CN114266269A - Bearing fault diagnosis method and system, storage medium and equipment - Google Patents
Bearing fault diagnosis method and system, storage medium and equipment Download PDFInfo
- Publication number
- CN114266269A CN114266269A CN202111340215.4A CN202111340215A CN114266269A CN 114266269 A CN114266269 A CN 114266269A CN 202111340215 A CN202111340215 A CN 202111340215A CN 114266269 A CN114266269 A CN 114266269A
- Authority
- CN
- China
- Prior art keywords
- layer
- channels
- channel
- vibration
- data
- 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
Images
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention relates to a bearing fault diagnosis method, a system, a storage medium and equipment, which are characterized in that the following steps are executed by computer equipment to obtain vibration signal data s of N sensors at the fault occurrence position1,s2,…sNConstructing an attention module ATT, and increasing the weight of effective information; preprocessing data passing through an attention module, converting the acquired vibration signal data of N dimensions into a vibration image according to a set rule, and calculating a pixel value of the image; establishing a harmonic layer, performing characteristic fusion on the vibration signals of the N channels, and fully considering the signal characteristics of the N channels; and constructing a feature extractor by using the convolutional layer, and outputting a prediction result. The invention is provided withAnd extracting vibration signals of N dimensions, fusing the vibration signals, and increasing the weight of effective information by using an attention mechanism to ensure that the deep network has good robustness.
Description
Technical Field
The invention relates to the technical field of bearing fault diagnosis, in particular to a bearing fault diagnosis method, a bearing fault diagnosis system, storage equipment and equipment.
Background
In recent years, bearings have been widely used as important parts in smart manufacturing. Once damaged during the operation of the machine, the bearings cause the failure of the transmission part of the rotary machine, which in turn causes vibrations affecting the whole production line, with economic losses and safety risks. Therefore, the bearing fault diagnosis intelligent manufacturing industry plays a significant role, and a good fault diagnosis method can bring huge economic benefits and safety guarantee. Deep learning is used as a method for fusing feature extraction and classification, and an end-to-end solution is provided for bearing fault diagnosis.
Considering that the conventional fault diagnosis method only samples the vibration signal of one dimension, all effective information cannot be well utilized. Meanwhile, the conventional fault diagnosis method cannot well select information useful for tasks for given vibration signals, so that the diagnosis precision is not high.
Disclosure of Invention
The invention provides a bearing fault diagnosis method, a bearing fault diagnosis system, a storage medium and a bearing fault diagnosis device, which can solve the technical problem that the existing method cannot fully utilize effective information attached in a vibration signal.
In order to achieve the purpose, the invention adopts the following technical scheme:
a bearing fault diagnosis method is implemented by computer equipment,
s1: acquiring vibration signal data s of N sensors at fault occurrence position1,s2,…sNConstructing an attention module ATT, and increasing the weight of effective information;
s2: preprocessing data passing through an attention module, converting the acquired vibration signal data of N dimensions into a vibration image according to a set rule, and calculating a pixel value of the image;
s3: establishing a harmonic layer, performing characteristic fusion on the vibration signals of the N channels, and fully considering the signal characteristics of the N channels;
s4: and constructing a feature extractor by using the convolutional layer, and outputting a prediction result.
Further, the step S1 specifically includes the following steps S11 to S12:
s11: dividing an original signal into a plurality of segments with the length of M multiplied by N at random;
s12: constructing an attention mechanism ATT, and enabling the signal segments to pass through the attention mechanism; the attention mechanism is constructed as follows:
the signal received by each sensor forms a channel, assuming the input to the attention module is the channel combination a ═ a1,a2,a3,…,ac],(ai∈Rω×1) Carrying out normalization processing;
wherein E and Var represent expectation and variance, respectively; the attention module ATT compresses the global time information into channel descriptors by using the global average pool layer Avgpool and generates a channel statistics vector q, q ∈ R1×cAnd c is the number of input channels;
then, a channel recalibration vector q' is generated:
q′=σ(F″(δ(F′(q))))
wherein, δ is a ReLu activation function, F 'and F' respectively represent convolution operation with the number of channels being 1 and the convolution kernel being 1 × 1, and σ represents a Sigmoid function; q's'iThe value of (b) represents the importance of the ith channel;
m is the result of the importance reassignment of feature A, M1A new value representing the importance of the first channel after reassignment;
the final output is:
AATT=A+M。
further, the step S2 specifically includes the following steps S21 to S22:
s21: dividing the signal passing through the attention mechanism module to obtain a plurality of s (k, i), wherein k represents the number of channels, and the value is 1, 2 and 3; i represents the length of the signal, the length is M multiplied by N, and M multiplied by N is the length of the signal; and calculating S (k, i);
S(k,i)=s(k,i)·s(k,i)
s22: calculating a pixel value H (M, N) of the feature image, wherein M is 1, 2, …, 3M, and N is 1, 2, …, N;
wherein k takes the values:
further, the step S3 specifically includes the following steps S31 to S32:
s31: treating the reconciliation layer as a convolution operation with a 1 x 1 convolution kernel and an overall feature map with each feature map fused;
s32: the output of the reconciliation layer is used for N-channel information fusion; the convolution kernel is 1 xk, the step length is 1 xk, k represents the number of sensor channels, and in the network structure diagram, k is made to be 3, so that the data of N channels can be effectively fused; n channel data of the same time node in the reconciliation layer are merged, and data of different time nodes are not merged in the reconciliation layer.
Further, the step S4 specifically includes the following steps S41 to S43:
s41: establishing a convolutional neural network layer;
let x be an element of RdThe jth feature map may be described as:
cj=x*wj+bj
wherein, wjRepresents the jth filter, bjRepresents a deviation term;
s42: establishing a pooling layer; here, the maximum pooling layer is used, which is mathematically expressed as:
wherein, cjRepresenting the input, r is the pool size,represents the maximum value within the corresponding pooling area;
s43: establishing a full connection layer and inputting a result; the mathematical expression of the fully-connected layer is:
fcl=σ(wl·fcl-1+bl)
wherein fclRepresenting the output characteristic of the 1 st fully-connected layer, wlAnd blRespectively representing weight and deviation, and sigma represents a nonlinear activation function;
finally, the prediction result is output by using the decision layer:
p(z)irepresenting the probability of the bearing being in a failed state.
On the other hand, the invention also discloses a bearing fault diagnosis system, which comprises the following units:
an attention module construction unit for acquiring vibration signal data s of the N sensors at the fault occurrence position1,s2,…sNConstructing an attention module ATT, and increasing the weight of effective information;
the calculation unit is used for preprocessing the data passing through the attention module, converting the acquired vibration signal data of N dimensions into a vibration image according to a set rule and calculating a pixel value of the image;
the characteristic fusion unit is used for establishing a harmonic layer, performing characteristic fusion on the vibration signals of the N channels and fully considering the signal characteristics of the N channels;
and the result output unit is used for constructing the feature extractor by utilizing the convolution layer and outputting the prediction result.
In a second aspect, the invention also discloses a computer-readable storage medium, in which a computer program is stored, which, when executed by a processor, causes the processor to perform the steps of the method as described above.
In yet another aspect, the present invention also discloses a computer device comprising a memory and a processor, the memory storing a computer program, the computer program, when executed by the processor, causing the processor to perform the steps of the method as described above.
According to the technical scheme, the bearing fault diagnosis method achieves the aim of improving the fault diagnosis prediction precision by efficiently utilizing the effective information attached to the vibration signals, samples the vibration signals in multiple dimensions to utilize more effective information, and simultaneously adds an attention mechanism to extract essence and remove dregs from the vibration signals so as to fully utilize the effective information and improve the fault diagnosis precision.
The invention extracts and fuses the vibration signals of N dimensions, and increases the weight of effective information by using an attention mechanism, so that the deep network has good robustness.
Drawings
FIG. 1 is a flow chart of a method of the present invention;
FIG. 2 is a schematic diagram of a network model of the method of the present invention;
FIG. 3 is a schematic diagram of the internal structure of the ATT;
FIG. 4 is a schematic view of a reconciliation layer;
fig. 5 is a schematic diagram of feature visualization on a CWRU public data set according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention.
As shown in fig. 1 and fig. 2, the bearing fault diagnosis method according to the present embodiment includes the following steps:
step 1: acquiring vibration signal data s of N sensors at fault occurrence position1,s2,…sNAnd constructing an attention module ATT and increasing the weight of the effective information.
Step 2: and preprocessing the data passing through the attention module, converting the acquired vibration signal data of N dimensions into a vibration image according to a certain rule, and calculating the pixel value of the image.
And step 3: and establishing a harmonic layer, performing characteristic fusion on the vibration signals of the N channels, and fully considering the signal characteristics of the N channels.
And 4, step 4: and constructing a feature extractor by using the convolutional layer, and outputting a prediction result.
The method comprises the following specific steps:
further, the above step S1: acquiring vibration signal data s of N sensors at fault occurrence position1,s2,…sNAnd constructing an attention module and increasing the weight of the effective information.
The method specifically comprises the following subdivision steps S11-S12:
s11: the original signal is divided and randomly divided into a plurality of segments with the length of M multiplied by N.
S12: an attention mechanism ATT is constructed and the signal fragment is passed through the attention mechanism. The attention mechanism is constructed as follows:
the signal received by each sensor forms a channel, assuming the input to the attention module is the channel combination a ═ a1,a2,a3,…,ac],(ai∈Rω×1) And carrying out normalization processing.
Where E and Var represent expectation and variance, respectively. The attention module ATT compresses the global time information into channel descriptors by using the global average pool layer Avgpool and generates a channel statistics vector q (q ∈ R)1×cAnd c is the number of input channels).
Then, a channel recalibration vector q' is generated:
q′=σ(F″(δ(F′(q))))
wherein, δ is a ReLu activation function, F 'and F' respectively represent convolution operation with the number of channels being 1 and the convolution kernel being 1 × 1, and σ represents a Sigmoid function. q's'iThe value of (b) represents the importance of the ith channel.
M is the result of the importance reassignment of feature A, M1Indicating the new value of the first channel after the reassignment of the importance.
Finally, the invention utilizes the residual error network, introduces the residual connection, and improves the feasibility of optimization while keeping the original information. The final output is:
AATT=A+M;
the structure of the ATT module is shown in fig. 3 below.
Further, the above step S2: and preprocessing the data passing through the attention module, converting the acquired vibration signal data of N dimensions into a vibration image according to a certain rule, and calculating the pixel value of the image. The method specifically comprises the following subdivision steps S21-S22:
s21: dividing the signal passing through the attention mechanism module to obtain a plurality of s (k, i), wherein k represents the number of channels, and the value is 1, 2 and 3; i represents the length of the signal, the length being M × N, M × N being the length of the signal. And calculates S (k, i).
S(k,i)=s(k,i)·s(k,i)
S22: and calculating a pixel value H (M, N) of the characteristic image, wherein M is 1, 2, …, 3M, and N is 1, 2, …, N.
Wherein k takes the values:
further, the above step S3: and establishing a harmonic layer, performing characteristic fusion on the vibration signals of the N channels, and fully considering the signal characteristics of the N channels. The method specifically comprises the following subdivision steps S31-S32:
s31: the reconciliation layer can be viewed as a convolution operation with a 1 x 1 convolution kernel and an overall feature map that fuses each feature map.
S32: the output of the reconciliation layer may be used for N-channel information fusion. The convolution kernel is 1 × k, the step size is 1 × k, and k represents the number of sensor channels (in the network structure diagram, k is 3), so that the data of N channels can be effectively fused. N channel data of the same time node in the reconciliation layer are fused, and data of different time nodes are not fused in the reconciliation layer; an example of a reconciliation layer is shown below in FIG. 4.
Further, the above step S4: and constructing a feature extractor by using the convolutional layer, and outputting a prediction result. The method specifically comprises the following subdivision steps S41-S43:
s41: and establishing a convolutional neural network layer. Let x be an element of RdThe jth feature map may be described as:
cj=x*wj+bj
wherein, wjRepresents the jth filter, bjRepresenting the deviation term.
S42: and establishing a pooling layer. Here, the maximum pooling layer is used, which is mathematically expressed as:
wherein, cjRepresenting the input, r is the pool size,representing the maximum value within the corresponding pooling area.
S43: and establishing a full connection layer and inputting a result. The mathematical expression of the fully-connected layer is:
fcl=σ(wl·fcl-1+bl)
wherein fclRepresenting the output characteristic of the 1 st fully-connected layer, wlAnd blRespectively, weight and deviation, and sigma represents a nonlinear activation function.
And finally, outputting the prediction result by utilizing the decision layer.
p(z)iRepresenting the probability of the bearing being in a failed state.
The method of the invention has good performance on a CWRU bearing test data set, and the following figure 5 shows a characteristic visualization schematic diagram of the method on a CWRU public data set, so that the method can well divide normal bearings and bearings with different types of faults. The numbers 0-9 represent normal data and inner ring, outer ring, ball failure at different speeds, respectively, only three of which are labeled in the legend.
On the other hand, the invention also discloses a bearing fault diagnosis system, which comprises the following units:
an attention module construction unit for acquiring vibration signal data s of the N sensors at the fault occurrence position1,s2,…sNConstructing an attention module ATT, and increasing the weight of effective information;
the calculation unit is used for preprocessing the data passing through the attention module, converting the acquired vibration signal data of N dimensions into a vibration image according to a set rule and calculating a pixel value of the image;
the characteristic fusion unit is used for establishing a harmonic layer, performing characteristic fusion on the vibration signals of the N channels and fully considering the signal characteristics of the N channels;
and the result output unit is used for constructing the feature extractor by utilizing the convolution layer and outputting the prediction result.
It is understood that the system provided by the embodiment of the present invention corresponds to the method provided by the embodiment of the present invention, and the explanation, the example and the beneficial effects of the related contents can refer to the corresponding parts in the method.
In a second aspect, the invention also discloses a computer-readable storage medium, in which a computer program is stored, which, when executed by a processor, causes the processor to perform the steps of the method as described above.
In yet another aspect, the present invention also discloses a computer device comprising a memory and a processor, the memory storing a computer program, the computer program, when executed by the processor, causing the processor to perform the steps of the method as described above.
The embodiment of the application also provides an electronic device, which comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete mutual communication through the communication bus,
a memory for storing a computer program;
the processor is used for realizing the bearing fault diagnosis method when executing the program stored in the memory, and the method comprises the following steps:
s1: acquiring vibration signal data s _1, s _2 and … s _ N of N sensors at a fault occurrence position, constructing an attention module ATT, and increasing the weight of effective information;
s2: preprocessing data passing through an attention module, converting the acquired vibration signal data of N dimensions into a vibration image according to a set rule, and calculating a pixel value of the image;
s3: establishing a harmonic layer, performing characteristic fusion on the vibration signals of the N channels, and fully considering the signal characteristics of the N channels;
s4: and constructing a feature extractor by using the convolutional layer, and outputting a prediction result.
The communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus may be divided into an address bus, a data bus, a control bus, etc.
The communication interface is used for communication between the electronic equipment and other equipment.
The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), or other Programmable logic devices, discrete Gate or transistor logic devices, or discrete hardware components.
In yet another embodiment provided by the present application, a computer-readable storage medium is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any of the above-mentioned bearing fault diagnosis methods.
In yet another embodiment provided by the present application, there is also provided a computer program product containing instructions which, when run on a computer, cause the computer to perform any of the bearing fault diagnosis methods of the above embodiments.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the application to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. A bearing fault diagnosis method is characterized in that: the following steps are performed by a computer device,
s1: acquiring vibration signal data s of N sensors at fault occurrence position1,s2,...sNConstructing an attention module ATT, and increasing the weight of effective information;
s2: preprocessing data passing through an attention module, converting the acquired vibration signal data of N dimensions into a vibration image according to a set rule, and calculating a pixel value of the image;
s3: establishing a harmonic layer, performing characteristic fusion on the vibration signals of the N channels, and fully considering the signal characteristics of the N channels;
s4: and constructing a feature extractor by using the convolutional layer, and outputting a prediction result.
2. The bearing fault diagnosis method according to claim 1, characterized in that: the step S1 specifically includes the following subdivided steps S11 to S12:
s11: dividing an original signal into a plurality of segments with the length of M multiplied by N at random;
s12: constructing an attention mechanism ATT, and enabling the signal segments to pass through the attention mechanism; the attention mechanism is constructed as follows:
the signal received by each sensor forms a channel, assuming the input to the attention module is the channel combination a ═ a1,a2,a3,...,ac],(ai∈Rω×1) Carrying out normalization processing;
wherein E and Nar represent expectation and variance, respectively; the attention module ATT compresses the global time information into channel descriptors by using the global average pool layer Avgpool and generates a channel statistics vector q, q ∈ R1×cAnd c is the number of input channels;
then, a channel recalibration vector q' is generated:
q′=σ(F″(δ(F′(q))))
wherein δ is the ReLu activation function, and F' respectively represent the number of channels as 1And convolution operation with convolution kernel of 1 × 1, where σ represents Sigmoid function; q's'iThe value of (b) represents the importance of the ith channel;
m is the result of the importance re-assignment of the characteristic a, and M1 represents the new value of the first channel after the importance re-assignment;
the final output is:
AATT=A+M。
3. the bearing fault diagnosis method according to claim 2, characterized in that: the step S2 specifically includes the following steps S21 to S22:
s21: dividing the signal passing through the attention mechanism module to obtain a plurality of s (k, i), wherein k represents the number of channels, and the value is 1, 2 and 3; i represents the length of the signal, the length is M multiplied by N, and M multiplied by N is the length of the signal; and calculating S (k, i);
S(k,i)=s(k,i)·s(k,i)
s22: calculating pixel values H (M, N) of the feature image, wherein M is 1, 2,. multidot.3M, N is 1, 2,. multidot.n;
wherein k takes the values:
4. the bearing fault diagnosis method according to claim 3, characterized in that: the step S3 specifically includes the following subdivided steps S31 to S32:
s31: treating the reconciliation layer as a convolution operation with a 1 x 1 convolution kernel and an overall feature map with each feature map fused;
s32: the output of the reconciliation layer is used for N-channel information fusion; the convolution kernel is 1 xk, the step length is 1 xk, k represents the number of sensor channels, and in the network structure diagram, k is made to be 3, so that the data of N channels can be effectively fused; n channel data of the same time node in the reconciliation layer are merged, and data of different time nodes are not merged in the reconciliation layer.
5. The bearing fault diagnosis method according to claim 4, characterized in that: the step S4 specifically includes the following steps S41 to S43:
s41: establishing a convolutional neural network layer;
let x be an element of RdThe jth feature map may be described as:
cj=x*wj+bj
wherein, wjRepresents the jth filter, bjRepresents a deviation term;
s42: establishing a pooling layer; here, the maximum pooling layer is used, which is mathematically expressed as:
wherein, cjRepresenting the input, r is the pool size,represents the maximum value within the corresponding pooling area;
s43: establishing a full connection layer and inputting a result; the mathematical expression of the fully-connected layer is:
fcl=σ(wl·fcl-1+bl)
wherein fclRepresenting the output characteristic of the 1 st fully-connected layer, wlAnd blRespectively representing weight and deviation, and sigma represents a nonlinear activation function;
finally, the prediction result is output by using the decision layer:
p(z)irepresenting the probability of the bearing being in a failed state.
6. A bearing fault diagnostic system characterized by: the method comprises the following units:
an attention module construction unit for acquiring vibration signal data s of the N sensors at the fault occurrence position1,s2,...sNConstructing an attention module ATT, and increasing the weight of effective information;
the calculation unit is used for preprocessing the data passing through the attention module, converting the acquired vibration signal data of N dimensions into a vibration image according to a set rule and calculating a pixel value of the image;
the characteristic fusion unit is used for establishing a harmonic layer, performing characteristic fusion on the vibration signals of the N channels and fully considering the signal characteristics of the N channels;
and the result output unit is used for constructing the feature extractor by utilizing the convolution layer and outputting the prediction result.
7. A computer-readable storage medium, storing a computer program which, when executed by a processor, causes the processor to carry out the steps of the method according to any one of claims 1 to 5.
8. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method according to any one of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111340215.4A CN114266269A (en) | 2021-11-12 | 2021-11-12 | Bearing fault diagnosis method and system, storage medium and equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111340215.4A CN114266269A (en) | 2021-11-12 | 2021-11-12 | Bearing fault diagnosis method and system, storage medium and equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114266269A true CN114266269A (en) | 2022-04-01 |
Family
ID=80825187
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111340215.4A Pending CN114266269A (en) | 2021-11-12 | 2021-11-12 | Bearing fault diagnosis method and system, storage medium and equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114266269A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114993677A (en) * | 2022-05-11 | 2022-09-02 | 山东大学 | Rolling bearing fault diagnosis method and system based on unbalanced small sample data |
-
2021
- 2021-11-12 CN CN202111340215.4A patent/CN114266269A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114993677A (en) * | 2022-05-11 | 2022-09-02 | 山东大学 | Rolling bearing fault diagnosis method and system based on unbalanced small sample data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021204010A1 (en) | Time series anomaly detection method and apparatus, and computer device and storage medium | |
WO2022227388A1 (en) | Log anomaly detection model training method, apparatus and device | |
CN106874253A (en) | Recognize the method and device of sensitive information | |
CN109492104B (en) | Training method, classification method, system, device and medium of intention classification model | |
US20180365594A1 (en) | Systems and methods for generative learning | |
US20230117973A1 (en) | Data processing method and apparatus | |
CN111553215A (en) | Personnel association method and device, and graph convolution network training method and device | |
CN113761219A (en) | Knowledge graph-based retrieval method and device, electronic equipment and storage medium | |
CN114143049A (en) | Abnormal flow detection method, abnormal flow detection device, storage medium and electronic equipment | |
CN111062431A (en) | Image clustering method, image clustering device, electronic device, and storage medium | |
US20240135698A1 (en) | Image classification method, model training method, device, storage medium, and computer program | |
CN114266269A (en) | Bearing fault diagnosis method and system, storage medium and equipment | |
CN115907970A (en) | Credit risk identification method and device, electronic equipment and storage medium | |
Jin et al. | An unsupervised multi‐focus image fusion method based on Transformer and U‐Net | |
Sun et al. | Neighborhood graph embedding interpretable fault diagnosis network based on local and non-local information balanced under imbalanced samples | |
CN112257959A (en) | User risk prediction method and device, electronic equipment and storage medium | |
CN116739154A (en) | Fault prediction method and related equipment thereof | |
CN111079930A (en) | Method and device for determining quality parameters of data set and electronic equipment | |
CN108400887B (en) | Method and device for identifying structure of symbol network | |
Sun et al. | Randomized nonlinear two-dimensional principal component analysis network for object recognition | |
CN114997327A (en) | Target object classification method and device, storage medium and electronic equipment | |
CN115409262A (en) | Railway data center key performance index trend prediction method and abnormity identification method | |
CN114723019A (en) | Convolution calculation method and device based on photon calculation chip | |
CN114510609A (en) | Method, device, equipment, medium and program product for generating structure data | |
CN114397306A (en) | Power grid grading ring hypercomplex category defect multi-stage model joint detection method |
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 |