CN114429152A - Rolling bearing fault diagnosis method based on dynamic index antagonism self-adaption - Google Patents
Rolling bearing fault diagnosis method based on dynamic index antagonism self-adaption Download PDFInfo
- Publication number
- CN114429152A CN114429152A CN202111677762.1A CN202111677762A CN114429152A CN 114429152 A CN114429152 A CN 114429152A CN 202111677762 A CN202111677762 A CN 202111677762A CN 114429152 A CN114429152 A CN 114429152A
- Authority
- CN
- China
- Prior art keywords
- domain
- sample
- fault diagnosis
- loss
- discriminator
- 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 103
- 238000000034 method Methods 0.000 title claims abstract description 66
- 238000005096 rolling process Methods 0.000 title claims abstract description 27
- 230000008485 antagonism Effects 0.000 title claims abstract description 21
- 238000009826 distribution Methods 0.000 claims abstract description 104
- 238000012549 training Methods 0.000 claims abstract description 41
- 230000008569 process Effects 0.000 claims abstract description 12
- 230000003042 antagnostic effect Effects 0.000 claims abstract description 7
- 238000005259 measurement Methods 0.000 claims description 35
- 238000004364 calculation method Methods 0.000 claims description 18
- 230000006870 function Effects 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 11
- 238000012360 testing method Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 9
- 230000006978 adaptation Effects 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000011524 similarity measure Methods 0.000 claims description 6
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims description 4
- 230000001133 acceleration Effects 0.000 claims description 4
- 230000003862 health status Effects 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 238000012546 transfer Methods 0.000 abstract description 2
- 238000013508 migration Methods 0.000 description 17
- 230000005012 migration Effects 0.000 description 17
- 230000000694 effects Effects 0.000 description 9
- 230000006872 improvement Effects 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 6
- 238000013526 transfer learning Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003889 chemical engineering Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
- G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Molecular Biology (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Signal Processing (AREA)
- Probability & Statistics with Applications (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a rolling bearing fault diagnosis method based on dynamic index antagonism self-adaptation, which comprises the following steps of: collecting vibration data of a bearing in operation under different working conditions; taking the source domain characteristics and the mixed domain sample characteristics as input, training a classifier and a domain discriminator in an antagonistic way, optimizing a characteristic extractor, and calculating loss; constructing a target function of a bearing fault diagnosis model by using loss, searching for optimal parameters until the bearing fault diagnosis model is completed, and reducing the edge distribution and condition distribution difference of a source domain sample and a target domain sample by using a dynamic index adjustment factor in a training process; and inputting the target domain sample into a bearing fault diagnosis model, and outputting a bearing fault diagnosis result. The invention can accurately and quantitatively measure the proportion of the edge distribution and the condition distribution in the whole data distribution, thereby enabling the model to more pointedly transfer the data sets under different working conditions and realizing accurate fault diagnosis.
Description
Technical Field
The invention relates to the technical field of mechanical fault diagnosis, in particular to a rolling bearing fault diagnosis method based on dynamic index antagonism self-adaptation.
Background
With the development of industry, more and more rotating mechanical machines are used in production and life. Rolling bearings are one of the most important key components in rotating machines, and their state is directly related to whether such rotating machines can operate normally. The fault diagnosis is a comprehensive technology and is an important measure for ensuring the safe and reliable operation of mechanical equipment. Therefore, the diagnosis of the rolling bearing fault, especially the early fault analysis, and the realization of the rapid and accurate bearing fault monitoring are of great significance to the normal work and the safe production of mechanical equipment. The traditional fault diagnosis method, such as time domain statistical analysis, wavelet transformation, sparse representation and Fourier spectrum analysis, can realize accurate fault diagnosis. However, conventional fault diagnosis requires extensive experience and extensive prior knowledge of engineers. For example, in a bearing fault diagnosis method based on wavelet denoising, an appropriate wavelet basis must be manually selected. To overcome this limitation, coupled with the advent and continued advancement of artificial intelligence techniques, many researchers have turned their attention to intelligent fault diagnosis techniques.
The intelligent fault diagnosis is the application of machine learning theories such as artificial neural networks, support vector machines and deep neural networks in machine fault diagnosis. Among them, the deep neural network is widely used due to its excellent performance in feature extraction. However, the accuracy of fault diagnosis based on deep neural networks depends to a large extent on the number of samples involved in training, but in actual practice, the fault signals collected by rolling bearings at different speeds and loads are also different. While a full signature fault signal may be obtained in the laboratory under some conditions, it is not practical to obtain a large number of signature fault samples covering all conditions when the bearing is operating under variable conditions. Therefore, it has become a new challenge to diagnose data different from its operational status using existing flagged fault data. In this context, transfer learning becomes a new solution.
The migration learning can utilize similarities between data, tasks, or models to apply models and knowledge learned from old domains to new domains. The field adaptation is an important research direction of the transfer learning, aims at scenes with the same task in different fields, and is a direct push transfer learning. Data processed by the unsupervised field self-adaption problem does not contain a target domain label, so how to align data distribution of a source domain and a target domain so as to realize fault diagnosis of target domain unlabeled data by means of source domain labeled data becomes a new difficulty.
Most of the existing fault diagnosis methods focus on edge distribution of alignment features, and neglect alignment of distribution in class, namely condition distribution. These methods assume that the conditional distributions are automatically aligned during the alignment of the edge distributions. This assumption is not valid and has a negative effect on the model. A few methods that consider conditional distributions, such as the joint distribution alignment method, assume that edge distribution alignment and conditional distribution alignment have the same weight in overall data distribution alignment, which obviously cannot be applied to all data distributions.
Disclosure of Invention
The invention aims to provide a rolling bearing fault diagnosis method based on dynamic index antagonism self-adaptation, which can accurately and quantitatively measure the proportion of edge distribution and condition distribution in overall data distribution, so that a model can more specifically transfer data sets under different working conditions, and accurate fault diagnosis is realized.
In order to solve the technical problem, the invention provides a rolling bearing fault diagnosis method based on dynamic index antagonism self-adaptation, which comprises the following steps:
s1: acquiring vibration data of a bearing in operation under different working conditions to obtain a source domain sample and a target domain sample;
s2: inputting the source domain samples into a feature extractor to obtain source domain features; inputting the source domain sample and the target domain sample into a feature extractor together to obtain mixed domain sample features;
s3: taking the source domain characteristics and the mixed domain sample characteristics as input, training a classifier and a domain discriminator in an antagonistic way, optimizing a characteristic extractor, and calculating the loss of the classifier and the domain discriminator; wherein, the domain discriminator comprises a global domain discriminator and a local domain discriminator;
s4: passing similarity measures for global discriminator loss and local discriminator loss
Calculating global measurement and local measurement of distribution distance between a source domain sample and a target domain sample to obtain an index dynamic adjustment factor, and redefining the loss of the domain discriminator by using the index dynamic adjustment factor, wherein the global measurement and the local measurement respectively correspond to edge distribution and condition distribution difference;
s5: constructing an objective function of a bearing fault diagnosis model by using classifier loss and redefined domain discriminator loss, and searching optimal parameters of the objective function through training of a source domain sample with a label and a target domain sample without a label until the bearing fault diagnosis model is completed, wherein the edge distribution and condition distribution difference of the source domain sample and the target domain sample is reduced by using a dynamic index adjustment factor in the training process;
s6: and inputting the target domain sample into the finished bearing fault diagnosis model, and outputting a bearing fault diagnosis result.
As a further improvement of the present invention, in step S1, the bearing health status is different for each condition, the bearing vibration data of the different health status for each condition is used as a transferable data field, the data field is attached with a field label, the source field sample and the target field sample are selected from the data field, and the source field sample is attached with a fault type label.
As a further improvement of the method, an acceleration sensor is used for collecting vibration signals of the bearing in operation under each working condition, a source domain data set and a target domain data set are constructed, short-time Fourier transform is used for processing the source domain data set and the target domain data set, two-dimensional processing is carried out, and processed multi-source domain samples and processed target domain samples are output.
As a further improvement of the present invention, in said step S3, the source domain features are inputted into the classifier for supervised training to obtain the predicted source domain label and the classifier loss, wherein the supervised training is to obtain the classifier loss by calculating the cross entropy loss of the predicted source domain label and the source domain true label
yiIs a data true failure tag, C (G (x)i) Is a fault label predicted by the classifier, LcThe cross entropy loss of the two is calculated.
As a further improvement of the present invention, in the step S3:
inputting the mixed domain sample into a global area discriminator for training to obtain a predicted domain label and a predicted global area loss
Wherein d iskDomain tags for data trueness, Dg(G(xk) Represents a predicted domain label, LDgCalculating the cross entropy loss of the two;
inputting the characteristics of the mixed samples into a classifier to obtain the probability distribution of the target domain fault category prediction labels; inputting the characteristics of the mixed samples into a plurality of local domain discriminators to obtain a plurality of domain discrimination prediction labels, calculating cross entropy loss of each domain with a real domain label, and multiplying and summing the cross entropy loss and the probability distribution of the target domain fault type prediction label to obtain a final local domain loss calculation formula:
calculating local area loss, wherein H represents the number of fault label types,
is a domain discriminator associated with class h,
is that
The corresponding cross-entropy loss is then taken into account,
representing the probability distribution of the presence of the kth sample over the h class, dkIs a domain label of the data reality.
As a further improvement of the present invention, the step S4 specifically includes:
for global and local losses, passing through
Calculating, i.e. using global difference metric formulae
And local variance measure
Calculating global measurement and local measurement of distribution distance between two domains;
is converted into an exponential dynamic adjustment factor omega expressed as
Adjusting the difference between the edge distribution and the condition distribution of the two domains by using an index dynamic adjustment factor;
considering the exponential dynamics adjustment factor, the final domain discriminator loss is defined as:
as a further improvement of the present invention, the step S5 specifically includes:
according to classifier loss LySum-field discriminator loss LDEstablishing an objective function of a bearing fault diagnosis model, namely calculating the total loss, wherein the proportional change of the two follows a formula
I.e. the final total loss calculation is given by the formula L ═ Ly-λLD；
Calculating the total loss once in each epoch, and optimizing the established feature extractor, classifier, global domain discriminator and local domain discriminator by using the Adam algorithm for the total loss;
determining the number of times of model training by using a predefined epoch number, and according to a predefined step length and a step length attenuation formula
And obtaining the step length of each step, and circulating the whole model for a plurality of times according to the step length to obtain the trained model.
As a further improvement of the present invention, the step S6 specifically includes:
processing the target domain data set by using short-time Fourier transform to obtain a target domain sample picture;
and inputting the target domain sample picture into the trained fault diagnosis model to obtain a final fault diagnosis result.
A rolling bearing fault diagnosis system based on dynamic index antagonism self-adaptation comprises:
the acquisition module is used for acquiring vibration data of the bearing during operation under different working conditions to obtain a source domain sample and a target domain sample;
the characteristic extraction module is used for inputting the source domain samples into the characteristic extractor to obtain source domain characteristics; inputting the source domain sample and the target domain sample into a feature extractor together to obtain mixed domain sample features;
the classification calculation module is used for taking the source domain characteristics and the mixed domain sample characteristics as input, training a classifier and a domain discriminator in an antagonistic way, optimizing a characteristic extractor and calculating the loss of the classifier and the domain discriminator; wherein, the domain discriminator comprises a global domain discriminator and a local domain discriminator;
an exponential dynamic module for passing similarity measures for global discriminator loss and local discriminator loss
Calculating global measurement and local measurement of distribution distance between a source domain sample and a target domain sample to obtain an index dynamic adjustment factor, and redefining the loss of the domain discriminator by using the index dynamic adjustment factor, wherein the global measurement and the local measurement respectively correspond to edge distribution and condition distribution difference;
the training module is used for constructing an objective function of the bearing fault diagnosis model by using classifier loss and redefined domain discriminator loss, searching the optimal parameters of the objective function through the training of a source domain sample with a label and a target domain sample without a label until the bearing fault diagnosis model is completed, wherein the edge distribution and condition distribution difference of the source domain sample and the target domain sample are reduced by using a dynamic index adjustment factor in the training process;
and the test module is used for inputting the target domain sample into the finished bearing fault diagnosis model and outputting a bearing fault diagnosis result.
As a further improvement of the present invention, the classification calculation module includes:
the classification module is used for obtaining a prediction label by utilizing the characteristics obtained by the characteristic extraction module so as to classify fault types and calculating cross entropy loss with a real label;
the global area identification module is used for carrying out domain classification on the samples by utilizing the mixed sample characteristics to obtain a prediction label of each sample characteristic, and calculating the cross entropy loss of the global area identifier by utilizing the prediction label and the real label of each sample characteristic;
and the local area identification module is used for carrying out domain classification on the samples by utilizing the mixed sample characteristics to obtain a prediction label of each sample characteristic, classifying all the sample characteristics by utilizing the classifier to obtain the pseudo label probability distribution of the target domain sample, and calculating the cross entropy loss of the local area identifier by utilizing the prediction label of each sample characteristic, the real label and the pseudo label probability distribution.
The invention has the beneficial effects that: the invention can autonomously screen the characteristics more suitable for migration through the countermeasure among the characteristic extractor, the classifier and the domain discriminator to obtain a better fault diagnosis result; the method can measure the distribution difference between the source domain and the target domain better by measuring the distribution difference between the source domain and the target domain indirectly by loss without explicitly specifying the distance measurement between the source domain and the target domain, thereby obtaining better migration effect; the method utilizes the index dynamic factor to stably, quantitatively and dynamically adjust the proportion of the edge distribution and the condition distribution difference in the overall data distribution difference, and the index function can effectively reduce data collapse caused by calculation defects (except zero) in the quantitative calculation process, so that the model is more stable; the dynamic quantitative adjustment can enable the model to be better aligned with the data distribution between the source domain and the target domain, so that the model still has a good diagnosis effect when facing a variable working condition task, and can be widely applied to fault diagnosis tasks of complex systems of machinery, electricians, chemical engineering, aviation and the like under variable working conditions.
Drawings
FIG. 1 is a flow chart of a method of bearing fault diagnosis of the present invention;
FIG. 2 is a flow chart of a method embodying the present invention;
FIG. 3 is a test chart of a rolling bearing data generation test stand according to an embodiment of the present invention;
FIG. 4 is a fault diagnosis confusion matrix under a migration task according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating visualization of fault diagnosis features in a migration task according to an embodiment of the present invention;
FIG. 6 is a block diagram illustrating the system architecture of the present invention.
Detailed Description
The present invention is further described below in conjunction with the following figures and specific examples so that those skilled in the art may better understand the present invention and practice it, but the examples are not intended to limit the present invention.
Referring to fig. 1, the invention provides a rolling bearing fault diagnosis method based on dynamic index antagonism self-adaptation, comprising the following steps:
s1: acquiring vibration data of a bearing in operation under different working conditions to obtain a source domain sample and a target domain sample;
s2: inputting the source domain samples into a feature extractor to obtain source domain features; inputting the source domain sample and the target domain sample into a feature extractor together to obtain mixed domain sample features;
s3: taking the source domain characteristics and the mixed domain sample characteristics as input, training a classifier and a domain discriminator in an antagonistic way, optimizing a characteristic extractor, and calculating the loss of the classifier and the domain discriminator; wherein, the domain discriminator comprises a global domain discriminator and a local domain discriminator;
s4: passing similarity measures for global discriminator loss and local discriminator loss
Calculating global measurement and local measurement of distribution distance between a source domain sample and a target domain sample to obtain an index dynamic adjustment factor, and redefining the loss of the domain discriminator by using the index dynamic adjustment factor, wherein the global measurement and the local measurement respectively correspond to edge distribution and condition distribution difference;
s5: constructing an objective function of a bearing fault diagnosis model by using classifier loss and redefined domain discriminator loss, and searching optimal parameters of the objective function through training of a source domain sample with a label and a target domain sample without a label until the bearing fault diagnosis model is completed, wherein the edge distribution and condition distribution difference of the source domain sample and the target domain sample is reduced by using a dynamic index adjustment factor in the training process;
s6: and inputting the target domain sample into the finished bearing fault diagnosis model, and outputting a bearing fault diagnosis result.
The method of the invention uses an acceleration sensor to collect vibration signals of a bearing in operation under each working condition, constructs a source domain data set and a target domain data set, outputs a source domain sample and a target domain sample by short-time Fourier transform and two-dimensional processing, and can be helpful for better extracting required characteristics by a model; the method can autonomously screen the characteristics more suitable for migration through the countermeasures among the characteristic extractor, the classifier and the domain discriminator, and obtain a better fault diagnosis result. The countermeasure self-adaptation can not explicitly specify the distance measurement between the source domain and the target domain, and can better measure the distribution difference between the source domain and the target domain by utilizing the loss indirect measurement to measure the distribution difference between the source domain and the target domain, thereby obtaining better migration effect; the proportion of the edge distribution and the condition distribution difference in the whole data distribution difference is stably, quantitatively and dynamically adjusted by using the index dynamic factor, and the index function can effectively reduce data collapse caused by calculation defects (except zero) in the quantitative calculation process, so that the model is more stable; the dynamic quantitative adjustment can enable the model to better align the data distribution between the source domain and the target domain, so that the model still has good diagnosis effect when facing a variable working condition task
Examples
In this embodiment, based on the above method, a more detailed description is given by referring to an embodiment, a transferable data set including seven healthy bearings under six different conditions is constructed from the collected vibration signals, and a bearing fault model is trained, with reference to fig. 2, the specific operation steps are as follows:
step S101: acquiring vibration signals of a bearing in operation under six working conditions by using an acceleration sensor, and constructing a source domain data set and a target domain data set;
the signals collected by the test bed shown in fig. 3 are used to construct a transferable data set of the bearing containing seven health states under six different working conditions, and each data set has different condition distribution and edge distribution. The test bearing was set to three single faults (inner ring fault, roller fault and outer ring fault) and four compound faults (inner ring + outer ring fault, inner ring + roller fault, roller + outer ring fault, roller + inner ring + outer ring fault). Thus, seven health states were obtained, as listed in table 1. There are 200 training samples and 200 test samples per state of health. Each sample consists of 1024 sampling points.
Table 1 seven kinds of faulty bearing information for each domain:
according to six different working conditions, specifically set up as table 2, 12 different migration tasks were established, and the migration tasks and their abbreviations are provided in table 3.
Table 2 migration task condition settings:
table 3 all migration tasks and abbreviations:
step S102: outputting a source domain sample and a target domain sample by utilizing short-time Fourier transform and two-dimensional processing;
a Short Time Fourier Transform (STFT) is performed on the samples. The STFT samples are converted into time-frequency domain signals from time-domain signals, the converted signals simultaneously contain time-domain information and frequency-domain information, and abundant information can help the model to better perform fault diagnosis.
The generated samples are reshaped from 3 × 1024 to 3 × 32 × 32, and for the source domain, a fault class label is appended according to the fault class to which it belongs. Adding a domain label to all data sets containing the fault category label;
step S103: inputting the source domain samples into a feature extractor and a classifier, and training the feature extractor by using a supervised method; inputting the target domain sample and the source domain sample into the feature extractor together to obtain mixed sample features;
the feature extractor is constructed according to the parameters of table 4, and is essentially a deep residual network consisting essentially of a plurality of rolling blocks, pooling layers, and a fully-connected neural network.
Table 4 structural parameters of the feature learner:
and taking the source domain sample as an input of a feature extractor, and obtaining a source domain feature representation through the feature extractor.
And constructing a classifier, wherein the classifier is essentially a neural network consisting of a full connection layer and a softmax layer. The supervised method trains the feature extractor, namely, the classifier loss is obtained by calculating the cross entropy loss of the predicted source domain label and the source domain real label
Wherein y isiIs a data true failure tag, C (G (x)i) Is a fault label predicted by the classifier, LcThe cross entropy loss of the two is calculated.
And inputting the target domain sample and the source domain sample into the feature extractor together to obtain mixed sample features.
Step S104, using a global area classifier and a local area classifier to take the mixed sample characteristics as the input of the global area classifier and the local area classifier, using a supervised method to carry out countermeasure training on the characteristic extractor and the domain discriminator, and outputting a prediction domain label and a prediction domain loss;
computing in accordance with global loss using a global classifier, the mixed sample features and the domain labelsFormula (II)
The global area loss is calculated.
Wherein d iskDomain tags for data trueness, Dg(G(xk) Is) represents a predicted domain label,
the cross entropy loss of the two is calculated.
Inputting the mixed sample characteristics into the classifier to obtain the probability distribution of the target domain fault category prediction label; inputting the mixed sample characteristics into the local domain discriminators to obtain a plurality of domain discrimination prediction labels, calculating cross entropy loss of each domain with a real domain label, multiplying and summing the cross entropy loss and the probability distribution of the target domain fault type prediction label to obtain a final local domain loss calculation formula
Local area loss is calculated.
Where H represents the number of faulty label categories,
is a domain discriminator associated with class h,
is that
The corresponding cross-entropy loss is then taken into account,
representing the probability distribution of the presence of the kth sample over the h class, dkIs a domain label of the data reality.
S105, adjusting the global loss and the local loss by using the index dynamic factor so as to adjust the influence of edge distribution and condition distribution on the overall training of the model and realize the adjustment of the domain loss;
for the global and local losses, passing
Calculating, in particular using global difference metric formulae
And local variance measure
A global metric and a local metric for the distribution distance between the two domains are calculated.
By using an exponential dynamic adjustment factor omega, it can be expressed as
Because the edge distribution difference can be indirectly measured by using the global distance measurement, and the condition distribution difference can be indirectly measured by using the local distance measurement, when the influence of the difference between the edge distribution and the condition distribution of the two domains on the model parameters is adjusted, the importance of the two distributions under the condition of inputting data can be adjusted by using the index dynamic adjustment factor.
Considering the exponential dynamics adjustment factor, the final domain discriminator loss may be defined as
S106, constructing an objective function of the bearing fault diagnosis model by using classifier loss and domain loss, reducing the edge distribution and condition distribution difference of the source domain sample characteristics and the target domain sample characteristics by using an antagonistic adaptive training strategy, optimizing the model by using an Adam algorithm, and iterating according to a preset epoch number and step length until the bearing fault diagnosis model is trained;
according to the classification loss LyAnd said domain discriminator loss LDCalculating the total loss, and the proportion change of the two follows the formula
I.e. the final total loss calculation is given by the formula L ═ Ly-λLD. And calculating the total loss once every epoch, and optimizing the built feature extractor, classifier, global domain discriminator and local domain discriminator by using the Adam algorithm. In the optimization process, because the domain discriminator is opposite to the optimization targets of the feature extractor and the classifier, the model can simultaneously complete opposite training purposes in one training by utilizing the gradient inversion layer.
Determining the number of model training times by using the predefined epoch number of 100, and according to the predefined step length of 0.001 and the step length attenuation formula
And obtaining the step size of each step. And circulating the whole model for 100 times according to the step length to obtain the trained model.
Step S107, inputting the target domain data set into the bearing fault diagnosis model, matching fault types of the fault diagnosis characteristics, and outputting a bearing fault diagnosis result of the target domain data set;
processing the target domain data set by using the short-time Fourier transform to obtain a target domain sample picture;
and inputting the target domain sample picture into the trained fault diagnosis model to obtain a final fault diagnosis result.
Table 5 fault diagnosis accuracy for the inventive method and variants under each migration task:
| migration tasks | Method for producing a composite material | Variant 1 | Variant 2 | Variant 3 |
| D1-> |
100 | 100 | 100 | 100 |
| D2-> |
100 | 99.78 | 100 | 100 |
| D1-> |
100 | 97.21 | 100 | 100 |
| D3-> |
100 | 91.43 | 88.64 | 100 |
| D1-> |
100 | 100 | 100 | 99.93 |
| D4-> |
100 | 99.78 | 100 | 100 |
| D1-> |
100 | 100 | 100 | 85.36 |
| D5-> |
100 | 66.86 | 99.93 | 100 |
| D1-> |
100 | 100 | 68 | 100 |
| D6-> |
100 | 82.64 | 100 | 100 |
| Average rate of |
100 | 93.77 | 95.66 | 98.53 |
Fig. 4, fig. 5 and table 5 show the fault diagnosis confusion matrix, the fault diagnosis feature visualization condition and the fault diagnosis accuracy of each variant method under a certain migration task, respectively. Through experimental data verification, the rolling bearing fault diagnosis method based on dynamic index antagonism self-adaptation carries out fault diagnosis according to the process, under the data condition of 1400 source domain samples and 1400 target domain samples and under the condition of load change, rotating speed change and load rotating speed change, the method can reach 100% of fault diagnosis accuracy after 100 iterations, which shows that the method can stably realize migration fault diagnosis of the rolling bearing under variable working conditions, and the classification accuracy can meet the actual application requirements.
In summary, the invention discloses a rolling bearing fault diagnosis method based on dynamic index antagonism self-adaptation, which adopts a feature extractor to extract features and adopts distance measurement based on domain loss to evaluate domain difference. The superiority and robustness of the method are proved through experimental verification and comparison with several methods. Some of the results of the invention are summarized below: 1) through the countermeasure among the feature extractor, the classifier and the domain discriminator, the features more suitable for migration can be autonomously screened, and a better fault diagnosis result is obtained. 2) Distance measurement between a source domain and a target domain is not specified explicitly, and distribution difference between the source domain and the target domain is measured indirectly by loss, so that the distribution difference between the source domain and the target domain can be measured better, and better migration effect is obtained; 3) the method utilizes the index dynamic factor to stably, quantitatively and dynamically adjust the proportion of the edge distribution and the condition distribution difference in the overall data distribution difference, and the index function can effectively reduce data collapse caused by calculation defects (except zero) in the quantitative calculation process, so that the model is more stable; the dynamic quantitative adjustment can enable the model to better align the data distribution between the source domain and the target domain, so that the model still has good diagnosis effect when facing the variable working condition task.
In the method provided by the embodiment, the data sets under six different working conditions are collected as the source domain data set and the target domain data set to train and test the bearing fault diagnosis model, the training strategies of mutual confrontation of the feature extractor, the classifier and the domain discriminator are adopted, and the gradient turning layer is utilized to optimize the three simultaneously, so that the training efficiency can be effectively improved, and the possibility of model collapse is reduced; the importance of the model optimization by the edge distribution and the condition distribution is stably and dynamically adjusted by adopting the index dynamic self-adaptive factors, and the requirement of the distribution self-adaptation of the transfer learning is better met, so that the knowledge learned by the source domain is better utilized by the target domain, and the egg knife has good fault diagnosis effect. Compared with other variant methods, the invention has the advantages of high average fault diagnosis accuracy rate of 100%, good feature extraction effect, good stability and capability of processing fault diagnosis of variable working conditions.
The invention also provides a rolling bearing fault diagnosis system based on dynamic index antagonism self-adaptation, which comprises the following steps:
the acquisition module 100 is used for acquiring vibration data of the bearing during operation under different working conditions to obtain a source domain sample and a target domain sample;
the feature extraction module 300 is configured to input the source domain samples into the feature extractor to obtain source domain features; inputting the source domain sample and the target domain sample into a feature extractor together to obtain mixed domain sample features;
a classification calculation module 400, configured to use the source domain features and the mixed domain sample features as inputs, perform a training on a classifier and a domain discriminator against each other, optimize a feature extractor, and calculate losses of the classifier and the domain discriminator; wherein, the domain discriminator comprises a global domain discriminator and a local domain discriminator;
an exponent dynamic module 500 for passing similarity measures for global discriminator loss and local discriminator loss
Calculating global measurement and local measurement of distribution distance between a source domain sample and a target domain sample to obtain an index dynamic adjustment factor, and redefining the loss of the domain discriminator by using the index dynamic adjustment factor, wherein the global measurement and the local measurement respectively correspond to edge distribution and condition distribution difference;
a training module 600, configured to construct an objective function of a bearing fault diagnosis model by using classifier loss and redefined domain discriminator loss, and train and search an optimal parameter of the objective function through a source domain sample with a label and a target domain sample without a label until the bearing fault diagnosis model is completed, where edge distribution and condition distribution differences of the source domain sample and the target domain sample are reduced by using a dynamic index adjustment factor in a training process;
and the test module 700 is used for inputting the target domain sample into the finished bearing fault diagnosis model and outputting a bearing fault diagnosis result.
Further, the classification calculation module 400 includes:
the classification module 401 is configured to obtain a prediction tag by using the features obtained by the feature extraction module, so as to classify fault types, and calculate cross entropy loss with a real tag;
a global area identification module 402, configured to perform domain classification on the samples by using the mixed sample features to obtain a prediction label of each sample feature, and calculate cross entropy loss of the global area identifier by using the prediction label and the real label of each sample feature;
the local domain identification module 403 is configured to perform domain classification on the samples by using the mixed sample features to obtain a prediction label of each sample feature, perform classification on all sample features by using the classifier to obtain a pseudo label probability distribution of the target domain sample, and calculate cross entropy loss of the local domain identifier by using the prediction label of each sample feature together with the true label and the pseudo label probability distribution.
The system is configured to implement the foregoing bearing fault diagnosis method, and therefore a specific implementation manner in the bearing fault diagnosis system may be found in the foregoing embodiment parts of the bearing fault diagnosis method, for example, the acquisition module 100 and the processing module 200, the feature extraction module 300 and the classification module 400, the global area identification module 500 and the local area identification module 600, the index dynamic module 700, and the test module 800 are respectively configured to implement steps S101 to S107 in the above bearing fault diagnosis method, so that the specific implementation manner thereof may refer to descriptions of corresponding embodiments of each part, and is not described herein again.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention. The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The above-mentioned embodiments are merely preferred embodiments for fully illustrating the present invention, and the scope of the present invention is not limited thereto. The equivalent substitution or change made by the technical personnel in the technical field on the basis of the invention is all within the protection scope of the invention. The protection scope of the invention is subject to the claims.
Claims (10)
1. A rolling bearing fault diagnosis method based on dynamic index antagonism self-adaptation is characterized in that: the method comprises the following steps:
s1: acquiring vibration data of a bearing in operation under different working conditions to obtain a source domain sample and a target domain sample;
s2: inputting the source domain samples into a feature extractor to obtain source domain features; inputting the source domain sample and the target domain sample into a feature extractor together to obtain mixed domain sample features;
s3: taking the source domain characteristics and the mixed domain sample characteristics as input, training a classifier and a domain discriminator in an antagonistic way, optimizing a characteristic extractor, and calculating the loss of the classifier and the domain discriminator; wherein, the domain discriminator comprises a global domain discriminator and a local domain discriminator;
s4: passing similarity measures for global discriminator loss and local discriminator loss
Calculating global measurement and local measurement of distribution distance between the source domain sample and the target domain sample to obtain an index dynamic adjustment factor, and redefining the loss of the domain discriminator by using the index dynamic adjustment factor, wherein the global measurement and the local measurement respectively correspond to edge distribution and condition distribution difference;
s5: constructing an objective function of a bearing fault diagnosis model by using classifier loss and redefined domain discriminator loss, and searching optimal parameters of the objective function through training of a source domain sample with a label and a target domain sample without a label until the bearing fault diagnosis model is completed, wherein the edge distribution and condition distribution difference of the source domain sample and the target domain sample is reduced by using a dynamic index adjustment factor in the training process;
s6: and inputting the target domain sample into the finished bearing fault diagnosis model, and outputting a bearing fault diagnosis result.
2. The rolling bearing fault diagnosis method based on dynamic index antagonism adaptation according to claim 1, characterized in that: in step S1, the bearing health status in each condition is different, the bearing vibration data in the different health status in each condition is used as a transferable data field, the data field is attached with a field label, the source field sample and the target field sample are selected from the data field, and the source field sample is attached with a fault type label.
3. The rolling bearing fault diagnosis method based on dynamic index antagonism adaptation as claimed in claim 2, characterized in that: the method comprises the steps of collecting vibration signals of a bearing in operation under each working condition by using an acceleration sensor, constructing a source domain data set and a target domain data set, processing the source domain data set and the target domain data set by using short-time Fourier transform, performing two-dimensional processing, and outputting processed multi-source domain samples and target domain samples.
4. The rolling bearing fault diagnosis method based on dynamic index antagonism adaptation as claimed in claim 2, characterized in that: inputting the source domain features into a classifier to perform supervised training in the step S3 to obtain predicted source domain labels and classifier losses, wherein the supervised training is to obtain the classifier losses by calculating cross entropy losses of the predicted source domain labels and the source domain true labels
yiIs a data true failure tag, C (G (x)i) Is a fault label predicted by the classifier, LcThe cross entropy loss of the two is calculated.
5. The rolling bearing fault diagnosis method based on dynamic index antagonism adaptation according to claim 4, characterized in that: in the step S3:
inputting the mixed domain sample into a global area discriminator for training to obtain a predicted domain label and a predicted global area loss
Wherein d iskDomain tags for data trueness, Dg(G(xk) Is) represents a predicted domain label,
both are calculatedCross entropy loss;
inputting the characteristics of the mixed samples into a classifier to obtain the probability distribution of the target domain fault category prediction labels; inputting the characteristics of the mixed samples into a plurality of local domain discriminators to obtain a plurality of domain discrimination prediction labels, calculating cross entropy loss of each domain with a real domain label, and multiplying and summing the cross entropy loss and the probability distribution of the target domain fault type prediction label to obtain a final local domain loss calculation formula:
calculating local area loss, wherein H represents the number of fault label types,
is a domain discriminator associated with class h,
is that
The corresponding cross-entropy loss is then taken into account,
representing the probability distribution of the presence of the kth sample over the h class, dkIs a domain label of the data reality.
6. The rolling bearing fault diagnosis method based on dynamic index antagonism adaptation according to claim 5, characterized in that: the step S4 specifically includes:
for global and local losses, passing through
Calculating, i.e. using global difference metric formulae
And local variance measure
Calculating global measurement and local measurement of distribution distance between two domains;
is converted into an exponential dynamic adjustment factor omega expressed as
Adjusting the difference between the edge distribution and the condition distribution of the two domains by using an index dynamic adjustment factor;
considering the exponential dynamics adjustment factor, the final domain discriminator loss is defined as:
7. the rolling bearing fault diagnosis method based on dynamic index antagonism adaptation according to claim 6, characterized in that: the step S5 specifically includes:
according to classifier loss LySum-field discriminator loss LDEstablishing an objective function of a bearing fault diagnosis model, namely calculating the total loss, wherein the proportional change of the two follows a formula
I.e. the final total loss calculation is given by the formula L ═ Ly-λLD；
Calculating the total loss once in each epoch, and optimizing the established feature extractor, classifier, global domain discriminator and local domain discriminator by using the Adam algorithm for the total loss;
determining the number of times of model training by using a predefined epoch number, and according to a predefined step length and a step length attenuation formula
Obtaining the step length of each step, and circulating the whole model for a plurality of times according to the step length to obtain the trainedThe model of (1).
8. The rolling bearing fault diagnosis method based on dynamic index antagonism adaptation as claimed in claim 2, characterized in that: the step S6 specifically includes:
processing the target domain data set by using short-time Fourier transform to obtain a target domain sample picture;
and inputting the target domain sample picture into the trained fault diagnosis model to obtain a final fault diagnosis result.
9. A rolling bearing fault diagnosis system based on dynamic index antagonism self-adaptation is characterized in that: the method comprises the following steps:
the acquisition module is used for acquiring vibration data of the bearing during operation under different working conditions to obtain a source domain sample and a target domain sample;
the characteristic extraction module is used for inputting the source domain samples into the characteristic extractor to obtain source domain characteristics; inputting the source domain sample and the target domain sample into a feature extractor together to obtain mixed domain sample features;
the classification calculation module is used for taking the source domain characteristics and the mixed domain sample characteristics as input, training a classifier and a domain discriminator in an antagonistic way, optimizing a characteristic extractor and calculating the loss of the classifier and the domain discriminator; wherein, the domain discriminator comprises a global domain discriminator and a local domain discriminator;
an exponential dynamic module for passing similarity measures for global discriminator loss and local discriminator loss
Calculating global measurement and local measurement of distribution distance between the source domain sample and the target domain sample to obtain an index dynamic adjustment factor, and redefining the loss of the domain discriminator by using the index dynamic adjustment factor, wherein the global measurement and the local measurement respectively correspond to edge distribution and condition distribution difference;
the training module is used for constructing an objective function of the bearing fault diagnosis model by using classifier loss and redefined domain discriminator loss, searching the optimal parameters of the objective function through the training of a source domain sample with a label and a target domain sample without a label until the bearing fault diagnosis model is completed, wherein the edge distribution and condition distribution difference of the source domain sample and the target domain sample are reduced by using a dynamic index adjustment factor in the training process;
and the test module is used for inputting the target domain sample into the finished bearing fault diagnosis model and outputting a bearing fault diagnosis result.
10. The dynamic index antagonism-based adaptive rolling bearing fault diagnosis system according to claim 9, wherein: the classification calculation module includes:
the classification module is used for obtaining a prediction label by utilizing the characteristics obtained by the characteristic extraction module so as to classify fault types and calculating cross entropy loss with a real label;
the global area identification module is used for carrying out domain classification on the samples by utilizing the mixed sample characteristics to obtain a prediction label of each sample characteristic, and calculating the cross entropy loss of the global area identifier by utilizing the prediction label and the real label of each sample characteristic;
and the local area identification module is used for carrying out domain classification on the samples by utilizing the mixed sample characteristics to obtain a prediction label of each sample characteristic, classifying all the sample characteristics by utilizing the classifier to obtain the pseudo label probability distribution of the target domain sample, and calculating the cross entropy loss of the local area identifier by utilizing the prediction label of each sample characteristic, the real label and the pseudo label probability distribution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111677762.1A CN114429152A (en) | 2021-12-31 | 2021-12-31 | Rolling bearing fault diagnosis method based on dynamic index antagonism self-adaption |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111677762.1A CN114429152A (en) | 2021-12-31 | 2021-12-31 | Rolling bearing fault diagnosis method based on dynamic index antagonism self-adaption |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114429152A true CN114429152A (en) | 2022-05-03 |
Family
ID=81310973
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111677762.1A Pending CN114429152A (en) | 2021-12-31 | 2021-12-31 | Rolling bearing fault diagnosis method based on dynamic index antagonism self-adaption |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114429152A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115773883A (en) * | 2022-11-07 | 2023-03-10 | 苏州大学 | Double-generator cooperative domain confronting bearing fault diagnosis method and system |
| CN116521493A (en) * | 2022-12-02 | 2023-08-01 | 北京小米移动软件有限公司 | Fault detection method, device and storage medium |
| CN116878885A (en) * | 2023-09-04 | 2023-10-13 | 佛山科学技术学院 | Bearing fault diagnosis method based on self-adaptive joint domain adaptive network |
| CN117349749A (en) * | 2023-10-09 | 2024-01-05 | 石家庄铁道大学 | Multi-source domain bearing fault diagnosis method based on mixed convolution |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103605781A (en) * | 2013-11-29 | 2014-02-26 | 苏州大学 | Implicit expression chapter relationship type inference method and system |
| CN106227988A (en) * | 2016-07-11 | 2016-12-14 | 浪潮(北京)电子信息产业有限公司 | A kind of relative error calculation optimization method |
| CN111651937A (en) * | 2020-06-03 | 2020-09-11 | 苏州大学 | Method for diagnosing similar self-adaptive bearing fault under variable working conditions |
| CN111829782A (en) * | 2020-07-16 | 2020-10-27 | 苏州大学 | Fault diagnosis method based on adaptive manifold embedding dynamic distribution alignment |
| CN112183581A (en) * | 2020-09-07 | 2021-01-05 | 华南理工大学 | Semi-supervised mechanical fault diagnosis method based on self-adaptive migration neural network |
| CN112633339A (en) * | 2020-12-14 | 2021-04-09 | 华中科技大学 | Bearing fault intelligent diagnosis method, bearing fault intelligent diagnosis system, computer equipment and medium |
| CN112629863A (en) * | 2020-12-31 | 2021-04-09 | 苏州大学 | Bearing fault diagnosis method for dynamic joint distribution alignment network under variable working conditions |
| CN112683532A (en) * | 2020-11-25 | 2021-04-20 | 西安交通大学 | Cross-working condition countermeasure diagnostic method for bearing |
| CN113723491A (en) * | 2021-08-24 | 2021-11-30 | 哈尔滨理工大学 | Rolling bearing with semi-supervised model contrast migration and gear fault diagnosis method |
-
2021
- 2021-12-31 CN CN202111677762.1A patent/CN114429152A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103605781A (en) * | 2013-11-29 | 2014-02-26 | 苏州大学 | Implicit expression chapter relationship type inference method and system |
| CN106227988A (en) * | 2016-07-11 | 2016-12-14 | 浪潮(北京)电子信息产业有限公司 | A kind of relative error calculation optimization method |
| CN111651937A (en) * | 2020-06-03 | 2020-09-11 | 苏州大学 | Method for diagnosing similar self-adaptive bearing fault under variable working conditions |
| CN111829782A (en) * | 2020-07-16 | 2020-10-27 | 苏州大学 | Fault diagnosis method based on adaptive manifold embedding dynamic distribution alignment |
| CN112183581A (en) * | 2020-09-07 | 2021-01-05 | 华南理工大学 | Semi-supervised mechanical fault diagnosis method based on self-adaptive migration neural network |
| CN112683532A (en) * | 2020-11-25 | 2021-04-20 | 西安交通大学 | Cross-working condition countermeasure diagnostic method for bearing |
| CN112633339A (en) * | 2020-12-14 | 2021-04-09 | 华中科技大学 | Bearing fault intelligent diagnosis method, bearing fault intelligent diagnosis system, computer equipment and medium |
| CN112629863A (en) * | 2020-12-31 | 2021-04-09 | 苏州大学 | Bearing fault diagnosis method for dynamic joint distribution alignment network under variable working conditions |
| CN113723491A (en) * | 2021-08-24 | 2021-11-30 | 哈尔滨理工大学 | Rolling bearing with semi-supervised model contrast migration and gear fault diagnosis method |
Non-Patent Citations (2)
| Title |
|---|
| JINYANG JIAO 等: "A mixed adversarial adaptation network for intelligent fault diagnosis", 《JOURNAL OF INTELLIGENT MANUFACTURING》 * |
| 沈长青 等: "基于自适应流形嵌入动态分布对齐的轴承故障诊断", 《电子测量与仪器学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115773883A (en) * | 2022-11-07 | 2023-03-10 | 苏州大学 | Double-generator cooperative domain confronting bearing fault diagnosis method and system |
| CN116521493A (en) * | 2022-12-02 | 2023-08-01 | 北京小米移动软件有限公司 | Fault detection method, device and storage medium |
| CN116521493B (en) * | 2022-12-02 | 2024-02-13 | 北京小米移动软件有限公司 | Fault detection method, device and storage medium |
| CN116878885A (en) * | 2023-09-04 | 2023-10-13 | 佛山科学技术学院 | Bearing fault diagnosis method based on self-adaptive joint domain adaptive network |
| CN116878885B (en) * | 2023-09-04 | 2023-12-19 | 佛山科学技术学院 | Bearing fault diagnosis method based on self-adaptive joint domain adaptive network |
| CN117349749A (en) * | 2023-10-09 | 2024-01-05 | 石家庄铁道大学 | Multi-source domain bearing fault diagnosis method based on mixed convolution |
| CN117349749B (en) * | 2023-10-09 | 2024-03-15 | 石家庄铁道大学 | Multi-source domain bearing fault diagnosis method based on mixed convolution |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110361176B (en) | Intelligent fault diagnosis method based on multitask feature sharing neural network | |
| CN114429152A (en) | Rolling bearing fault diagnosis method based on dynamic index antagonism self-adaption | |
| CN110110768B (en) | Rolling bearing fault diagnosis method based on parallel feature learning and multiple classifiers | |
| CN111709448A (en) | Mechanical fault diagnosis method based on migration relation network | |
| CN112257530B (en) | Rolling bearing fault diagnosis method based on blind signal separation and support vector machine | |
| CN110647830B (en) | Bearing fault diagnosis method based on convolutional neural network and Gaussian mixture model | |
| CN105678343B (en) | Hydropower Unit noise abnormality diagnostic method based on adaptive weighted group of sparse expression | |
| CN111562108A (en) | Rolling bearing intelligent fault diagnosis method based on CNN and FCMC | |
| CN114004252A (en) | Bearing fault diagnosis method, device and equipment | |
| CN109781411A (en) | A kind of combination improves the Method for Bearing Fault Diagnosis of sparse filter and KELM | |
| CN116894187A (en) | Gear box fault diagnosis method based on deep migration learning | |
| CN114358124B (en) | New fault diagnosis method for rotary machinery based on deep countermeasure convolutional neural network | |
| CN112926642B (en) | Multi-source domain self-adaptive intelligent mechanical fault diagnosis method and system | |
| CN115358259A (en) | Self-learning-based unsupervised cross-working-condition bearing fault diagnosis method | |
| CN114091504A (en) | Rotary machine small sample fault diagnosis method based on generation countermeasure network | |
| Chen et al. | Multi-expert attention network with unsupervised aggregation for long-tailed fault diagnosis under speed variation | |
| CN114169377A (en) | G-MSCNN-based fault diagnosis method for rolling bearing in noisy environment | |
| CN115587290A (en) | Aero-engine fault diagnosis method based on variational self-coding generation countermeasure network | |
| CN114091349A (en) | Multi-source field self-adaption based rolling bearing service life prediction method | |
| CN115165366A (en) | Variable working condition fault diagnosis method and system for rotary machine | |
| Wei et al. | WSAFormer-DFFN: A model for rotating machinery fault diagnosis using 1D window-based multi-head self-attention and deep feature fusion network | |
| CN115407753B (en) | Industrial fault diagnosis method for multi-variable weighting integrated learning | |
| Guan et al. | Fault diagnosis of rolling bearing with imbalanced small sample scenarios | |
| CN114383846B (en) | Bearing composite fault diagnosis method based on fault label information vector | |
| Brandao et al. | Fault Diagnosis of Rotary Machines Using Machine Learning |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220503 |