CN113761466A - Method and device for constructing vibration signal order ratio spectrum of rotary machine - Google Patents

Method and device for constructing vibration signal order ratio spectrum of rotary machine Download PDF

Info

Publication number
CN113761466A
CN113761466A CN202111057898.2A CN202111057898A CN113761466A CN 113761466 A CN113761466 A CN 113761466A CN 202111057898 A CN202111057898 A CN 202111057898A CN 113761466 A CN113761466 A CN 113761466A
Authority
CN
China
Prior art keywords
order component
time
probability density
constructing
order
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
Application number
CN202111057898.2A
Other languages
Chinese (zh)
Other versions
CN113761466B (en
Inventor
冯志鹏
于欣楠
陈小旺
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.)
University of Science and Technology Beijing USTB
Original Assignee
University of Science and Technology Beijing USTB
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 University of Science and Technology Beijing USTB filed Critical University of Science and Technology Beijing USTB
Priority to CN202111057898.2A priority Critical patent/CN113761466B/en
Publication of CN113761466A publication Critical patent/CN113761466A/en
Application granted granted Critical
Publication of CN113761466B publication Critical patent/CN113761466B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
    • 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
    • G01M13/02—Gearings; Transmission mechanisms
    • G01M13/021—Gearings
    • 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
    • G01M13/02—Gearings; Transmission mechanisms
    • G01M13/028—Acoustic or vibration analysis
    • 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
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Biology (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Acoustics & Sound (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

The invention discloses a method and a device for constructing a scale spectrum of a vibration signal of rotary machinery, wherein the method comprises the following steps: extracting instantaneous rotating speed from time-frequency distribution of the vibration signal by constructing a probability density function of the rotating speed; constructing a demodulation phase function of each order component according to the extracted instantaneous rotating speed, mapping a time-varying frequency component into a constant frequency component by using the demodulation phase function, and identifying and extracting each order component related to the rotating speed; and calculating the amplitude envelope average value of each proposed order component in the time domain, and replacing the Fourier transform amplitude value of each proposed order component by using the calculated amplitude envelope average value corresponding to each order component, thereby constructing a vibration signal order spectrum. The technical scheme provided by the invention improves the resolution of the order spectrum, enhances the readability of the order spectrum and can meet the actual engineering requirements. The method has great engineering application value for state monitoring and fault diagnosis of the rotary machine.

Description

Method and device for constructing vibration signal order ratio spectrum of rotary machine
Technical Field
The invention relates to the technical field of fault diagnosis of rotary machines, in particular to a method and a device for constructing a scale spectrum of a vibration signal of a rotary machine.
Background
In engineering practice, rotary machines often work under time-varying rotation speed and time-varying load conditions, so that vibration signals have complexity and non-stationarity, and difficulty is brought to state monitoring and fault feature extraction. Order ratio spectrum analysis provides an effective method for identifying time-varying characteristics. Firstly, the instantaneous rotating speed is obtained through a speed measuring device. The signal is then angular resampled according to the instantaneous speed. And finally, carrying out Fourier transform on the equal-angle sampling signals to obtain a ratio spectrum. The order ratio spectrum analysis can accurately identify the time-varying characteristic frequency, so that the method is widely applied to fault diagnosis of the rotating machinery under the non-stationary condition.
However, conventional order ratio spectra suffer from three major inherent drawbacks. First, accurate rotational speed information requires additional speed measuring devices. The accurate rotating speed is the key of order ratio spectrum analysis and is the basis of subsequent angular domain resampling. The rotational speed information is typically obtained by a rotational speed measuring device, such as a tachometer or a rotary encoder. However, in practical applications, there are many inconveniences and cost limitations to additionally installing a tachometer or an encoder. Chinese patent application No. CN 111780980a discloses a diesel engine speed extraction method based on vibration signal envelope cepstrum analysis. After low-pass filtering is carried out on a vibration signal, cepstrum analysis is carried out on an envelope signal of the vibration signal; and (5) corresponding the curve passing time of the cepstrum with the working frequency of the cylinder, and calculating the rotating speed of the diesel engine. Although this method extracts the rotational speed information directly from the vibration signal, the following difficulties need to be overcome. First, there are various harmonics in the signal. Secondly, the signal contains the resonance response of the machine structure. Finally, the actual signal also contains background noise. These factors may lead to errors in the instantaneous speed extraction, thereby misleading subsequent signal analysis.
Secondly, the conventional ratio spectrum has frequency interference which is independent of the rotation speed. Background noise and even interference of the resonant response are present in the actual signal. Chinese patent application No. CN 104700119 a discloses an electroencephalogram independent component extraction method based on convolution blind source separation, which utilizes a frequency domain instantaneous blind source separation module to separate frequency domain instantaneous mixed signals. The method is only effective for signals without frequency domain aliasing, and the actual signals have the characteristics of multi-component and nonlinear time variation. Under the working condition of time-varying rotating speed, the instantaneous frequency track on the time-frequency plane changes along with time. When the time-frequency structure of the signal is projected onto the frequency axis, a frequency domain overlapping phenomenon occurs. For such signals, filtering directly in the frequency domain results in loss or confusion of information and failure to obtain a complete single component. Furthermore, without a priori knowledge of the signal, it is difficult to estimate the instantaneous frequency and the component related to the rotation speed cannot be identified.
Third, conventional order ratio spectra have energy leakage. The conventional order ratio spectrum converts the equal time sampling signal into an equal angle sampling signal through angular domain resampling. For signals under the time-varying working condition, after angular domain resampling, the amplitude envelope is still time-varying, so that energy leakage occurs in Fourier transformation of the angular domain resampling signals.
In view of the above problems, no effective solution has been proposed.
Disclosure of Invention
The invention provides a method and a device for constructing a vibration signal order ratio spectrum of a rotary machine, which aim to solve the three inherent defects of the conventional order ratio spectrum, namely the technical problems of needing tachometer/rotary encoder hardware, being interfered by frequency components irrelevant to the rotating speed and energy leakage.
In order to solve the technical problems, the invention provides the following technical scheme:
in one aspect, the invention provides a method for constructing a scale spectrum of a vibration signal of a rotating machine, which comprises the following steps:
extracting instantaneous rotating speed from time-frequency distribution of the vibration signal by constructing a probability density function of the rotating speed;
constructing a demodulation phase function of each order component according to the extracted instantaneous rotating speed, mapping a time-varying frequency component into a constant frequency component by using the demodulation phase function, and identifying and extracting each order component related to the rotating speed;
and calculating the amplitude envelope average value of each extracted order component in the time domain, and replacing the Fourier transform amplitude value of each extracted order component by using the calculated amplitude envelope average value corresponding to each order component, thereby constructing a scale spectrum.
Further, the extracting the instantaneous rotation speed from the time-frequency distribution of the vibration signal by constructing a probability density function of the rotation speed includes:
constructing time-frequency distribution of the vibration signals, and performing pre-whitening treatment to generate whitened time-frequency distribution of the signals;
constructing a probability density function of the rotating speed according to the rotating speed and the order range based on the whitening time-frequency distribution;
based on the principle of prior continuity, smoothing the constructed probability density function;
and extracting the frequency corresponding to the maximum probability density value at each moment in each time step based on the probability density function after the smoothing processing as the instantaneous rotating speed at the moment.
Further, the expression of the whitening time-frequency distribution a (f, t) is:
Figure BDA0003255288900000021
wherein S (f, t) is a short-time Fourier transform spectrum of the signal,<>trepresenting the average operator, Sn(f) Denotes the value at n% after S (f, t) at all times is sorted in ascending order at frequency f.
Further, based on the whitening time-frequency distribution, according to the rotating speed and the order range, a probability density function of the rotating speed is constructed, which comprises:
considering each column of a (f, t) as the temporal spectrum a (f) for each time step; taking into account the maximum frequency fmaxAnd a plurality of potential orders HiScaling A (f) and setting the maximum rotation speed omegamaxAnd a minimum rotation speed omegaminThe value between is truncated to a probability density function omega | Hi](ii) a Wherein [ omega ] Hi]Expressed as:
Figure BDA0003255288900000031
wherein ω represents a frequency value; xiiIs a normalization factor, expressed as:
Figure BDA0003255288900000032
multiplying the probability density functions of all potential orders to obtain a combined probability density function omega, and taking the combined probability density function omega as the probability density function of the rotating speed.
Further, the smoothing the constructed probability density function based on the principle of a priori continuity includes:
on the basis of prior continuity, the probability density function at the time j + k is convoluted with a Gaussian function to predict the probability density function at the time j, so that the probability density function has continuity;
wherein the probability density function at time j is represented as:
Figure BDA0003255288900000033
wherein [ omega ]j]j+kA probability density function representing the moment j predicted by the probability density function at the moment j + k; [ omega ]j+k]Represents the probability density function at time j + k, [ omega ]j∣Ωj+k]Is a gaussian function and is expressed as:
Figure BDA0003255288900000034
wherein σk＝|γkΔt|，ΔtFor a step of time in the time spectrum,
Figure BDA0003255288900000035
representing normal distribution with the mean value of omega and the variance of sigma, and gamma is the acceleration of the instantaneous rotating speed;
multiplying the probability density functions at the moment j predicted by all the adjacent points to obtain the probability density function after smoothing treatment, wherein the probability density function is expressed as
Figure BDA0003255288900000041
Wherein [ omega ]j]sRepresenting the probability density function after final smoothing; k is the maximum range of the given probability density function for predicting the moment j; k represents the number of time steps adjacent to time j.
Further, constructing a demodulation phase function of each order component according to the extracted instantaneous rotating speed, mapping a time-varying frequency component to a constant frequency component by using the demodulation phase function, and identifying and extracting each order component related to the rotating speed, wherein the demodulation phase function comprises the following steps:
constructing a demodulation phase function of each order component according to the extracted instantaneous rotating speed, and mapping a time-varying frequency component into a constant frequency component according to the demodulation phase function;
separating each order component with constant frequency through a band-pass filter, constructing a reverse demodulation phase function, and recovering each separated order component into an original order component through the reverse demodulation phase function;
performing phase randomization on each separated order component to generate a plurality of phase-randomized signals; respectively carrying out entropy calculation on the original order component and the signal after phase randomization;
and identifying whether the extracted original order component is true or not by comparing the size of the entropy value corresponding to the original order component and the signal after the phase randomization, and if the extracted original order component is true, reserving the extracted original order component, and if the extracted original order component is false, discarding the extracted original order component.
Further, the constructing a demodulation phase function of each order component according to the extracted instantaneous rotation speed and mapping the time-varying frequency component to a constant frequency component according to the demodulation phase function includes:
according to known instantaneous speed of rotationw (t), calculating the instantaneous frequency w of the ith orderi(t) ═ w (t) · i; the ith order time-varying frequency component xi(t) and demodulation function exp [ -jvi(t)]Multiplying and mapping it to a constant frequency component ui(t)：
ui(t)＝x(t)exp[-jvi(t)]
Wherein v isi(t)＝-∫[wi(t)-ω0]dt，ω0Is a constant frequency value; x (t) is the original time-varying frequency component.
Further, the formula for calculating the entropy value is as follows:
Figure BDA0003255288900000042
wherein D [ f (x) ] represents a signal entropy value; (x) is the power spectrum of the signal;
the identifying whether the extracted original order component is true by comparing the size of the entropy value corresponding to the original order component and the signal after the phase randomization, and if so, retaining the extracted original order component, and if not, discarding the extracted original order component, including:
if the entropy value of the original order component is larger than the entropy value of the signal after the phase randomization of 95%, determining that the original order component is true, otherwise, determining that the original order component is false; if true, the data is retained, and if false, the data is discarded.
Further, calculating an average value of the amplitude envelopes of the extracted order components in the time domain, and replacing the Fourier transform amplitude of the extracted order components with the calculated average value of the amplitude envelopes corresponding to each order component, thereby constructing an order spectrum, comprising:
for each extracted order component, calculating the amplitude envelope by the following formula:
Ai(t)＝|xi(t)+jHT[xi(t)]|
wherein A isi(t) representing the amplitude envelope of each order component; HT [ ·]Representing a hilbert transform;
calculation of AiAverage value M of (t)iAs the magnitude of the corresponding order component;
with the order i as abscissa, the corresponding mean value MiFor the ordinate, an enhanced order spectrum is constructed.
In another aspect, the present invention further provides a rotating mechanical vibration signal ratio spectrum constructing apparatus, including:
the instantaneous rotating speed extracting module is used for extracting the instantaneous rotating speed from the time-frequency distribution of the vibration signal by constructing a probability density function of the rotating speed;
the order component extraction module is used for constructing a demodulation phase function of each order component according to the instantaneous rotating speed extracted by the instantaneous rotating speed extraction module, mapping a time-varying frequency component into a constant frequency component by using the demodulation phase function, and identifying and extracting each order component related to the rotating speed;
and the order spectrum construction module is used for calculating the amplitude envelope average value of each order component extracted by the order component extraction module in a time domain, and replacing the Fourier transform amplitude value of each order component by using the calculated amplitude envelope average value corresponding to each order component so as to construct an order spectrum.
In yet another aspect, the present invention also provides an electronic device comprising a processor and a memory; wherein the memory has stored therein at least one instruction that is loaded and executed by the processor to implement the above-described method.
In yet another aspect, the present invention also provides a computer-readable storage medium having at least one instruction stored therein, the instruction being loaded and executed by a processor to implement the above method.
The technical scheme provided by the invention has the beneficial effects that at least:
(1) the invention directly extracts instantaneous rotating speed information from the vibration signal without a tachometer/encoder, thereby solving the problems of inconvenient hardware installation and cost limitation.
(2) The invention effectively eliminates the interference of frequency irrelevant to the rotating speed and improves the readability of the order spectrum.
(3) The invention effectively avoids energy leakage caused by Fourier transform and enhances resolution and amplitude precision.
(4) The order ratio spectrum construction method provided by the invention has higher adaptability in the aspect of revealing the frequency structure of the vibration signal, and is suitable for the vibration signal analysis of the rotating machinery with high non-stationarity and complexity.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flow chart of a method for constructing a scale spectrum of a vibration signal of a rotating machine according to an embodiment of the present invention;
FIG. 2 is a waveform schematic diagram of a wear signal of a gear ring of a planetary gear box in a wind field;
FIG. 3 is a schematic Fourier spectrum of a wear signal of a planetary gearbox ring gear in a wind field;
FIG. 4 is a schematic diagram of time-frequency distribution of a wear signal of a gear ring of a planetary gear box in a wind field;
FIG. 5 is a schematic diagram of a conventional scale spectrum of a wear signal of a ring gear of a planetary gearbox in a wind farm;
FIG. 6 is a schematic diagram of a rotation speed curve extracted from time-frequency distribution of a planetary gearbox in a wind field by a rotating mechanical vibration signal order spectrum construction method provided by an embodiment of the invention;
FIG. 7 is a schematic diagram of an enhanced scale spectrum of a planetary gearbox in a wind field constructed by a method for constructing a scale spectrum of a vibration signal of a rotating machine according to an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
First embodiment
The embodiment provides a method for constructing a rotating machinery vibration signal order ratio spectrum suitable for a time-varying rotating speed working condition, which is used for constructing an enhanced order ratio spectrum. The execution flow of the method is shown in fig. 1, and comprises the following steps:
s1, extracting the instantaneous rotating speed from the time-frequency distribution of the vibration signal by constructing a probability density function of the rotating speed;
s2, constructing a demodulation phase function of each order component according to the extracted instantaneous rotating speed, mapping the time-varying frequency component into a constant frequency component by using the demodulation phase function, and identifying and extracting each order component related to the rotating speed;
and S3, calculating the average value of the amplitude envelopes of the extracted order components in the time domain, and replacing the Fourier transform amplitude of the average value of the amplitude envelopes corresponding to each order component by the calculated average value of the amplitude envelopes, thereby constructing an order spectrum.
Further, in this embodiment, the step S1 includes the following steps:
s11, constructing the time-frequency distribution of the vibration signal, and carrying out pre-whitening processing on the constructed time-frequency distribution of the vibration signal to generate whitening time-frequency distribution A (f, t) of the signal so as to remove the structural response; a (f, t) is represented by:
Figure BDA0003255288900000061
wherein S (f, t) is a short-time Fourier transform spectrum of the signal,<>trepresenting the average operator, Sn(f) Denotes the value at n% after S (f, t) at all times is sorted in ascending order at frequency f.
S12, constructing a probability density function of the instantaneous rotating speed according to the rotating speed and the order range based on A (f, t);
specifically, in this embodiment, the implementation process of S12 is specifically as follows:
regarding each column of the time-frequency distribution A (f, t) as an instantaneous spectrum A (f) of each time step; taking into account the maximum frequency fmaxAnd a plurality of potential orders HiScaling A (f) and setting the maximum rotation speed omegamaxAnd a minimum rotation speed omegaminThe value between is intercepted as a probability density function omega |Hi](ii) a Wherein [ omega ] Hi]Expressed as:
Figure BDA0003255288900000071
where ω represents a frequency value.
ξiIs a normalization factor, expressed as:
Figure BDA0003255288900000072
multiplying the probability density functions of all potential orders to obtain a combined probability density function Ω, which is expressed as:
Figure BDA0003255288900000073
wherein [ Ω ] represents a combined probability density function.
The combined probability density function omega is the probability density function of the rotating speed.
S13, based on the priori continuity principle, smoothing the constructed probability density function, so as to avoid step change of the rotating speed;
specifically, in this embodiment, the implementation process of S13 is specifically as follows:
based on the prior continuity, the probability density function at the time j + k is predicted by convolving the probability density function with the Gaussian function, so that the probability density function has continuity. Where K represents the number of time steps adjacent to time j, K ∈ [ -K, K ]. The K value can be selected in a user-defined mode according to requirements, the larger the K value is, the better the continuity of the probability density function is, and meanwhile, the calculated amount is increased;
wherein the probability density function at time j is represented as:
Figure BDA0003255288900000074
wherein [ omega ]j]j+kRepresenting the probability density function at time j predicted by the probability density function at time j + k.
[Ωj+k]Represents the probability density function at time j + k, [ omega ]j∣Ωj+k]Is a gaussian function and is expressed as:
Figure BDA0003255288900000081
wherein σk＝|γkΔt|，ΔtFor a step of time in the time spectrum,
Figure BDA0003255288900000082
representing normal distribution with the mean value of omega and the variance of sigma, and gamma is the acceleration of the instantaneous rotating speed; omegaj+kRepresenting the probability density function at time j + k.
Multiplying the probability density functions at the moment j predicted by all the adjacent points to obtain the final probability density function after smoothing treatment, which is expressed as
Figure BDA0003255288900000083
Where K is the maximum range of the probability density function for a given predicted time j. [ omega ]j]sRepresenting the probability density function after the final smoothing process.
S14, based on the smoothed probability density function, extracts the frequency corresponding to the maximum probability density value at each time as the instantaneous rotational speed at that time.
Specifically, in this embodiment, the implementation process of S14 is specifically as follows:
the frequency at the maximum of the probability density function is extracted in each time step as the instantaneous rotational speed w (t).
Further, in this embodiment, the step S2 includes the following steps:
s21, constructing a demodulation phase function of each order component according to the extracted instantaneous rotating speed, and mapping a time-varying frequency component into a constant frequency component according to the demodulation phase function;
specifically, in this embodiment, the implementation process of S21 is specifically as follows:
according to the known instantaneous rotating speed w (t), calculating the instantaneous frequency w of the ith orderi(t) ═ w (t) · i; the ith order time-varying frequency component xi(t) and demodulation function exp [ -jvi(t)]Multiplying and mapping it to a constant frequency component ui(t)：
ui(t)＝x(t)exp[-jvi(t)]
Wherein v isi(t)＝-∫[wi(t)-ω0]dt，ω0Is a constant frequency value. x (t) represents the original signal.
S22, separating each order component u with constant frequency by a band-pass filteri(t) constructing an inverse demodulation phase function, and restoring each separated order component to the original order component x by the inverse demodulation phase functioni(t)：
xi(t)＝ui(t)exp[jvi(t)]
Wherein, exp [ jvi(t)]Is an inverse demodulation function.
S23, phase randomizing each separated order component to generate multiple phase randomized signals Si(t); for original order component xi(t) signal s randomized with phasei(t) respectively performing entropy calculation;
wherein, the formula of calculating entropy value can be expressed as:
Figure BDA0003255288900000091
where f (x) is the power spectrum of the signal. D [ f (x) ] represents the signal entropy value.
And S24, identifying whether the extracted original order component is true or not by comparing the size of the entropy value corresponding to the original order component and the signal after the phase randomization, and if the extracted original order component is true, reserving the extracted original order component, and if the extracted original order component is false, discarding the extracted original order component.
Specifically, in this embodiment, the implementation process of S24 is specifically as follows:
if the entropy value of the original order component is larger than the entropy value of the signal after the phase randomization of 95%, determining that the original order component is true, otherwise, determining that the original order component is false; if true, the data is retained, and if false, the data is discarded.
S25, repeating the steps S21-S24 for other target order components, and acquiring all real order components.
Further, in this embodiment, the step S3 includes the following steps:
s31, calculating the amplitude envelope of each extracted order component by the following formula:
Ai(t)＝|xi(t)+jHT[xi(t)]|
wherein HT [ ·]Representing a hilbert transform; a. thei(t) represents the amplitude envelope of each order component.
S32, calculating AiAverage value M of (t)iAs the magnitude of the corresponding order component;
s33, with the order i as abscissa, corresponding average value MiFor the ordinate, an enhanced order spectrum is constructed.
The method of the invention is explained in detail below with reference to fig. 2 to 7 by way of an example of application. Wherein, fig. 2 and 3 are a waveform and a Fourier frequency spectrum of a wear signal of a planet gear box ring gear of a certain wind field. FIG. 4 is a time-frequency distribution of the ring gear wear signal, and it can be seen that frequency domain overlap occurs for each frequency component due to the change in rotational speed of the planetary gearbox over time. Fig. 5 is an analysis result of a conventional order spectrum, and it can be seen that due to the problems of interference of frequencies independent of the rotation speed and energy leakage, aliasing of each order component is severe, and it is difficult to clearly show the order component of the signal. In contrast, the method of the present invention is used to construct an enhanced order ratio spectrum, comprising the following steps:
(1) constructing time-frequency distribution of vibration signals, and performing pre-whitening treatment;
(2) constructing a probability density function of the rotating speed according to the rotating speed and the order ratio range thereof;
(3) based on the prior continuity principle, the probability density function is smoothed, and the step change of the rotating speed is avoided.
(4) And extracting the frequency corresponding to the maximum probability density at each moment as the instantaneous rotating speed at the moment.
Fig. 6 shows an instantaneous rotation speed curve directly extracted from time-frequency distribution by the method of the present invention, and as can be seen from fig. 6, the instantaneous rotation speed curve extracted by the method of the present invention reveals a change rule of a rotation speed with time.
(5) Constructing a demodulation phase function of each order component according to the instantaneous rotating speed, and mapping a time-varying frequency component into a constant frequency component according to the demodulation phase function;
(6) separating each order component by applying a band-pass filter, and restoring the order component into the original order component by using an inverse demodulation phase function;
(7) and identifying whether each extracted order component is true or not by comparing the entropy value after phase randomization, if true, keeping, and if false, abandoning.
(8) Calculating the average value of the amplitude envelopes of the components of each order as the amplitude of each order;
(9) and constructing an enhanced order ratio spectrum by taking the order as an abscissa and the corresponding amplitude as an ordinate.
Where fig. 7 is an enhanced ratio spectrum constructed using the method of the present invention, it can be seen that the method of the present invention eliminates interference with frequencies that are not related to the rotational speed and avoids energy leakage, thereby clearly showing the order components of the signals and their amplitudes. Second-stage mesh band f in fig. 7m2Nearby finds out the characteristic frequency f of the gear ring faultr2And related sidebands, successfully diagnose gear faults.
In summary, the order spectrum construction method for the vibration signals of the rotating machinery provided by the invention identifies and extracts the components of each order, eliminates the interference of the frequency irrelevant to the rotating speed, does not need angular domain resampling, and replaces the Fourier transform amplitude with the average amplitude envelope in the time domain of the components of each order, thereby avoiding the inherent energy leakage problem of the Fourier transform of the amplitude time-varying signals, and clearly displaying the order composition and the amplitude of the signals.
Second embodiment
The embodiment provides a rotating machinery vibration signal order ratio spectrum construction device, which comprises:
the instantaneous rotating speed extracting module is used for extracting the instantaneous rotating speed from the time-frequency distribution of the vibration signal by constructing a probability density function of the rotating speed;
the order component extraction module is used for constructing a demodulation phase function of each order component according to the instantaneous rotating speed extracted by the instantaneous rotating speed extraction module, mapping a time-varying frequency component into a constant frequency component by using the demodulation phase function, and identifying and extracting each order component related to the rotating speed;
and the order spectrum construction module is used for calculating the amplitude envelope average value of each order component extracted by the order component extraction module in a time domain, and replacing the Fourier transform amplitude value of each order component by using the calculated amplitude envelope average value corresponding to each order component so as to construct an order spectrum.
The rotating mechanical vibration signal scale spectrum construction device of the present embodiment corresponds to the rotating mechanical vibration signal scale spectrum construction method of the first embodiment described above; the functions realized by the functional modules in the rotating mechanical vibration signal order ratio spectrum construction device of the embodiment correspond to the flow steps in the rotating mechanical vibration signal order ratio spectrum construction method of the first embodiment one by one; therefore, it is not described herein.
Third embodiment
The present embodiment provides an electronic device, which includes a processor and a memory; wherein the memory has stored therein at least one instruction that is loaded and executed by the processor to implement the method of the first embodiment.
The electronic device may have a relatively large difference due to different configurations or performances, and may include one or more processors (CPUs) and one or more memories, where at least one instruction is stored in the memory, and the instruction is loaded by the processor and executes the method.
Fourth embodiment
The present embodiment provides a computer-readable storage medium, in which at least one instruction is stored, and the instruction is loaded and executed by a processor to implement the method of the first embodiment. The computer readable storage medium may be, among others, ROM, random access memory, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like. The instructions stored therein may be loaded by a processor in the terminal and perform the above-described method.
Furthermore, it should be noted that the present invention may be provided as a method, apparatus or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied in the medium.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.
Finally, it should be noted that while the above describes a preferred embodiment of the invention, it will be appreciated by those skilled in the art that, once the basic inventive concepts have been learned, numerous changes and modifications may be made without departing from the principles of the invention, which shall be deemed to be within the scope of the invention. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.

Claims (10)

1. A method for constructing a scale spectrum of a vibration signal of a rotating machine is characterized by comprising the following steps:
extracting instantaneous rotating speed from time-frequency distribution of the vibration signal by constructing a probability density function of the rotating speed;
constructing a demodulation phase function of each order component according to the extracted instantaneous rotating speed, mapping a time-varying frequency component into a constant frequency component by using the demodulation phase function, and identifying and extracting each order component related to the rotating speed;
and calculating the amplitude envelope average value of each extracted order component in the time domain, and replacing the Fourier transform amplitude value of each extracted order component by using the calculated amplitude envelope average value corresponding to each order component, thereby constructing a scale spectrum.
2. The method of constructing a scale spectrum of a vibration signal of a rotating machine according to claim 1, wherein the extracting an instantaneous rotation speed from a time-frequency distribution of the vibration signal by constructing a probability density function of the rotation speed comprises:
constructing time-frequency distribution of the vibration signals, and performing pre-whitening treatment to generate whitened time-frequency distribution of the signals;
constructing a probability density function of the rotating speed according to the rotating speed and the order range based on the whitening time-frequency distribution;
based on the principle of prior continuity, smoothing the constructed probability density function;
and extracting the frequency corresponding to the maximum probability density value at each moment in each time step based on the probability density function after the smoothing processing as the instantaneous rotating speed at the moment.
3. The method of constructing a scale spectrum of a rotating machine vibration signal as claimed in claim 2, wherein said whitening time-frequency distribution a (f, t) is expressed as:
Figure FDA0003255288890000011
wherein S (f, t) is a short-time Fourier transform spectrum of the signal,<>trepresenting the average operator, Sn(f) Denotes the value at n% after S (f, t) at all times is sorted in ascending order at frequency f.
4. The method of claim 3, wherein constructing a probability density function of the rotation speed according to the rotation speed and the order range based on the whitened time-frequency distribution comprises:
considering each column of a (f, t) as the temporal spectrum a (f) for each time step; taking into account the maximum frequency fmaxAnd a plurality of potential orders HiScaling A (f) and setting the maximum rotation speed omegamaxAnd a minimum rotation speed omegaminThe value between is truncated to a probability density function omega | Hi](ii) a Wherein [ omega ] Hi]Expressed as:
Figure FDA0003255288890000012
wherein ω represents a frequency value; xiiIs a normalization factor, expressed as:
Figure FDA0003255288890000021
multiplying the probability density functions of all potential orders to obtain a combined probability density function omega, and taking the combined probability density function omega as the probability density function of the rotating speed.
5. The method for constructing the scale spectrum of the vibration signal of the rotary machine according to claim 4, wherein the smoothing of the constructed probability density function based on the a priori continuity principle comprises:
on the basis of prior continuity, the probability density function at the time j + k is convoluted with a Gaussian function to predict the probability density function at the time j, so that the probability density function has continuity;
wherein the probability density function at time j is represented as:
Figure FDA0003255288890000022
wherein [ omega ]j]j+kA probability density function representing the moment j predicted by the probability density function at the moment j + k; [ omega ]j+k]Represents the probability density function at time j + k, [ omega ]j∣Ωj+k]Is a gaussian function and is expressed as:
Figure FDA0003255288890000023
wherein σk＝|γkΔt|，ΔtFor a step of time in the time spectrum,
Figure FDA0003255288890000024
representing normal distribution with the mean value of omega and the variance of sigma, and gamma is the acceleration of the instantaneous rotating speed;
multiplying the probability density functions at the moment j predicted by all the adjacent points to obtain the probability density function after smoothing treatment, wherein the probability density function is expressed as
Figure FDA0003255288890000025
Wherein [ omega ]j]sRepresenting the probability density function after final smoothing; k is the maximum range of the given probability density function for predicting the moment j; k represents the number of time steps adjacent to time j.
6. The method of constructing a scale spectrum of a vibration signal of a rotating machine according to claim 1, wherein constructing a demodulation phase function for each order component based on the extracted instantaneous rotation speed, and mapping a time-varying frequency component to a constant frequency component using the demodulation phase function, and identifying and extracting each order component related to the rotation speed comprises:
constructing a demodulation phase function of each order component according to the extracted instantaneous rotating speed, and mapping a time-varying frequency component into a constant frequency component according to the demodulation phase function;
separating each order component with constant frequency through a band-pass filter, constructing a reverse demodulation phase function, and recovering each separated order component into an original order component through the reverse demodulation phase function;
performing phase randomization on each separated order component to generate a plurality of phase-randomized signals; respectively carrying out entropy calculation on the original order component and the signal after phase randomization;
and identifying whether the extracted original order component is true or not by comparing the size of the entropy value corresponding to the original order component and the signal after the phase randomization, and if the extracted original order component is true, reserving the extracted original order component, and if the extracted original order component is false, discarding the extracted original order component.
7. The method of constructing a scale spectrum of a rotating machine vibration signal according to claim 6, wherein constructing a demodulation phase function for each order component according to the extracted instantaneous rotation speed and mapping a time-varying frequency component to a constant frequency component according to the demodulation phase function comprises:
according to the known instantaneous rotating speed w (t), calculating the instantaneous frequency w of the ith orderi(t) ═ w (t) · i; the ith order time-varying frequency component xi(t) and demodulation function exp [ -jvi(t)]Multiplying and mapping it to a constant frequency component ui(t)：
ui(t)＝x(t)exp[-jvi(t)]
Wherein v isi(t)＝-∫[wi(t)-ω0]dt，ω0Is a constant frequency value; x (t) is the original time-varying frequency component.
8. The method of constructing a scale spectrum of a vibration signal of a rotating machine according to claim 7, wherein the formula of entropy calculation is:
Figure FDA0003255288890000031
wherein D [ f (x) ] represents a signal entropy value; (x) is the power spectrum of the signal;
the identifying whether the extracted original order component is true by comparing the size of the entropy value corresponding to the original order component and the signal after the phase randomization, and if so, retaining the extracted original order component, and if not, discarding the extracted original order component, including:
if the entropy value of the original order component is larger than the entropy value of the signal after the phase randomization of 95%, determining that the original order component is true, otherwise, determining that the original order component is false; if true, the data is retained, and if false, the data is discarded.
9. The method for constructing a scale spectrum of a rotating machine vibration signal according to claim 1, wherein the step of calculating the average value of the amplitude envelopes of the extracted order components in the time domain and replacing the fourier transform amplitude of each order component with the calculated average value of the amplitude envelope corresponding to the order component comprises:
for each extracted order component, calculating the amplitude envelope by the following formula:
Ai(t)＝|xi(t)+jHT[xi(t)]|
wherein A isi(t) representing the amplitude envelope of each order component; HT [ ·]Representing a hilbert transform;
calculation of AiAverage value M of (t)iAs the magnitude of the corresponding order component;
with the order i as abscissa, the corresponding mean value MiFor the ordinate, an enhanced order spectrum is constructed.
10. A rotating machinery vibration signal order ratio spectrum construction device is characterized by comprising:
the instantaneous rotating speed extracting module is used for extracting the instantaneous rotating speed from the time-frequency distribution of the vibration signal by constructing a probability density function of the rotating speed;
the order component extraction module is used for constructing a demodulation phase function of each order component according to the instantaneous rotating speed extracted by the instantaneous rotating speed extraction module, mapping a time-varying frequency component into a constant frequency component by using the demodulation phase function, and identifying and extracting each order component related to the rotating speed;
and the order spectrum construction module is used for calculating the amplitude envelope average value of each order component extracted by the order component extraction module in a time domain, and replacing the Fourier transform amplitude value of each order component by using the calculated amplitude envelope average value corresponding to each order component so as to construct an order spectrum.
CN202111057898.2A 2021-09-09 2021-09-09 Method and device for constructing vibration signal order ratio spectrum of rotary machine Active CN113761466B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111057898.2A CN113761466B (en) 2021-09-09 2021-09-09 Method and device for constructing vibration signal order ratio spectrum of rotary machine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111057898.2A CN113761466B (en) 2021-09-09 2021-09-09 Method and device for constructing vibration signal order ratio spectrum of rotary machine

Publications (2)

Publication Number Publication Date
CN113761466A true CN113761466A (en) 2021-12-07
CN113761466B CN113761466B (en) 2022-08-02

Family

ID=78794485

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111057898.2A Active CN113761466B (en) 2021-09-09 2021-09-09 Method and device for constructing vibration signal order ratio spectrum of rotary machine

Country Status (1)

Country Link
CN (1) CN113761466B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114353927A (en) * 2021-12-28 2022-04-15 嘉兴市特种设备检验检测院 Wireless vibration probe
CN117309387A (en) * 2023-09-01 2023-12-29 北京科技大学 Separation demodulation method and device for multiband time-varying modulation component of gear box
CN117691561A (en) * 2024-01-31 2024-03-12 华中科技大学 Secondary equipment cooperative protection method for resonance overvoltage
CN117309387B (en) * 2023-09-01 2024-06-04 北京科技大学 Separation demodulation method and device for multiband time-varying modulation component of gear box

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108398260A (en) * 2018-01-10 2018-08-14 浙江大学 The fast evaluation method of gear-box instantaneous angular velocity based on mixing probabilistic method
CN109520738A (en) * 2018-10-25 2019-03-26 桂林电子科技大学 Rotating machinery Fault Diagnosis of Roller Bearings based on order spectrum and envelope spectrum
US20190339163A1 (en) * 2017-01-18 2019-11-07 Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. Apparatus for analyzing or monitoring a rotating element and corresponding method
CN112035789A (en) * 2020-08-12 2020-12-04 北京科技大学 Fault diagnosis method for gear transmission system of wind driven generator

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190339163A1 (en) * 2017-01-18 2019-11-07 Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. Apparatus for analyzing or monitoring a rotating element and corresponding method
CN108398260A (en) * 2018-01-10 2018-08-14 浙江大学 The fast evaluation method of gear-box instantaneous angular velocity based on mixing probabilistic method
CN109520738A (en) * 2018-10-25 2019-03-26 桂林电子科技大学 Rotating machinery Fault Diagnosis of Roller Bearings based on order spectrum and envelope spectrum
CN112035789A (en) * 2020-08-12 2020-12-04 北京科技大学 Fault diagnosis method for gear transmission system of wind driven generator

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
ZHIPENG FENG ET AL.: "Adaptive iterative generalized demodulation for nonstationary complex signal analysis: Principle and application in rotating machinery fault diagnosis", 《MECHANICAL SYSTEMS AND SIGNAL PROCESSING》 *
冯志鹏 等: "旋转机械振动故障诊断理论与技术进展综述", 《振动与冲击》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114353927A (en) * 2021-12-28 2022-04-15 嘉兴市特种设备检验检测院 Wireless vibration probe
CN114353927B (en) * 2021-12-28 2024-04-05 嘉兴市特种设备检验检测院 Wireless vibration probe
CN117309387A (en) * 2023-09-01 2023-12-29 北京科技大学 Separation demodulation method and device for multiband time-varying modulation component of gear box
CN117309387B (en) * 2023-09-01 2024-06-04 北京科技大学 Separation demodulation method and device for multiband time-varying modulation component of gear box
CN117691561A (en) * 2024-01-31 2024-03-12 华中科技大学 Secondary equipment cooperative protection method for resonance overvoltage
CN117691561B (en) * 2024-01-31 2024-04-26 华中科技大学 Secondary equipment cooperative protection method for resonance overvoltage

Also Published As

Publication number Publication date
CN113761466B (en) 2022-08-02

Similar Documents

Publication Publication Date Title
Li et al. Multi-dimensional variational mode decomposition for bearing-crack detection in wind turbines with large driving-speed variations
Moshrefzadeh et al. The Autogram: An effective approach for selecting the optimal demodulation band in rolling element bearings diagnosis
CN113761466B (en) Method and device for constructing vibration signal order ratio spectrum of rotary machine
Zhao et al. A tacho-less order tracking technique for large speed variations
Borghesani et al. A new procedure for using envelope analysis for rolling element bearing diagnostics in variable operating conditions
Zhang et al. Enhancement of time-frequency post-processing readability for nonstationary signal analysis of rotating machinery: Principle and validation
Hu et al. Bearing fault diagnosis based on an improved morphological filter
Qu et al. A new acoustic emission sensor based gear fault detection approach
Djebala et al. Application of the wavelet multi-resolution analysis and Hilbert transform for the prediction of gear tooth defects
Bechhoefer et al. Processing for improved spectral analysis
Borghesani et al. Cyclostationary analysis with logarithmic variance stabilisation
Barbini et al. Phase editing as a signal pre-processing step for automated bearing fault detection
Rodopoulos et al. A parametric approach for the estimation of the instantaneous speed of rotating machinery
CN109029999B (en) Rolling bearing fault diagnosis method based on enhanced modulation bispectrum analysis
Zhang et al. Improved local cepstrum and its applications for gearbox and rolling bearing fault detection
Zhang et al. A two-step denoising strategy for early-stage fault diagnosis of rolling bearings
Lin et al. A review and strategy for the diagnosis of speed-varying machinery
Zhao et al. Vibration health monitoring of rolling bearings under variable speed conditions by novel demodulation technique
CN112857804B (en) Rolling bearing fault diagnosis method, device, medium and computer equipment
Barbini et al. Amplitude-cyclic frequency decomposition of vibration signals for bearing fault diagnosis based on phase editing
Sahu et al. Fault diagnosis of rolling bearing based on an improved denoising technique using complete ensemble empirical mode decomposition and adaptive thresholding method
Ding et al. Multiple instantaneous frequency ridge based integration strategy for bearing fault diagnosis under variable speed operations
Babouri et al. Application of the cyclostationarity analysis in the detection of mechanical defects: comparative study
CN116519301A (en) Bearing fault diagnosis method and system based on time-frequency envelope spectrum peak analysis
Huang et al. Short-time kurtogram for bearing fault feature extraction under time-varying speed conditions

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant