CN110792613B - Method for extracting weak signal modulation characteristics of centrifugal pump - Google Patents

Method for extracting weak signal modulation characteristics of centrifugal pump Download PDF

Info

Publication number
CN110792613B
CN110792613B CN201910880935.6A CN201910880935A CN110792613B CN 110792613 B CN110792613 B CN 110792613B CN 201910880935 A CN201910880935 A CN 201910880935A CN 110792613 B CN110792613 B CN 110792613B
Authority
CN
China
Prior art keywords
signal
matrix
modulation
centrifugal pump
weak
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.)
Active
Application number
CN201910880935.6A
Other languages
Chinese (zh)
Other versions
CN110792613A (en
Inventor
宋永兴
刘竞婷
张林华
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.)
Shandong Jianzhu University
Original Assignee
Shandong Jianzhu University
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 Shandong Jianzhu University filed Critical Shandong Jianzhu University
Priority to CN201910880935.6A priority Critical patent/CN110792613B/en
Publication of CN110792613A publication Critical patent/CN110792613A/en
Application granted granted Critical
Publication of CN110792613B publication Critical patent/CN110792613B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
    • F04D—NON-POSITIVE-DISPLACEMENT PUMPS
    • F04D15/00—Control, e.g. regulation, of pumps, pumping installations or systems
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
    • F04D—NON-POSITIVE-DISPLACEMENT PUMPS
    • F04D15/00—Control, e.g. regulation, of pumps, pumping installations or systems
    • F04D15/0088—Testing machines
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
    • F04D—NON-POSITIVE-DISPLACEMENT PUMPS
    • F04D27/00—Control, e.g. regulation, of pumps, pumping installations or pumping systems specially adapted for elastic fluids
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
    • F04D—NON-POSITIVE-DISPLACEMENT PUMPS
    • F04D27/00—Control, e.g. regulation, of pumps, pumping installations or pumping systems specially adapted for elastic fluids
    • F04D27/001—Testing thereof; Determination or simulation of flow characteristics; Stall or surge detection, e.g. condition monitoring

Landscapes

  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • General Engineering & Computer Science (AREA)
  • Structures Of Non-Positive Displacement Pumps (AREA)

Abstract

The invention discloses a method for extracting weak signal modulation characteristics of a centrifugal pump, which comprises the following steps: the method comprises the steps of signal filtering pretreatment, second-order cumulant matrix construction, modulated signal component extraction and weak modulation characteristic extraction, and finally weak signal modulation characteristics of the centrifugal pump are obtained. The method for extracting the weak signal of the centrifugal pump can obtain the modulation characteristics of the centrifugal pump, and has important theoretical significance on condition monitoring, fault diagnosis and cavitation identification of the centrifugal pump.

Description

Method for extracting weak signal modulation characteristics of centrifugal pump
Technical Field
The invention relates to the field of signal processing of rotating machinery, in particular to a method for extracting weak signal modulation characteristics of a centrifugal pump.
Background
The centrifugal pump is a typical rotating machine and widely applied to industrial production, and because the rotating speed is high, the structure is complex, and the internal flow field flows in a three-position unsteady manner, the accurate judgment of the working state and the fault type of the centrifugal pump is particularly important. Centrifugal pumps have typical rotating components and therefore, during their operation, there is a typical modulation of the signals of vibration, noise, pressure pulsations, etc. generated. In addition, the working environment of the centrifugal pump is generally complex, so that interference exists in monitoring signals of the centrifugal pump, and therefore the modulation signal characteristics are weak, and the research on the extraction method of the weak signal modulation characteristics of the centrifugal pump is particularly important.
In the process of implementing the invention, the inventor finds that at least the following disadvantages and shortcomings exist in the prior art:
the algorithm for extracting the characteristics of the monitoring signal of the rotary machine is widely researched by scholars at home and abroad, and the main signal demodulation methods comprise envelope demodulation, spectral kurtosis analysis, cyclostationary analysis and the like. In practical application, the application of envelope demodulation is the most extensive, and the algorithm is mostly adopted in a field feature extraction mode, but the method has poor anti-noise performance and is difficult to extract the modulation feature of a weak signal. The spectral kurtosis analysis method is developed on the basis of an envelope demodulation algorithm, the identification of a resonance frequency band is emphasized, and the extraction of modulation characteristics is still envelope demodulation, so that the algorithm reduces the interference existing in a demodulation spectrum, but the extraction capability of weak signals is still poor. The cyclostationary analysis method is a high-order feature extraction algorithm, has good anti-noise performance, but has low calculation efficiency, is difficult to apply to online monitoring and diagnosis of the features of the centrifugal pump, and has more interference frequency features in a demodulation spectrum. Therefore, the performance of the existing weak signal feature extraction algorithm needs to be improved.
Disclosure of Invention
The invention provides a method for extracting the modulation characteristics of a weak signal of a centrifugal pump, which improves the extraction capability of the modulation characteristics of the weak signal by constructing high-order statistic of a monitoring signal, meets the requirements in practical application and is described in detail in the following:
a method for extracting weak signal modulation characteristics of a centrifugal pump comprises the following steps:
step 1: pre-processing the signal by filtering;
step 2: constructing a second-order cumulant matrix;
and step 3: extracting modulation signal components;
and 4, step 4: and extracting weak modulation characteristics.
In the step 1, the signal filtering preprocessing method is to perform high-pass filtering processing on the monitoring signal, because the modulating signal mostly occurs in a high-frequency region and the signal-to-noise ratio of the high-frequency signal is high, filtering low-frequency noise is beneficial to reducing interference and improving the signal-to-noise ratio of the characteristic signal.
In step 2, the specific process of constructing the second-order cumulant matrix is as follows:
step 2-1, solving second-order cumulant for the monitoring signals according to different delay times at the same time;
step 2-2, solving cumulant corresponding to the same delay time at different times;
and 2-3, constructing an accumulation matrix for the obtained accumulation according to two dimensions of time and delay time.
In step 3, the specific process of modulated signal component extraction is as follows:
step 3-1, solving a covariance matrix and decomposing eigenvalues according to the second-order cumulant matrix determined in the step 2 to obtain eigenvalues and eigenvectors of the second-order cumulant matrix;
step 3-2, selecting a feature vector according to the feature value obtained by feature value decomposition and a selection criterion;
and 3-3, reconstructing the weak modulation signal component by using the characteristic vector and the characteristic value.
In step 4, the specific process of weak modulation feature extraction is as follows: and F, carrying out Fourier transform spectrum analysis according to the weak modulation signal component obtained in the step 3.
The technical scheme provided by the invention has the beneficial effects that: the invention realizes the extraction of the weak signal modulation characteristics of the centrifugal pump, provides a foundation for the online monitoring and fault diagnosis of the centrifugal pump, solves the fault identification capability of the centrifugal pump under weak faults, and has important theoretical and engineering values for flow state characterization and fault identification of the centrifugal pump.
Drawings
FIG. 1 is a schematic diagram of a method for extracting weak signal modulation characteristics of a centrifugal pump according to the present invention;
FIG. 2 is a schematic diagram of a centrifugal pump monitoring raw timing signal;
FIG. 3 is a schematic diagram of a timing signal after filtering processing of a centrifugal pump monitoring signal;
FIG. 4 is a schematic diagram of a distribution of eigenvalues;
FIG. 5 is a diagram illustrating weak signal feature extraction results;
FIG. 6 is a diagram illustrating the weak signal feature envelope demodulation extraction result;
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention are described in further detail below.
Aiming at the problems in the background art, the invention provides a method for extracting the modulation characteristics of the weak signal of the centrifugal pump, which realizes the extraction of the modulation characteristics of the weak signal of the centrifugal pump and solves the problems of the extraction of the modulation characteristics under the condition of strong interference, and the details are described as follows:
and S01, preprocessing the filtering of the signal.
For centrifugal pumps monitoring signals such as vibrations and noise, there is a significant modulation in the monitoring signal due to the presence of rotating components in the centrifugal pump. According to the characteristics of the modulation signal of the centrifugal pump, the modulation frequency band of the centrifugal pump mostly exists in a high-frequency resonance region, so for the monitoring signal, the method firstly filters out a low-frequency region by adopting high-pass filtering. For the raw noise time domain signal monitored by the centrifugal pump (as shown in fig. 2), a high pass filtering above 300Hz is used in this demonstration (as shown in fig. 3).
And S02, constructing a second-order cumulant matrix.
In the periodic operation process of the rotating component of the centrifugal pump, a typical modulation signal is generated, and the characteristic of remarkable cyclostationarity exists, so that the second-order cumulant of the modulation signal has remarkable periodicity and can reflect the modulation characteristic of the centrifugal pump. From the resulting filtered signal, the second order cumulant is solved as follows.
Figure GDA0003070694010000031
In the formula, R2(t, τ) is the second order statistic of the monitoring signal, i.e. the conjugate operator, s (t) is the monitoring time sequence signal, and τ is the time delay. According to R2The periodicity of (t, τ) may be divided into a second order steady-state signal and a second order cyclostationary signal.
Figure GDA0003070694010000041
In the formula, T2The period of the second-order accumulation amount is represented.
The low-frequency modulation characteristic frequency of the weak modulation signal of the centrifugal pump can be obtained through a second-order statistical function R2The period of (t, τ) is obtained.
Figure GDA0003070694010000042
In the formula, alpha is the low-frequency modulation characteristic frequency of the weak modulation signal of the centrifugal pump.
And constructing a second-order cumulant matrix of the monitoring signals according to the characteristics of the second-order statistic function of the centrifugal pump cyclostationary signals.
Figure GDA0003070694010000043
In the formula, RMx(t, τ) is the second order cumulant matrix of the monitored signal, t1,t2,...,tnFor monitoring the instantaneous time of the signal, τ1,τ2,...,τmTime delay of the second order cumulative amount.
And S03, extracting the modulation signal component.
According to the periodic characteristics of the second-order cumulant function of the cyclostationary signal, the established second-order cumulant matrix of the monitoring signal contains the characteristics of the modulation signal, and the second-order cumulants under different delay times contain the same modulation signal component, so that the modulation signal component in the second-order cumulant matrix can be extracted by means of data dimension reduction, and the enhancement of the modulation signal component is realized. In the invention, the extraction of the weak modulation signal is realized by using a eigenvalue solution method. According to the constructed second-order cumulant matrix, solving the corresponding covariance function as follows.
RMCcov＝cov(RMx(t,τ)) (6)
And decomposing the obtained covariance function by using the eigenvalue to obtain a corresponding eigenvalue matrix and eigenvector matrix.
[V,U]＝eig(RMCcov) (7)
In the formula, V is an eigenvalue matrix of matrix decomposition, and U is an eigenvector matrix of matrix decomposition. In the present invention, the number of constituent reconstructions is determined using the distribution of eigenvalues, the criteria for eigenvalue selection being as follows.
Figure GDA0003070694010000051
In the formula, λiRepresenting eigenvalues in an eigenvalue matrix, k representing the total number of selected eigenvalues and m representing the total number of decomposed eigenvalues.
And reconstructing the modulation signal components according to the determined reconstruction number k of the modulation signal components to realize the extraction of the modulation signal components.
PMCi(t)＝RMx(t,τ)ui (9)
In the formula，PMCi(t) denotes the reconstructed modulated signal principal component.
And S04, extracting weak modulation characteristics.
And extracting the modulation characteristics of the weak signal in the monitoring signal by using a spectrum analysis means according to the extracted modulation signal component. In the invention, a fast Fourier transform method is used for realizing weak signal modulation characteristic extraction, and an MATLAB program is as follows.
Pi(f)＝fft(PMCi(t)) (10)
In view of the above, it is desirable to provide,
the mechanical characteristic that the centrifugal pump has a periodic rotating component is utilized, and the monitored weak signal has the cyclostationarity, so that the weak signal modulation characteristic extraction (shown in figure 5) is realized by constructing a second-order cumulant matrix, and the problem of poor capability of extracting the weak signal modulation characteristic by envelope demodulation (shown in figure 6) is solved, so that the method for extracting the weak signal modulation characteristic can obtain a better extraction effect.
Those skilled in the art will appreciate that the drawings are only schematic illustrations of preferred embodiments, and the above-described embodiments of the present invention are merely provided for description and do not represent the merits of the embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (1)

1. A method for extracting weak signal modulation characteristics of a centrifugal pump is characterized by comprising the following steps:
step 1: and (3) filtering and preprocessing of signals: filtering a low-frequency area of the monitoring signal by adopting high-pass filtering;
step 2: constructing a second-order cumulant matrix;
and (3) solving the second-order cumulant according to the signals in the step 1:
Figure FDA0003096171660000011
in the formula, R2(t, τ) is the second order statistic of the monitoring signal, i.e. the conjugate operator, s (t) is the monitoring time sequence signal, τ is the time delay amount;
the second order cumulant matrix is constructed as follows:
Figure FDA0003096171660000012
in the formula, RMx(t, τ) is the second order cumulant matrix of the monitored signal, t1,t2,...,tnFor monitoring the instantaneous time of the signal, τ1,τ2,...,τmA time delay that is a second order cumulative amount;
and step 3: extracting modulation signal components;
step 3-1: carrying out covariance matrix solution and eigenvalue decomposition according to the second-order cumulant matrix in the step 2 to obtain an eigenvalue and an eigenvector of the covariance matrix;
the covariance function is: RMCcov＝cov(RMx(t,τ))，
And decomposing the obtained covariance function by using an eigenvalue: [ V, U ]]＝eig(RMCcov)，
In the formula, V is an eigenvalue matrix of matrix decomposition, and U is an eigenvector matrix of matrix decomposition;
step 3-2: determining the quantity of component reconstruction by using the distribution of the characteristic values, wherein the specific criteria are as follows:
Figure FDA0003096171660000013
in the formula, λiRepresenting eigenvalues in an eigenvalue matrix, k representing the total number of selected eigenvalues, m representing the total number of decomposed eigenvalues;
step 3-3: reconstructing the modulation signal components according to the determined component reconstruction quantity to realize the extraction of the modulation signal components:
PMCi(t)＝RMx(t,τ)ui，
in the formula, PMCi(t) denotes the reconstructed modulated Signal principal component, uiRepresenting a feature vector;
and 4, step 4: weak modulation feature extraction: and (4) carrying out Fourier transform spectrum analysis according to the reconstructed modulation signal principal component obtained in the step (3) to extract the modulation characteristics of the weak signal.
CN201910880935.6A 2019-09-18 2019-09-18 Method for extracting weak signal modulation characteristics of centrifugal pump Active CN110792613B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910880935.6A CN110792613B (en) 2019-09-18 2019-09-18 Method for extracting weak signal modulation characteristics of centrifugal pump

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910880935.6A CN110792613B (en) 2019-09-18 2019-09-18 Method for extracting weak signal modulation characteristics of centrifugal pump

Publications (2)

Publication Number Publication Date
CN110792613A CN110792613A (en) 2020-02-14
CN110792613B true CN110792613B (en) 2021-07-06

Family

ID=69427260

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910880935.6A Active CN110792613B (en) 2019-09-18 2019-09-18 Method for extracting weak signal modulation characteristics of centrifugal pump

Country Status (1)

Country Link
CN (1) CN110792613B (en)

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101729157B (en) * 2009-12-11 2016-02-17 南京航空航天大学 Method for separating vibration signal blind sources under a kind of strong noise environment
CN102519582B (en) * 2011-12-22 2013-11-06 南京航空航天大学 Blind source separation method of aeroengine vibration signal

Also Published As

Publication number Publication date
CN110792613A (en) 2020-02-14

Similar Documents

Publication Publication Date Title
Lei et al. A review on empirical mode decomposition in fault diagnosis of rotating machinery
Wang et al. Fault diagnosis of diesel engine based on adaptive wavelet packets and EEMD-fractal dimension
CN111238808B (en) Compound fault diagnosis method for gearbox based on empirical mode decomposition and improved variational mode decomposition
CN110806315B (en) Gearbox composite fault diagnosis method based on inversion editing
Shao et al. Multi-fault feature extraction and diagnosis of gear transmission system using time-frequency analysis and wavelet threshold de-noising based on EMD
CN108152037A (en) Method for Bearing Fault Diagnosis based on ITD and improvement shape filtering
CN111769810B (en) Fluid mechanical modulation frequency extraction method based on energy kurtosis spectrum
Zhao et al. Probabilistic principal component analysis assisted new optimal scale morphological top-hat filter for the fault diagnosis of rolling bearing
CN110173439A (en) A kind of nascent recognition methods of pump cavitation based on balanced squared envelope spectrum
CN112668518A (en) VMSST time-frequency analysis method for vibration fault signal
Xu et al. A novel parameter-adaptive VMD method based on grey wolf optimization with minimum average mutual information for incipient fault detection
Zhang et al. Gearbox fault diagnosis using multiscale sparse frequency-frequency distributions
Sun et al. Application of wavelet soft threshold denoising algorithm based on EMD decomposition in vibration signals
CN110792613B (en) Method for extracting weak signal modulation characteristics of centrifugal pump
CN115434872A (en) Wind turbine generator gearbox composite fault diagnosis method based on AVMD and improved RSSD
CN113033304B (en) Multi-resonance-band amplitude demodulation analysis method for overcoming frequency domain overlapping interference
CN113899548A (en) Rolling bearing fault diagnosis method based on harmonic kurtosis spectrum
CN114992033A (en) NLM-CEEMDAN-based hydroelectric generating set signal denoising method
CN111582248B (en) SVD-based gearbox signal noise reduction method
CN112364767B (en) High-pressure circulating pump mechanical signal feature extraction method based on fast spectrum correlation
CN111503010B (en) Method for extracting broadband radiation noise signal characteristics of centrifugal pump
CN115014765A (en) Method for extracting fault characteristics of rolling bearing retainer through acoustic signals
Ma et al. Fault Feature Extraction of Rolling Bearing Based on Improved VMD-SVD Joint Denoising and Synchrosqueezing Wavelet Transform
CN110765867A (en) Method for extracting low-frequency characteristics of radiation noise of propeller
Li et al. Incipient fault diagnosis of the planetary gearbox based on improved variational mode decomposition and frequency-weighted energy operator

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