CN116881694A - Torsional vibration measurement method based on time-frequency ridge line extraction and numerical integration - Google Patents
Torsional vibration measurement method based on time-frequency ridge line extraction and numerical integration Download PDFInfo
- Publication number
- CN116881694A CN116881694A CN202310855937.6A CN202310855937A CN116881694A CN 116881694 A CN116881694 A CN 116881694A CN 202310855937 A CN202310855937 A CN 202310855937A CN 116881694 A CN116881694 A CN 116881694A
- Authority
- CN
- China
- Prior art keywords
- frequency
- time
- signal
- torsional vibration
- vibration
- 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
Links
- 230000010354 integration Effects 0.000 title claims abstract description 46
- 238000000605 extraction Methods 0.000 title claims abstract description 30
- 238000000691 measurement method Methods 0.000 title claims description 6
- 238000004458 analytical method Methods 0.000 claims abstract description 53
- 238000000034 method Methods 0.000 claims abstract description 43
- 230000001133 acceleration Effects 0.000 claims abstract description 34
- 238000006243 chemical reaction Methods 0.000 claims abstract description 25
- 238000001228 spectrum Methods 0.000 claims abstract description 25
- 238000009826 distribution Methods 0.000 claims abstract description 13
- 238000009499 grossing Methods 0.000 claims abstract description 12
- 238000010183 spectrum analysis Methods 0.000 claims description 9
- 230000009466 transformation Effects 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 7
- 230000001360 synchronised effect Effects 0.000 claims description 7
- 230000006835 compression Effects 0.000 claims description 5
- 238000007906 compression Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 238000012805 post-processing Methods 0.000 claims description 4
- 108010076504 Protein Sorting Signals Proteins 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 abstract 1
- 238000012544 monitoring process Methods 0.000 description 9
- 238000012360 testing method Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 4
- 230000003595 spectral effect Effects 0.000 description 4
- 230000002776 aggregation Effects 0.000 description 3
- 238000004220 aggregation Methods 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000013016 damping Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000001125 extrusion Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000035559 beat frequency Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000002427 irreversible effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a torsional vibration measuring method based on time-frequency ridge line extraction and numerical integration, which relates to the technical field of torsional vibration signal extraction of rotary machinery. The method comprises the following steps: and acquiring vibration acceleration signals of the outer casing of the rotary machine by using a vibration acceleration sensor, and performing preprocessing such as downsampling and the like for time-frequency analysis. And obtaining a high-resolution time frequency spectrum by using a high-resolution time frequency analysis method, extracting the frequency conversion information of the rotating shaft required to be analyzed in time frequency distribution, and converting the frequency conversion information into an angular velocity signal. Based on the physical characteristics of torsional vibration, carrying out numerical integration on the angular velocity signal, and carrying out trending treatment and smoothing operation to finally obtain a torsional vibration signal.
Description
Technical Field
The invention relates to the technical field of torsional vibration signal conversion and extraction of rotary machinery, in particular to a torsional vibration measurement method based on time-frequency ridge line extraction and numerical integration.
Background
In order to ensure safe and stable operation of the rotating machine, it is important to monitor and diagnose the condition thereof. Vibration has been attracting attention as one of the key indicators of health monitoring of rotary machine rotor systems. In practice, dampers are often installed to increase lateral damping to quickly dampen and attenuate the hazards presented by lateral vibrations. However, torsional vibration of the shafting is a special form of unit vibration, and torsional resonance is difficult to damp once excited due to low torsional damping in the rotor system. The essence of shafting torsional vibration is that the elastic rotating shaft has instantaneous angular velocities with different sizes and phases along each axial cross section when the unit is in operation, so that reciprocating torsional motion in a rotating direction is formed, and torsional impact or alternating stress is generated. Severe torsional vibrations may cause irreversible damage or failure such as coupling fatigue damage, shaft cracking, fan blade failure, etc., severely threatening the safe operation of the rotating machinery.
The current special instruments such as strain gauge telemetry system, rotary shaft encoder/gear teeth, laser vibration meter, etc. can realize real-time monitoring of torsional vibration. The principle of the torsion test is mainly by monitoring the torsional stress or instantaneous angular velocity of the shaft. With the refinement of the sensor design and manufacturing level and the iterative improvement of the corresponding algorithm, the torsional vibration test has reached a higher accuracy. However, for some complex rotary machines, torsional vibration monitoring is not considered throughout the design phase, and therefore, the conditions for installing a torsional vibration monitoring professional instrument are lacking. In addition, most rotating machines can only install vibration sensors on the surface of a casing or a housing to sense signals of the rotor system vibration response transmitted through a complex path and reaching the casing, so that intensive research and digital signal processing methods suitable for demodulation and reconstruction of casing vibration signals are required, and signal components only including rotor torsional vibration information are extracted from casing vibration signals including multiple vibration response information to realize extraction and analysis of torsional signals.
The time-frequency reconstruction is a method suitable for extracting multiple vibration components of the rotary machine. Early time-frequency analysis methods such as wavelet transformation and short-time Fourier transformation are mainly suitable for analyzing linear and stable vibration signals, and due to the influence of the Haisenberg uncertainty, the time-frequency analysis results have the defects of low time-frequency resolution and energy divergence, and the extraction of multi-component vibration signals cannot be accurately realized. The synchronous compression transformation is used as a time-frequency post-processing method, so that the aggregation of time frequency spectrum is effectively improved, the accuracy of time frequency reconstruction is improved, and the method has great application potential in the aspect of torsional vibration extraction.
Disclosure of Invention
In view of the above, the invention provides a torsional vibration measuring method based on time-frequency ridge line extraction and numerical integration, which can successfully extract the rotation frequency information of a rotating shaft by using a case vibration signal under the keyless phase condition and adopting a high-resolution time-frequency analysis method, and the torsional vibration signal can be obtained by performing corresponding numerical integration operation based on the rotation frequency information of the rotating shaft because the physical vibration parameters representing the torsional vibration are angular displacement (or called angle), angular velocity and angular acceleration. By the method, the machine can be additionally monitored in the case of not installing a special torsional vibration tester such as a gear disc or an encoder, and the measurement information of the casing vibration sensor is obviously utilized, so that a convenient method is provided for torsional vibration monitoring.
In order to achieve the above purpose, the technical scheme of the invention is to provide a torsional vibration measuring method based on time-frequency ridge line extraction and numerical integration, which comprises the following steps:
s1, acquiring an external casing or external shell vibration signal to be analyzed by using a vibration acceleration sensor;
s2, after downsampling to a required analysis range, performing time-frequency analysis on the vibration acceleration signal by using a high-resolution time-frequency analysis method to obtain a high-resolution time-frequency spectrum;
s3, extracting the frequency conversion information of the rotating shaft required to be analyzed in the time-frequency distribution, and converting the frequency conversion information into an angular velocity signal;
s4, carrying out numerical integration on the angular velocity signal based on physical characteristics of torsional vibration;
s5, removing trend and smoothing the logarithmic integral result to obtain a torsional vibration signal;
s6, performing spectrum analysis on the obtained torsional vibration signals, and analyzing the accuracy of the extraction method.
Further, in S1, the vibration acceleration sensor is used to collect the vibration signal of the outer casing or the outer housing to be analyzed, including the following steps: the vibration acceleration sensor is arranged on an outer casing of the rotary machine in a magnetic attraction mode or a bracket mounting mode, and the sampling frequency f of an original signal is set s At least higher than the frequency fr to be analyzed max Is a multiple of 2.56.
Further, the original signal to be analyzed acquired by the vibration acceleration sensor is subjected to downsampling treatment, specifically: the vibration acceleration sensor acquires an original signal to be analyzed and evaluated, the original signal is subjected to downsampling according to the required analysis frequency requirement, and the subsequent calculated amount is reduced while the required analysis frequency band is reserved. And intercepting the part to be analyzed from the down-sampled signal as a vibration acceleration signal required by time-frequency analysis for analysis.
In S2, further, a high-resolution time-frequency analysis method is used to perform time-frequency analysis on the vibration acceleration signal, specifically:
the down-sampled acceleration vibration signal is used as a signal to be analyzed, short-time Fourier transformation is firstly adopted to conduct time-frequency analysis, a Gaussian window is selected as a window function, the length parameter of the window function is set to be hlength, and a one-dimensional time sequence signal is expanded to a two-dimensional time-frequency space, so that a time spectrum with relatively low time-frequency resolution is obtained. Post-processing is carried out on the frequency spectrum base at the moment, the time-frequency resolution is improved by utilizing second-order synchronous compression conversion, and instantaneous frequency operators are respectively calculatedGroup delay operator->Frequency modulation rate estimation operator->Finally, calculating second-order instantaneous frequency operator omega 2 (t,ω)＝ω 0 (t,ω)+c 0 (t,ω)[t-t 0 (t,ω)]Where t represents time, ω represents frequency, G (t, ω) represents short-time fourier transform time spectrum, i represents imaginary unit, and the blurred time-frequency energy is redistributed to intermediate frequency estimation in the second-order instantaneous frequency direction, and finally a high-resolution time spectrum is obtained. The second order synchronous compression transformation algorithm can be expressed asWhere η represents the reassigned frequency position and δ represents the dirac distribution symbol.
Further, in S3, the frequency conversion information of the rotating shaft required to be analyzed in the time-frequency distribution is extracted, which specifically includes: the algorithm firstly determines a time-frequency ridge line value at the initial moment through a time-frequency maximum amplitude, then searches forward and backward, and the output result comprises a time-frequency coefficient of a ridge line index and energy of returned ridges.
Further, based on the physical relationship between the rotational speed and the angular velocity, and the physical vibration parameters characterizing the torsional vibration are the angular displacement (or referred to as angle), the angular velocity, and the angular acceleration, the extracted frequency conversion information is converted into an angular velocity signal.
Further, in S4, based on the physical characteristics of the torsional vibration, the angular velocity signal is numerically integrated, specifically: inputting the angular velocity signal sequence, and integrating discrete point values of the extracted angular velocity signal by using an accumulated trapezoidal value integration method so as to obtain a value integration result of the angular velocity signal.
Further, in S5, the trend removal and smoothing are performed on the logarithmic integral result to obtain a torsional vibration signal, which specifically includes: inputting the numerical integration result of the angular velocity signal, performing polynomial trend removal operation on the numerical integration result, and removing the optimal linear fitting line from the sequence data of the integration result, thereby eliminating the influence of offset generated during numerical integration on the later calculation, and removing the trend from the numerical integration result can concentrate analysis on the fluctuation of the data trend. And finally, smoothing by adopting a moving average method to smooth the extracted torsional vibration signals.
Further, in S6, spectrum analysis is performed on the obtained torsional vibration signal, and the accuracy of the analysis and extraction method is specifically: performing fast Fourier transform on the extracted torsional vibration time domain signal to obtain a frequency spectrum of the torsional vibration time domain signal, judging whether the frequency component of the torsional vibration time domain signal is consistent with a theoretical value or not, and knowing the accuracy of the adopted method; and similarly, carrying out envelope spectrum analysis on the extracted torsional vibration time domain signal, and judging whether the frequency components of the torsional vibration time domain signal are consistent with the theoretical values so as to verify the accuracy of the adopted method.
The beneficial effects are that:
1. the invention provides a torsional vibration measuring method based on time-frequency ridge line extraction and numerical integration. And obtaining a high-resolution time frequency spectrum by using a high-resolution time frequency analysis method, extracting the frequency conversion information of the rotating shaft required to be analyzed in time frequency distribution, and converting the frequency conversion information into an angular velocity signal. Based on the physical characteristics of torsional vibration, carrying out numerical integration on the angular velocity signal, and carrying out trending treatment and smoothing operation to finally obtain a torsional vibration signal. The invention completes the work of time-frequency analysis, ridge line extraction, signal conversion, numerical integration and the like of the rotating shaft to be analyzed, successfully extracts the torsional vibration signal of the rotating shaft to be analyzed from the casing vibration signal containing multiple vibration response information, and can be used for the feature extraction and analysis of the rotor torsional vibration signal so as to guide the fault diagnosis work of a rotor system.
2. The embodiment of the invention provides a torsional vibration signal extraction method based on a rotary machine outer casing vibration signal, which can accurately extract a torsional vibration signal of a required rotating shaft, can carry out additional torsional vibration monitoring on a machine under the condition that a special torsional vibration tester such as a gear disc or an encoder is not installed, can serve for monitoring the state of rotor torsional vibration on site under the condition of no key phase or under the condition that the special torsional vibration tester is not installed, solves the difficult problem of difficult rotor torsional vibration extraction under the condition of no key phase or under the condition that the special torsional vibration tester is not installed in engineering application, and can provide a simple and convenient method for torsional vibration monitoring.
Drawings
FIG. 1 is a schematic flow chart of a torsional vibration measurement method based on time-frequency ridge line extraction and numerical integration;
FIG. 2 is a waveform diagram of vibration time domain of an outer casing of a rotary machine according to an embodiment of the present invention;
FIG. 3 is a time-frequency distribution diagram obtained by a high-resolution time-frequency analysis method according to an embodiment of the present invention;
FIG. 4 shows the frequency conversion signal and the angular velocity signal extracted in the embodiment of the invention;
FIG. 5 is a graph showing the results of the integrated result, the trending and the smoothing operation of the torsional vibration signal according to the embodiment of the present invention;
fig. 6 shows the spectral and envelope spectral results of a torsional vibration signal in an embodiment of the present invention.
Detailed Description
The invention is described in detail below with reference to the accompanying drawings and examples.
The invention provides a torsional vibration measuring method based on time-frequency ridge line extraction and numerical integration, which has the flow shown in figure 1 and comprises the following steps:
s1, acquiring an external casing or external shell vibration signal to be analyzed by using a vibration acceleration sensor;
s2, after downsampling to a required analysis range, performing time-frequency analysis on the vibration acceleration signal by using a high-resolution time-frequency analysis method to obtain a high-resolution time-frequency spectrum;
s3, extracting the frequency conversion information of the rotating shaft required to be analyzed in the time-frequency distribution, and converting the frequency conversion information into an angular velocity signal;
s4, carrying out numerical integration on the angular velocity signal based on physical characteristics of torsional vibration;
s5, removing trend and smoothing the logarithmic integral result to obtain a torsional vibration signal;
s6, performing spectrum analysis on the obtained torsional vibration signals, and analyzing the accuracy of the extraction method.
The method comprises the steps of carrying out time-frequency analysis on an external casing vibration acceleration signal by a high-resolution time-frequency analysis method to obtain a high-resolution time-frequency spectrum, extracting the frequency conversion information of a rotating shaft required to be analyzed in time-frequency distribution according to the result of the time-frequency spectrum, converting the frequency conversion information into an angular velocity signal, carrying out numerical integration on the angular velocity signal based on the physical characteristics of torsional vibration, and finally obtaining a time domain signal of the torsional vibration signal by means of trending and smoothing operation, thereby effectively realizing the extraction of the torsional vibration signal, being applicable to the feature extraction and analysis of the torsional vibration signal and further guiding the fault diagnosis work of a rotor system.
Examples:
the data of this embodiment are obtained in a field test of an accessory gearbox that has ten shafts in total. And a BK4519 acceleration sensor is used for measuring a vibration acceleration signal of the outer casing of the accessory gearbox, the sensor is fixed on a designed bracket, and the bracket is fixed on the outer casing through bolts. In the test process, the rotating speed is stabilized at 90% working condition, the rotating frequency information of each rotating shaft is shown in the following table, the 2-axis is the rotating shaft of the input end, the rotating speed is 11984rpm, and the sampling frequency is 128000Hz. In the test, the second-order torsional vibration fixed frequency of the 2 axis and the 4 axis rotation frequency at 90 load are different by about 1.3Hz, so that beat frequency phenomenon is generated, the 4 axis rotation frequency fluctuates, and the scheme example considers the extraction of the torsional vibration signal of the 4 axis.
Table 1 frequency conversion information of each shaft of accessory gearbox
A torsional vibration measurement method based on time-frequency ridge line extraction and numerical integration is shown in figure 1, and comprises the following specific steps:
s1, mounting a vibration acceleration sensor on an outer casing of a rotary machine in a magnetic attraction mode or a bracket mounting mode, and setting the sampling frequency f of an original signal s At least higher than the frequency f to be analyzed max Is a multiple of 2.56. The vibration acceleration sensor acquires an original signal to be analyzed and evaluated, the original signal is subjected to downsampling according to the required analysis frequency requirement, and the original signal is reduced while the required analysis frequency band is reservedAnd (5) subsequent calculated amount. And intercepting the part to be analyzed from the down-sampled signal as a vibration acceleration signal required by time-frequency analysis for analysis.
In this embodiment, the sampling frequency is 128000Hz, the maximum analysis frequency of 4-axis rotation frequency is approximately below 500Hz, so that a section of stable working condition is selected for analysis according to the required maximum analysis frequency of the analysis rotation shaft down-sampled to 1000 Hz. The time interception is 10s, and the number of data points after downsampling is 10000.
S2, taking the downsampled acceleration vibration signal as a signal to be analyzed, firstly adopting short-time Fourier transform to perform time-frequency analysis, selecting a Gaussian window as a window function, setting the length parameter of the window function as hlength, expanding a one-dimensional time sequence signal to a two-dimensional time-frequency space, and obtaining a time frequency spectrum with relatively low time-frequency resolution. And carrying out post-processing on the frequency spectrum basis, calculating a second-order instantaneous frequency operator to obtain accurate instantaneous frequency information, carrying out second-order synchronous extrusion transformation on the obtained short-time Fourier transformation time-frequency result, and redistributing the blurred time-frequency energy to intermediate frequency estimation in the instantaneous frequency direction to finally obtain the high-resolution time-frequency spectrum.
In this embodiment, the time length of the intercepted signal is 10s, and the number of sampling points obtained after downsampling is 10000, so that a gaussian window is selected as a window function of time-frequency analysis, a window length parameter hlength is set to 2000, and time-frequency distribution with better time-frequency aggregation is obtained by utilizing second-order synchronous extrusion transformation.
S3, calculating a ridge line approximate value of the rotation frequency of the rotating shaft required to be analyzed in the high-resolution time spectrum by adopting a punishment function method, firstly determining a time-frequency ridge line value at the initial moment by a time-frequency maximum amplitude, then searching forwards and backwards, and outputting a result comprising a time-frequency coefficient of a ridge line index and energy returned to the ridge. Based on the physical relationship between the rotational speed and the angular velocity, and the physical vibration parameters characterizing the torsional vibration are angular displacement (or referred to as angle), angular velocity, and angular acceleration, the extracted rotational frequency information is converted into an angular velocity signal.
In the embodiment of the invention, a punishment function method is adopted to calculate the ridge of the rotation frequency of the rotation shaft needed to be analyzed in the time-frequency distributionThe line approximation value is obtained by firstly determining a time-frequency ridge line value at the initial moment through a time-frequency maximum amplitude value, then searching forward or backward, outputting a result comprising a vector of ridge line indexes and energy returned to the ridge, and the formula can be expressed asWherein IF (n-1) represents a time-frequency ridge value determined at a previous time, A [ n, m]Representing the spectrum at high resolution, w is a penalty factor. The extracted conversion information is converted into an angular velocity signal based on a physical relationship between the rotational speed and the angular velocity.
S4, inputting an angular velocity signal sequence, and performing discrete point number value integration on the extracted angular velocity signal by using an accumulated trapezoidal value integration method so as to obtain a value integration result of the angular velocity signal.
In the embodiment of the invention, discrete point number integration is performed on the extracted angular velocity signal by using an integrated trapezoidal value integration method, and the integrated integration of uniform intervals is mainly obtained, and for the same sequence a (n), the calculation process can be expressed as
S5, removing trend and smoothing the logarithmic integral result to obtain a torsional vibration signal.
In the embodiment of the invention, the numerical integration result of the angular velocity signal is input, polynomial trend removal operation is carried out on the numerical integration result, the optimal linear fitting line is removed from the sequence data of the integration result, so that the influence of offset generated during numerical integration on post calculation is eliminated, and analysis can be concentrated on fluctuation of the data trend by deleting the trend from the numerical integration result. And finally, smoothing the extracted torsional vibration signal by adopting a sliding average method, wherein the sliding window length is 2001 in the sliding average method.
S6, performing spectrum analysis on the obtained torsional vibration signals, and analyzing the accuracy of the extraction method, wherein the method specifically comprises the following steps: performing fast Fourier transform on the extracted torsional vibration time domain signal to obtain a frequency spectrum of the torsional vibration time domain signal, judging whether the frequency component of the torsional vibration time domain signal is consistent with a theoretical value or not, and knowing the accuracy of the adopted method; and similarly, carrying out envelope spectrum analysis on the extracted torsional vibration time domain signal, judging whether the frequency components of the torsional vibration time domain signal are consistent with the theoretical values, and knowing the accuracy of the adopted method.
The time domain waveform of the vibration acceleration signal of the outer casing of the accessory gearbox, which is measured by the embodiment, is shown in fig. 2, and the time domain waveform is obtained by downsampling and then intercepting data for analysis. In this example, the time spectrum obtained by the high resolution time-frequency analysis method is shown in fig. 3, and the time-frequency distribution of fig. 3 has better time-frequency aggregation, and can be used for the accurate extraction and separation of the subsequent ridge line. Fig. 4 shows the frequency conversion signal and the angular velocity signal extracted in the embodiment of the invention, which shows that the invention can effectively extract the change trend of the frequency conversion along with time under the condition of no key phase. FIG. 5 is an integration result using numerical integration and a trended, smoothed torsional vibration signal that can be used for feature extraction and analysis of torsional vibrations to guide the fault diagnosis of the rotor system. Fig. 6 is a spectral and envelope spectral result of a torsional signal used to verify the accuracy of the method.
In summary, the above embodiments are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. The torsional vibration measurement method based on time-frequency ridge line extraction and numerical integration is characterized by comprising the following steps of:
s1, acquiring an external casing or external shell vibration signal to be analyzed by using a vibration acceleration sensor;
s2, after downsampling to a required analysis range, performing time-frequency analysis on the vibration acceleration signal by using a high-resolution time-frequency analysis method to obtain a high-resolution time-frequency spectrum;
s3, extracting the frequency conversion information of the rotating shaft required to be analyzed in the time-frequency distribution, and converting the frequency conversion information into an angular velocity signal;
s4, carrying out numerical integration on the angular velocity signal based on physical characteristics of torsional vibration;
s5, removing trend and smoothing the logarithmic integral result to obtain a torsional vibration signal;
s6, performing spectrum analysis on the obtained torsional vibration signals, and analyzing the accuracy of the extraction method.
2. The method according to claim 1, wherein in S1, the step of acquiring the vibration signal of the outer casing or the outer casing to be analyzed by using the vibration acceleration sensor includes the steps of:
the vibration acceleration sensor is arranged on an outer casing of the rotary machine in a magnetic attraction mode or a bracket mounting mode, and the sampling frequency f of an original signal is set s At least higher than the frequency fr to be analyzed max Is a multiple of 2.56.
3. The method of claim 2, wherein the down-sampling process is performed on the raw signal to be analyzed acquired by the vibration acceleration sensor, specifically:
acquiring an original signal to be analyzed and evaluated by a vibration acceleration sensor, carrying out downsampling treatment on the original signal according to the required analysis frequency requirement, and reducing the subsequent calculated amount while keeping the required analysis frequency band; and intercepting the part to be analyzed from the down-sampled signal as a vibration acceleration signal required by time-frequency analysis for analysis.
4. A method according to claim 1, 2 or 3, wherein in S2, the vibration acceleration signal is subjected to time-frequency analysis by using a high-resolution time-frequency analysis method, specifically:
taking the down-sampled acceleration vibration signal as a signal to be analyzed, firstly adopting short-time Fourier transform to perform time-frequency analysis, selecting a Gaussian window as a window function, setting the length parameter of the window function as hlength, expanding a one-dimensional time sequence signal to a two-dimensional time-frequency space, and obtaining a time frequency spectrum with relatively low time-frequency resolution; post-processing is carried out on the frequency spectrum base at the moment, the time-frequency resolution is improved by utilizing second-order synchronous compression conversion, and instantaneous frequency operators are respectively calculatedGroup delay operator->Frequency modulation rate estimation operator->Finally, calculating second-order instantaneous frequency operator omega 2 (t,ω)＝ω 0 (t,ω)+c 0 (t,ω)[t-t 0 (t,ω)]Where t represents time, ω represents frequency, G (t, ω) represents short-time fourier transform time spectrum, i represents imaginary unit, and the blurred time-frequency energy is redistributed to intermediate frequency estimation in the second-order instantaneous frequency direction, and finally a high-resolution time spectrum is obtained. The second order synchronous compression transformation algorithm can be expressed asWhere η represents the reassigned frequency position and δ represents the dirac distribution symbol.
5. The method of claim 4, wherein in S3, the extracting the frequency conversion information of the rotation axis required to be analyzed in the time-frequency distribution specifically includes:
the algorithm firstly determines a time-frequency ridge line value at the initial moment through a time-frequency maximum amplitude, then searches forward and backward, and the output result comprises a time-frequency coefficient of a ridge line index and energy of returned ridges.
6. The method of claim 5, wherein the extracted frequency conversion information is converted into an angular velocity signal based on a physical relationship between rotational speed and angular velocity, and the physical vibration parameters characterizing torsional vibration are angular displacement (or referred to as angle), angular velocity, and angular acceleration.
7. The method according to claim 5 or 6, wherein in S4, the angular velocity signal is integrated numerically based on the physical characteristics of the torsional vibrations, in particular:
inputting the angular velocity signal sequence, and integrating discrete point values of the extracted angular velocity signal by using an accumulated trapezoidal value integration method so as to obtain a value integration result of the angular velocity signal.
8. The method of claim 7, wherein in S5, the digital integrated result is trended and smoothed to obtain a torsional vibration signal, specifically:
inputting a numerical integration result of the angular velocity signal, performing polynomial trend removal operation on the numerical integration result, and removing an optimal linear fitting line from sequence data of the integration result, so that the influence of offset generated during numerical integration on later calculation is eliminated, and analysis can be concentrated on fluctuation of the data trend by deleting the trend from the numerical integration result; and finally, smoothing by adopting a moving average method to smooth the extracted torsional vibration signals.
9. The method according to claim 1, wherein in S6, the obtained torsional vibration signal is subjected to a spectrum analysis, and the accuracy of the extracted method is analyzed, specifically:
performing fast Fourier transform on the extracted torsional vibration time domain signal to obtain a frequency spectrum of the torsional vibration time domain signal, judging whether the frequency component of the torsional vibration time domain signal is consistent with a theoretical value or not, and knowing the accuracy of the adopted method; and similarly, carrying out envelope spectrum analysis on the extracted torsional vibration time domain signal, and judging whether the frequency components of the torsional vibration time domain signal are consistent with the theoretical values so as to verify the accuracy of the adopted method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310855937.6A CN116881694A (en) | 2023-07-13 | 2023-07-13 | Torsional vibration measurement method based on time-frequency ridge line extraction and numerical integration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310855937.6A CN116881694A (en) | 2023-07-13 | 2023-07-13 | Torsional vibration measurement method based on time-frequency ridge line extraction and numerical integration |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116881694A true CN116881694A (en) | 2023-10-13 |
Family
ID=88254355
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310855937.6A Pending CN116881694A (en) | 2023-07-13 | 2023-07-13 | Torsional vibration measurement method based on time-frequency ridge line extraction and numerical integration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116881694A (en) |
-
2023
- 2023-07-13 CN CN202310855937.6A patent/CN116881694A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Rolling element bearing fault diagnosis via fault characteristic order (FCO) analysis | |
| CN110346591B (en) | Determining machine speed based on vibration spectrogram | |
| EP2237002B1 (en) | Method for analysing vibration in rotor blades | |
| CN103502827B (en) | For the situation method and apparatus of monitoring machine electric system | |
| JP5757936B2 (en) | Method for detecting structural anomalies in a mechanical assembly including a rotating member | |
| CN110763462B (en) | Time-varying vibration signal fault diagnosis method based on synchronous compression operator | |
| Rzadkowski et al. | Analysis of middle bearing failure in rotor jet engine using tip-timing and tip-clearance techniques | |
| US20200217753A1 (en) | Method and apparatus for identifying gear tooth numbers in a gearbox | |
| CN112345827B (en) | Pattern distinguishing method for spectrum frequency family | |
| US6697759B2 (en) | Method of determining speed of rotation of a motor and a computer software product to carry out the method | |
| CN105628176A (en) | Rotation machinery torsional vibration signal collection analysis method | |
| Lin et al. | A review and strategy for the diagnosis of speed-varying machinery | |
| CN102721462A (en) | Method for quickly computing Bode plot and Nyquist plot of rotary mechanical vehicle starting and parking processes | |
| CN113565584A (en) | Time-frequency filtering method for leaf-end timing signals | |
| AU2001248388A1 (en) | Method of determining speed of rotation of a motor and a computer software product to carry out the method | |
| Song et al. | Multispectral balanced automatic fault diagnosis for rolling bearings under variable speed conditions | |
| CN116881694A (en) | Torsional vibration measurement method based on time-frequency ridge line extraction and numerical integration | |
| CN109916624B (en) | Hilbert yellow-based fatigue failure diagnosis method for ball screw pair | |
| Zhu et al. | Fault diagnosis method for rolling element bearings under variable speed based on TKEO and Fast-SC | |
| Roy et al. | Amplitude demodulation of instantaneous angular speed for fault detection in multistage gearbox | |
| Hou et al. | Estimation of the defect width on the outer race of a rolling element bearing under time‐varying speed conditions | |
| CN114486252B (en) | Rolling bearing fault diagnosis method of vector mode maximum envelope | |
| CN112781723B (en) | Harmonic component detection method based on frequency spectrum variance | |
| Nataraj et al. | Experimental investigation of misalignment and looseness in rotor bearing system using bartlett power spectral density | |
| Thanagasundram et al. | Autoregressive based diagnostics scheme for detection of bearing faults |
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 |