, respectively A0 and S0 mode Lamb wave.
In second step S2, according to the mode prior information of supersonic guide-wave obtained, covering is constructed respectively not by mode
Dictionary D is propagated with the son under propagation distancei, and be combined to obtain multi-modal composite propagation dictionary D=[D1, D2..., DP].It is excellent
Selection of land, the present embodiment obtain different propagation distance x by following guided waves propagation model using analytic method1, x2..., xMUnder response
Signal
Dm=[gm(x1), gm(x2) ..., gm(xM)]T
In formula, DmM-th of mode propagation dictionary of table, gm(xj) expression propagation distance be xjM-th of mode under single mode
Response signal gm(xj, t) and the column vector g that is formed after discretem(xj)=[gm(xj, t1), gm(xj, t2) ..., gm(xj, tN)]T,
Middle gm(xj, t1), gm(xj, t2) ..., gm(xj, tN) respectively indicate signal gm(xj, t) the 1st, the 2nd, n-th discrete point.This reality
It applies in example, x1Value be preferably 200mm, xMValue be preferably 900mm, the value of M is preferably 701, x1, x2..., xMFor in 200-
It is obtained between 900mm with interval 1mm uniform sampling；A0 mode is indicated when m=1, m=2 indicates S0 mode；Time point t1=0, tN
=0.25ms, N=2500, t1, t2..., tNTo be obtained between time 0-0.25ms with 1 μ s uniform sampling of interval.
Signal gm(xj, t) and it is determined by following formula
In formula, S (ω) is the frequency domain form of original pumping signal s (t), km(ω) is that frequency-wave number of m-th of mode is bent
Line.Wherein k1(ω) is frequency-wave number curve of A0 mode Lamb wave in measured structure, k2(ω) is A0 mode in measured structure
The frequency-wavenumber curve of Lamb wave.
In third step S3, response signal r (t) is subjected to rarefaction representation at dictionary D, establishes sparse representation model
min||w||0S.t.r=Dw=[D1, D2..., DP][W1, W2... WP]T
In formula, r is by the discrete rear column vector r=[r (t formed of signal r (t)1), r (t2) ..., r (tN)]T, N is signal
Points, subscript T indicate the transposition operation of vector or matrix, w=[W1, W2... WP]TW is weight system of the signal r at dictionary D
It counts, wherein wiCorresponding to sub- dictionary DiWeight coefficient, s.t. expression be constrained in.
In four steps S4, solved using management loading algorithm with drag
R=Dw+n
In formula, n is system noise.By management loading algorithm, weight vectors w is obtained, and by its corresponding son
Weight coefficient w1, w2... wPIt is separated one by one by mode.Referring to Fig. 5, for the weight system obtained after step S4 in the present embodiment
Number schematic diagram.
In 5th step S5, former multi-modal composite propagation dictionary D is pressed into mode decomposition, the single mode obtained under each mode passes
Broadcast dictionary D1, D2..., DP.Each single mode is propagated into dictionary DiWith corresponding sub- weight coefficient wiIt is multiplied, m-th of mould can be obtained
The single mode signal component r of statem=DmWm.Referring to figure 6 and figure 7, the A0 that is obtained after step S5 respectively in the present embodiment and
The weight coefficient schematic diagram of S0 mode Lamb wave.Referring to Fig. 8 and Fig. 9, respectively obtained after step S5 in the present embodiment
The time domain waveform of A0 and S0 mode Lamb wave ingredient.So far, the present embodiment completes the separation of A0, S0 both modalities which, obtains
Respective mode ingredient component.
Comparison diagram 4 and Fig. 8, Fig. 9 are as it can be seen that the present invention can by the former signal integrity of two mode of aliasing in the time domain
It separates, has obtained the single mode component signal of respective mode ingredient, realize multi-modal separation and extraction.And the present invention
Not by the interference of the noise contribution in original aliasing signal.
As shown in Figure 10, the separation system for implementing the method includes,
Measured structure is provided to propagate supersonic guide-wave, measured structure be equipped with excitation supersonic guide-wave driver and
The collector for the supersonic guide-wave propagated through measured structure is acquired,
Signal exciting unit is configured to issue supersonic guide-wave,
Signal amplification unit is configured to amplify amplifying unit one end connection signal exciting unit of the waveform signal, separately
One end connects the driver in measured structure,
Signal acquisition unit is configured to signal acquisition unit one end connection measured structure of acquisition supersonic guide-wave response signal
In collector, other end connection signal processing unit,
Signal processing unit obtains the single mode signal of each mode based on the supersonic guide-wave response signal.
In one embodiment of the separation system, signal processing unit includes,
Multi-modal composite propagation dictionary creation unit is configured to based on supersonic guide-wave mode obtained, by mode point
The son that Gou Jian do not cover under different propagation distance propagates dictionary DiAnd it is combined to obtain multi-modal composite propagation dictionary；
Sparse representation model generation unit, be configured to based on response signal r (t) carried out at dictionary D rarefaction representation with
Establish sparse representation model；
Signal processing unit is configured with management loading algorithm solving model r=Dw+n, obtain weight to
W is measured, and by its corresponding sub- weight coefficient W1, W2... WPIt is separated one by one by mode；
Computing unit is configured to multi-modal composite propagation dictionary D described in mode decomposition, obtains the single mode under each mode
Propagate dictionary D1, D2..., DP, each single mode is propagated into dictionary DiWith corresponding sub- weight coefficient wiIt is multiplied to obtain m-th of mode
Single mode signal component.
In one embodiment of the separation system, driver and collector are PZT piezoelectric ceramic piece, at the signal
Reason unit is digital signal processor, application-specific integrated circuit ASIC or on-site programmable gate array FPGA, the signal processing list
Member includes memory, and the memory includes one or more read only memory ROM, random access memory ram, flash
Device or Electrical Erasable programmable read only memory EEPROM.
In one embodiment, separation system further includes display unit and wireless telecom equipment, and wireless telecom equipment includes
4G/GPRS or the Internet communication module.
The present invention can be separated multi-modal signal by mode, obtain multiple single mode ingredient supersonic guide-wave ingredients
Subsignal.
Although embodiment of the present invention is described in conjunction with attached drawing above, the invention is not limited to above-mentioned
Specific embodiments and applications field, above-mentioned specific embodiment are only schematical, directiveness, rather than restricted
's.Those skilled in the art are under the enlightenment of this specification and in the range for not departing from the claims in the present invention and being protected
In the case where, a variety of forms can also be made, these belong to the column of protection of the invention.
Claims (10)
1. a kind of multi-modal separation method of supersonic guide-wave based on management loading, the described method comprises the following steps:
In first step (S1), it is applied to measured structure using signal s (t) as excitation, the supersonic guide-wave for obtaining measured structure is rung
Induction signal r (t), it includes P supersonic guide-wave mode that the supersonic guide-wave response signal, which has altogether, wherein symbol t indicates that the time becomes
Amount；
In second step (S2), according to supersonic guide-wave mode obtained, constructed respectively by mode under covering different propagation distance
Son propagate dictionary Di, and be combined to obtain multi-modal composite propagation dictionary D=[D1, D2..., DP]；
In third step (S3), supersonic guide-wave response signal r (t) is subjected to rarefaction representation at multi-modal composite propagation dictionary D,
Establish sparse representation model:
min ||w||0S.t.r=Dw=[D1, D2..., DP][w1, w2... WP]T,
In formula, r is by the discrete rear column vector r=[r (t formed of signal r (t)1), r (t2) ..., r (tN)]T, N is signal points,
Subscript T indicates the transposition operation of vector or matrix, w=[W1, W2... WP]T, w is column vector r in multi-modal composite propagation dictionary D
Under weight vectors, wherein wiCorresponding to sub- dictionary DiWeight coefficient, s.t. expression be constrained in；
In four steps (S4), solved using management loading algorithm with drag: r=Dw+n,
In formula, n is system noise, obtains weight vectors w by management loading algorithm, and by its corresponding sub- weight system
Number W1, W2... WPIt is separated one by one by mode；
In 5th step (S5), the multi-modal composite propagation dictionary D is pressed into mode decomposition, the single mode obtained under each mode passes
Broadcast dictionary D1, D2..., DP, each single mode is propagated into dictionary D1, D2..., DPWith corresponding sub- weight coefficient wiIt is multiplied to obtain
The single mode signal component of i mode stores obtained single mode signal component.
2. according to the method described in claim 1, wherein, it is preferred that in second step (S2), the multi-modal composite propagation
Dictionary D is by obtaining measurement method, finite element simulation method or analytic method.
3. according to the method described in claim 2, wherein, the measurement method is the position by changing excitation and receiving sensor
It sets, excitation motivates identical signal with former in former measured structure and same environment, measures its response and as matrix
Column vector forms a matrix in order and obtains composite propagation dictionary.
4. according to the method described in claim 2, wherein, the finite element simulation method is the ultrasound by establishing measured structure
Guided waves propagation finite element model, after being applied to identical original excitation signal, emulation obtains the response under different propagation distance
Signal and as matrix column vector, forms a matrix in order and obtains composite propagation dictionary.
5. according to the method described in claim 1, wherein, the analytic method is to obtain difference by following guided waves propagation model
Propagation distance x1, x2..., xMUnder response signal,
Dm=[gm(x1), gm(x2) ..., gm(xM)]T
In formula, DmM-th of mode propagation dictionary of table, gm(xj) expression propagation distance be xjM-th of mode under single modal response
Signal gm(xj, t) and the column vector that is formed after discrete, gm(xj)=[gm(xj, t1), gm(xj, t2) ..., gm(xj, tN)]T, wherein
gm(xj, t1), gm(xj, t2) ..., gm(xj, tN) respectively indicate signal gm(xj, t) the 1st, the 2nd, n-th discrete point, signal gm
(xj, t) and it is determined by following formula,In formula, S (ω) is former sharp
Encourage the frequency domain form of signal s (t), km(ω) is the frequency-wavenumber curve of m-th of mode, and ω is angular frequency, xMFor maximum propagation
Distance.
6. according to the method described in claim 5, wherein, maximum propagation distance xMMiddle acquisition duration T including original signal with should
Guided wave maximum propagation speed vmaxProduct, i.e. xM=Tvmax。
7. according to the method described in claim 1, wherein, supersonic guide-wave includes Lamb wave.
8. a kind of separation system for implementing any one of claim 1-7 the method, the separation system include,
Measured structure is provided to propagate supersonic guide-wave, and measured structure is equipped with driver and the acquisition of excitation supersonic guide-wave
The collector for the supersonic guide-wave propagated through measured structure,
Signal exciting unit is configured to issue supersonic guide-wave,
Signal amplification unit is configured to amplify amplifying unit one end connection signal exciting unit of the waveform signal, the other end
The driver in measured structure is connected,
Signal acquisition unit is configured in signal acquisition unit one end connection measured structure of acquisition supersonic guide-wave response signal
Collector, other end connection signal processing unit,
Signal processing unit obtains the single mode signal of each mode based on the supersonic guide-wave response signal.
9. separation system according to claim 8, wherein signal processing unit includes,
Multi-modal composite propagation dictionary creation unit, is configured to based on supersonic guide-wave mode obtained, distinguishes structure by mode
The son built under covering different propagation distance propagates dictionary DiAnd it is combined to obtain multi-modal composite propagation dictionary；
Sparse representation model generation unit is configured to carry out rarefaction representation at dictionary D based on response signal r (t) to establish
Sparse representation model；
Signal processing unit is configured with management loading algorithm solving model r=Dw+n, obtains weight vectors w,
And by its corresponding sub- weight coefficient W1, W2... WPIt is separated one by one by mode；
Computing unit is configured to multi-modal composite propagation dictionary D described in mode decomposition, and the single mode obtained under each mode is propagated
Dictionary D1, D2..., DP, each single mode is propagated into dictionary DiWith corresponding sub- weight coefficient wiIt is multiplied to obtain the list of m-th of mode
Mode signals ingredient.
10. separation system according to claim 8, wherein driver and collector are PZT piezoelectric ceramic piece, the letter
Number processing unit is digital signal processor, application-specific integrated circuit ASIC or on-site programmable gate array FPGA, at the signal
Managing unit includes memory, and the memory includes one or more read only memory ROM, random access memory ram, quick flashing
Memory or Electrical Erasable programmable read only memory EEPROM.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910183672.3A CN109871824B (en) | 2019-03-11 | 2019-03-11 | Ultrasonic guided wave multi-mode separation method and system based on sparse Bayesian learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910183672.3A CN109871824B (en) | 2019-03-11 | 2019-03-11 | Ultrasonic guided wave multi-mode separation method and system based on sparse Bayesian learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109871824A true CN109871824A (en) | 2019-06-11 |
| CN109871824B CN109871824B (en) | 2022-06-21 |
Family
ID=66920347
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910183672.3A Active CN109871824B (en) | 2019-03-11 | 2019-03-11 | Ultrasonic guided wave multi-mode separation method and system based on sparse Bayesian learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109871824B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110542723A (en) * | 2019-09-17 | 2019-12-06 | 哈尔滨工业大学 | guided wave signal sparse decomposition and damage positioning-based two-stage damage position identification method |
| CN114240899A (en) * | 2021-12-20 | 2022-03-25 | 中国石油大学(华东) | Guided wave frequency dispersion curve modal tracking method and system based on image registration |
| CN116046893A (en) * | 2023-02-22 | 2023-05-02 | 华南理工大学 | Ultrasonic guided wave signal enhancement method and system based on multi-mode recognition-fusion |
| CN116124902A (en) * | 2023-02-03 | 2023-05-16 | 哈尔滨工业大学 | Diagnostic method for ultrasonic guided wave damage positioning accuracy based on Bayes uncertainty quantification |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017165774A1 (en) * | 2016-03-25 | 2017-09-28 | Quad Analytix Llc | Systems and methods for multi-modal automated categorization |
| CN107607628A (en) * | 2017-08-11 | 2018-01-19 | 西安交通大学 | A kind of supersonic guide-wave frequency dispersion compensation method and its application based on rarefaction representation |
| CN108474755A (en) * | 2015-11-20 | 2018-08-31 | 集成动态电子解决方案公司 | Time compresses sensing system |
| CN109145418A (en) * | 2018-08-07 | 2019-01-04 | 暨南大学 | A kind of external load real-time identification method based on truncation response sparse decomposition method |
-
2019
- 2019-03-11 CN CN201910183672.3A patent/CN109871824B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108474755A (en) * | 2015-11-20 | 2018-08-31 | 集成动态电子解决方案公司 | Time compresses sensing system |
| WO2017165774A1 (en) * | 2016-03-25 | 2017-09-28 | Quad Analytix Llc | Systems and methods for multi-modal automated categorization |
| CN107607628A (en) * | 2017-08-11 | 2018-01-19 | 西安交通大学 | A kind of supersonic guide-wave frequency dispersion compensation method and its application based on rarefaction representation |
| CN109145418A (en) * | 2018-08-07 | 2019-01-04 | 暨南大学 | A kind of external load real-time identification method based on truncation response sparse decomposition method |
Non-Patent Citations (2)
| Title |
|---|
| XIN PENG等: ""Multimode Process Monitoring and Fault Detection: A Sparse Modeling and Dictionary Learning Method"", 《IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS》 * |
| 文方青等: ""基于块稀疏贝叶斯学习的多任务压缩感知重构算法"", 《物理学报》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110542723A (en) * | 2019-09-17 | 2019-12-06 | 哈尔滨工业大学 | guided wave signal sparse decomposition and damage positioning-based two-stage damage position identification method |
| CN110542723B (en) * | 2019-09-17 | 2020-04-24 | 哈尔滨工业大学 | Guided wave signal sparse decomposition and damage positioning-based two-stage damage position identification method |
| CN114240899A (en) * | 2021-12-20 | 2022-03-25 | 中国石油大学(华东) | Guided wave frequency dispersion curve modal tracking method and system based on image registration |
| CN114240899B (en) * | 2021-12-20 | 2024-04-09 | 中国石油大学(华东) | Guided wave dispersion curve modal tracking method and system based on image registration |
| CN116124902A (en) * | 2023-02-03 | 2023-05-16 | 哈尔滨工业大学 | Diagnostic method for ultrasonic guided wave damage positioning accuracy based on Bayes uncertainty quantification |
| CN116124902B (en) * | 2023-02-03 | 2023-08-18 | 哈尔滨工业大学 | Diagnostic method for ultrasonic guided wave damage positioning accuracy |
| CN116046893A (en) * | 2023-02-22 | 2023-05-02 | 华南理工大学 | Ultrasonic guided wave signal enhancement method and system based on multi-mode recognition-fusion |
| CN116046893B (en) * | 2023-02-22 | 2023-08-04 | 华南理工大学 | Ultrasonic guided wave signal enhancement method and system based on multi-mode recognition-fusion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109871824B (en) | 2022-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109871824A (en) | The multi-modal separation method of supersonic guide-wave and its system based on management loading | |
| Devenport et al. | Sound radiation from real airfoils in turbulence | |
| CN106960068B (en) | Rapid modal damping ratio calculation method based on pulse excitation response frequency spectrum | |
| Ishizaka et al. | Direct determination of vocal tract wall impedance | |
| CN108871647A (en) | Method for measuring the axial force of bolt | |
| CN106908755A (en) | Wireless acoustic sensor network pushes the sound bearing method of estimation of contracting gas leakage | |
| Ahlefeldt et al. | Road to acquisition: Preparing a MEMS microphone array for measurement of fuselage surface pressure fluctuations | |
| Schultz et al. | Modal decomposition method for acoustic impedance testing in square ducts | |
| Fedotov et al. | Numerical simulation of the processes in the normal incidence tube for high acoustic pressure levels | |
| CN104729677B (en) | A kind of time-domain digital weighted method of nonstationary noise signal | |
| Caresta et al. | Vibration of fluid loaded conical shells | |
| Dos Santos et al. | Inflow turbulence distortion for airfoil leading-edge noise prediction for large turbulence length scales for zero-mean loading | |
| Miao et al. | Experimental investigation on pulsating pressure of a cone-cylinder-hemisphere model under different flow velocities | |
| Doebler et al. | Stability of sonic boom metrics regarding signature distortions from atmospheric turbulence | |
| CN110045014A (en) | Lamb wave frequency dispersion removing method and its system based on Bayesian learning | |
| Waxler et al. | A theoretical treatment of the long range propagation of impulsive signals under strongly ducted nocturnal conditions | |
| DiPassio et al. | Direction of arrival estimation of an acoustic wave using a single structural vibration sensor | |
| Rindel et al. | Comparisons between computer simulations of room acoustical parameters and those measured in concert halls | |
| JP6886890B2 (en) | Decay time analysis methods, instruments, and programs | |
| Stinson et al. | Transverse pressure distributions in a simple model ear canal occluded by a hearing aid test fixture | |
| Caldwell Jr et al. | Characterizing wave behavior in a beam experiment by using complex orthogonal decomposition | |
| Brambley et al. | The critical layer in sheared flow | |
| Sisto et al. | Crucial 3-D viscous hydrodynamic contributions to the theoretical modeling of the cochlear response | |
| CN105301112A (en) | Method for measuring and calculating visco-elastic dynamic mechanical parameters of rubber-like damping material | |
| Van Walstijn et al. | Adaptive calibration of a three-microphone system for acoustic waveguide characterization under time-varying conditions |
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 |