CN114858430A - Mechanical seal acoustic emission signal noise reduction method based on LCD-wavelet new threshold - Google Patents

Mechanical seal acoustic emission signal noise reduction method based on LCD-wavelet new threshold Download PDF

Info

Publication number
CN114858430A
CN114858430A CN202210414301.3A CN202210414301A CN114858430A CN 114858430 A CN114858430 A CN 114858430A CN 202210414301 A CN202210414301 A CN 202210414301A CN 114858430 A CN114858430 A CN 114858430A
Authority
CN
China
Prior art keywords
signal
isc
wavelet
acoustic emission
mechanical seal
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.)
Pending
Application number
CN202210414301.3A
Other languages
Chinese (zh)
Inventor
陆俊杰
刘柱
马雨润
张炜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University of Science and Technology ZUST
Original Assignee
Zhejiang University of Science and Technology ZUST
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University of Science and Technology ZUST filed Critical Zhejiang University of Science and Technology ZUST
Priority to CN202210414301.3A priority Critical patent/CN114858430A/en
Publication of CN114858430A publication Critical patent/CN114858430A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/15—Correlation function computation including computation of convolution operations

Abstract

The invention discloses a mechanical seal acoustic emission signal denoising method based on an LCD-wavelet new threshold, which decomposes an acoustic emission signal X (t) through local feature scale decomposition to obtain a plurality of ISC components arranged from high to low according to frequency; respectively calculating the cross-correlation coefficient of each ISC component and the original signal obtained by local feature scale decomposition; constructing a noisy component screening rule based on the combination of the frequency spectrum analysis and the cross-correlation coefficient, and screening a pure component and a noisy component from the ISC components; performing combined reconstruction on the screened noise-containing components, and performing noise reduction processing on the combined reconstructed noise-containing components; and reconstructing the screened pure signal components and the noise-reduced combined noise-containing components to obtain the signals subjected to noise reduction by the LCD-wavelet new threshold. The invention can effectively solve the technical problems that the mechanical seal is in a low-speed unstable state, the acoustic emission signal has more interference components, and the condition monitoring and fault diagnosis are not facilitated.

Description

Mechanical seal acoustic emission signal noise reduction method based on LCD-wavelet new threshold
Technical Field
The invention relates to a mechanical seal acoustic emission signal denoising method based on an LCD-wavelet new threshold, and belongs to the field of mechanical fault diagnosis.
Background
Compared with the common vibration signal, the acoustic emission signal has the unique advantages in the early rub-impact fault detection of the rotary machine due to the sensitive response, wide frequency band range and rich frequency components. However, in the detection of acoustic emission signals of a rotary mechanical seal, the acoustic emission signals are often interfered by actual noise and even annihilated, so that the useful acoustic emission signals are difficult to identify, and therefore, when the collected acoustic emission signals are analyzed, noise reduction processing is required.
In an actual signal denoising application, the hard threshold function is discontinuous in the whole wavelet domain, there are discontinuities, and only wavelet coefficients smaller than the threshold are processed, while those wavelet coefficients larger than the threshold are not processed, so that the denoised signal has a large variance; the soft threshold function is continuous in the wavelet domain, although there is no discontinuity problem, its derivative is not continuous, and in practical application, it is often to perform first derivative or even higher derivative operation on the signal, and when the soft threshold function is denoised, those wavelet coefficients which are greater than the threshold are compressed by a constant value, which is practically not in accordance with the trend that the noise component gradually decreases as the wavelet coefficients increase.
Disclosure of Invention
Aiming at the defects and defects of the prior art, the signal denoising method based on the combination of the LCD and the wavelet new threshold is provided, is used for preprocessing the acoustic emission signals of the mechanical seal, and can effectively solve the technical problems that the mechanical seal is in a low-speed unstable state, the acoustic emission signals have more interference components, and the state monitoring and fault diagnosis are not facilitated.
A mechanical seal acoustic emission signal denoising method based on an LCD-wavelet new threshold comprises the following steps:
(1) collecting mechanical seal acoustic emission signals through a mechanical seal acoustic emission signal collection test platform;
(2) decomposing the mechanical seal acoustic emission signals X (t) through local feature scale decomposition to obtain a plurality of ISC components which are arranged from high to low according to frequency;
(3) respectively calculating the cross-correlation coefficient of each ISC component and the original signal obtained by local feature scale decomposition;
(4) carrying out FFT spectrum analysis on the ISC components, constructing a noisy component screening rule based on the combination of the spectrum analysis and the cross-correlation coefficient, and screening pure components and noisy components from the ISC components;
(5) combining and reconstructing the screened noise-containing components by using a signal reconstruction technology, and performing noise reduction processing on the combined and reconstructed noise-containing components by using an improved wavelet new threshold noise reduction method;
(6) and reconstructing the screened pure signal components and the noise-reduced combined noise-containing components to obtain the signals subjected to noise reduction by the LCD-wavelet new threshold.
Further, in step (1), the mechanical seal acoustic emission signal acquisition test platform comprises a mechanical seal test bed, a variable frequency starter, an acoustic emission signal collector, a mechanical seal test piece, an air supply system and an industrial personal computer. The acoustic emission signals of the mechanical seal can be completely collected so as to facilitate the subsequent noise reduction treatment.
Further, in the step (2), a specific method for obtaining the ISC component is as follows:
(2-1) taking all extreme points of x (t), carrying out interval division on x (t) by using adjacent extreme point pairs, and converting x (t) by using linear transformation in the interval of the interval division of the extreme point;
(2-2) suppose that the original signal consists of the baseline signal Lt and the residual signal P 1 (t) separating the two from the original signal x (t), using ISC component criterion to pair the separated P 1 (t) performing discrimination;
(2-3) removing ISC from the original signal x (t) 1 And (t) circularly calculating n times, wherein an ISC component meeting a component criterion can be obtained each time.
The LCD decomposition method separates the ISC components from the original signal in order of frequency from high to low, decomposes the signal into components having different frequency scales, and thus sufficiently mines information.
Further, in the step (2), the expression that satisfies the ISC component is:
Figure BDA0003604744480000021
in the formula: ISC i (t) ISC component is obtained after each cycle, r n The final residual component;
the residual component rn in the decomposition result reflects the variation trend of the signal, and the noise reduction effect can be achieved to a certain degree by eliminating the residual component. In addition, compared with a signal decomposition method using spline difference, such as Empirical Mode Decomposition (EMD), the LCD method separates a baseline signal from an original signal through linear transformation, and has less calculation amount and better performance in suppressing endpoint effect and modal aliasing.
Further, in the step (4), the expression of the cross-correlation coefficient is:
Figure BDA0003604744480000022
in the formula: r (i) is the cross-correlation coefficient; x (i, j) is the ISC component; y (j) is a vibration signal component; m is the signal length; n is the number of ISC components; since the source signal component is a component including noise, and the decomposed ISC component includes noise and true information. The introduction of the cross-correlation coefficient R enables more reasonable selection of the appropriate ISC component for further noise reduction.
Further, in the step (5), a threshold function formula of the wavelet new threshold denoising method is as follows:
Figure BDA0003604744480000031
in the formula: w j,k Is a wavelet coefficient;
Figure BDA0003604744480000032
are approximate wavelet coefficients; taking thr as a threshold value
Figure BDA0003604744480000033
N is the signal length;
Figure BDA0003604744480000034
the invention has the beneficial effects that: the signal denoising method based on the combination of the LCD and the wavelet new threshold value is used for preprocessing the acoustic emission signal of the mechanical seal, and can effectively solve the technical problems that the acoustic emission signal has more interference components and is not beneficial to state monitoring and fault diagnosis when the mechanical seal is in a low-speed unstable state.
The LCD method of the invention separates the ISC components from the original signal in sequence from high to low frequency, decomposes the signal into components with different frequency scales, and fully mines the information. The residual component rn in the decomposition result reflects the variation trend of the signal, and the noise reduction effect can be achieved to a certain degree by eliminating the residual component. In addition, compared with a signal decomposition method using spline difference, such as Empirical Mode Decomposition (EMD), the LCD method separates a baseline signal from an original signal through linear transformation, and has less calculation amount and better performance in suppressing an endpoint effect and modal aliasing.
The wavelet threshold filtering denoising method of the invention considers that all the wavelet coefficients corresponding to the useful signals contain the important information of the signals, and the amplitude of the signals is larger, but the number is smaller; the wavelet coefficients corresponding to the noise signals are distributed uniformly, and the number of the wavelet coefficients is more, but the amplitude of the wavelet coefficients is smaller. In the application of actual signal noise reduction, the hard threshold function is discontinuous in the whole wavelet domain, a discontinuity point exists, only wavelet coefficients smaller than the threshold are processed, and wavelet coefficients larger than the threshold are not processed, so that a noise reduction signal has larger variance; the soft threshold function is continuous in the wavelet domain, although the problem of discontinuity point does not exist, the derivative of the soft threshold function is discontinuous, and in practical application, the first derivative and even the high-order derivative operation processing is often carried out on the signal.
Drawings
FIG. 1 is a flow chart of noise reduction of mechanical seal acoustic emission signals;
FIG. 2 is a component diagram of a mechanical seal acoustic emission signal acquisition system;
FIG. 3 is an exploded schematic view of an LCD;
FIG. 4 is an exploded view of the LCD;
FIG. 5 is a raw signal plot of a mechanical seal acoustic emission signal;
FIG. 6 is an ISC component diagram for LCD decomposition;
FIG. 7 is a diagram of FFT spectral analysis;
FIG. 8 is a graph of combined noisy ISC components;
FIG. 9 is a graph of noisy ISC components after wavelet new threshold denoising;
FIG. 10 is a waveform diagram of the signal after LCD-wavelet new threshold denoising;
Detailed Description
In order to make those skilled in the art better understand the technical solution of the present invention, the following further describes the technical solution of the present invention with reference to the drawings and the embodiment.
According to the flow chart shown in fig. 1, a signal noise reduction test is performed on a section of acoustic emission signals in the starting process collected by the mechanical seal test bed through the mechanical seal acoustic emission signal collection test platform shown in fig. 2. The original signal is shown in fig. 5, in this embodiment, the selected pressure is 0.1Mpa, and the specific noise reduction steps are as follows:
1. firstly, local characteristic scale (LCD) decomposition is carried out on an original signal, and the decomposition principle of the LCD is shown in figure 3;
all extreme points (Tk, Xk) of x (t), k being 1,2,3, …, M, where M is the number of extreme points, are taken. Dividing x (t) into intervals by using adjacent extreme value point pairs, and converting x (t) by using linear transformation in the interval of the division of the extreme value points:
Figure BDA0003604744480000041
in the formula (I), the compound is shown in the specification,
Figure BDA0003604744480000042
is a baseline signal segment; k is the number of linear transformations performed.
According to the sequence of extreme point division, the base line signal segments are combined in sequence, and the base line signal L is obtained through calculation t In the formula (2-1),
L k+1 ＝aA k+1 +(1-a)X k+1 (2-2)
in the formula, the value of the parameter a is generally 0.5.
The original signal is composed of a baseline signal L t And a residual signal P 1 (t), separating the two from the original signal x (t). Utilizing ISC component criterion to judge P1(t) obtained by separation, and enabling ISC if criterion condition is met 1 (t)＝P 1 (t) of (d). Otherwise P will be 1 (t) as an initial signal, recalculating the steps (1) and (2), and cycling k times to obtain an ISC component Pk which is recorded as ISC 1 (t)。
Removal of ISC from original signal x (t) 1 And (t), circularly calculating the step (1) and the step (2) n times, wherein an intrinsic scale component meeting a component criterion can be obtained each time. Computing the final residual component r as a termination condition n Monotonous, or less than a threshold set before signal decomposition. Thus obtaining
Figure BDA0003604744480000051
By LCD decomposition, 10 ISC components can be obtained, as shown in fig. 6.
2. Using the short-time fourier method, spectra of 10 ISC components are made and subjected to spectral analysis, as shown in fig. 7;
3. solving the cross-correlation coefficient between each ISC component and the original signal, as shown in table 1;
the cross-correlation coefficient is expressed as:
Figure BDA0003604744480000052
wherein R is a cross-correlation coefficient; x (i, j) is the ISC component; y (j) is a vibration signal component; m is the signal length; and N is the number of ISC components.
Table 1:
Figure BDA0003604744480000053
4. noise-containing ISC components (ISC1, ISC2, ISC3, ISC7, ISC8, ISC9 and ISC10) are screened out by combining cross-correlation coefficients and spectrum analysis, and wavelet threshold denoising is performed by combining the noise-containing components. Before and after noise reduction, as shown in fig. 8 and 9;
Figure BDA0003604744480000061
in the formula:
Figure BDA0003604744480000062
W j,k is a wavelet coefficient;
Figure BDA0003604744480000063
approximate wavelet coefficients; taking thr as a threshold value
Figure BDA0003604744480000064
N is the signal length;
5. and reconstructing the pure component and the noise-reduced signal by using a signal reconstruction technology to realize noise reduction of the mechanical seal acoustic emission signal, wherein the noise-reduced signal is shown in figure 10.
6. And outputting the signal subjected to noise reduction.
In order to check the decomposition effect of the LCD, the following simulation signal x (t) is decomposed using the LCD method. The simulation signal consists of a frequency modulation and amplitude modulation signal x1(t), an amplitude modulation signal x2(t) and a sinusoidal signal x3(t), the sampling frequency is 1000Hz, and the sampling time is 1 s;
x1(t)＝[1+0.5sin(5πt)]cos[300πt+2cos(10πt)]
x2(t)＝2e -2t sin(60πt)
x3(t)＝sin(10πt)
x(t)＝x1(t)+x2(t)+x3(t)
the decomposed effect of the LCD is shown in fig. 4. It can be seen that the LCD method separates the baseline signal from the original signal by linear transformation, which is less computationally intensive and superior in terms of suppressing endpoint effects and modal aliasing.
The embodiment can be used for solving the problems that the generated acoustic emission signal is disordered and unstable and contains a large amount of noise and the like when the mechanical seal is started and stopped.
The above is only a preferred embodiment of the invention, and the scope of the invention is not limited to the above embodiments, and all technical solutions belonging to the inventive idea belong to the scope of the invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.

Claims (6)

1. A mechanical seal acoustic emission signal noise reduction method based on an LCD-wavelet new threshold is characterized by comprising the following steps:
(1) collecting mechanical seal acoustic emission signals through a mechanical seal acoustic emission signal collection test platform;
(2) decomposing the mechanical seal acoustic emission signals X (t) through local feature scale decomposition to obtain a plurality of ISC components which are arranged from high to low according to frequency;
(3) respectively calculating the cross correlation coefficient of each ISC component and the original signal obtained by local feature scale decomposition;
(4) carrying out FFT spectrum analysis on the ISC components, constructing a noisy component screening rule based on the combination of the spectrum analysis and the cross-correlation coefficient, and screening pure components and noisy components from the ISC components;
(5) combining and reconstructing the screened noise-containing components by using a signal reconstruction technology, and performing noise reduction processing on the combined and reconstructed noise-containing components by using an improved wavelet new threshold noise reduction method;
(6) and reconstructing the screened pure signal components and the noise-reduced combined noise-containing components to obtain the signals subjected to noise reduction by the LCD-wavelet new threshold.
2. The method for denoising the mechanical seal acoustic emission signal based on the LCD-wavelet new threshold value according to claim 1, wherein in the step (1), the mechanical seal acoustic emission signal acquisition test platform comprises a mechanical seal test bed, a variable frequency starter, an acoustic emission signal collector, a mechanical seal test piece, an air supply system and an industrial personal computer.
3. The method for denoising mechanical seal acoustic emission signals based on LCD-wavelet new threshold value according to claim 1, wherein in the step (2), the specific method for obtaining ISC components is:
(2-1) taking all extreme points of x (t), carrying out interval division on x (t) by using adjacent extreme point pairs, and converting x (t) by using linear transformation in the interval of the extreme point division;
(2-2) suppose that the original signal consists of the baseline signal Lt and the residual signal P 1 (t) separating the two from the original signal x (t), using ISC component criterion to pair the separated P 1 (t) performing discrimination;
(2-3) removing ISC from the original signal x (t) 1 And (t) circularly calculating n times, wherein an ISC component meeting a component criterion can be obtained each time.
4. The method for denoising LCD-wavelet based mechanical seal acoustic emission signals according to claim 1, wherein in step (2), the ISC component expression is satisfied as:
Figure FDA0003604744470000021
in the formula: ISC i (t) ISC component is obtained after each cycle, r n The final residual component.
5. The method for denoising mechanical seal acoustic emission signals based on LCD-wavelet new threshold value according to claim 1, wherein in the step (4), the expression of the cross-correlation coefficient is:
Figure FDA0003604744470000022
in the formula: r (i) is a cross-correlation coefficient; x (i, j) is the ISC component; y (j) is a vibration signal component; m is the signal length; n- -number of ISC components.
6. The method for denoising mechanical seal acoustic emission signals based on LCD-wavelet new threshold as claimed in claim 1, wherein in the step (5), the formula of the threshold function of the wavelet new threshold denoising method is:
Figure FDA0003604744470000023
in the formula: w j，h Is a wavelet coefficient;
Figure FDA0003604744470000024
approximate wavelet coefficients; taking thr as a threshold value
Figure FDA0003604744470000025
N is the signal length;
Figure FDA0003604744470000026
CN202210414301.3A 2022-04-20 2022-04-20 Mechanical seal acoustic emission signal noise reduction method based on LCD-wavelet new threshold Pending CN114858430A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210414301.3A CN114858430A (en) 2022-04-20 2022-04-20 Mechanical seal acoustic emission signal noise reduction method based on LCD-wavelet new threshold

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210414301.3A CN114858430A (en) 2022-04-20 2022-04-20 Mechanical seal acoustic emission signal noise reduction method based on LCD-wavelet new threshold

Publications (1)

Publication Number Publication Date
CN114858430A true CN114858430A (en) 2022-08-05

Family

ID=82631326

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210414301.3A Pending CN114858430A (en) 2022-04-20 2022-04-20 Mechanical seal acoustic emission signal noise reduction method based on LCD-wavelet new threshold

Country Status (1)

Country Link
CN (1) CN114858430A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116989937A (en) * 2023-09-25 2023-11-03 苏州俊煌机械科技有限公司 Detection method and device for mechanical sealing element

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116989937A (en) * 2023-09-25 2023-11-03 苏州俊煌机械科技有限公司 Detection method and device for mechanical sealing element
CN116989937B (en) * 2023-09-25 2023-12-22 苏州俊煌机械科技有限公司 Detection method and device for mechanical sealing element

Similar Documents

Publication Publication Date Title
CN110598166B (en) Wavelet denoising method for adaptively determining wavelet layering progression
He et al. Detection of signal transients using independent component analysis and its application in gearbox condition monitoring
CN109932624B (en) Cable partial discharge period narrow-band interference denoising method based on Gaussian scale space
JP2014206403A (en) Rolling bearing diagnostic system
CN111769810A (en) Fluid mechanical modulation frequency extraction method based on energy kurtosis spectrum
Belkadi et al. A robust QRS detection approach using stationary wavelet transform
CN112926014A (en) Rolling bearing acoustic signal multiband fusion fault diagnosis method based on RLS and RSSD
CN110596458B (en) DEMON spectrum harmonic line spectrum and fundamental frequency automatic estimation method
CN114858430A (en) Mechanical seal acoustic emission signal noise reduction method based on LCD-wavelet new threshold
CN104778342B (en) A kind of heart sound feature extracting method based on wavelet singular entropy
CN110420022B (en) P wave detection method based on dual-density wavelet transform
CN110909827A (en) Noise reduction method suitable for fan blade sound signals
CN115655719A (en) Bearing vibration signal staged noise reduction method and bearing fault identification method
CN114486256B (en) CNG compressor rolling bearing fault feature extraction method
CN112766044B (en) Method and device for analyzing longitudinal and transverse wave speeds of loose sample and computer storage medium
CN113567129A (en) CEEMD-based noise reduction method for train bearing vibration signal
CN112903296B (en) Rolling bearing fault detection method and system
CN111505710B (en) High-speed rail seismic source seismic signal extraction method based on equidistant spectrum characteristics
Tayachi et al. A New Approach for Detection of Gear Defects using a Discrete Wavelet Transform and Fast Empirical Mode Decomposition
CN112764108B (en) Novel seismic data noise suppression algorithm based on improved empirical wavelet transformation
CN114569140A (en) Spindle wave extraction method, system, computer device, storage medium and program product
CN110477916B (en) Method and device for identifying motion starting time of surface electromyographic signal
He et al. Sparsity-assisted signal representation for rotating machinery fault diagnosis using the tunable Q-factor wavelet transform with overlapping group shrinkage
CN112766203A (en) Method and device for determining fault signal, storage medium and computer equipment
Yang The detection of bearing incipient fault with maximal overlap discrete wavelet packet transform and sparse code shrinkage denoising

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