CN113155466A - Bearing fault visual vibration detection method and system - Google Patents
Bearing fault visual vibration detection method and system Download PDFInfo
- Publication number
- CN113155466A CN113155466A CN202110564916.XA CN202110564916A CN113155466A CN 113155466 A CN113155466 A CN 113155466A CN 202110564916 A CN202110564916 A CN 202110564916A CN 113155466 A CN113155466 A CN 113155466A
- Authority
- CN
- China
- Prior art keywords
- fault
- signal
- bearing
- full
- visual 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.)
- Granted
Links
- 230000000007 visual effect Effects 0.000 title claims abstract description 147
- 238000001514 detection method Methods 0.000 title claims abstract description 51
- 238000000034 method Methods 0.000 claims abstract description 54
- 238000004458 analytical method Methods 0.000 claims abstract description 27
- 238000004364 calculation method Methods 0.000 claims description 23
- 230000000877 morphologic effect Effects 0.000 claims description 22
- 238000000354 decomposition reaction Methods 0.000 claims description 19
- 238000001228 spectrum Methods 0.000 claims description 15
- 238000000926 separation method Methods 0.000 claims description 12
- 238000007476 Maximum Likelihood Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 abstract description 4
- 239000010410 layer Substances 0.000 description 79
- 238000003745 diagnosis Methods 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 239000002356 single layer Substances 0.000 description 5
- 239000000126 substance Substances 0.000 description 5
- 230000003321 amplification Effects 0.000 description 4
- 238000003199 nucleic acid amplification method Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 230000001052 transient effect Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000013016 damping Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000000691 measurement method Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- JZKMIPDOAWBAHE-NVQRDWNXSA-N (2r,3r,4s,5r)-2-[6-[cyclohexyl(prop-2-enyl)amino]purin-9-yl]-5-(hydroxymethyl)oxolane-3,4-diol Chemical compound O[C@@H]1[C@H](O)[C@@H](CO)O[C@H]1N1C2=NC=NC(N(CC=C)C3CCCCC3)=C2N=C1 JZKMIPDOAWBAHE-NVQRDWNXSA-N 0.000 description 1
- 235000005459 Digitaria exilis Nutrition 0.000 description 1
- 240000008570 Digitaria exilis Species 0.000 description 1
- 235000010575 Pueraria lobata Nutrition 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000010276 construction Methods 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
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000002904 solvent Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
Landscapes
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a bearing fault visual vibration detection method and a system, wherein the method comprises the following steps: collecting fault video information of a bearing; acquiring a full-layer global visual vibration signal of the bearing according to the fault video information of the bearing; calculating to obtain a fault signal-to-noise ratio of the bearing according to the global visual vibration signal of the whole layer, and calculating to obtain a generalized Kurtosis operator based on the fault signal-to-noise ratio according to the fault signal-to-noise ratio; separating the global visual vibration signal of the whole layer, and calculating according to a generalized Kurtosis operator to obtain an impact component of the global visual vibration signal of the whole layer; and carrying out fault analysis on the bearing according to the impact component. The invention can realize non-contact detection of bearing faults, improve the automation degree, accuracy and scene universality of fault detection, effectively simplify the fault detection process and save the labor cost.
Description
Technical Field
The invention relates to the technical field of fault detection, in particular to a visual vibration detection method and system for bearing faults.
Background
The visual vibration detection method for bearing faults adopts video information shot by a camera to automatically detect bearing impact faults, and has the advantages of non-contact, automation, high accuracy, wide applicable scene and the like.
When the bearing is locally damaged, periodic impact components often appear in a vibration signal of the bearing, but for early bearing failure, the impact components are weak, are easily submerged in the vibration of the bearing and random noise of the bearing and are not easy to detect, and the abnormal impact components, namely an engine failure signal, often appear along with the normal resonance frequency of an engine. Therefore, if the resonance signal and the periodic impact in the bearing vibration signal can be effectively separated, and the time interval between impacts can be identified, an important basis can be provided for bearing fault detection.
The traditional vibration fault detection technology is that a vibration sensor is additionally arranged on the surface of a bearing seat to acquire a vibration signal, and the contact mode needs manual installation, consumes time and labor. Non-contact detection methods such as laser sensors are expensive.
Recently, Justin Chen et al have proposed a visual vibration measurement method (PVE based on phase estimation) based on video motion amplification, wherein the method directly uses a camera to shoot a video, obtains a vibration waveform by extracting local phase information in the video, and calculates a vibration frequency on the basis of the vibration waveform, and the method is also applicable to vibration detection of mechanical equipment [ Time-varying motion filtering for video-based non-stationary vibration measurement, 2020 ].
In addition, because the bearing structure is complex, the excitation is numerous and the frequency range is wide, various excitations are reflected in the vibration of the surface of the engine through corresponding transmission and coupling, so that the measured surface vibration signal is very complex and has nonlinearity, non-Gaussian property and cycle stability. Meanwhile, most of abnormal signals at the initial stage of the fault are weak energy signals, which are often submerged in strong background noise, and the traditional theoretical method for extracting fault characteristics has limitation.
At present, in the bearing fault impact extraction, the main methods are spectral kurtosis [ Antoni J, random R B ] the spectral kurtosis: application to the simulation and diagnostics of rotation mechanics [ J ]. Mechanical Systems and Signal Processing, 2006, 20(2):308 plus 331 ]; LeiY G, Lin J, He Z J, et al. application of an improved reconstruction method for fault diagnosis of rolling element bearings [ J ]. Mechanical Systems and Signal Processing, 2011, 25(5): 1738-, 129(4), 458-470, etc., which all have certain effect on the extraction of the impact component of the fault signal of the rolling bearing. The kurtosis can reflect the convexity and flatness of the peak top of the signal probability density function and is very sensitive to a large amplitude. When the probability is increased, the signal kurtosis is rapidly increased, and the impact information in the signal is favorably detected.
Starck et al [ Starck J L, moudeden Y, Robin J. Morphological Component Analysis [ C ]. Proceedings of SPIE, 2005, 59(14): 1-15. ] proposed Morphological Component Analysis (MCA) based on sparse representation of signals and Morphological diversity and developed its extended algorithm Generalized Morphological Component (GMCA) based thereon. The main idea of morphological component analysis is to use morphological differences of signal components (which can be sparsely represented by different dictionaries) for separation [ lie, Zhanning, Schwann. The method is first applied to image processing [ Starck J L, Elad M, Donoho D.Redundant multiscale transformations and the hair application for morphological component determination [ J ]. Advances in Imaging and Electron Physics, 2004, 132(82): 287-384; in the line M, Starck J L, query P.Simultaneous vehicle and future image inpainting using morphological Analysis (MCA) [ J ]. Journal on Applied and comparative harmonic Analysis ACHA, 2005, 19(3): 340. 358.] and brain signal separation [ Yong X Y, Ward R K, Birch G E.generalized morphological Analysis for a source section and an artifact removal [ C ]. proceedings of the 4th International IEEE interference Analysis, Tukey, 2009, 343. ] EEG, the field of GMCA methods was recently introduced and used for the diagnosis of mechanical failures, Tukey, 2009, 343. ] EEG, and for the diagnosis of mechanical failures, more recent failures than the field of vibration Analysis [ 31. J. ] and vibration impact bearing Analysis [ 12. J. ] and for the improved Analysis of mechanical failures and vibration bearing. However, the GMCA method requires simultaneous acquisition of multiple sensor signals, which increases the difficulty of engineering application in certain situations.
In summary, when the bearing is locally damaged, a periodic impact component often appears in a vibration signal, but for an early bearing failure, the impact component is weak, is easily submerged in the vibration of the bearing and random noise thereof, and is not easy to detect. According to the morphological difference between a resonance component (a smooth part) and an impact component (a detail part) in a signal, the MCA method adopts different overcomplete dictionaries to respectively carry out sparse representation (each component can be sparsely represented by only one dictionary), so that the separation of the resonance component and the impact component in the signal is realized, then the impact component is analyzed, and the engine fault is diagnosed according to the time interval between impacts. In addition, the problem that the Kurtosis criterion is sensitive to the impact characteristic abnormity is considered, so that a visual engine fault detection technology based on generalized Kurtosis-adaptive resonance sparse decomposition is provided by using a visual vibration detection method based on PVE and combining a morphological component analysis method and generalized Kurtosis, a continuous oscillation period component and a transient impact component are separated, and fault diagnosis is carried out according to the impact component and an impact interval. The visual vibration detection method based on kurtosis-adaptive resonance sparse decomposition is mainly researched and applied to bearing fault detection. Mechanical fault detection often requires detection of transient impact components, for example using resonance sparse decomposition to separate continuous oscillation period components from transient impact components [ Selesnick, 2011 ]. In the existing fault diagnosis based on resonance sparse decomposition [ a rotating machinery local fault diagnosis method based on a sparse decomposition optimization algorithm, application publication number, CN109813547A ], signals are collected by an acceleration sensor, and a sample set for dictionary learning needs to be constructed. Compared with the existing fault diagnosis based on resonance sparse decomposition, the method has the advantages that the vibration waveform is acquired in a non-contact mode by utilizing the video information shot by the camera, a specific sample set is not required to be constructed, the impact wavelet dictionary based on the dynamic model is constructed in a self-adaptive mode, on the basis, the optimal rank rho is selected by combining the generalized kurtosis of the visual vibration signal, the MCA equation is solved, and the impact component and the resonance component are separated. In other words, compared with the existing bearing fault detection method, the method has the advantages that the vibration fault data acquisition mode and the dictionary construction mode are different in nature, and the method has obvious advantages. In addition, in the existing patent [ a rotating mechanical vibration measurement method based on micro-motion amplification, application publication No. CN 111307487a ], and [ a rotating mechanical rotor modal shape global measurement device based on video and method-application publication No. CN201811279166 ] extract a vibration signal by using a motion amplification method based on video, and the purpose is to realize vibration measurement. The method extracts the visual vibration signals in a phase-based mode, and aims to perform kurtosis-adaptive resonance sparse decomposition on the visual vibration signals in extraction so as to realize bearing fault analysis. In other words, the present patent is substantially different from the prior video-based motion amplification method in technical use and purpose.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the related art. Therefore, a first objective of the present invention is to provide a visual vibration detection method for bearing faults, so as to implement non-contact detection of bearing faults, improve automation degree, accuracy and scene universality of fault detection, simplify fault detection process, and save labor cost.
The second purpose of the invention is to provide a bearing fault visual vibration detection system.
In order to achieve the above object, a first aspect of the present invention provides a method for detecting visual vibration of a bearing fault, including the following steps: collecting fault video information of a bearing; acquiring a full-layer global visual vibration signal of the bearing according to the fault video information of the bearing; calculating to obtain a fault signal-to-noise ratio of the bearing according to the full-layer global visual vibration signal, and calculating to obtain a generalized Kurtosis operator based on the fault signal-to-noise ratio according to the fault signal-to-noise ratio; separating the whole-layer global visual vibration signal, and calculating according to the generalized Kurtosis operator to obtain an impact component of the whole-layer global visual vibration signal; and carrying out fault analysis on the bearing according to the impact component.
Optionally, the step of obtaining a full-layer global visual vibration signal of the bearing according to the bearing fault video information includes: decomposing the fault video frame by adopting a complex controllable pyramid to obtain a vibration amplitude signal and a vibration phase signal of each frame of the fault video; calculating the local phase difference signal of each single point in each frame of video image; extracting a bearing seat area from the fault video, and extracting the local phase difference signal and the vibration amplitude signal of each single point in the bearing seat area from a vibration amplitude signal and a vibration phase signal of each frame of the fault video; and weighting according to the local phase difference signal and the vibration amplitude signal of each single point in the bearing seat area to obtain the full-layer global visual vibration signal.
Optionally, the step of calculating a fault signal-to-noise ratio of the bearing according to the full-layer global visual vibration signal, and calculating a generalized Kurtosis operator based on the fault signal-to-noise ratio according to the fault signal-to-noise ratio includes: calculating according to the full-layer global visual vibration signal to obtain a frequency energy spectrum of the generalized Kurtosis operator; estimating the fault characteristic frequency of the full-layer global visual vibration signal by a maximum likelihood method according to the frequency energy spectrum; calculating to obtain a fault signal-to-noise ratio of the bearing according to the fault characteristic frequency; and selecting a corresponding fault parameter when the fault signal-to-noise ratio is maximum, and obtaining the generalized Kurtosis operator according to the fault parameter.
Optionally, the step of separating the full-layer global visual vibration signal includes: establishing a dynamic model, and constructing an impact wavelet dictionary based on the dynamic model to sparsely represent impact components in the full-layer global visual vibration signal; after the impact component is sparsely represented, separating the full-layer global visual vibration signal to obtain the impact component.
Optionally, the step of obtaining the impulse component of the full-layer global visual vibration signal by calculation according to the generalized Kurtosis operator includes: performing 5-layer decomposition on the full-layer global visual vibration signal by using the impact wavelet of the dynamic model, and selecting the generalized Kurtosis operator of the 5 th-layer visual vibration signal; and calculating the morphological parameters corresponding to the generalized Kurtosis operator of the 5 th-layer visual vibration signal according to the generalized Kurtosis operator, and determining a corresponding optimal rank to obtain the impact component by using an iterative threshold method.
In order to achieve the above object, a second aspect of the present invention provides a bearing fault visual vibration detection system, including: the acquisition module is used for acquiring fault video information of the bearing; the acquisition module is used for acquiring a full-layer global visual vibration signal of the bearing according to the fault video information of the bearing; the first calculation module is used for calculating to obtain a fault signal-to-noise ratio of the bearing according to the full-layer global visual vibration signal and calculating to obtain a generalized Kurtosis operator based on the fault signal-to-noise ratio according to the fault signal-to-noise ratio; the separation module is used for separating the full-layer global visual vibration signal; the second calculation module is used for calculating and obtaining the impact component of the full-layer global visual vibration signal according to the generalized Kurtosis operator; and the analysis module is used for carrying out fault analysis on the bearing according to the impact component.
Optionally, the obtaining module is specifically configured to: decomposing the fault video frame by adopting a complex controllable pyramid to obtain a vibration amplitude signal and a vibration phase signal of each frame of the fault video; calculating the local phase difference signal of each single point in each frame of video image; extracting a bearing seat area from the fault video, and extracting the local phase difference signal and the vibration amplitude signal of each single point in the bearing seat area from a vibration amplitude signal and a vibration phase signal of each frame of the fault video; and weighting according to the local phase difference signal and the vibration amplitude signal of each single point in the bearing seat area to obtain the full-layer global visual vibration signal.
Optionally, the first computing module comprises: the first calculation submodule is used for calculating and obtaining a frequency energy spectrum of the generalized Kurtosis operator according to the full-layer global visual vibration signal; the estimation submodule is used for estimating the fault characteristic frequency of the full-layer global visual vibration signal through a maximum likelihood method according to the frequency energy spectrum; the second calculation submodule is used for calculating the fault signal-to-noise ratio of the bearing according to the fault characteristic frequency; and the selection submodule is used for selecting the corresponding fault parameter when the fault signal-to-noise ratio is maximum, and obtaining the generalized Kurtosis operator according to the fault parameter.
Optionally, the separation module is specifically configured to: establishing a dynamic model, and constructing an impact wavelet dictionary based on the dynamic model to sparsely represent impact components in the full-layer global visual vibration signal; after the impact component is sparsely represented, separating the full-layer global visual vibration signal to obtain the impact component.
Optionally, the second calculating module is specifically configured to: performing 5-layer decomposition on the full-layer global visual vibration signal by using the impact wavelet of the dynamic model, and selecting the generalized Kurtosis operator of the 5 th-layer visual vibration signal; and calculating the morphological parameters corresponding to the generalized Kurtosis operator of the 5 th-layer visual vibration signal according to the generalized Kurtosis operator, and determining a corresponding optimal rank to obtain the impact component by using an iterative threshold method.
The invention has at least the following advantages:
according to the invention, the bearing visual vibration signal is acquired in a non-contact and phase-based manner, the impact wavelet dictionary based on the dynamic model is combined with the generalized kurtosis operator of the visual vibration signal to separate the impact component and the resonance component in the vibration signal, and the bearing is subjected to fault detection analysis according to the impact component, so that the non-contact detection of the bearing fault can be realized, the automation degree, the accuracy and the scene universality of the fault detection are improved, the fault detection process can be effectively simplified, and the labor cost is saved.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
Fig. 1 is a flowchart of a bearing fault visual vibration detection method according to an embodiment of the present invention;
FIG. 2 is a general flowchart of a bearing fault visual vibration detection method according to an embodiment of the present invention;
FIG. 3 is a flow chart of phase-based visual vibration signal extraction according to an embodiment of the present invention;
fig. 4 is a schematic view of a bearing seat region of a bearing fault table according to an embodiment of the present invention;
FIG. 5 is a waveform diagram of a phase-based visual vibration signal according to an embodiment of the present invention;
fig. 6 is a flowchart of a generalized Kurtosis operator selection algorithm based on a fault signal-to-noise ratio according to an embodiment of the present invention;
FIG. 7 is a waveform of an impulse component isolated from an adaptive MCA based on generalized Kurtosis according to an embodiment of the present invention;
fig. 8 is a waveform diagram of a resonance component isolated by an adaptive MCA based on generalized Kurtosis according to an embodiment of the present invention;
fig. 9 is a block diagram of a bearing fault visual vibration detection system according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
The bearing fault visual vibration detection method and system of the present embodiment are described below with reference to the accompanying drawings.
Referring to fig. 1, a method for detecting visual vibration of a bearing fault according to this embodiment includes the following steps:
and step S1, acquiring fault video information of the bearing.
Step S2, acquiring a full-layer global visual vibration signal of the bearing according to the fault video information of the bearing;
specifically, as shown in fig. 2 and 3, an engine video is captured by a camera to acquire fault video information of the bearing. The method for acquiring the full-layer global visual vibration signal of the bearing according to the bearing fault video information comprises the following steps:
and step S21, performing frame-by-frame decomposition on the fault video by adopting the complex controllable pyramid to obtain a vibration amplitude signal and a vibration phase signal of each frame of fault video.
In step S22, the local phase difference signal of each single point in each frame of video image is calculated.
Step S23, extracting a bearing seat region from the fault video, and extracting a local phase difference signal and a vibration amplitude signal of each single point in the bearing seat region from the vibration amplitude signal and the vibration phase signal of each frame of the fault video.
And step S24, weighting according to the local phase difference signal and the vibration amplitude signal of each single point in the bearing seat area to obtain a full-layer global visual vibration signal.
It should be noted that the method implemented in this step is a phase-based engine visual vibration signal extraction method, and the purpose is to extract an engine vibration signal from the fault video information. According to the input engine video, each frame of video is decomposed by adopting a complex controllable pyramid (CSP), and amplitude and phase information is separated, wherein the obtained local phase information represents the vibration condition of a single point. Because the amplitude in the complex controllable pyramid decomposition result represents texture information, the single-layer global signal can be obtained by weighting by combining the amplitude square of all the points and the local phase difference signal. And correcting the single-layer global signal, removing irrelevant interference, and weighting each layer to obtain a full-layer global visual vibration signal. Compared with a single-layer single-point local signal, the full-layer global visual vibration signal obtained in the embodiment can better represent the vibration condition of the engine.
Specifically, as shown in fig. 2 and 3, the camera captures engine video information, decomposes each frame of video I (r, θ, x, y, t) using CSP, each scale r and each direction θ being represented as amplitude a and phase
The complex number image of (2):
further, after the vibration amplitude signal and the vibration phase signal are obtained through separation, each frame t and the fixed frame t are calculated0The phase difference of (a);
then, as shown in fig. 4, extracting a bearing seat region in a fault video, extracting a local phase difference signal and a vibration amplitude signal of each single point in the bearing seat region from a vibration amplitude signal and a vibration phase signal of each frame of fault video, and weighting to obtain a single-layer and single-scale global signal by combining amplitude squares and the local phase difference signals of all points:
modifying the single-layer global signal to obtain a modified signal phii(ri,θi,t-ti) (ii) a Wherein the content of the first and second substances,
where i denotes the index in the full scale direction, Φ0(r0,θ0T) denotes a reference frame of fixed scale and orientation;
as shown in fig. 5, the global visual vibration signal of the whole layer can be obtained by weighting each layer:
δx(t)＝∑iΦi(ri,θi,t-ti) (5)
and step S3, calculating to obtain a fault signal-to-noise ratio of the bearing according to the full-layer global visual vibration signal, and calculating to obtain a generalized Kurtosis operator based on the fault signal-to-noise ratio according to the fault signal-to-noise ratio.
The method comprises the following steps of calculating to obtain the fault signal-to-noise ratio of a bearing according to a full-layer global visual vibration signal, and calculating to obtain a generalized Kurtosis operator based on the fault signal-to-noise ratio according to the fault signal-to-noise ratio, wherein the steps comprise:
and step S31, calculating according to the global visual vibration signal of the whole layer to obtain the frequency energy spectrum of the generalized Kurtosis operator.
And step S32, estimating the fault characteristic frequency of the full-layer global visual vibration signal through a maximum likelihood method according to the frequency energy spectrum.
And step S33, calculating the fault signal-to-noise ratio of the bearing according to the fault characteristic frequency.
And step S34, selecting the corresponding fault parameter when the fault signal-to-noise ratio is maximum, and obtaining the generalized Kurtosis operator according to the fault parameter.
It should be noted that the method implemented in this step is a generalized Kurtosis operator selection algorithm based on a fault signal-to-noise ratio, and the purpose is to select a zeta parameter most sensitive to the impact characteristics of the global visual vibration signal of the whole layer. Calculating a frequency energy spectrum P (K (zeta, x (t)) of the generalized Kurtosis according to the global visual vibration signal of the whole layer; estimating the fault characteristic frequency of the full-layer global visual vibration signal by using a maximum likelihood method; on the basis, according to the obtained fault frequency of each generalized Kurtosis sequence visual vibration signal, calculating a fault signal-to-noise ratio corresponding to the generalized Kurtosis sequence, namely the signal-to-noise ratio of fault characteristic frequency energy and interference component energy: and analyzing the signal-to-noise ratio curve of the obtained global visual vibration signal generalized Kurtosis of the whole layer, and selecting a zeta value with the maximum signal-to-noise ratio, namely a fault parameter.
Specifically, as shown in fig. 2 and 6, the generalized Kurtosis of the full-layer global visual vibration signal can be expressed as:
K(ζ,δx(t))＝E{|δx(t)|2ζ}/E2{|δx(t)|ζ} (6)
zeta is a constant, zeta belongs to [2,5], the value is taken every 0.1, and the frequency energy spectrum P (K (zeta, x (t)) of generalized Kurtosis is calculated according to the global visual vibration signal of the whole layer;
then, according to the frequency energy spectrum, the maximum likelihood method is applied to estimate the fault characteristic frequency of the global visual vibration signal of the whole layer
Where m denotes the index of the generalized Kurtosis sequence.
Calculating a fault signal-to-noise ratio corresponding to the generalized Kurtosis sequence, namely the signal-to-noise ratio of fault characteristic frequency energy and interference component energy according to the obtained fault characteristic frequency of the generalized Kurtosis full-layer global visual vibration signal, and obtaining the fault signal-to-noise ratio of the bearing and the SNR corresponding to the relational expression thereofm(signal-to-noise ratio) curve:
wherein f isdRepresenting the theoretical failure frequency value.
Further, according to the obtained SNRmCurve, selecting the fault with the largest signal-to-noise ratioAnd obtaining a generalized Kurtosis operator K (zeta, delta x (t)) according to the fault parameter zeta.
And step S4, separating the global visual vibration signal of the whole layer, and calculating according to the generalized Kurtosis operator to obtain the impact component of the global visual vibration signal of the whole layer.
Wherein the step of separating the full-layer global visual vibration signal comprises:
and step S41, establishing a dynamic model, and constructing an impact wavelet dictionary based on the dynamic model to sparsely represent the impact component in the full-layer global visual vibration signal.
And step S42, after the impact component is sparsely represented, separating the full-layer global visual vibration signal to obtain the impact component.
The method comprises the following steps of calculating and obtaining the impact component of the full-layer global visual vibration signal according to a generalized Kurtosis operator, wherein the step of calculating and obtaining the impact component of the full-layer global visual vibration signal according to the generalized Kurtosis operator comprises the following steps:
and step S43, performing 5-layer decomposition on the full-layer global visual vibration signal by using the impact wavelet of the dynamic model, and selecting a generalized Kurtosis operator of the 5 th-layer visual vibration signal.
And step S44, calculating corresponding morphological parameters according to the generalized Kurtosis operator of the 5 th-layer visual vibration signal, and determining a corresponding optimal rank so as to obtain an impact component by using an iterative threshold method.
It should be noted that the method implemented in this step is a visual vibration signal separation method based on adaptive Morphological Component Analysis (MCA), and the purpose of the method is to separate a resonance component, an impact component, and a noise component from a vibration signal. Specifically, an impact wavelet dictionary based on a dynamic model is constructed by combining the dynamic model to sparsely represent impact components in a signal; and constructing a local discrete cosine transform and discrete sine transform dictionary for sparsely representing resonance components in the signal. On the basis, the vibration signal is separated by using the MCA method to obtain an impact component Y containing fault information1Resonant component Y including the vibration of the engine itself2And the difference between the sum of the two components and the original fault signal, i.e. the noise component N, then using the dynamicsThe impact wavelet of the model carries out 5-layer decomposition on the whole-layer global visual vibration signal, selects a generalized Kurtosis operator of a 5 th-layer signal (basically only comprising an impact component), calculates a morphological parameter alpha corresponding to the generalized Kurtosis operator, determines an optimal rank rho, and further solves the impact component and the resonance component by using an iterative threshold method.
Specifically, a kinetic model is established:
where T represents the wavelet support set of the time domain, ωdDenotes the center frequency of the wavelet and ξ denotes the decay rate of the wavelet.
Further, a local discrete cosine transform and a discrete sine transform dictionary are constructed for sparsely representing resonance components in the signal. Then, as shown in fig. 7 and 8, the vibration signal is separated by the MCA method to obtain the impulse component N including the failure information1Resonant component Y including the vibration of the engine itself2And the difference between the sum of the two components and the original fault signal, namely noise component N:
δx＝Y1+Y2+N (10)
solving y by using iterative threshold value method1And y2The method specifically comprises the following steps:
according to the MCA algorithm, an objective function is established:
solving for y using an iterative thresholding method1And y2：
Wherein the content of the first and second substances,
represents Y2The estimation error of the previous iteration.
Wherein the content of the first and second substances,
and n1Respectively represent
And the impact wavelet coefficients of the dynamic model of the noise. To accurately estimate y1It is desirable to devise a method for accurate removal
Denotes y2An estimation error of (a), and y2As is the resonant component. Therefore, when
By using a baseWhen the impact wavelet coefficients of the dynamic model are represented,
is lower than r1Rank of (i.e.)
Wherein, determining
For the purpose of accurately estimating the optimum rank in the equation (12)
Here, let y1In connection with the generalized Kurtosis operator, the generalized Kurtosis operator is expressed as:
where α represents a generalized morphological parameter. In the embodiment, one morphological parameter α corresponding to the optimal rank ρ can be given
To make it nearest to y1The optimum rank ρ is written as an objective function in the form:
further, 5-layer decomposition can be performed on the visual vibration signal by using the impact wavelet of the dynamic model, a generalized Kurtosis operator of a 5-layer signal (basically only including an impact component) is selected, a morphological parameter alpha corresponding to the generalized Kurtosis operator is calculated, and the optimal rank rho is determined according to an equation (17).
The impact component may then be solved as follows
In particular, the amount of the solvent to be used,
the solution can be obtained by an optimization method, as follows:
wherein | x | Y calculationFIs the Frobenious norm of x. By using singular-value decomposition (SVD), r1Can be decomposed into r1＝U∑VTWhere, Σ ═ diag (σ)1,…,σn)。
Further, according to Eckart-Young-Mirsky's theorem, formula (18) can be decomposed into:
wherein eta isρ(∑)＝diag(σ1,…,σn,0,…,0)。
Further, y1Is roughly estimated
The following were used:
final product of
Can be determined by a hard threshold as follows:
wherein λ is1MAD/0.6745, MAD stands for
Median absolute value.
Further, the resonance component can be solved as follows
According to equation (12):
is similar to r1，r2Can be defined as:
wherein the content of the first and second substances,
and n2Are respectively as
And DCT coefficients of noise. Here threshold λ2Is proportional to
Error RTThe definition is as follows:
from this, we can estimate
Wherein the content of the first and second substances,
can be calculated by equation (19). At this time, the threshold λ2Can be set as follows:
further, the final
Can be determined by a hard threshold as follows:
and step S5, carrying out fault analysis on the bearing according to the impact component.
Specifically, after solving the impact component in the vibration signal, the bearing can be subjected to fault analysis according to the impact component.
In order to achieve the above object, the present embodiment further provides a visual vibration detection system for bearing failure, as shown in fig. 9, the visual vibration detection system for bearing failure 10 includes an acquisition module 100, an acquisition module 200, a first calculation module 300, a separation module 400, a second calculation module 500, and an analysis module 600.
The acquisition module 100 is used for acquiring fault video information of the bearing; the obtaining module 200 is configured to obtain a full-layer global visual vibration signal of the bearing according to the fault video information of the bearing; the first calculation module 300 is used for calculating to obtain a fault signal-to-noise ratio of the bearing according to the full-layer global visual vibration signal, and calculating to obtain a generalized Kurtosis operator based on the fault signal-to-noise ratio according to the fault signal-to-noise ratio; the separation module 400 is configured to separate the full-layer global visual vibration signal; the second calculation module 500 is configured to calculate an impulse component of the full-layer global visual vibration signal according to the generalized Kurtosis operator; the analysis module 600 is configured to perform fault analysis on the bearing according to the impact component.
In an embodiment of the present invention, the obtaining module 200 is specifically configured to: performing frame-by-frame decomposition on the fault video by adopting a complex controllable pyramid to obtain a vibration amplitude signal and a vibration phase signal of each frame of fault video; calculating the local phase difference signal of each single point in each frame of video image; extracting a bearing seat area from the fault video, and extracting the local phase difference signal and the vibration amplitude signal of each single point in the bearing seat area from the vibration amplitude signal and the vibration phase signal of each frame of fault video; and weighting according to the local phase difference signal and the vibration amplitude signal of each single point in the bearing seat area to obtain a full-layer global visual vibration signal.
In one embodiment of the present invention, the first calculation module 300 includes: a first calculation submodule 301, an estimation submodule 302, a second calculation submodule 303 and a selection submodule 304.
The first calculation submodule 301 is configured to calculate a frequency energy spectrum of a generalized Kurtosis operator according to the full-layer global visual vibration signal; the estimation submodule 302 is used for estimating the fault characteristic frequency of the full-layer global visual vibration signal according to the frequency energy spectrum by a maximum likelihood method; the second calculation submodule 303 is used for calculating a fault signal-to-noise ratio of the bearing according to the fault characteristic frequency; the selecting submodule 304 is configured to select a fault parameter corresponding to the maximum fault signal-to-noise ratio, and obtain a generalized Kurtosis operator according to the fault parameter.
In an embodiment of the present invention, the separation module 400 is specifically configured to: establishing a dynamic model, and constructing an impact wavelet dictionary based on the dynamic model to sparsely represent impact components in a full-layer global visual vibration signal; after the impact component is sparsely represented, separating the full-layer global visual vibration signal to obtain the impact component.
In an embodiment of the present invention, the second calculating module is specifically configured to: performing 5-layer decomposition on the full-layer global visual vibration signal by using an impact wavelet of a dynamic model, and selecting a generalized Kurtosis operator of the 5 th-layer visual vibration signal; and calculating the corresponding morphological parameters according to the generalized Kurtosis operator of the 5 th layer of visual vibration signals, and determining the corresponding optimal rank to obtain the impact component by using an iterative threshold method.
It should be noted that, for a specific implementation of the bearing fault visual vibration detection system in this embodiment, reference may be made to the specific implementation of the bearing fault visual vibration detection method, and details are not described here again.
According to the bearing fault visual vibration detection method and system, the bearing visual vibration signals are acquired in a non-contact and phase-based mode, the impact wavelet dictionary based on the dynamic model is combined with the generalized kurtosis operator of the visual vibration signals to separate the impact component and the resonance component in the vibration signals, and the bearing is subjected to fault detection analysis according to the impact component, so that non-contact detection of bearing faults can be achieved, the automation degree, the accuracy and the scene universality of fault detection are improved, the fault detection process can be effectively simplified, and the labor cost is saved.
It should be noted that the logic and/or steps represented in the flowcharts or otherwise described herein, such as an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (10)
1. A bearing fault visual vibration detection method is characterized by comprising the following steps:
collecting fault video information of a bearing;
acquiring a full-layer global visual vibration signal of the bearing according to the fault video information of the bearing;
calculating to obtain a fault signal-to-noise ratio of the bearing according to the full-layer global visual vibration signal, and calculating to obtain a generalized Kurtosis operator based on the fault signal-to-noise ratio according to the fault signal-to-noise ratio;
separating the whole-layer global visual vibration signal, and calculating according to the generalized Kurtosis operator to obtain an impact component of the whole-layer global visual vibration signal;
and carrying out fault analysis on the bearing according to the impact component.
2. The method for detecting bearing fault visual vibration according to claim 1, wherein the step of obtaining the full-layer global visual vibration signal of the bearing according to the bearing fault video information comprises:
decomposing the fault video frame by adopting a complex controllable pyramid to obtain a vibration amplitude signal and a vibration phase signal of each frame of the fault video;
calculating the local phase difference signal of each single point in each frame of video image;
extracting a bearing seat area from the fault video, and extracting the local phase difference signal and the vibration amplitude signal of each single point in the bearing seat area from a vibration amplitude signal and a vibration phase signal of each frame of the fault video;
and weighting according to the local phase difference signal and the vibration amplitude signal of each single point in the bearing seat area to obtain the full-layer global visual vibration signal.
3. The bearing fault visual vibration detection method according to claim 2, wherein the step of obtaining the fault signal-to-noise ratio of the bearing by calculation according to the full-layer global visual vibration signal, and obtaining the generalized Kurtosis operator based on the fault signal-to-noise ratio by calculation according to the fault signal-to-noise ratio comprises:
calculating according to the full-layer global visual vibration signal to obtain a frequency energy spectrum of the generalized Kurtosis operator;
estimating the fault characteristic frequency of the full-layer global visual vibration signal by a maximum likelihood method according to the frequency energy spectrum;
calculating to obtain a fault signal-to-noise ratio of the bearing according to the fault characteristic frequency;
and selecting a corresponding fault parameter when the fault signal-to-noise ratio is maximum, and obtaining the generalized Kurtosis operator according to the fault parameter.
4. The bearing fault visual vibration detection method of claim 3, wherein said step of separating said full-layer global visual vibration signal comprises:
establishing a dynamic model, and constructing an impact wavelet dictionary based on the dynamic model to sparsely represent impact components in the full-layer global visual vibration signal;
after the impact component is sparsely represented, separating the full-layer global visual vibration signal to obtain the impact component.
5. The method for detecting visual vibration of bearing faults as claimed in claim 4, wherein the step of calculating the impulse component of the full-layer global visual vibration signal according to the generalized Kurtosis operator comprises:
performing 5-layer decomposition on the full-layer global visual vibration signal by using the impact wavelet of the dynamic model, and selecting the generalized Kurtosis operator of the 5 th-layer visual vibration signal;
and calculating the morphological parameters corresponding to the generalized Kurtosis operator of the 5 th-layer visual vibration signal according to the generalized Kurtosis operator, and determining a corresponding optimal rank to obtain the impact component by using an iterative threshold method.
6. A bearing fault visual vibration detection system, comprising:
the acquisition module is used for acquiring fault video information of the bearing;
the acquisition module is used for acquiring a full-layer global visual vibration signal of the bearing according to the fault video information of the bearing;
the first calculation module is used for calculating to obtain a fault signal-to-noise ratio of the bearing according to the full-layer global visual vibration signal and calculating to obtain a generalized Kurtosis operator based on the fault signal-to-noise ratio according to the fault signal-to-noise ratio;
the separation module is used for separating the full-layer global visual vibration signal;
the second calculation module is used for calculating and obtaining the impact component of the full-layer global visual vibration signal according to the generalized Kurtosis operator;
and the analysis module is used for carrying out fault analysis on the bearing according to the impact component.
7. The bearing fault visual vibration detection system of claim 6, wherein the acquisition module is specifically configured to:
decomposing the fault video frame by adopting a complex controllable pyramid to obtain a vibration amplitude signal and a vibration phase signal of each frame of the fault video;
calculating the local phase difference signal of each single point in each frame of video image;
extracting a bearing seat area from the fault video, and extracting the local phase difference signal and the vibration amplitude signal of each single point in the bearing seat area from a vibration amplitude signal and a vibration phase signal of each frame of the fault video;
and weighting according to the local phase difference signal and the vibration amplitude signal of each single point in the bearing seat area to obtain the full-layer global visual vibration signal.
8. The bearing fault visual vibration detection system as claimed in claim 7, wherein said first calculation module comprises:
the first calculation submodule is used for calculating and obtaining a frequency energy spectrum of the generalized Kurtosis operator according to the full-layer global visual vibration signal;
the estimation submodule is used for estimating the fault characteristic frequency of the full-layer global visual vibration signal through a maximum likelihood method according to the frequency energy spectrum;
the second calculation submodule is used for calculating the fault signal-to-noise ratio of the bearing according to the fault characteristic frequency;
and the selection submodule is used for selecting the corresponding fault parameter when the fault signal-to-noise ratio is maximum, and obtaining the generalized Kurtosis operator according to the fault parameter.
9. The bearing fault visual vibration detection system of claim 8, wherein the separation module is specifically configured to:
establishing a dynamic model, and constructing an impact wavelet dictionary based on the dynamic model to sparsely represent impact components in the full-layer global visual vibration signal;
after the impact component is sparsely represented, separating the full-layer global visual vibration signal to obtain the impact component.
10. The bearing fault visual vibration detection system of claim 9, wherein the second calculation module is specifically configured to:
performing 5-layer decomposition on the full-layer global visual vibration signal by using the impact wavelet of the dynamic model, and selecting the generalized Kurtosis operator of the 5 th-layer visual vibration signal;
and calculating the morphological parameters corresponding to the generalized Kurtosis operator of the 5 th-layer visual vibration signal according to the generalized Kurtosis operator, and determining a corresponding optimal rank to obtain the impact component by using an iterative threshold method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110564916.XA CN113155466B (en) | 2021-05-24 | 2021-05-24 | Bearing fault visual vibration detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110564916.XA CN113155466B (en) | 2021-05-24 | 2021-05-24 | Bearing fault visual vibration detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113155466A true CN113155466A (en) | 2021-07-23 |
| CN113155466B CN113155466B (en) | 2024-04-23 |
Family
ID=76877083
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110564916.XA Active CN113155466B (en) | 2021-05-24 | 2021-05-24 | Bearing fault visual vibration detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113155466B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113670612A (en) * | 2021-08-16 | 2021-11-19 | 西南交通大学 | Rolling bearing fault diagnosis method based on weighted combined envelope spectrum |
| CN113820133A (en) * | 2021-09-28 | 2021-12-21 | 华中科技大学 | Sparse reconstruction method and system for bearing vibration signals |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120027733A (en) * | 2010-09-13 | 2012-03-22 | 한국수력원자력 주식회사 | Rotating machinery fault diagnostic method and system using support vector machines |
| EP2581725A2 (en) * | 2011-10-13 | 2013-04-17 | General Electric Company | Methods and systems for automatic rolling-element bearing fault detection |
| CN104316323A (en) * | 2014-10-24 | 2015-01-28 | 西安交通大学 | Method for confirming optimal resonance frequency band based on period target |
| CN104819766A (en) * | 2015-05-13 | 2015-08-05 | 西安交通大学 | Envelope demodulation frequency band determination method based on harmonic-to-noise ratio |
| AU2015201595A1 (en) * | 2015-03-27 | 2016-10-13 | Tensor Systems Pty Ltd | Vibration measurement and analysis |
| CN107876425A (en) * | 2017-11-10 | 2018-04-06 | 华东交通大学 | A kind of bearing defect detecting system device of view-based access control model |
| KR101879385B1 (en) * | 2017-10-10 | 2018-07-18 | 한국발전기술주식회사 | Signal processing apparatus for vibration supervisory |
| CN109029993A (en) * | 2018-06-20 | 2018-12-18 | 中国计量大学 | In conjunction with the bearing fault detection algorithm of genetic algorithm optimization parameter and machine vision |
| CN110440902A (en) * | 2019-08-29 | 2019-11-12 | 合肥工业大学 | A kind of contactless microvibration vision measuring method |
| CN111060315A (en) * | 2019-11-28 | 2020-04-24 | 南京航空航天大学 | Mechanical fault diagnosis method based on vision |
| CN111275007A (en) * | 2020-02-24 | 2020-06-12 | 山东大学 | Bearing fault diagnosis method and system based on multi-scale information fusion |
| CN112629854A (en) * | 2020-11-25 | 2021-04-09 | 西安交通大学 | Bearing fault classification method based on neural network attention mechanism |
-
2021
- 2021-05-24 CN CN202110564916.XA patent/CN113155466B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120027733A (en) * | 2010-09-13 | 2012-03-22 | 한국수력원자력 주식회사 | Rotating machinery fault diagnostic method and system using support vector machines |
| EP2581725A2 (en) * | 2011-10-13 | 2013-04-17 | General Electric Company | Methods and systems for automatic rolling-element bearing fault detection |
| CN104316323A (en) * | 2014-10-24 | 2015-01-28 | 西安交通大学 | Method for confirming optimal resonance frequency band based on period target |
| AU2015201595A1 (en) * | 2015-03-27 | 2016-10-13 | Tensor Systems Pty Ltd | Vibration measurement and analysis |
| CN104819766A (en) * | 2015-05-13 | 2015-08-05 | 西安交通大学 | Envelope demodulation frequency band determination method based on harmonic-to-noise ratio |
| KR101879385B1 (en) * | 2017-10-10 | 2018-07-18 | 한국발전기술주식회사 | Signal processing apparatus for vibration supervisory |
| CN107876425A (en) * | 2017-11-10 | 2018-04-06 | 华东交通大学 | A kind of bearing defect detecting system device of view-based access control model |
| CN109029993A (en) * | 2018-06-20 | 2018-12-18 | 中国计量大学 | In conjunction with the bearing fault detection algorithm of genetic algorithm optimization parameter and machine vision |
| CN110440902A (en) * | 2019-08-29 | 2019-11-12 | 合肥工业大学 | A kind of contactless microvibration vision measuring method |
| CN111060315A (en) * | 2019-11-28 | 2020-04-24 | 南京航空航天大学 | Mechanical fault diagnosis method based on vision |
| CN111275007A (en) * | 2020-02-24 | 2020-06-12 | 山东大学 | Bearing fault diagnosis method and system based on multi-scale information fusion |
| CN112629854A (en) * | 2020-11-25 | 2021-04-09 | 西安交通大学 | Bearing fault classification method based on neural network attention mechanism |
Non-Patent Citations (3)
| Title |
|---|
| XUENAN LIU 等: "Heart Rate Detection From Facial Videos Using A Frequencyconstrained Multilayer Sparse Coding", 《 2020 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP)》 * |
| 张文义;于德介;陈向民;: "基于信号共振稀疏分解与能量算子解调的轴承故障诊断方法", 中国电机工程学报, no. 20 * |
| 齐鹏;范玉刚;冯早;: "敏感SVD和EEMD的故障诊断方法研究", 传感器与微系统, no. 02 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113670612A (en) * | 2021-08-16 | 2021-11-19 | 西南交通大学 | Rolling bearing fault diagnosis method based on weighted combined envelope spectrum |
| CN113670612B (en) * | 2021-08-16 | 2023-03-14 | 西南交通大学 | Rolling bearing fault diagnosis method based on weighted combined envelope spectrum |
| CN113820133A (en) * | 2021-09-28 | 2021-12-21 | 华中科技大学 | Sparse reconstruction method and system for bearing vibration signals |
| CN113820133B (en) * | 2021-09-28 | 2022-07-12 | 华中科技大学 | Sparse reconstruction method and system for bearing vibration signals |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113155466B (en) | 2024-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ding et al. | Time–frequency manifold sparse reconstruction: A novel method for bearing fault feature extraction | |
| Mishra et al. | Rolling element bearing defect diagnosis under variable speed operation through angle synchronous averaging of wavelet de-noised estimate | |
| Pan et al. | A novel deep learning network via multiscale inner product with locally connected feature extraction for intelligent fault detection | |
| Wang et al. | Fault diagnosis of diesel engine based on adaptive wavelet packets and EEMD-fractal dimension | |
| He et al. | Sparse representation based on local time–frequency template matching for bearing transient fault feature extraction | |
| CN113155466A (en) | Bearing fault visual vibration detection method and system | |
| CN108414240B (en) | Method and device for detecting abnormal vibration of machine | |
| CN114266339A (en) | Rolling bearing fault diagnosis method based on small sample and GAF-DCGAN | |
| CN114112400A (en) | Mechanical bearing fault diagnosis method based on multi-angle information fusion | |
| CN111735523A (en) | Vehicle weight detection method and device based on video identification and storage medium | |
| CN117390380A (en) | Data analysis method in oil-residue separation system | |
| CN111353400A (en) | Whole scene vibration intensity atlas analysis method based on visual vibration measurement | |
| CN116363121A (en) | Computer vision-based inhaul cable force detection method, system and device | |
| CN109708877B (en) | Mechanical fault analysis method based on wavelet fuzzy recognition and image analysis theory | |
| CN114755010A (en) | Rotary machine vibration fault diagnosis method and system | |
| CN117828447A (en) | Aeroengine bearing fault diagnosis method based on spectrogram wavelet transformation and multi-scale permutation entropy | |
| CN113447111B (en) | Visual vibration amplification method, detection method and system based on morphological component analysis | |
| CN110160642B (en) | Propeller noise line spectrum reconstruction method under small sample condition | |
| CN116295790A (en) | Frequency detection method and system based on inter-frame phase difference of bridge inhaul cable characteristic region | |
| CN113128474B (en) | Structural modal identification method based on computer vision and variational modal decomposition | |
| CN115657140A (en) | Magnetic anomaly detection method and system based on structured Hankel total variation regularization | |
| Aiordachioaie et al. | Change Detection by Feature Extraction and Processing from Time-Frequency Images | |
| CN114705427A (en) | Rolling bearing intelligent fault diagnosis method under strong noise environment | |
| Yang et al. | Basis pursuit‐based intelligent diagnosis of bearing faults | |
| Bao et al. | Motion magnification for quantifying aeroelastic modes from high-speed videos |
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 |