CN112504441A - Vibration acceleration signal segmentation and integration method based on important information reconstruction - Google Patents
Vibration acceleration signal segmentation and integration method based on important information reconstruction Download PDFInfo
- Publication number
- CN112504441A CN112504441A CN202011472767.6A CN202011472767A CN112504441A CN 112504441 A CN112504441 A CN 112504441A CN 202011472767 A CN202011472767 A CN 202011472767A CN 112504441 A CN112504441 A CN 112504441A
- Authority
- CN
- China
- Prior art keywords
- frequency
- signal
- integration
- amplitude
- low
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- 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
Abstract
A vibration acceleration signal segmentation integration method based on important information reconstruction comprises the following steps; determining a critical point of frequency domain segmentation through a signal-to-noise threshold, performing forward exploration by taking the maximum value of the type I frequency and the type II frequency as an initial point, and taking the frequency with the frequency domain amplitude being smaller than the signal-to-noise threshold for the first time as the critical point; determining whether the type II frequency exists or not through comparison of track chaos degrees of front and back phase planes of the signal added by the chaotic oscillator; in a low-frequency band smaller than the critical point, all III-class frequencies larger than a signal-to-noise threshold value in the low-frequency band are obtained by a tip-pulling screening method; correcting amplitude and phase information of checked class I, class II and low-frequency band class III frequency points; reconstructing a low-frequency band signal according to the amplitude and the phase of the corrected frequency and integrating; performing frequency domain integration on the high-frequency band signals which are greater than or equal to the critical point; and combining the low-frequency-band integration result with the high-frequency-band integration result. The invention realizes the reconstruction and integration of the signals belonging to the three types of monitoring frequency sets.
Description
Technical Field
The invention relates to the technical field of vibration signal processing, in particular to a vibration acceleration signal segmentation and integration method based on important information reconstruction.
Background
The measuring device of the vibration signal comprises an acceleration sensor, a speed sensor and a displacement sensor, but the speed sensor and the displacement sensor are limited in use due to the restriction of environment or installation conditions, and the acceleration sensor is widely applied due to small volume, light weight and convenient installation. However, for many rotating machines, not only the vibration acceleration signal, but also the velocity or displacement signal is important, which involves the problem of obtaining a first or second integral of the velocity or displacement signal from the acceleration signal.
In the field of software integration of acceleration signals, there are generally two large directions, i.e. time domain integration and frequency domain integration. The time domain integration is mainly to perform numerical integration on the signal subjected to direct current removal in the time domain through a trapezoidal formula, a Simpson formula and the like, but the result is greatly influenced by noise and trend terms, and the accumulation of errors is caused by two times of integration. The frequency domain integration firstly converts the acceleration frequency spectrum into a speed or displacement frequency spectrum in a frequency domain, and then obtains the speed or displacement frequency spectrum through inverse Fourier transform, but the frequency domain integration is particularly sensitive to low-frequency components, when low-frequency noise is obvious, the frequency domain integration can seriously amplify low-frequency noise signals, and even covers useful signals at middle and high frequencies, so that the integrated signals can hardly be used. In addition, another scholars provides an accurate reconstruction integration method for a fault rotor system, wherein signals of frequency conversion, frequency multiplication and frequency division and multiplication are extracted and integrated, and the integration result is more accurate for the extracted frequency signals.
Therefore, it is necessary to provide a software integration method for acceleration signals, which can not only effectively improve the accuracy of the integration result, but also cover the amount of the original signal information as much as possible to avoid the loss of the effective information.
Disclosure of Invention
In order to overcome the technical problems, the invention aims to provide a vibration acceleration signal segmentation and integration method based on important information reconstruction, which can effectively improve the accuracy of an integration result, and can cover the original signal information quantity as much as possible so as to avoid the loss of effective information, thereby realizing the reconstruction and integration of signals to which three types of monitoring frequency sets belong.
In order to achieve the purpose, the technical scheme adopted by the invention and the beneficial effects of the invention are as follows:
a vibration acceleration signal segmentation integration method based on important information reconstruction comprises the following steps;
a) determining a critical point of frequency domain segmentation through a signal-to-noise threshold, performing forward exploration by taking the maximum value of the type I frequency and the type II frequency as an initial point, and taking the frequency with the frequency domain amplitude being smaller than the signal-to-noise threshold for the first time as the critical point;
b) determining whether the type II frequency exists or not through comparison of track chaos degrees of front and back phase planes of the signal added by the chaotic oscillator;
c) in a low-frequency band smaller than the critical point, all III-class frequencies larger than a signal-to-noise threshold value in the low-frequency band are obtained by a tip-pulling screening method;
d) correcting amplitude and phase information of checked class I, class II and low-frequency band class III frequency points;
e) reconstructing a low-frequency band signal according to the amplitude and the phase of the corrected frequency and integrating;
f) performing frequency domain integration on the high-frequency band signals which are greater than or equal to the critical point;
g) and combining the low-frequency-band integration result with the high-frequency-band integration result.
Further, the class I frequency represents a frequency with definite frequency and significantly higher amplitude than noise, such as 1-3 frequency multiplication and the like; the type II frequency represents a frequency which is determined but not determined to be present, such as a fault characteristic frequency or certain high frequency multiplication components of a rolling bearing; class iii frequencies represent frequencies where the frequency is uncertain and uncertain as to whether it occurs, such as certain frequency division multiple components and other unknown frequency components.
Further, the method for calculating the signal noise threshold in step a) is as follows:
1) arranging the amplitudes of the signal frequency spectrums in a reverse order;
2) truncating the tail 3/4 of the amplitude sequence to obtain a sequence ai，i＝1,2,…,n；
3) Calculating a signal-to-noise threshold value sigma by the formula
Further, the critical point in step a) is a separation frequency dividing the frequency domain into a low frequency band and a high frequency band, so that the integral of the whole signal is divided into the integral of the low frequency band and the integral of the high frequency band.
Further, the chaotic oscillator in the step b) is formed by a Homles type Duffing equation, and the type II frequency to be detected is used as the periodic perturbation power frequency in the oscillator.
Further, the calculation method of the phase plane trajectory chaos S in step b) is as follows:
1) calculating the maximum value Ma and the minimum value Mi of a sequence (one of the two is selected) formed by x or y of the phase plane trajectory;
2) judging from the starting point of the sequence, and taking a point of the sequence as a new sequence until the sequence is finished when the value of the point of the sequence is in a Ma-0.08 (Ma-Mi) -Ma interval or Mi-Mi +0.08 (Ma-Mi) interval;
3) and carrying out empirical mode decomposition on the new sequence, and taking the number of eigenmode functions obtained by decomposition as the chaos degree S of the phase plane trajectory.
The comparison of the track chaos of the front phase plane and the rear phase plane of the signal added by the chaotic oscillator in the step b) means that when the frequency of the periodic perturbation force in the chaotic oscillator is the frequency f to be measured, if the formula S is satisfied1≤S 02, then it is verified that a signal with frequency f is present in the original signal, in which S1For the degree of disorder of the phase plane trajectory after addition of the signal, S0The degree of misordering of the phase plane trajectories before adding the signal.
Further, the tip-pulling screening method in the step c) comprises the following steps:
1) setting the amplitudes of the class I frequency, the confirmed class II frequency, 3 frequency points on the left side of each frequency and 3 frequency points on the right side of each frequency as 0 in a low frequency band;
2) finding out the maximum amplitude and the corresponding frequency in the rest low-frequency-band amplitudes, if the amplitude is greater than a signal-to-noise threshold value, dividing the frequency into the III-class frequency of the low frequency band, and simultaneously setting the amplitude of the frequency and 3 points on the left side and 3 points on the right side of the frequency as 0;
3) repeating the step 2) until the found maximum amplitude is less than or equal to the signal-to-noise threshold.
Further, the method for correcting the amplitude and phase of the verified kernel frequency f in the step e) comprises the following steps:
1) finding out the corresponding position i and amplitude A of the frequency f in the frequency spectrumiAnd phase thetai；
2) Calculating the spectral line moving length dL according to the formula
dL＝α(2Ai-1-Ai)/(Ai+Ai-1)-(1-α)(2Ai+1-Ai)/(Ai+Ai+1) In the formula, Ai-1Is the amplitude, A, of the first line to the left of frequency f in the spectrumi+1Is the amplitude of the first spectral line to the right of frequency f in the spectrum, when Ai-1>Ai+1When A is 1, alpha isi-1≤Ai+1When α is 0;
3) calculating the corrected frequency F, wherein the formula is F ═ (i-1-dL) × fs/N, fs is the sampling frequency, and N is the number of sampling points;
4) calculating the corrected amplitude A, wherein the formula is that A is (1-dL)2)*Ai/sinc(dL)；
5) Calculating the corrected phase theta, wherein the formula is that theta is 180 (theta)i+πdL)/π。
Further, in the step f), the frequency domain integration of the high-frequency band signal is performed, the lower limit cut-off frequency is the frequency of the critical point, the upper limit cut-off frequency can be optionally selected according to the analysis requirement, and the maximum limit does not exceed the sampling frequency divided by 2.56.
Further, the merging of the integration results in step g) is to add the integration result sequence of the low frequency band and the integration result sequence of the high frequency band at any same time in the time domain.
The invention has the beneficial effects that:
according to the vibration acceleration signal segmentation and integration method based on important information reconstruction, provided by the invention, starting from the defect that frequency domain integration is particularly sensitive to low-frequency noise, a signal is segmented into a low-frequency band and a high-frequency band, the frequency reflecting the characteristics of the signal is found out and corrected through methods such as chaotic oscillator and tip-pulling search in the low-frequency band, a frequency domain integration mode is adopted in the high-frequency band, and finally, the final integrated signal is obtained by combining the integration results of the low-frequency band and the high-frequency band. Therefore, the low-frequency trend term generated by time domain integration is avoided, the strong interference of low-frequency noise in frequency domain integration is bypassed, finally reserved signals cover useful signals as much as possible, and the accuracy of speed and displacement signals obtained by integration is improved. In the fields of vibration monitoring and fault diagnosis of rotary machinery, three signals of acceleration, speed and displacement and corresponding characteristic values are often needed for comprehensive monitoring and diagnosis of equipment, and more accurate speed and displacement signals undoubtedly improve the accuracy of equipment state monitoring, alarming and fault diagnosis.
Drawings
FIG. 1 is a flow chart of a vibration acceleration signal segmentation and integration method based on important information reconstruction.
FIG. 2 is a graph of raw acceleration and frequency spectrum for an embodiment.
FIG. 3 is the result of empirical mode decomposition of the trajectory x sequence before adding the signal to calculate the phase plane trajectory chaos.
Fig. 4 shows the result of empirical mode decomposition of the trajectory x sequence after adding the signal and calculating the phase plane trajectory chaos.
Fig. 5 is a comparison of the velocity profile and frequency spectrum obtained by the integration method of the present invention with theoretical values.
Fig. 6 is a comparison of the displacement curve and the frequency spectrum obtained by the integration method of the present invention with theoretical values.
Fig. 7 is a comparison of the velocity curve and the frequency spectrum obtained by the frequency domain integration method with theoretical values.
Fig. 8 is a comparison of a displacement curve and a frequency spectrum obtained by a frequency domain integration method with theoretical values.
Fig. 9 is a comparison of the velocity profile and frequency spectrum obtained by the time domain integration method with theoretical values.
Fig. 10 is a comparison of a displacement curve and a frequency spectrum obtained by a time domain integration method with theoretical values.
Detailed Description
The invention is further described in the following with reference to the accompanying drawings and examples.
The flow of the vibration acceleration signal segmentation integration method based on important information reconstruction provided by the invention is shown in figure 1, and comprises the following steps:
1) and determining a critical point of the frequency domain segmentation through a signal-to-noise threshold, performing forward exploration by taking the maximum value of the class I frequency and the class II frequency as an initial point, and taking the frequency with the frequency domain amplitude being smaller than the signal-to-noise threshold for the first time as the critical point.
The class I frequency, the class II frequency and the class III frequency contained in the constructed case signal are respectively as follows:
class i frequency: 1.0X, 2.0X, 3.0X;
class ii frequency: 0.61X, 0.78X, 1.56X, 2.69X, 4.0X, 5.0X;
class iii frequency: 1.21X, 2.42X, 4.71X, 7.28X, 10.93X, 14.88X, 18.48X, 23.75X.
Where X refers to the frequency of the frequency conversion, here 24.8 Hz. The raw acceleration curve and spectrum used for the example are shown in fig. 2.
According to the calculation mode of the signal-to-noise threshold value: a) arranging the amplitudes of the signal frequency spectrums in a reverse order; b) truncating the tail 3/4 of the amplitude sequence to obtain a sequence aiI ═ 1,2, …, n; c) calculating a signal-to-noise threshold value sigma by the formula
A value of 0.0447 was obtained.
The maximum value of the class I frequency and the class II frequency is 5.0X, namely 124Hz, and the critical point of the frequency domain section obtained by combining the signal-to-noise threshold value sigma is 125.5 Hz.
2) And determining whether the type II frequency exists or not by comparing the track chaos degrees of the front phase plane and the rear phase plane of the signal added by the chaotic oscillator.
The chaotic oscillator is formed by a Homles type Duffing equation, and the type II frequency to be detected is used as the periodic perturbation power frequency in the oscillator.
The calculation method of the phase plane trajectory chaos degree S comprises the following steps: a) calculating the maximum value Ma and the minimum value Mi of a sequence (one of the two is selected) formed by x or y of the phase plane trajectory; b) judging from the starting point of the sequence, and taking a point of the sequence as a new sequence until the sequence is finished when the value of the point of the sequence is in a Ma-0.08 (Ma-Mi) -Ma interval or Mi-Mi +0.08 (Ma-Mi) interval; c) and carrying out empirical mode decomposition on the new sequence, and taking the number of eigenmode functions obtained by decomposition as the chaos degree S of the phase plane trajectory.
Taking 1.56X in class II frequency as an example, the degree of misordering (S) of the phase plane trajectories before and after adding the signal is calculated by the method0And S1) 5 and 0, respectively, satisfy the formula S1≤S 02, the original signal contains a periodic signal with a frequency of 1.56X. The results of empirical mode decomposition of the x-sequence of the trajectory when calculating the phase plane trajectory chaos before and after adding the signal are shown in fig. 3 and 4, respectively. And sequentially adopting the method to identify all the II type frequencies.
3) And in the low frequency band smaller than the critical point, all the III-class frequencies larger than the signal-to-noise threshold value in the low frequency band are obtained by a tip-pulling screening method.
The tip-pulling screening method comprises the following steps: a) setting the amplitudes of the class I frequency, the confirmed class II frequency, 3 frequency points on the left side of each frequency and 3 frequency points on the right side of each frequency as 0 in a low frequency band; b) finding out the maximum amplitude and the corresponding frequency in the rest low-frequency-band amplitudes, if the amplitude is greater than a signal-to-noise threshold value, dividing the frequency into the III-class frequency of the low frequency band, and simultaneously setting the amplitude of the frequency and 3 points on the left side and 3 points on the right side of the frequency as 0; c) repeating step b) until the found maximum amplitude is less than or equal to the signal-to-noise threshold. All low band (0Hz to critical point) class III frequencies are thus obtained.
4) And correcting the amplitude and phase information of the checked class I, class II and low-frequency-band class III frequency points.
The method for correcting the checked frequency comprises the following steps: a) finding out the corresponding position i and amplitude A of the frequency f in the frequency spectrumiAnd phase thetai(ii) a b) Calculating the spectral line shift length dL, wherein the formula is dL ═ alpha (2A)i-1-Ai)/(Ai+Ai-1)-(1-α)(2Ai+1-Ai)/(Ai+Ai+1) In the formula, Ai-1Is the amplitude, A, of the first line to the left of frequency f in the spectrumi+1Is the amplitude of the first spectral line to the right of frequency f in the spectrum, when Ai-1>Ai+1When A is 1, alpha isi-1≤Ai+1When α is 0; c) calculating the corrected frequency F, and the formula is as follows: f ═ i-1-dL × fs/N, where fs is the sampling frequency and N is the number of sampling points; d) calculating the corrected amplitude A, wherein the formula is that A is (1-dL)2)*Ai(dL); e) calculating the corrected phase theta, wherein the formula is that theta is 180 (theta)i+πdL)/π。
5) The low band signal is reconstructed from the corrected frequency amplitude and phase and integrated.
6) And performing frequency domain integration on the high-frequency band signal which is greater than or equal to the critical point.
And (3) integrating the high-frequency band signal in a frequency domain, wherein the lower limit cut-off frequency is the frequency of a critical point, the upper limit cut-off frequency can be selected optionally according to the analysis requirement, and the maximum cut-off frequency does not exceed the sampling frequency divided by 2.56. Here the upper cut-off frequency is taken to be the maximum, i.e. the sampling frequency divided by 2.56.
7) And combining the low-frequency-band integration result with the high-frequency-band integration result.
The integration result combination is to add the integration result sequence of the low frequency band and the integration result sequence of the high frequency band at any one same time in the time domain.
The velocity curve and the frequency spectrum obtained by the integration method according to the present invention are shown in fig. 5, and the displacement curve and the frequency spectrum are shown in fig. 6, and are compared with the theoretical integration result in the figure.
For comparison, fig. 7 shows a velocity curve and a frequency spectrum obtained by a frequency domain integration method compared with theoretical values, and fig. 8 is a displacement curve and a frequency spectrum; fig. 9 shows a comparison of the velocity curve and the frequency spectrum obtained by the time domain integration method with theoretical values, and fig. 10 is a displacement curve and a frequency spectrum. For visual comparison, only the first 400 points of the data were plotted in all curves and spectra.
First, the result of frequency domain integration is observed, wherein the lower limit cut-off frequency used by the frequency domain integration is 5Hz, and no matter the frequency domain integration is carried out to speed or the frequency domain integration is carried out to displacement, the amplified trace of the noise at the low frequency can be obviously seen from the frequency domain, and the larger the integration times, the larger the noise amplitude is, and even the noise amplitude exceeds the main frequency. When the frequency is higher, the influence of noise on integration is smaller, and the amplitude of the noise after integration is smaller; conversely, the lower the frequency, the greater the effect of the noise on the integration and the greater the amplitude of the noise after integration. Therefore, when the lower limit cutoff frequency is slightly changed, the influence on the integration result is also very significant.
Observing the result of time domain integration, wherein a first-order trend term is eliminated from the integration result after the time domain integration, and in a velocity spectrogram, the noise amplitude is not obvious, but the difference between an integration velocity curve and a theoretical velocity curve is larger in the time domain; after the second integration to the displacement, the noise amplitude at the low frequency begins to appear, and the error between the integral displacement curve and the theoretical displacement curve in the time domain is increased more and more.
Focusing on the result of the integration method used by the invention, it can be seen from the spectrogram that the integration result of the low frequency band has almost no noise because the signal reconstruction method is adopted in the low frequency band according to the frequency after screening and confirmation. And the frequency domain integration is adopted in the high frequency band, after the frequency is greater than the critical point, noise with very low amplitude is gradually shown, because the influence of the noise with higher frequency in the frequency domain integration is smaller, the influence of the high frequency noise on the final result is very little, and in the displacement spectrum after twice integration, the noise at the high frequency can not be seen almost. It can also be seen from the velocity and displacement curves that the first and second integration results obtained by the integration method of the present invention have high goodness of fit with the theoretical value, and the error is far smaller than the frequency domain integration and time domain integration results.
Therefore, the vibration acceleration signal segmentation and integration method based on important information reconstruction provided by the invention starts from the defect that frequency domain integration is sensitive to low-frequency noise, divides the signal into a low-frequency band and a high-frequency band, finds out and corrects the frequency reflecting the signal characteristics in the low-frequency band through methods such as chaotic oscillator and tip-pulling search, and obtains the final integrated signal by combining the low-frequency band and high-frequency band integration results in a frequency domain integration mode in the high-frequency band. Therefore, the low-frequency trend term generated by time domain integration is avoided, the strong interference of low-frequency noise in frequency domain integration is bypassed, and finally the reserved signal covers a useful signal as much as possible.
Claims (10)
1. A vibration acceleration signal segmentation integration method based on important information reconstruction is characterized by comprising the following steps;
a) determining a critical point of frequency domain segmentation through a signal-to-noise threshold, performing forward exploration by taking the maximum value of the type I frequency and the type II frequency as an initial point, and taking the frequency with the frequency domain amplitude being smaller than the signal-to-noise threshold for the first time as the critical point;
b) determining whether the type II frequency exists or not through comparison of track chaos degrees of front and back phase planes of the signal added by the chaotic oscillator;
c) in a low-frequency band smaller than the critical point, all III-class frequencies larger than a signal-to-noise threshold value in the low-frequency band are obtained by a tip-pulling screening method;
d) correcting amplitude and phase information of checked class I, class II and low-frequency band class III frequency points;
e) reconstructing a low-frequency band signal according to the amplitude and the phase of the corrected frequency and integrating;
f) performing frequency domain integration on the high-frequency band signals which are greater than or equal to the critical point;
g) and combining the low-frequency-band integration result with the high-frequency-band integration result.
2. The vibration acceleration signal segmentation and integration method based on important information reconstruction is characterized in that the type I frequency represents the frequency with definite frequency and amplitude which is significantly higher than the frequency of noise, the type II frequency represents the frequency with definite but uncertain occurrence, and the type III frequency represents the frequency with uncertain and uncertain occurrence.
3. The vibration acceleration signal segmentation and integration method based on important information reconstruction as claimed in claim 1, wherein the signal noise threshold in step a) is calculated by:
1) arranging the amplitudes of the signal frequency spectrums in a reverse order;
2) truncating the tail 3/4 of the amplitude sequence to obtain a sequence ai，i＝1,2,…,n；
3) Calculating a signal-to-noise threshold value sigma by the formula
4. The vibration acceleration signal segmentation and integration method based on important information reconstruction as claimed in claim 1, wherein the critical point in step a) is a separation frequency dividing a frequency domain into a low frequency band and a high frequency band, so that the integral of the whole signal is divided into the integral of the low frequency band and the integral of the high frequency band.
5. The method according to claim 1, wherein the chaotic oscillator in the step b) is formed by a holles Duffing equation, and the class ii frequency to be detected is used as the periodic perturbation power frequency in the oscillator.
6. The method according to claim 1, wherein the phase plane trajectory chaos S in step b) is calculated by:
1) calculating a maximum value Ma and a minimum value Mi of a sequence formed by x or y of the phase plane trajectory;
2) judging from the starting point of the sequence, and taking a point of the sequence as a new sequence until the sequence is finished when the value of the point of the sequence is in a Ma-0.08 (Ma-Mi) -Ma interval or Mi-Mi +0.08 (Ma-Mi) interval;
3) and carrying out empirical mode decomposition on the new sequence, and taking the number of eigenmode functions obtained by decomposition as the chaos degree S of the phase plane trajectory.
7. The method for segmenting and integrating the vibration acceleration signal based on important information reconstruction as claimed in claim 1, wherein the comparison of the track chaos degrees of the front and rear phase planes of the signal added by the chaotic oscillator in the step b) means that when the frequency of the periodic perturbation force in the chaotic oscillator is the frequency f to be measured, if the formula S is satisfied1≤S02, then it is verified that a signal with frequency f is present in the original signal, in which S1For the degree of disorder of the phase plane trajectory after addition of the signal, S0The degree of misordering of the phase plane trajectories before adding the signal.
8. The vibration acceleration signal segmentation and integration method based on important information reconstruction as claimed in claim 1, wherein the tip extraction screening method in step c) comprises the following steps:
1) setting the amplitudes of the class I frequency, the confirmed class II frequency, 3 frequency points on the left side of each frequency and 3 frequency points on the right side of each frequency as 0 in a low frequency band;
2) finding out the maximum amplitude and the corresponding frequency in the rest low-frequency-band amplitudes, if the amplitude is greater than a signal-to-noise threshold value, dividing the frequency into the III-class frequency of the low frequency band, and simultaneously setting the amplitude of the frequency and 3 points on the left side and 3 points on the right side of the frequency as 0;
3) repeating the step 2) until the found maximum amplitude is less than or equal to the signal-to-noise threshold.
9. The vibration acceleration signal segmentation and integration method based on important information reconstruction as claimed in claim 1, wherein the method for correcting the amplitude and phase of the verified frequency f in step e) is as follows:
1) finding out the corresponding position i and amplitude A of the frequency f in the frequency spectrumiAnd phase thetai；
2) Calculating the spectral line shift length dL, wherein the formula is dL ═ alpha (2A)i-1-Ai)/(Ai+Ai-1)-(1-α)(2Ai+1-Ai)/(Ai+Ai+1) In the formula, Ai-1Is the amplitude, A, of the first line to the left of frequency f in the spectrumi+1Is the amplitude of the first spectral line to the right of frequency f in the spectrum, when Ai-1>Ai+1When A is 1, alpha isi-1≤Ai+1When α is 0;
3) calculating the corrected frequency F, wherein the formula is F ═ (i-1-dL) × fs/N, fs is the sampling frequency, and N is the number of sampling points;
4) calculating the corrected amplitude A, wherein the formula is that A is (1-dL)2)*Ai/sinc(dL)；
5) Calculating the corrected phase theta, wherein the formula is that theta is 180 (theta)i+πdL)/π。
10. The vibration acceleration signal segmentation and integration method based on important information reconstruction as claimed in claim 1, wherein the frequency domain integration of the high-frequency band signal in step f) is performed by taking a lower-limit cut-off frequency as a frequency of a critical point, and an upper-limit cut-off frequency is optional according to analysis requirements and does not exceed a sampling frequency divided by 2.56 at most;
the merging of the integration results in the step g) is to add the integration result sequence of the low frequency band and the integration result sequence of the high frequency band at any one same time in the time domain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011472767.6A CN112504441B (en) | 2020-12-15 | 2020-12-15 | Vibration acceleration signal segmentation and integration method based on important information reconstruction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011472767.6A CN112504441B (en) | 2020-12-15 | 2020-12-15 | Vibration acceleration signal segmentation and integration method based on important information reconstruction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112504441A true CN112504441A (en) | 2021-03-16 |
| CN112504441B CN112504441B (en) | 2022-12-13 |
Family
ID=74973264
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011472767.6A Active CN112504441B (en) | 2020-12-15 | 2020-12-15 | Vibration acceleration signal segmentation and integration method based on important information reconstruction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112504441B (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB939976A (en) * | 1959-04-22 | 1963-10-16 | Globe Ind Inc | Accelerometer |
| US3672229A (en) * | 1961-02-06 | 1972-06-27 | Aga Ab | Apparatus for indicating errors in inclination for inertial navigation |
| JPH09150709A (en) * | 1995-11-30 | 1997-06-10 | Nec Home Electron Ltd | Judging method for collision of vehicle and collision judging device |
| KR20010063969A (en) * | 1999-12-24 | 2001-07-09 | 이계안 | Sensor error minimum method of accident data recorder for vehicle |
| US20040243346A1 (en) * | 2003-04-05 | 2004-12-02 | Mugler Dale H. | Robust DSP integrator for accelerometer signals |
| CN1971620A (en) * | 2005-11-21 | 2007-05-30 | 清华大学 | Image rebuilding system and method using straight track scanning |
| CN102346809A (en) * | 2011-06-30 | 2012-02-08 | 中国人民解放军理工大学工程兵工程学院 | Method for converting blasting-vibration acceleration into velocity |
| US20150185069A1 (en) * | 2012-08-07 | 2015-07-02 | Atlas Elektronik Gmbh | Method and apparatus for determining a frequency line pattern within at least one amplitude spectrum |
| CN105182405A (en) * | 2015-03-04 | 2015-12-23 | 中石化石油工程地球物理有限公司胜利分公司 | Design method for frequency domain low frequency compensation scanning signal |
| CN106950047A (en) * | 2017-03-09 | 2017-07-14 | 西安交通大学 | The visual analysis method of vibration acceleration signal frequency spectrum |
| WO2018029392A1 (en) * | 2016-08-10 | 2018-02-15 | Maria Del Carmen Ovejero Mayoral | Method for the automation of the dose calibration, reconstruction and verification of complex radiotherapy treatments, integrated into one environment, and system for implementing same |
| CN108764073A (en) * | 2018-05-14 | 2018-11-06 | 华南理工大学 | A kind of acceleration filter of combination spectrum energy form fitting is made an uproar and integration method |
| CN108760318A (en) * | 2018-08-31 | 2018-11-06 | 西安热工研究院有限公司 | A kind of wind energy conversion system pitch variable bearings fault detection method |
| CN109029886A (en) * | 2018-07-17 | 2018-12-18 | 浙江大学 | A kind of shake table acceleration frequency response function measurement method |
| CN111458017A (en) * | 2020-01-07 | 2020-07-28 | 天津大学 | Vibration speed frequency domain reconstruction method based on generalized trend term error control |
-
2020
- 2020-12-15 CN CN202011472767.6A patent/CN112504441B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB939976A (en) * | 1959-04-22 | 1963-10-16 | Globe Ind Inc | Accelerometer |
| US3672229A (en) * | 1961-02-06 | 1972-06-27 | Aga Ab | Apparatus for indicating errors in inclination for inertial navigation |
| JPH09150709A (en) * | 1995-11-30 | 1997-06-10 | Nec Home Electron Ltd | Judging method for collision of vehicle and collision judging device |
| KR20010063969A (en) * | 1999-12-24 | 2001-07-09 | 이계안 | Sensor error minimum method of accident data recorder for vehicle |
| US20040243346A1 (en) * | 2003-04-05 | 2004-12-02 | Mugler Dale H. | Robust DSP integrator for accelerometer signals |
| CN1971620A (en) * | 2005-11-21 | 2007-05-30 | 清华大学 | Image rebuilding system and method using straight track scanning |
| CN102346809A (en) * | 2011-06-30 | 2012-02-08 | 中国人民解放军理工大学工程兵工程学院 | Method for converting blasting-vibration acceleration into velocity |
| US20150185069A1 (en) * | 2012-08-07 | 2015-07-02 | Atlas Elektronik Gmbh | Method and apparatus for determining a frequency line pattern within at least one amplitude spectrum |
| CN105182405A (en) * | 2015-03-04 | 2015-12-23 | 中石化石油工程地球物理有限公司胜利分公司 | Design method for frequency domain low frequency compensation scanning signal |
| WO2018029392A1 (en) * | 2016-08-10 | 2018-02-15 | Maria Del Carmen Ovejero Mayoral | Method for the automation of the dose calibration, reconstruction and verification of complex radiotherapy treatments, integrated into one environment, and system for implementing same |
| CN106950047A (en) * | 2017-03-09 | 2017-07-14 | 西安交通大学 | The visual analysis method of vibration acceleration signal frequency spectrum |
| CN108764073A (en) * | 2018-05-14 | 2018-11-06 | 华南理工大学 | A kind of acceleration filter of combination spectrum energy form fitting is made an uproar and integration method |
| CN109029886A (en) * | 2018-07-17 | 2018-12-18 | 浙江大学 | A kind of shake table acceleration frequency response function measurement method |
| CN108760318A (en) * | 2018-08-31 | 2018-11-06 | 西安热工研究院有限公司 | A kind of wind energy conversion system pitch variable bearings fault detection method |
| CN111458017A (en) * | 2020-01-07 | 2020-07-28 | 天津大学 | Vibration speed frequency domain reconstruction method based on generalized trend term error control |
Non-Patent Citations (6)
| Title |
|---|
| XINKAI FAN: "Coordinated Control of DC Grid and Offshore Wind Farms to Improve Rotor-Angle Stability", 《IEEE TRANSACTIONS ON POWER SYSTEMS 》 * |
| ZHU YONG: "Useful information reconstruction-based vibration acceleration signal integral method for faulty rotating machinery", 《CHINA MECHANICAL ENGINEERING》 * |
| 吴涛等: "姿态检测系统中的数据融合算法综述", 《电子科技》 * |
| 孙焕纯: "关于非线性结构振动方程的求解──二级近似加速度逐步积分法", 《大连理工大学学报》 * |
| 温广瑞等: "基于精确信息重构的故障转子系统振动加速度信号积分方法", 《机械工程学报》 * |
| 陈太聪等: "基于频谱能量形态拟合的加速度积分方法研究", 《振动与冲击》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112504441B (en) | 2022-12-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107505135B (en) | Rolling bearing composite fault extraction method and system | |
| US5744723A (en) | Method for determining rotational speed from machine vibration data | |
| CN110617964A (en) | Synchronous compression transformation order ratio analysis method for fault diagnosis of rolling bearing | |
| CN104502099B (en) | Cyclic frequency extraction method for characteristic components of transient conditions of gearbox | |
| EP2005125B1 (en) | A method and a system for monitoring the condition and operation of periodically moving objects | |
| CN110763462B (en) | Time-varying vibration signal fault diagnosis method based on synchronous compression operator | |
| JP6236402B2 (en) | A system for acquiring vibration signals of rotary motors | |
| Rodopoulos et al. | A parametric approach for the estimation of the instantaneous speed of rotating machinery | |
| CN101251445B (en) | Method for analysis of fractal characteristic of rotating machinery bump-scrape acoustic emission signal | |
| EP3093669A1 (en) | Tachometer systems and methods of determining the rotation speed of a wheel of a landing gear of an aircraft | |
| CN103116032A (en) | Method and device for acquiring rotating speed of wind generating set | |
| CN111323233B (en) | Local mean decomposition method for low-speed rotating machine fault diagnosis | |
| CN110573845B (en) | Method for detecting defects in a vibration sensor, associated device and computer program | |
| CN103674550A (en) | Real-time mixing denoising method for rolling bearing static monitoring signal | |
| CN103221791B (en) | Method for monitoring a rotary element belonging to a mechanical transmission of a wind turbine | |
| CN112504441B (en) | Vibration acceleration signal segmentation and integration method based on important information reconstruction | |
| CN112465068A (en) | Rotating equipment fault feature extraction method based on multi-sensor data fusion | |
| CN105300688A (en) | RMS-based self-adaptive quick evaluating method for rotating speed of gearbox | |
| US10775209B2 (en) | Control circuit and method for checking the plausibility of a rotor position angle | |
| CN114781466B (en) | Fault diagnosis method and system based on harmonic fundamental frequency of rotary mechanical vibration signal | |
| CN115014765B (en) | Method for extracting fault characteristics of rolling bearing retainer through acoustic signals | |
| CN112345247A (en) | Fault diagnosis method and device for rolling bearing | |
| CN112781723B (en) | Harmonic component detection method based on frequency spectrum variance | |
| CN115683644A (en) | Double-source beat vibration characteristic identification method for aircraft engine | |
| CN110866519B (en) | Rolling bearing fault diagnosis method based on Fourier decomposition and multiscale permutation entropy deviation value |
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 |