CN103198053A - Instantaneous wavelet bicoherence method based on phase randomization - Google Patents
Instantaneous wavelet bicoherence method based on phase randomization Download PDFInfo
- Publication number
- CN103198053A CN103198053A CN201310091704XA CN201310091704A CN103198053A CN 103198053 A CN103198053 A CN 103198053A CN 201310091704X A CN201310091704X A CN 201310091704XA CN 201310091704 A CN201310091704 A CN 201310091704A CN 103198053 A CN103198053 A CN 103198053A
- Authority
- CN
- China
- Prior art keywords
- small echo
- formula
- instantaneous
- bicoherence
- random
- 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
Images
Landscapes
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention discloses an instantaneous wavelet bicoherence method based on phase randomization. The instantaneous wavelet bicoherence method based on the phase randomization comprises the following steps: firstly, selecting a mother wavelet function for continuous wavelet transform; secondly, performing continuous wavelet transform on a signal to be analyzed to obtain a time and frequency domain expression form of the signal; thirdly, computing an instantaneous wavelet bispectrum at each scale, computing the instantaneous wavelet bispectrum again, repeating computation, and summing computed results to take an expected instantaneous wavelet bispectrum based on the phase randomization; fourthly, computing the obtained instantaneous wavelet bispectrum based on the phase randomization, computing the obtained instantaneous wavelet bispectrum based on the phase randomization, and performing integration at a scale of s2 to obtain the instantaneous wavelet bispectrum based on the phase randomization; sixthly, computing to obtain wavelet bicoherence based on the phase randomization; and finally, computing to obtain instantaneous wavelet bicoherence based on the phase randomization. During the computation of the wavelet bicoherence, the amplitude and phase information are taken into account at the same time, so that defects of the conventional method such as low accuracy are avoided.
Description
Technical field
The invention belongs to the mechanical signal processing technology field of mechanical system fault diagnosis and control, particularly a kind of based on phase place instantaneous small echo bicoherence method at random.
Background technology
Two spectrums are proposed to use by Brillinger and Hinich the earliest.It is as the effective ways of system responses abnormal characteristic (the abnormal feature of system responses is the nonlinear sign of system) identification, and the signal that is widely used in fields such as geophysics, biological doctor's electricity, mechanical industry and radar signal is handled.In mechanical field, when there is fault in physical construction, the impact that produces owing to faults such as local damages will cause non-stationary, abnormal and the nonlinear characteristic of structural response signal.All there are the quadratic phase coupling usually in this class non-stationary, abnormal signal, and therefore, two spectrums and bicoherence method often are applied to this class Signal Processing and analysis.Traditional double spectrum and bicoherence method are based on all that Fast Fourier Transform (FFT) and Short Time Fourier Transform realize, defective such as have a little less than the signal transient characteristic recognition capability.For overcoming this defective, the two spectrums of small echo with signal transient characteristic recognition capability are suggested and widespread use with the bicoherence method.This method combines two spectrums and bicoherence method, and nonlinear characteristic detectability advantage and continuous wavelet transform thereby can obtain more efficient, objective testing result in actual applications to the advantage of signal transient feature recognition capability efficiently.
Usually, the two spectrums of small echo and the bicoherence of calculating signal all comprise following steps, at first, be the data segment of N equal time (establishing the time interval is T) with division of signal to be analyzed, then, each data segment is carried out the calculating of the two spectrums of small echo and bicoherence, last, every section result of calculation summation is got average as final result of calculation.When coherence time of signal itself with respect to the time interval T that gets more in short-term, the phase component of each data segment is separate.Yet, for most mechanical signal, all have long coherence time usually, therefore, the method for calculating the two spectrums of small echo and bicoherence by simple dividing data section can cause quadratic phase modulation result false on the spectrogram.Simultaneously, when signal existed quadratic phase coupling and non-quadratic phase coupling information simultaneously on identical bifrequency, the two spectrums of traditional small echo can not effectively be identified with the bicoherence computing method.Above problem all can significantly reduce the accuracy of result of calculation.Therefore, press for a kind of otherwise effective technique more to improve the accuracy that detects.
Summary of the invention
In order to overcome the shortcoming of above-mentioned prior art, the object of the present invention is to provide a kind ofly based on phase place instantaneous small echo bicoherence method at random, when calculating the small echo bicoherence, consider its amplitude and phase information simultaneously, avoided the classic method defects of low accuracy.
In order to achieve the above object, the technical scheme taked of the present invention is:
A kind of based on phase place instantaneous small echo bicoherence method at random, may further comprise the steps:
Step 1: select to be used for the mother wavelet function of continuous wavelet transform, when the non-stationary signal in the mechanical signal was detected, female small echo adopted the Morlet small echo shown in the formula (1),
Wherein: σ---decay factor;
The female little wave frequency of f---Morlet;
Step 2: adopt formula (2) to treat analytic signal and make continuous wavelet transform, the time-frequency domain expression-form of picked up signal:
Wherein: ψ (t)---selected female small echo;
X (t)---signal to be analyzed;
The s---scale factor;
The t---time factor;
*--conjugation is got in-expression;
Step 3: calculate two spectrums of instantaneous small echo under each yardstick according to formula (3):
B
W,T(s
1,s
2,t)=W
ψ(s
1,t)W
ψ(s
2,t)W
ψ *(s,t) (3)
Wherein: s, s
1With s
2Satisfy relation
Owing to calculate the two spectrums of instantaneous small echo for plural number, so formula (3) also can be expressed as the form of formula (4)
Wherein: A (s
1, s
2, t)---bifrequency (s
1, s
2) time the two spectral amplitude ratio of instantaneous small echo;
Step 4: will be calculated the two spectrums of instantaneous small echo by the instantaneous small echo quarter-phase substitution formula (5) that formula (3) calculate:
Wherein: the R---interval is the sequence of random variables of [π, π];
The instantaneous small echo quarter-phase that---by formula (3) calculates;
Step 5: step 4 is repeated 150 times, and the result that will calculate summation, get expectation and obtain based on the two spectrums of phase place instantaneous small echo at random, as shown in Equation (6):
Wherein: E{.}---represents to get desired operation;
Step 6: will be obtained based on the two spectrums of phase place small echo at random as shown in Equation (7) in time interval T upper integral by composing based on phase place instantaneous small echo at random is two that formula (6) calculate:
Step 7: will be by composing based on phase place instantaneous small echo at random is two that formula (6) calculate, at yardstick s
2On carry out integration and obtain based on the two spectrums of phase place instantaneous small echo at random as shown in Equation (8):
Step 8: will be calculated by the formula of substitution as a result (9) that formula (7) calculate based on phase place small echo bicoherence at random:
Step 9: will be calculated by the formula of substitution as a result (10) that formula (8) calculate based on phase place instantaneous small echo bicoherence at random:
Advantage of the present invention is: traditional small echo bicoherence computing formula (11) is;
W
ψ(s, t)---yardstick is the continuous wavelet transform of s;
B
W, T(s
1, s
2)---be that traditional two spectrums of small echo are calculated, as shown in Equation (12),
Wherein: *--conjugation is got in-expression,
A (s
1, s
2, t)---bifrequency is (s
1, s
2) time the two spectral amplitude ratio of instantaneous small echo;
The small echo bicoherence that is calculated by formula (11) is yardstick s
1With s
2Function, span is 0 to 1.Desirable small echo bicoherence computing formula need satisfy following two conditions:
If I. signal is at bifrequency (s
1, s
2) locate to exist the quadratic phase coupling, then at bifrequency (s
1, s
2) locate small echo bicoherence value b
W, T(s
1, s
2)=1 is simultaneously at bifrequency (s
1, s
2) quarter-phase located
If II. signal is at bifrequency (s
1, s
2) locate not exist the quadratic phase coupling, then at bifrequency (s
1, s
2) locate the value b of small echo bicoherence
W, T(s
1, s
2)=0 is simultaneously at bifrequency (s
1, s
2) quarter-phase located
Usually, the traditional small echo bicoherence computing method requirement of II that all can not satisfy condition.Therefore, the present invention considers its amplitude and phase information simultaneously when calculating the small echo bicoherence, avoided defectives such as the classic method accuracy is low.
Description of drawings
Fig. 1 is the simulate signal time domain waveform.
Fig. 2 is simulate signal continuous wavelet transform spectrum.
Fig. 3 is based on phase place small echo bicoherence spectrum at random.
Fig. 4 is traditional small echo bicoherence spectrum.
Fig. 5 is based on phase place instantaneous small echo bicoherence spectrum at random.
Fig. 6 is traditional instantaneous small echo bicoherence spectrum.
Embodiment
Below in conjunction with drawings and Examples the present invention is described in detail.
Be example with the simulate signal, the expression formula of simulate signal as shown in Equation (13)
Wherein: ε (t)---average is zero white Gaussian noise;
f
i---coupling frequency (i=1,2).
The sample frequency of this simulate signal is 100Hz, wherein f
1=9Hz, f
2=19Hz, property is made an uproar than being 3dB.Be f at first and the 3rd ripple bag medium frequency
1And f
2Cosine wave (CW) satisfy the quadratic phase coupling condition, and second and the 4th ripple bag medium frequency are f
1And f
2Cosine wave (CW) do not satisfy the quadratic phase coupling condition.The time domain waveform of simulate signal as shown in Figure 1; For the coupling of two chi phase places in the identification signal, adopt the present invention that data are analyzed.
A kind of based on phase place instantaneous small echo bicoherence method at random, may further comprise the steps:
Step 1: select to be used for the mother wavelet function of continuous wavelet transform, when the non-stationary signal in the mechanical signal was detected, female small echo adopted the Morlet small echo shown in the formula (1),
Wherein: get decay factor σ=4, the frequency f=0.95Hz of female small echo is to satisfy the admissibility condition;
Step 2: adopt formula (2) that simulate signal x (t) is made continuous wavelet transform, according to the sample frequency of simulate signal, the yardstick of getting travels through whole Fourier analysis frequency, and the time-frequency domain expression-form of picked up signal obtains the time-frequency figure of signal as shown in Figure 2;
Wherein: ψ (t)---selected female small echo;
X (t)---simulate signal;
The s---scale factor;
The t---time factor;
*--conjugation is got in-expression;
Step 3: calculate two spectrums of instantaneous small echo under each yardstick according to formula (3), wherein getting frequency resolution is 0.25Hz, and obtains the two spectrum of the instantaneous small echo phase places under this yardstick
B
W,T(s
1,s
2,t)=W
ψ(s
1,t)W
ψ(s
2,t)W
ψ *(s,t) (3)
Step 4: will be calculated the two spectrums of instantaneous small echo by the instantaneous small echo quarter-phase substitution formula (5) that formula (3) calculate;
Wherein: the R---interval is the sequence of random variables of [π, π];
The instantaneous small echo quarter-phase that---by formula (3) calculates;
Step 5: step 4 is repeated 150 times, and the result that will calculate summation, get expectation and obtain based on the two spectrums of phase place instantaneous small echo at random, as shown in Equation (6);
Wherein: E{.}---represents to get desired operation;
Step 6: will be obtained based on the two spectrums of phase place small echo at random as shown in Equation (7) in time interval T upper integral by composing based on phase place instantaneous small echo at random is two that formula (6) calculate;
Step 7: will be by composing based on phase place instantaneous small echo at random is two that formula (6) calculate, at yardstick s
2On carry out integration and obtain based on the two spectrums of phase place instantaneous small echo at random as shown in Equation (8);
Step 8: will be calculated based on phase place small echo bicoherence at random by the formula of substitution as a result (9) that formula (7) calculate, result of calculation as shown in Figure 3;
The collection of illustrative plates that Fig. 4 obtains for traditional small echo bicoherence computing method, as can be seen from Figure 4, spectrogram exists at a plurality of frequencies place and false peak value occurs, is difficult to judge real quadratic phase coupling frequency composition.Yet, based on phase place instantaneous small echo bicoherence method at random only at the frequency content place (9Hz and 19Hz) that has the quadratic phase coupling peak value appears, as shown in Figure 3, therefore, can overcome classic method effectively under certain conditions based on phase place instantaneous small echo bicoherence method at random, the defective of the false quadratic phase coupling testing result that exists has improved the accuracy that detects;
Step 9: will be calculated by the formula of substitution as a result (10) that formula (8) calculate based on phase place instantaneous small echo bicoherence at random, result of calculation as shown in Figure 5;
The spectrogram that is calculated by the instantaneous bicoherence method of traditional small echo as shown in Figure 6, as seen from Figure 6, no matter ripple is surrounded by does not exist the quadratic phase coupling, testing result all demonstrates bigger small echo bicoherence coefficient.And as shown in Figure 5, only there are first and the 3rd the bigger small echo two-phase responsibility of ripple bag place appearance of quadratic phase coupling based on phase place instantaneous small echo bicoherence spectrum at random, and at second and the 4th ripple bag place not having the quadratic phase coupling, its small echo bicoherence coefficient is then very little.What this patent proposed can overcome classic method effectively based on phase place instantaneous small echo bicoherence computing method at random, when quadratic phase coupling exists simultaneously with non-quadratic phase coupling under identical bifrequency and the defective that can't differentiate, improve the accuracy that detects, be applicable to each associated field.
Claims (1)
1. one kind based on phase place instantaneous small echo bicoherence method at random, it is characterized in that, may further comprise the steps:
Step 1: select to be used for the mother wavelet function of continuous wavelet transform, when the non-stationary signal in the mechanical signal was detected, female small echo adopted the Morlet small echo shown in the formula (1),
Wherein: σ---decay factor;
The female little wave frequency of f---Morlet;
Step 2: adopt formula (2) to treat analytic signal and make continuous wavelet transform, the time-frequency domain expression-form of picked up signal:
X (t)---signal to be analyzed;
The s---scale factor;
The t---time factor;
*--conjugation is got in-expression;
Step 3: calculate two spectrums of instantaneous small echo under each yardstick according to formula (3):
B
W,T(s
1,s
2,t)=W
ψ(s
1,t)W
ψ(s
2,t)W
ψ *(s,t) (3)
Wherein: s, s
1With s
2Satisfy relation
Owing to calculate the two spectrums of instantaneous small echo for plural number, so formula (3) also can be expressed as the form of formula (4)
Wherein: A (s
1, s
2, t)---bifrequency (s
1, s
2) time the two spectral amplitude ratio of instantaneous small echo;
Step 4: will be calculated the two spectrums of instantaneous small echo by the instantaneous small echo quarter-phase substitution formula (5) that formula (3) calculate:
Wherein: the R---interval is the sequence of random variables of [π, π];
Step 5: step 4 is repeated 150 times, and the result that will calculate summation, get expectation and obtain based on the two spectrums of phase place instantaneous small echo at random, as shown in Equation (6):
Wherein: E{.}---represents to get desired operation;
Step 6: will be obtained based on the two spectrums of phase place small echo at random as shown in Equation (7) in time interval T upper integral by composing based on phase place instantaneous small echo at random is two that formula (6) calculate:
Step 7: will be by composing based on phase place instantaneous small echo at random is two that formula (6) calculate, at yardstick s
2On carry out integration and obtain based on the two spectrums of phase place instantaneous small echo at random as shown in Equation (8):
Step 8: will be calculated by the formula of substitution as a result (9) that formula (7) calculate based on phase place small echo bicoherence at random:
Step 9: will be calculated by the formula of substitution as a result (10) that formula (8) calculate based on phase place instantaneous small echo bicoherence at random:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310091704.XA CN103198053B (en) | 2013-03-21 | 2013-03-21 | A kind of instantaneous small echo bicoherence method random based on phase place |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310091704.XA CN103198053B (en) | 2013-03-21 | 2013-03-21 | A kind of instantaneous small echo bicoherence method random based on phase place |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103198053A true CN103198053A (en) | 2013-07-10 |
CN103198053B CN103198053B (en) | 2015-11-25 |
Family
ID=48720623
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310091704.XA Active CN103198053B (en) | 2013-03-21 | 2013-03-21 | A kind of instantaneous small echo bicoherence method random based on phase place |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103198053B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111610364A (en) * | 2020-06-02 | 2020-09-01 | 江苏方天电力技术有限公司 | Forced oscillation mode correlation analysis method based on bispectrum |
CN117347773A (en) * | 2023-12-05 | 2024-01-05 | 天津致新轨道交通运营有限公司 | Intelligent service method based on multi-equipment linkage |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6411914B1 (en) * | 1999-11-29 | 2002-06-25 | Goodrich Corporation | System and method for coherent signal detection using wavelet functions |
CN102937477A (en) * | 2012-11-06 | 2013-02-20 | 昆山北极光电子科技有限公司 | Bi-spectrum analysis method for processing signals |
-
2013
- 2013-03-21 CN CN201310091704.XA patent/CN103198053B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6411914B1 (en) * | 1999-11-29 | 2002-06-25 | Goodrich Corporation | System and method for coherent signal detection using wavelet functions |
CN102937477A (en) * | 2012-11-06 | 2013-02-20 | 昆山北极光电子科技有限公司 | Bi-spectrum analysis method for processing signals |
Non-Patent Citations (2)
Title |
---|
JING LIN ET AL: "FEATURE EXTRACTION BASED ON MORLET WAVELET AND ITS APPLICATION FOR MECHANICAL FAULT DIAGNOSIS", 《JOURNAL OF SOUND AND VIBRATION》, vol. 234, no. 1, 29 June 2000 (2000-06-29), pages 135 - 148 * |
林京等: "基于连续小波变换的奇异性检测与故障诊断", 《振动工程学报》, vol. 13, no. 4, 30 December 2000 (2000-12-30), pages 31 - 38 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111610364A (en) * | 2020-06-02 | 2020-09-01 | 江苏方天电力技术有限公司 | Forced oscillation mode correlation analysis method based on bispectrum |
CN117347773A (en) * | 2023-12-05 | 2024-01-05 | 天津致新轨道交通运营有限公司 | Intelligent service method based on multi-equipment linkage |
Also Published As
Publication number | Publication date |
---|---|
CN103198053B (en) | 2015-11-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Chen et al. | Intrinsic chirp component decomposition by using Fourier series representation | |
CN106226407B (en) | A kind of online preprocess method of ultrasound echo signal based on singular spectrum analysis | |
Obuchowski et al. | Selection of informative frequency band in local damage detection in rotating machinery | |
CN104820786A (en) | Method for analyzing instantly weighted synchronous extrusion wavelet bispectrum | |
CN110236573A (en) | The detection method and relevant apparatus of psychological pressure state | |
CN103743980A (en) | Power quality disturbance recognition and classification method based on PSO (Particle Swarm Optimization) for SVM (Support Vector Machine) | |
CN105572501B (en) | A kind of electrical energy power quality disturbance recognition methods based on SST transformation and LS-SVM | |
CN106546818A (en) | A kind of harmonic signal detection method based on DNL Mode Decomposition | |
KR101294681B1 (en) | Apparatus and method for processing weather signal | |
Wang et al. | Application of the dual-tree complex wavelet transform in biomedical signal denoising | |
CN106446868A (en) | Side channel signal feature extraction method based on EMD and singular value difference spectrum | |
CN105447464A (en) | Electric energy quality disturbance recognition and classification method based on PSO | |
CN105258940A (en) | Standardized multiwavelet and multiwavelet packet transformation method for mechanical failure quantitative extraction | |
CN103198053A (en) | Instantaneous wavelet bicoherence method based on phase randomization | |
CN106842141B (en) | A kind of high-order is repeatedly conjugated lagged product Intrapulse analysis method | |
CN104655965B (en) | A kind of phasor measurement method in power system | |
Niazy et al. | Performance evaluation of ensemble empirical mode decomposition | |
CN107688167B (en) | Multi-time-width linear frequency modulation pulse compression signal amplitude envelope curve generation method | |
Zeng et al. | Underwater sound classification based on Gammatone filter bank and Hilbert-Huang transform | |
CN103438983B (en) | Data processing method of signal random average spectrums | |
CN104535855A (en) | Electric energy quality disturbing signal detecting algorithm based on discrete orthogonal S transformation | |
EP2629089B1 (en) | Method and apparatus for wavelet-based propagation time measurement of ultrasonic pulses | |
Gao et al. | A Method Using EEMD and L-Kurtosis to detect faults in roller bearings | |
US8867862B1 (en) | Self-optimizing analysis window sizing method | |
CN104808643A (en) | Control circuit nonlinearity detection method based on improved bi-cepstrum analysis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |