CN117347054A - Motor bearing fault diagnosis method, system, computer and storage medium - Google Patents
Motor bearing fault diagnosis method, system, computer and storage medium Download PDFInfo
- Publication number
- CN117347054A CN117347054A CN202311630430.7A CN202311630430A CN117347054A CN 117347054 A CN117347054 A CN 117347054A CN 202311630430 A CN202311630430 A CN 202311630430A CN 117347054 A CN117347054 A CN 117347054A
- Authority
- CN
- China
- Prior art keywords
- signal
- motor bearing
- representing
- signals
- energy
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 78
- 238000000034 method Methods 0.000 title claims abstract description 68
- 238000003860 storage Methods 0.000 title claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 62
- 239000011159 matrix material Substances 0.000 claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 24
- 238000001228 spectrum Methods 0.000 claims abstract description 19
- 238000006243 chemical reaction Methods 0.000 claims abstract description 17
- 238000004364 calculation method Methods 0.000 claims abstract description 12
- 230000006870 function Effects 0.000 claims description 22
- 230000009467 reduction Effects 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 9
- 230000000737 periodic effect Effects 0.000 claims description 6
- 238000013519 translation Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 description 8
- 230000002159 abnormal effect Effects 0.000 description 6
- 230000001133 acceleration Effects 0.000 description 5
- 238000002149 energy-dispersive X-ray emission spectroscopy Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 238000012795 verification Methods 0.000 description 5
- 238000013527 convolutional neural network Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 238000000354 decomposition reaction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000003058 natural language processing Methods 0.000 description 3
- 238000010200 validation analysis Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 150000001875 compounds Chemical class 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000011946 reduction process Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000010485 coping Effects 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
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000011423 initialization method Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000010363 phase shift Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000002864 sequence alignment Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M17/00—Testing of vehicles
- G01M17/08—Railway vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/776—Validation; Performance evaluation
-
- 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)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Acoustics & Sound (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention provides a motor bearing fault diagnosis method, a motor bearing fault diagnosis system, a computer and a storage medium, wherein the motor bearing fault diagnosis method comprises the steps of obtaining a vibration signal of a motor bearing; constructing the original signals into two-dimensional signal data by utilizing a matrix formula, calculating the energy value of each group of the basic signals by utilizing a singular value energy difference spectrum according to an energy calculation formula, and separating the noise signals of the basic signals according to the energy values to obtain useful signals; processing the useful signal by using a conversion processing formula to obtain a two-dimensional time-frequency image, and carrying out modal sectionalization on the useful signal by using modal section mathematical definition to obtain a sectionalized signal; and creating and training a double-processing diagnosis model, and inputting the two-dimensional time-frequency image and the compartmentalization signal into the diagnosis model to obtain the fault type of the motor bearing and the position of the fault, thereby reducing the problem that noise interference caused by external environment affects the recognition rate.
Description
Technical Field
The invention belongs to the technical field of vehicle detection, and particularly relates to a motor bearing fault diagnosis method, a motor bearing fault diagnosis system, a motor bearing fault diagnosis computer and a motor bearing fault diagnosis storage medium.
Background
Traction motor bearings are important mechanical elements for supporting and bearing rotating parts of a traction motor (e.g., the motor rotor), which are quite common in trains. Traction motor bearings play a critical role in motor operation, allowing the rotating parts of the motor to rotate smoothly during operation while withstanding forces from motor loads and vibrations. The fault diagnosis of the traction motor bearing is beneficial to improving the reliability, availability and efficiency of the train, reducing the maintenance cost, improving the safety and prolonging the service life of equipment.
With the continuous development of artificial intelligence technology, traction motor bearing fault diagnosis has gradually progressed from traditional fault diagnosis to intelligent fault diagnosis. The traditional fault diagnosis method has high dependence on professional knowledge and working experience of service personnel, so subjective judgment can occur. Conventional fault diagnosis methods generally adopt periodic detection and maintenance equipment, and a great deal of funds are required to support work development every year. The conventional fault diagnosis method is generally divided into two parts, namely feature extraction and pattern recognition, wherein the feature extraction is difficult to fully extract, the diagnosis is carried out by relying on manual intervention and manual data analysis, and a long time is required for diagnosis, which can be a serious disadvantage in the case of taking measures rapidly. In summary, although the conventional method plays an important role in the past, with technological progress and development, the more intelligent fault diagnosis method meets the needs of the human at present.
However, when the existing intelligent fault diagnosis method is used for fault diagnosis, abnormal noise is usually associated with the bearing fault of the traction motor, and the abnormal noise can interfere with accurate detection of fault characteristics.
Disclosure of Invention
In order to solve the technical problems, the invention provides a motor bearing fault diagnosis method, a motor bearing fault diagnosis system, a computer and a storage medium system, which are used for solving the technical problems of lower detection accuracy caused by abnormal vibration and noise interference of the existing traction motor bearing fault diagnosis method.
In one aspect, the invention provides the following technical scheme, and a motor bearing fault diagnosis method comprises the following steps:
acquiring a vibration signal of a motor bearing, performing time domain processing analysis on the vibration signal, and taking the vibration signal after time domain processing which accords with preset vibration amplitude as an original signal;
constructing the original signals into two-dimensional signal data by utilizing a matrix formula, decomposing the two-dimensional signal data into a plurality of groups of basic signals, calculating the energy value of each group of basic signals by utilizing a singular value energy difference spectrum according to an energy calculation formula, and separating the noise signals of the basic signals according to the energy values to obtain useful signals;
processing the useful signal by using a conversion processing formula to obtain a two-dimensional time-frequency image, and carrying out modal sectionalization on the useful signal by using modal section mathematical definition to obtain a sectionalized signal;
and creating and training to obtain a double-processing diagnosis model, and inputting the two-dimensional time-frequency image and the compartmentalization signal into the diagnosis model to obtain the fault type of the motor bearing and the position of the fault.
Compared with the prior art, the beneficial effects of this application are: the problem of determining how to reconstruct the effective order of a matrix after the vibration signal of the traction motor bearing is decomposed is solved by the coordination of the singular value energy difference spectrum and an energy calculation formula, so that the problem of influence of noise interference brought by an external environment on the recognition rate is reduced; in addition, a two-dimensional time-frequency image is obtained by utilizing a conversion processing formula, so that the time-frequency characteristic signal can be fully reacted, the information contained in the vibration signal is reserved with high quality, the running states of the train traction motor bearing under different fault conditions are accurately represented, and meanwhile, the problem of uncertainty of the train traction motor bearing is further processed by utilizing a modal interval.
Further, the matrix formula includes:
in the method, in the process of the invention,for->Constructing a Hankel matrix>、/>、/>、/>Are data points representing motor bearing data, +.>Represents an upper limit corresponding to +.>=/>+/>+1,/>Representing the final value of the matrix abscissa, +.>Representing the final value of the matrix on the ordinate, +.>,/>Representing useful signal space, < > for>Representing noise signal space.
Further, the singular value energy differential spectrum is defined as
In the method, in the process of the invention,the sequence formed is called singular value energy difference spectrum,/->Representing singular values from i=1 to i= =>Square of>Representing singular values from i=2 to i= =>The square of +1, E, represents the signal energy.
Further, the conversion processing formula includes:
in the formula (i),representing time-frequency resolution, < >>Representing time in the time domain,/->Indicate frequency,/->Representing a>Periodic function of variation->Representing the imaginary part of the complex frequency, < >>Signal representing motor bearing after noise reduction, +.>Represents the translation amount->Representing a gaussian window function, +.>Representing the degree of movement of the gaussian window function; wherein,and->Representing the inverse of the absolute value of the frequency after adjustment of the parameter m,、/>and respectively representing the signal after noise reduction of the motor bearing and the Gaussian window function after adjustment after the parameter m is added.
Further, the energy calculation formula includes:
where E represents the signal energy,representing singular values from i=1 to i= =>Square of>Representing the total order, i.e. to +.>Until that point.
Further, the step of inputting the two-dimensional time-frequency image and the compartmentalized signal into the diagnostic model includes:
and inputting the two-dimensional time-frequency image into the diagnosis model, outputting the fault type of the motor bearing, inputting the compartmentalization signal into the diagnosis model, and outputting the position of the fault of the motor bearing.
Further, the dual-processing diagnostic model comprises an Embedding module, a Positional Encoder module, a Stacking module and an Output Head module.
In a second aspect, the present invention provides the following technical solutions, where the motor bearing fault diagnosis system includes:
the acquisition module is used for acquiring a vibration signal of the motor bearing, performing time domain processing analysis on the vibration signal, and taking the vibration signal after time domain processing which accords with preset vibration amplitude as an original signal;
the denoising module is used for constructing the original signals into two-dimensional signal data by utilizing a matrix formula, decomposing the two-dimensional signal data into a plurality of groups of basic signals, calculating the energy value of each group of basic signals by utilizing a singular value energy difference spectrum according to an energy calculation formula, and separating the noise signals of the basic signals according to the energy values to obtain useful signals;
the processing module is used for processing the useful signal by using a conversion processing formula to obtain a two-dimensional time-frequency image, and carrying out modal sectionalization on the useful signal by using modal section mathematical definition to obtain a sectionalized signal;
the detection module is used for creating and training a double-processing diagnosis model, and inputting the two-dimensional time-frequency image and the interval signal into the diagnosis model to obtain the fault type of the motor bearing and the position of the fault.
In a third aspect, the present invention provides a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the motor bearing fault diagnosis method as described above when executing the computer program.
In a fourth aspect, the present invention provides a storage medium having a computer program stored thereon, which when executed by a processor, implements a motor bearing fault diagnosis method as described above.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a motor bearing fault diagnosis method according to a first embodiment of the present invention;
FIG. 2 is a block diagram of a motor bearing fault diagnosis system according to a second embodiment of the present invention;
fig. 3 is a schematic hardware structure of a computer according to a third embodiment of the present invention.
Embodiments of the present invention will be further described below with reference to the accompanying drawings.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and intended to illustrate embodiments of the invention and should not be construed as limiting the invention.
In the description of the embodiments of the present invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like indicate orientations or positional relationships based on the orientation or positional relationships shown in the drawings, merely to facilitate description of the embodiments of the present invention and simplify description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless explicitly defined otherwise.
Example 1
In a first embodiment of the present invention, as shown in fig. 1, a motor bearing fault diagnosis method includes the following steps S01 to S04:
s01, obtaining a vibration signal of a motor bearing, performing time domain processing analysis on the vibration signal, and taking the vibration signal after time domain processing conforming to preset vibration amplitude as an original signal;
when the method is implemented, a fault diagnosis experiment platform for the traction motor bearing of the train is built, and the fault diagnosis experiment platform comprises the traction motor bearing, two piezoelectric acceleration sensors, a frequency converter, a data acquisition card, a PC (personal computer) and the like. Mounted on traction electricity using piezoelectric acceleration sensorThe vertical and horizontal directions of the machine bearing end cover are respectively provided with a data acquisition card, a traction motor bearing vibration signal acquired by a sensor is acquired by the data acquisition card and is transmitted to a PC for time domain processing analysis, and train traction motor bearing data with obvious vibration amplitude is selected as an original signal. Counting as。
In the embodiment, a fault diagnosis experiment platform for a train traction motor bearing is built and comprises a traction motor bearing, two piezoelectric acceleration sensors CAYD051V, a frequency converter G7R5/P011T4, a data acquisition card YE6231, a PC and the like. The piezoelectric acceleration sensor CAYD051V is arranged in the vertical direction and the horizontal direction of the traction motor bearing end cover, the vibration signals of the traction motor bearing are collected through the data acquisition card and transmitted to the PC for time domain processing analysis, and train traction motor bearing data with obvious vibration amplitude is selected as an original signal. Counting as. In this example implementation, 1024 points are defined as one set of data lengths, and 1000 sets are acquired for each state, so 5000 sets are total.
S02, constructing the original signal into two-dimensional signal data by utilizing a matrix formula, decomposing the two-dimensional signal data into a plurality of groups of basic signals, calculating the energy value of each group of basic signals by utilizing a singular value energy difference spectrum according to an energy calculation formula, and separating the noise signals of the basic signals according to the energy values to obtain useful signals;
it should be noted that, since the vibration signal (original signal) is usually a one-dimensional signal, SVD cannot be directly performed on the vibration signal, and a two-dimensional matrix needs to be constructed first; decomposing the two-dimensional signal data into a plurality of groups of basic signals for signal reconstruction, wherein the signal reconstruction refers to decomposing one signal (two-dimensional signal data) into a group of basic signals or modes, and reconstructing an original signal by utilizing the basic signals;
in the aspect of signal reconstruction, the determination of useful orders of singular values is particularly important, if more singular values are selected for signal reconstruction, the noise reduction is incomplete because a part of noise signals still remain in the noise reduced signals, and if fewer singular values are selected for signal reconstruction, useful signals are deleted, and incomplete information in the original vibration signals is caused. To solve this problem, a singular value energy difference spectrum is introduced, and the effective order of the reconstructed matrix after singular value decomposition is determined according to the contributions of the noise signal and the useful signal to the singular values.
In this embodiment, there are various methods for constructing a two-dimensional matrix from one-dimensional signals, such as: cyclic matrices, toeplitz matrices, hankel matrices, etc.; the Hankel matrix is constructed for one-dimensional vibration signals because of its zero phase shift and wavelet-like characteristics, which are the most widely used. So that the original signal is firstly subjected toConstructing Hankel matrix->The method comprises the steps of carrying out a first treatment on the surface of the Singular Value Decomposition (SVD) in combination with energy-dispersive spectroscopy (EDS) is an effective noise reduction method commonly used in image, audio and signal processing. The method combines the dimension reduction capability of SVD and the spectral feature analysis of EDS, can effectively reduce noise in signals, and simultaneously retains important features of the signals. SVD can be used to separate signal and noise components. By retaining the largest singular value (singular value with higher energy), the main signal component can be reconstructed while suppressing the noise component, thereby achieving the effect of noise reduction.
The matrix formula includes:
in the method, in the process of the invention,for->Constructing a Hankel matrix>、/>、/>、/>Are data points representing motor bearing data, +.>Represents an upper limit corresponding to +.>=/>+/>+1,/>Representing the final value of the matrix abscissa, +.>Representing the final value of the matrix on the ordinate, +.>,/>Representing useful signal space, < > for>Represents the noise signal space, wherein, when +.>In this case, the Hankel matrix has good noise reduction effect.
The energy calculation formula includes:
where E represents the signal energy,representing singular values from i=1 to i= =>Square of>Representing the total order, i.e. to +.>Until that point.
Therefore, defining a singular value energy differential spectrum and carrying out normalization processing;
the definition of the singular value energy differential spectrum is as follows
In the method, in the process of the invention,the sequence formed is called singular value energy difference spectrum,/->Representing singular values from i=1 to i= =>Square of>Representing singular values from i=2 to i= =>The square of +1, E, represents the signal energy.
The singular value energy of the useful signal is larger than the proportion of the noise signal, so that the demarcation between the noise signal and the useful signal can cause largerThe peak value fluctuates, and the singular value after the peak value is mainly generated by a noise signal, so that the singular value corresponding to the point can be found in the singular value energy differential spectrum, then the point is taken as the order of a reconstruction signal, the separation of the noise signal and the useful signal can be realized, and the purpose of noise reduction of the vibration signal of the bearing of the traction motor of the train is achieved. The data set after noise reduction optimization is as follows。
S03, processing the useful signal by using a conversion processing formula to obtain a two-dimensional time-frequency image, and carrying out modal sectionalization on the useful signal by using modal section mathematical definition to obtain a sectionalized signal;
the conversion processing formula includes:
in the formula (i),representing time-frequency resolution, < >>Representing time in the time domain,/->Indicate frequency,/->Representing a>Periodic function of variation->Representing the imaginary part of the complex frequency, < >>Signal representing motor bearing after noise reduction, +.>Represents the translation amount->Representing a gaussian window function, +.>Representing the degree of movement of the gaussian window function; wherein,and->Representing the inverse of the absolute value of the frequency after adjustment of the parameter m,、/>and respectively representing the signal after noise reduction of the motor bearing and the Gaussian window function after adjustment after the parameter m is added.
It should be noted that the conversion processing formula is based on the extension of the conventional time-frequency analysis method (such as short-time fourier transform and wavelet transform) of generalized S transform, and is better for processing non-stationary signals. The generalized S transformation mainly aims at transforming the one-dimensional vibration signal from a time domain to a time domain, so as to provide more detailed and comprehensive fault signal characteristics on the time domain. The principle of GST is based on the ideas of Short Time Fourier Transform (STFT) and Continuous Wavelet Transform (CWT). Its core concept is to perform local spectral processing on the signal at different points in time.
In the formula (I), the compound (II) is a compound (III),representing S transform +.>For time variable, time in the time domain, +.>Representing frequency, & lt>Described is a time +.>Periodic function of variation->Representing the imaginary part of the complex frequency, < >>For the signal after the train traction motor bearing falls the noise, translation volume is with +.>And (3) representing. />Representing a gaussian window function, +.>Indicating the degree of movement of the gaussian window function,and->Representing the inverse of the absolute value of the frequency. GST is modified on the S transform formula. The width of the Gaussian window is adjusted by adding the parameter m, so that the time-frequency resolution of S conversion is improved, and a conversion processing formula is obtained; GST is carried out on the one-dimensional vibration signal of the traction motor bearing after noise reduction, so that a two-dimensional time-frequency image is obtained. The vibration signal is imaged, so that the information contained in the vibration signal can be reserved with high quality, and meanwhile, the deep learning model has good recognition processing characteristics on the converted two-dimensional image data.
The current deep learning method is applied to the initial effect of fault diagnosis. However, studies have been made on the one-dimensional vibration signal input, and the method of imaging the vibration signal is not yet deep enough for fault diagnosis. The generalized S-transform (GST) is a signal analysis method, which is used for converting a time domain signal into a time-frequency domain image, and using GST to image a vibration signal, so that the information contained in the vibration signal can be reserved with high quality, and the vibration signal data can be optimized and preprocessed. Meanwhile, the deep learning model has good recognition processing characteristics on the converted two-dimensional image data, and the diagnosis accuracy is high.
The modal interval mathematical definition includes:
in the mode region Y of the device,upper bound (I)>Representing the lower bound, the interval in the modal interval is +.>. The Kaucher algorithm is a theoretical basis of a modal interval;
for the problems that uncertainty and the like possibly occur in the process of processing the train traction motor bearing, the data after noise reduction processing is converted into a mode interval form and is recorded as sum,/>In the formula, < >>For the lower limit of the bearing data set of the traction motor of the train, < + >>Is the lower limit of the bearing data set of the traction motor of the train.
Modal Interval Analysis (MIA) is one of the effective tools for coping with uncertainty problems. It uses intervals of modal parameters to represent the uncertainty range, rather than a single determined value. By taking into account the range in which different modal parameters may exist, MIA may provide more comprehensive information with greater reliability.
S04, creating and training to obtain a double-processing diagnosis model, and inputting the two-dimensional time-frequency image and the interval signal into the diagnosis model to obtain the fault type of the motor bearing and the position of the fault.
Specifically, the step of inputting the two-dimensional time-frequency image and the compartmentalized signal into the diagnostic model includes:
and inputting the two-dimensional time-frequency image into the diagnosis model, outputting the fault type of the motor bearing, inputting the compartmentalization signal into the diagnosis model, and outputting the position of the fault of the motor bearing.
Specifically, the dual-processing diagnostic model includes an encoding module, a Positional Encoder module, a Stacking module and an Output Head module.
In recent years, the transducer model has been used to increase the variety of colors in the field of deep learning. The transducer model is a classical Natural Language Processing (NLP) model due to its excellent sequence modeling capability and parallel computing advantages. The transducer model has also been successfully applied to fault diagnosis tasks, particularly for the processing of time series data or sensor data. Its biggest feature is the use of an attention mechanism to calculate its inputs and outputs, as well as to balance the processing power. The sequence structure of the traditional cyclic neural network (RNN) sequence alignment is not adopted, and the defect of limited receptive field of the Convolutional Neural Network (CNN) is avoided, so that the Convolutional Neural Network (CNN) can be used for capturing global information. The input of the original transducer in the NLP field is one-dimensional word sequence, while the picture is two-dimensional, so Vision Transformer is used for two-dimensional image input. The vibration signal of the traction motor bearing is converted into a two-dimensional time-frequency image, so that abundant characteristic information can be reserved, and the input of a Vision Transformer model is facilitated.
In this example, the diagnostic model for the double treatment is the MIA-Vision Transformer model;
and (3) carrying out coding processing on the vibration images of various state types of the train traction motor bearing after conversion, and carrying out proportion division on the various state types of the train traction motor bearing, wherein the vibration images are divided into training samples, verification samples and test samples. The MIA-Vision Transformer model comprises the following steps: the building module, the Positional Encoder module, the Stacking module, the Output Head module and the like form a complete MIA-Vision Transformer model together. The MIA-Vision Transformer model is a delay rise of the Vision Transformer model, vision Transformer is used for further analyzing and understanding abnormal conditions by combining MIA results in outputting various train traction motor bearing results in Vision Transformer model through sectioning the data parameters of the train traction motor bearing after noise reduction. Image features associated with modal parameter anomalies are identified to aid in determining fault type and location.
In this embodiment, the vibration images of various state types after the conversion of the train traction motor bearing are encoded, and the various state types of the train traction motor bearing are proportioned according to 7:2:1 are divided into training samples, validation samples and test samples. In this example implementation, the running states of the train traction motor bearing are divided into five states of an inner ring failure, a rolling body failure, a cage fracture failure, an outer ring failure and a normal traction motor bearing. When 1024 points of the current data acquisition have been specified as a group of data length, 1000 groups are acquired for each state, so 5000 groups are all acquired. Training set, validation set and test set partitioning according to 7:2:1, i.e. 3500 sets of training set, 1000 sets of validation set, 500 sets of test set. The MIA-Vision Transformer model comprises the following steps: the building module, the Positional Encoder module, the Stacking module, the Output Head module and the like form a complete MIA-Vision Transformer model together. The MIA-Vision Transformer model is a Vision Transformer model, the train traction motor bearing data after noise reduction is subjected to modal interval, the train traction motor bearing results in various states are output in the Vision Transformer model, the parameter of Vision Transformer is subjected to modal interval, and the abnormal situation is further analyzed and understood by combining the MIA result in the output of the Vision Transformer model. Image features associated with modal parameter anomalies are identified to aid in determining fault type and location.
MIA-Vision Transformer model optimization:
inputting a part of the image as a training sample and a verification sample into the MIA-Vision Transformer model for training, and dynamically adjusting the learning rate along with the training so as to ensure that the model can converge to the optimal performance; the proper weight initialization method has influence on the convergence speed and performance of the model, and the Xavier is selected for initialization, so that the training process is smoother; introducing a regularization method, and adding a Dropout layer to prevent the model from being over fitted; the depth (layer number) and the width (characteristic dimension) of the MIA-Vision Transformer model are adjusted, so that the performance and the training speed of the model reach the optimal state; and judging the training degree of the model through a loss curve, a accuracy curve and a T-SNE visualization, and when the training sample and the verification sample in the loss curve reach the minimum value after converging, basically no fluctuation exists and the training sample and the verification sample are overlapped, which indicates that the training is finished at the moment. In the accuracy curve training sample and the verification sample, a higher accuracy can be achieved, and the two curves are smooth and stable without large fluctuation, so that the training is finished. The same label sample in the T-SNE visualization is concentrated, and samples of different categories are scattered, so that the model training degree is good, and the training of the traction motor bearing MIA-Vision Transformer model is completed.
Identifying and diagnosing the running state of the motor bearing:
and taking the rest part of the image as a test sample, putting the part of the training sample into a trained traction motor bearing MIA-Vision Transformer model, and obtaining an output result through the model. In combination with the results of MIA, the anomalies are further analyzed and understood. Image features associated with modal parameter anomalies are identified and a fault type is determined.
Compared with the existing train traction motor bearing state identification method, the method has the advantages that EDS is introduced in the SVD noise reduction process to determine the order, the problem of how to determine the effective order of the reconstructed matrix after the vibration signals of the traction motor bearing are decomposed is solved, and therefore the problem that noise interference caused by the external environment affects the identification rate is reduced. The two-dimensional time-frequency image obtained by GST can fully react to the time-frequency characteristic signal, the information contained in the vibration signal is reserved with high quality, the running state of the train traction motor bearing under different fault conditions is accurately represented, and meanwhile, the problem of uncertainty of the train traction motor bearing is further processed by using the modal interval. Finally, feature extraction and pattern recognition classification problems are completed through an MIA-Vision Transformer model, and fault types and positions can be determined in an assisted mode through MIA.
In sum, through simultaneously collecting train traction motor bearing signals at different positions by using the double sensors, the time domain state is selected to be more obvious and used as a data set, so that the expressed information is richer and more real.
EDS is introduced in the SVD noise reduction process to determine the order, so that the problem of how to determine the effective order of the reconstructed matrix after decomposition is solved;
more hidden fault information can be mined on a traction motor bearing two-dimensional vibration image obtained by using GST through an MIA-Vision Transformer model, and meanwhile, the problems of incomplete and excessively complex feature extraction of a train traction motor bearing in the fault diagnosis process are effectively solved;
the abnormal condition can be further analyzed and processed through the combined use of MIA and Vision Transformer models, and image characteristics related to modal parameter abnormality are identified, so that the fault type is determined, and the problem of uncertainty in input of the train traction motor bearing is solved.
Example two
As shown in fig. 2, in a second embodiment of the present invention, there is provided a motor bearing failure diagnosis system including:
the acquisition module 10 is used for acquiring a vibration signal of the motor bearing, performing time domain processing analysis on the vibration signal, and taking the vibration signal after time domain processing which accords with preset vibration amplitude as an original signal;
the denoising module 20 is configured to construct the original signal into two-dimensional signal data by using a matrix formula, decompose the two-dimensional signal data into a plurality of groups of base signals, calculate energy values of each group of base signals by using a singular value energy difference spectrum and according to an energy calculation formula, and separate noise signals of the base signals according to the energy values to obtain useful signals;
the processing module 30 is configured to process the useful signal by using a conversion processing formula to obtain a two-dimensional time-frequency image, and perform modal compartmentalization on the useful signal by using a modal interval mathematical definition to obtain a compartmentalized signal;
the detection module 40 is configured to create and train a dual-processing diagnostic model, and input the two-dimensional time-frequency image and the compartmentalized signal into the diagnostic model to obtain a fault type of the motor bearing and a position where the fault is located.
In some alternative embodiments, the denoising module 20 includes:
a matrix unit, configured to, for the matrix formula, include:
in the method, in the process of the invention,for->Constructing a Hankel matrix>、/>、/>、/>Are data points representing motor bearing data, +.>Represents an upper limit corresponding to +.>=/>+/>+1,/>Representing the final value of the matrix abscissa, +.>Representing the final value of the matrix on the ordinate, +.>,/>Representing useful signal space, < > for>Representing noise signal space.
In some alternative embodiments, the denoising module 20 further includes:
singular units for the definition of the singular value energy difference spectrum as
In the method, in the process of the invention,the sequence formed is called singular value energy difference spectrum,/->Representing singular values from i=1 to i= =>Square of>Representing singular values from i=2 to i= =>The square of +1, E, represents the signal energy.
In some alternative embodiments, the processing module 30 includes:
the conversion unit is used for converting the processing formula and comprises the following steps:
in the formula (i),representing time-frequency resolution, < >>Representing time in the time domain,/->Indicate frequency,/->Representing a>Periodic function of variation->Representing the imaginary part of the complex frequency, < >>Signal representing motor bearing after noise reduction, +.>Represents the translation amount->Representing a gaussian window function, +.>Representing the degree of movement of the gaussian window function; wherein,and->Representing the inverse of the absolute value of the frequency after adjustment of the parameter m,、/>and respectively representing the signal after noise reduction of the motor bearing and the Gaussian window function after adjustment after the parameter m is added.
In some alternative embodiments, the denoising module 20 further includes:
the energy unit is used for the energy calculation formula and comprises the following components:
where E represents the signal energy,representing singular values from i=1 to i= =>Square of>Representing the total order, i.e. to +.>Until that point.
In some alternative embodiments, the detection module 40 includes:
the step of inputting the two-dimensional time-frequency image and the compartmentalized signal into the diagnostic model comprises the following steps of:
and inputting the two-dimensional time-frequency image into the diagnosis model, outputting the fault type of the motor bearing, inputting the compartmentalization signal into the diagnosis model, and outputting the position of the fault of the motor bearing.
In some alternative embodiments, the detection module 40 further comprises:
the diagnosis unit is used for the double-processing diagnosis model and comprises an Embedding module, a Positional Encoder module, a Stacking module and an Output Head module.
The motor bearing fault diagnosis system provided by the embodiment of the invention has the same implementation principle and technical effects as those of the embodiment of the method, and for the sake of brevity, reference is made to the corresponding content in the embodiment of the method.
Example III
As shown in fig. 3, in a third embodiment of the present invention, a computer is provided according to an embodiment of the present invention, including a memory 202, a processor 201, and a computer program stored in the memory 202 and executable on the processor 201, where the processor 201 implements the motor bearing fault diagnosis method as described above when executing the computer program.
In particular, the processor 201 may include a Central Processing Unit (CPU), or an application specific integrated circuit (Application Specific Integrated Circuit, abbreviated as ASIC), or may be configured to implement one or more integrated circuits of embodiments of the present application.
Memory 202 may include, among other things, mass storage for data or instructions. By way of example, and not limitation, memory 202 may comprise a Hard Disk Drive (HDD), floppy Disk Drive, solid state Drive (Solid State Drive, SSD), flash memory, optical Disk, magneto-optical Disk, tape, or universal serial bus (Universal Serial Bus, USB) Drive, or a combination of two or more of the foregoing. Memory 202 may include removable or non-removable (or fixed) media, where appropriate. The memory 202 may be internal or external to the data processing apparatus, where appropriate. In a particular embodiment, the memory 202 is a Non-Volatile (Non-Volatile) memory. In a particular embodiment, the Memory 202 includes Read-Only Memory (ROM) and random access Memory (Random Access Memory, RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (Programmable Read-Only Memory, abbreviated PROM), an erasable PROM (Erasable Programmable Read-Only Memory, abbreviated EPROM), an electrically erasable PROM (Electrically Erasable Programmable Read-Only Memory, abbreviated EEPROM), an electrically rewritable ROM (Electrically Alterable Read-Only Memory, abbreviated EAROM), or a FLASH Memory (FLASH), or a combination of two or more of these. The RAM may be Static Random-Access Memory (SRAM) or dynamic Random-Access Memory (Dynamic Random Access Memory DRAM), where the DRAM may be a fast page mode dynamic Random-Access Memory (Fast Page Mode Dynamic Random Access Memory FPMDRAM), extended data output dynamic Random-Access Memory (Extended Date Out Dynamic Random Access Memory EDODRAM), synchronous dynamic Random-Access Memory (Synchronous Dynamic Random-Access Memory SDRAM), or the like, as appropriate.
Memory 202 may be used to store or cache various data files that need to be processed and/or communicated, as well as possible computer program instructions for execution by processor 201.
The processor 201 implements the above-described motor bearing fault diagnosis method by reading and executing the computer program instructions stored in the memory 202.
In some of these embodiments, the computer may also include a communication interface 203 and a bus 200. As shown in fig. 3, the processor 201, the memory 202, and the communication interface 203 are connected to each other through the bus 200 and perform communication with each other.
The communication interface 203 is configured to enable communication between modules, apparatuses, units, and/or devices in embodiments of the present application. Communication interface 203 may also enable communication with other components such as: and the external equipment, the image/data acquisition equipment, the database, the external storage, the image/data processing workstation and the like are used for data communication.
Bus 200 includes hardware, software, or both, coupling components of a computer to each other. Bus 200 includes, but is not limited to, at least one of: data Bus (Data Bus), address Bus (Address Bus), control Bus (Control Bus), expansion Bus (Expansion Bus), local Bus (Local Bus). By way of example, and not limitation, bus 200 may include a graphics acceleration interface (Accelerated Graphics Port), abbreviated AGP, or other graphics Bus, an enhanced industry standard architecture (Extended Industry Standard Architecture, abbreviated EISA) Bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an industry standard architecture (Industry Standard Architecture, ISA) Bus, a radio Bandwidth (InfiniBand) interconnect, a low Pin Count (LO Pin Count, abbreviated LPC) Bus, a memory Bus, a micro channel architecture (Micro Channel Architecture, abbreviated MCa) Bus, a peripheral component interconnect (Peripheral Component Interconnect, abbreviated PCI) Bus, a PCI-Express (PCI-X) Bus, a serial advanced technology attachment (Serial Advanced Technology Attachment, abbreviated SATA) Bus, a video electronics standards Association local (Video Electronics Standards Association Local Bus, abbreviated VLB) Bus, or other suitable Bus, or a combination of two or more of these. Bus 200 may include one or more buses, where appropriate. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.
Example IV
In a fourth embodiment of the present invention, in combination with the above-mentioned motor bearing fault diagnosis method, the embodiment of the present invention provides a technical solution, a storage medium, where a computer program is stored, where the computer program is executed by a processor to implement the above-mentioned motor bearing fault diagnosis method.
Those of skill in the art will appreciate that the logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a sequence data table of executable instructions that may be considered to implement the logic functions, may 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 data) of the readable medium 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). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may 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.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.
The technical features of the above-described embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above-described embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples merely represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.
Claims (10)
1. A method of diagnosing a motor bearing failure, the method comprising:
acquiring a vibration signal of a motor bearing, performing time domain processing analysis on the vibration signal, and taking the vibration signal after time domain processing which accords with preset vibration amplitude as an original signal;
constructing the original signals into two-dimensional signal data by utilizing a matrix formula, decomposing the two-dimensional signal data into a plurality of groups of basic signals, calculating the energy value of each group of basic signals by utilizing a singular value energy difference spectrum according to an energy calculation formula, and separating the noise signals of the basic signals according to the energy values to obtain useful signals;
processing the useful signal by using a conversion processing formula to obtain a two-dimensional time-frequency image, and carrying out modal sectionalization on the useful signal by using modal section mathematical definition to obtain a sectionalized signal;
and creating and training to obtain a double-processing diagnosis model, and inputting the two-dimensional time-frequency image and the compartmentalization signal into the diagnosis model to obtain the fault type of the motor bearing and the position of the fault.
2. The motor bearing fault diagnosis method according to claim 1, wherein the matrix formula includes:
in the method, in the process of the invention,for->Constructing a Hankel matrix>、/>、/>、/>Are data points representing motor bearing data, +.>Represents an upper limit corresponding to +.>=/>+/>+1,/>Representing the final value of the matrix abscissa,representing the final value of the matrix on the ordinate, +.>,/>Representing useful signal space, < > for>Representing noise signal space.
3. The motor bearing fault diagnosis method according to claim 1, wherein the definition of the singular value energy differential spectrum is
In the method, in the process of the invention,the sequence formed is called singular value energy difference spectrum,/->Representing singular values from i=1 to i= =>Square of>Representing singular values from i=2 to i= =>The square of +1, E, represents the signal energy.
4. The motor bearing failure diagnosis method according to claim 1, wherein the conversion processing formula includes:
in the method, in the process of the invention,representing time-frequency resolution, < >>Representing time in the time domain,/->Indicate frequency,/->Representing a>Periodic function of variation->Representing the imaginary part of the complex frequency, < >>Signal representing motor bearing after noise reduction, +.>Represents the translation amount->Representing a gaussian window function, +.>Representing the degree of movement of the gaussian window function; wherein,and->Representing the inverse of the absolute value of the frequency after adjustment of the parameter m,、/>and respectively representing the signal after noise reduction of the motor bearing and the Gaussian window function after adjustment after the parameter m is added.
5. The motor bearing failure diagnosis method according to claim 1, characterized in that the energy calculation formula includes:
where E represents the signal energy,representing singular values from i=1 to i= =>Square of>Representing the total order, i.e. toUntil that point.
6. The motor bearing fault diagnosis method according to claim 1, wherein the step of inputting the two-dimensional time-frequency image and the compartmentalized signal into the diagnosis model comprises:
and inputting the two-dimensional time-frequency image into the diagnosis model, outputting the fault type of the motor bearing, inputting the compartmentalization signal into the diagnosis model, and outputting the position of the fault of the motor bearing.
7. The motor bearing fault diagnosis method according to claim 1, wherein the dual-process diagnosis model includes an assembly module, a Positional Encoder module, a Stacking module, and an Output Head module.
8. A motor bearing fault diagnosis system is characterized in that,
the acquisition module is used for acquiring a vibration signal of the motor bearing, performing time domain processing analysis on the vibration signal, and taking the vibration signal after time domain processing which accords with preset vibration amplitude as an original signal;
the denoising module is used for constructing the original signals into two-dimensional signal data by utilizing a matrix formula, decomposing the two-dimensional signal data into a plurality of groups of basic signals, calculating the energy value of each group of basic signals by utilizing a singular value energy difference spectrum according to an energy calculation formula, and separating the noise signals of the basic signals according to the energy values to obtain useful signals;
the processing module is used for processing the useful signal by using a conversion processing formula to obtain a two-dimensional time-frequency image, and carrying out modal sectionalization on the useful signal by using modal section mathematical definition to obtain a sectionalized signal;
the detection module is used for creating and training a double-processing diagnosis model, and inputting the two-dimensional time-frequency image and the interval signal into the diagnosis model to obtain the fault type of the motor bearing and the position of the fault.
9. A computer comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the motor bearing fault diagnosis method according to any one of claims 1 to 7 when executing the computer program.
10. A storage medium having stored thereon a computer program which, when executed by a processor, implements the motor bearing fault diagnosis method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311630430.7A CN117347054A (en) | 2023-12-01 | 2023-12-01 | Motor bearing fault diagnosis method, system, computer and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311630430.7A CN117347054A (en) | 2023-12-01 | 2023-12-01 | Motor bearing fault diagnosis method, system, computer and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117347054A true CN117347054A (en) | 2024-01-05 |
Family
ID=89355968
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311630430.7A Pending CN117347054A (en) | 2023-12-01 | 2023-12-01 | Motor bearing fault diagnosis method, system, computer and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117347054A (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104502126A (en) * | 2014-12-28 | 2015-04-08 | 华东交通大学 | Modal intervals-based high-speed train bogie fault diagnosis method |
CN111814404A (en) * | 2020-07-22 | 2020-10-23 | 华东交通大学 | Locomotive bogie bolt loosening fault detection method |
-
2023
- 2023-12-01 CN CN202311630430.7A patent/CN117347054A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104502126A (en) * | 2014-12-28 | 2015-04-08 | 华东交通大学 | Modal intervals-based high-speed train bogie fault diagnosis method |
CN111814404A (en) * | 2020-07-22 | 2020-10-23 | 华东交通大学 | Locomotive bogie bolt loosening fault detection method |
Non-Patent Citations (2)
Title |
---|
FENGYUN XIE 等: "Rolling Bearing Fault Diagnosis Based on SVD-GST Combined with Vision Transformer", 《ELECTRICS》, vol. 12, no. 16, pages 1 - 15 * |
博士学位论文编辑部: "《2007年上海大学博士学位论文 56 基于软计算机的故障诊断机理及其应用研究》", 30 September 2010, 上海大学出版社, pages: 96 - 97 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104819846B (en) | Rolling bearing sound signal fault diagnosis method based on short-time Fourier transform and sparse laminated automatic encoder | |
WO2022261805A1 (en) | Diesel engine gearbox fault diagnosis method | |
CN109211546B (en) | Rotary machine fault diagnosis method based on noise reduction automatic encoder and increment learning | |
CN112329914B (en) | Fault diagnosis method and device for buried transformer substation and electronic equipment | |
CN105973595A (en) | Diagnosis method of rolling bearing fault | |
CN105678343A (en) | Adaptive-weighted-group-sparse-representation-based diagnosis method for noise abnormity of hydroelectric generating set | |
CN117076935B (en) | Digital twin-assisted mechanical fault data lightweight generation method and system | |
Xu et al. | Hierarchical multiscale dense networks for intelligent fault diagnosis of electromechanical systems | |
CN111680665B (en) | Motor mechanical fault diagnosis method adopting current signals based on data driving | |
CN113160190A (en) | Method, equipment, device and storage medium for detecting abnormity of fastener of contact network | |
CN113390631A (en) | Fault diagnosis method for gearbox of diesel engine | |
CN117473414A (en) | Bearing fault position identification method based on low-noise time-frequency image | |
CN115876473A (en) | Bearing fault diagnosis method based on PWVD and DenseNet | |
CN117110744A (en) | Transformer fault diagnosis method and system based on voiceprint analysis | |
CN115166514A (en) | Motor fault identification method and system based on self-adaptive spectrum segmentation and denoising | |
CN114065809A (en) | Method and device for identifying abnormal sound of passenger car, electronic equipment and storage medium | |
CN113758709A (en) | Rolling bearing fault diagnosis method and system combining edge calculation and deep learning | |
CN117347054A (en) | Motor bearing fault diagnosis method, system, computer and storage medium | |
Yu et al. | Sparse time–frequency representation for the transient signal based on low-rank and sparse decomposition | |
CN106297768A (en) | Speech recognition method | |
CN117349657A (en) | Distributed data acquisition module and monitoring system for hydraulic engineering environment monitoring | |
CN113963719A (en) | Deep learning-based sound classification method and apparatus, storage medium, and computer | |
CN116796130A (en) | Bridge vibration low-frequency reconstruction denoising method, system, computer and storage medium | |
CN116484258A (en) | Elevator traction machine bearing fault diagnosis method | |
CN115901260A (en) | Rolling bearing fault diagnosis method based on ECA _ ResNet |
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 |