CN111982514A - Bearing fault diagnosis method based on semi-supervised deep belief network - Google Patents
Bearing fault diagnosis method based on semi-supervised deep belief network Download PDFInfo
- Publication number
- CN111982514A CN111982514A CN202010807298.2A CN202010807298A CN111982514A CN 111982514 A CN111982514 A CN 111982514A CN 202010807298 A CN202010807298 A CN 202010807298A CN 111982514 A CN111982514 A CN 111982514A
- Authority
- CN
- China
- Prior art keywords
- data
- semi
- belief network
- layer
- deep belief
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The application provides a bearing fault diagnosis method based on a semi-supervised deep belief network, which comprises the following steps of: collecting vibration signal data; performing wavelet transformation noise reduction on the vibration signal data and completing reconstruction; classifying the reconstructed signal data according to the labels, and extracting corresponding time domain characteristics; screening the classified signal data according to a set rule, and dividing the screened data into a training data set and a test data set; inputting a training data set into a semi-supervised deep belief network, and carrying out deep training on the network; inputting the test data set into a semi-supervised deep belief network, and carrying out fault classification judgment on the data in the test data set through a deeply trained model. The beneficial effect of this application: the judged bearing fault can be directly output by inputting the working data, and the automation degree is higher; the method is more suitable for bearing fault diagnosis under multiple working conditions, has stronger compatibility and universality, and improves the bearing fault diagnosis precision.
Description
Technical Field
The disclosure relates to the field of semi-supervised deep belief network learning, in particular to a bearing fault diagnosis method based on a semi-supervised deep belief network.
Background
Aiming at bearing fault diagnosis, the conventional bearing fault diagnosis method cannot automatically extract features at present, and the features need to be manually extracted and are judged depending on expert knowledge; the intelligent bearing fault diagnosis method mostly focuses on a fault diagnosis model with single load and single rotating speed, can automatically classify the fault types of the bearings, and does not need to manually extract features.
The traditional bearing fault diagnosis method needs manual feature extraction, wastes time and labor, results after feature extraction cannot be automatically classified, expert knowledge is needed for judgment, the labor cost is further increased, and the efficiency is low; the intelligent bearing fault diagnosis algorithm trains the model aiming at single load and single rotating speed, so that the bearing fault diagnosis under the change of multiple working conditions cannot be adapted, the diagnosis precision is low, in addition, the method also needs a large amount of data with fault type labels, and the cost for manufacturing the label data is high.
Disclosure of Invention
The bearing fault diagnosis method based on the semi-supervised deep belief network aims to solve the problems.
In a first aspect, the application provides a bearing fault diagnosis method based on a semi-supervised deep belief network, which includes the following steps:
collecting vibration signal data;
performing wavelet transformation noise reduction on the vibration signal data and completing reconstruction;
classifying the reconstructed signal data according to the labels, and extracting corresponding time domain characteristics;
screening the classified signal data according to a set rule, and dividing the screened data into a training data set and a test data set;
inputting a training data set into a semi-supervised deep belief network, and carrying out deep training on the network;
inputting the test data set into a semi-supervised deep belief network, and carrying out fault classification judgment on the data in the test data set through a deeply trained model.
According to the technical scheme provided by the embodiment of the application, the wavelet transformation denoising and reconstruction of the vibration signal data are specifically performed, and the method specifically comprises the following steps:
considering the vibration signal data x (t) as a space V in a subspace of the function space0Space V0Can be decomposed into V1And W1I.e. V0=V1+W1,V1Can be decomposed into V2And W2By analogy, Vj-1Can be decomposed into VjAnd WjJ is resolution, j is 0 or 1;
the signal decomposition and noise reduction process of the vibration signal data x (t) is as follows:
whereinIs a linear combination weight with a resolution of 0, P0x (t) is x (t) at V0A smooth approximation of (i.e. a profile of x (t) at a resolution of 0, phi0k(t) is a scale function;
whereinFor linear combining weights at resolution 1, P1x (t) is x (t) at V1A smooth approximation of (i.e. a profile of x (t) at a resolution of 1, phi1k(t) is a scale function;
whereinDiscrete details at resolution 1, D1x (t) is x (t) in W1Projection of1k(t) is a wavelet function;
P0x(t)=P1x(t)+D1x(t);
the decomposed and denoised signals are respectively:
the reconstructed signal is:
according to the technical scheme provided by the embodiment of the application, the label comprises fault position information and damage size information of the bearing, the fault position information comprises inner ring faults, rolling body faults and outer ring faults, and the damage size information comprises 0.178mm, 0.356mm and 0.533 mm; the labels comprise ten types, wherein nine types are labels simultaneously containing fault position information and damage size information, and the tenth type is labels of the bearing health state.
According to the technical scheme provided by the embodiment of the application, the extracting the corresponding time domain feature specifically comprises: and extracting a mean value, a mean square value, a root mean square value, an average amplitude value, a kurtosis value, a peak-peak value, a standard deviation, a variance, a skewness value, a pulse factor, a skewness factor, a waveform factor, a kurtosis coefficient, a margin coefficient, a kurtosis factor and a waveform entropy from the reconstructed signal data to form a total data set.
According to the technical scheme provided by the embodiment of the application, the screening of the classified signal data according to the set rule specifically comprises:
the total data set was screened using the Maximum Mean Difference (MMD) algorithm, given two distributions s and t, the MMD between them is defined as:
where E is the expectation for the allocation,to map raw data to a function of the Regenerated Kernel Hilbert Space (RKHS), when s and t distributions are the same, MMD2(s, t) ═ 0, and the kernel function associated with this map is k (x)s,xt)=<φ(xs),φ(xt)>;
Andthe empirical estimate of MMD, representing distribution s and distribution t, respectively, is as follows:
when the data is consistent with 0 being less than or equal to LM(Ds,Dt) If the condition is less than 0.16, the data is selected as the screened data, and if the condition is not met, the data is removed.
According to the technical scheme provided by the embodiment of the application, 80% of screened data is selected as a training set, 20% of the screened data is selected as a test set, 10% of the data in the training set is provided with a label, the rest 90% of the data in the training set is unlabeled data, and the label of the data in the training set is predicted and identified by a semi-supervised deep belief network.
According to the technical scheme provided by the embodiment of the application, the inputting of the training data set into the semi-supervised deep belief network to perform deep training on the network specifically comprises the following steps:
the semi-supervised deep belief network (SSDBN) is formed by stacking semi-supervised restricted Boltzmann machines (SSRBMs), the SSDBN is composed of an input layer, a plurality of hidden layers and an output layer, each layer is provided with a plurality of neurons, the neurons only have two states of activation and non-activation, the neurons of each layer are not connected with each other, the neurons in each layer are fully connected, information is transmitted by using an activation function, weight values are updated, and weight values are weighted, the SSRBMs are composed of a visible layer, a hidden layer and a supervision layer, and the energy function of the SSRBMs is defined as:
where v is a visible layer, h is a hidden layer, u is a supervisory layer, ψ ═ wij,pkj,ai,ck,bj) A is a visual layer bias value, b is a hidden layer bias value, c is a supervised layer bias value, λ is a weight parameter for controlling a ratio of supervised learning to unsupervised learning, w is a weight between the visual layer and the hidden layer, p is a weight between the supervised layers, and ψ is (w ═ isij,pkj,ai,ck,bj) And can be updated as:wherein tau is iteration times, eta is learning rate;
the activation function connecting the layers of the SSRBM is the Isigmoid function:
wherein a is a threshold, α is a slope, and satisfies:wherein alpha isminMinimum slope to work for the Isigmoid function.
According to the technical scheme provided by the embodiment of the application, the inputting of the test data set into the semi-supervised deep belief network and the fault classification and judgment of the data in the test data set through the deeply trained model specifically comprise: the semi-supervised deep belief network reserves the trained weight and bias matrix, the supervision layer refers to the label data to assist in predicting the label of the unlabelled data, the test data are transmitted from the input layer to the output layer according to the SSDBN after being input, and finally the test data are classified.
The invention has the beneficial effects that: the application provides a bearing fault diagnosis method based on a semi-supervised deep belief network, which can directly output the judged bearing fault by inputting working data and has higher automation degree; the method is more suitable for bearing fault diagnosis under multiple working conditions, has stronger compatibility and universality, and improves the bearing fault diagnosis precision; and the semi-supervised deep belief network needs a small amount of label data during model training, so that the training cost is reduced.
Drawings
FIG. 1 is a flow chart of a first embodiment of the present application;
fig. 2 is a diagram of a semi-supervised deep belief network in the first embodiment of the present application.
Detailed Description
In order that those skilled in the art will better understand the technical solutions of the present invention, the following detailed description of the present invention is provided in conjunction with the accompanying drawings, and the description of the present section is only exemplary and explanatory, and should not be construed as limiting the scope of the present invention in any way.
Fig. 1 shows a flow chart of a first embodiment of the present application, which includes the following steps:
and S1, collecting vibration signal data.
In this embodiment, the acceleration sensor is provided in the outer ring portion of the bearing of the rotary machine, and the magnetic adsorption type acceleration sensor is selected when the vibration amplitude of the bearing is large, and the adhesive attachment type acceleration sensor is selected when the vibration amplitude of the bearing is small.
And S2, performing wavelet transformation denoising on the vibration signal data and completing reconstruction.
The method specifically comprises the following steps: considering the vibration signal data x (t) as a space V in a subspace of the function space0Space V0Can be decomposed into V1And W1I.e. V0=V1+W1,V1Can be decomposed into V2And W2By analogy, Vj-1Can be decomposed into VjAnd WjJ is resolution, j is 0 or 1;
the signal decomposition and noise reduction process of the vibration signal data x (t) is as follows:
whereinIs a linear combination weight with a resolution of 0, P0x (t) is x (t) at V0A smooth approximation of (i.e. a profile of x (t) at a resolution of 0, phi0k(t) is a scale function;
whereinFor linear combining weights at resolution 1, P1x (t) is x (t) at V1A smooth approximation of (i.e. a profile of x (t) at a resolution of 1, phi1k(t) is a scale function;
whereinDiscrete details at resolution 1, D1x (t) is x (t) in W1Projection of1k(t) is a wavelet function;
P0x(t)=P1x(t)+D1x(t)
the decomposed and denoised signals are respectively:
the reconstructed signal is:
in the step, after wavelet transformation is carried out on original vibration signal data x (t), noise interference of vibration signals is reduced, and data are normalized.
And S3, classifying the reconstructed signal data according to the labels, and extracting corresponding time domain characteristics.
The fault data of the bearing is generally divided into three types, namely, fault position, damage size and load rotating speed, the label data set in the embodiment simultaneously comprises fault position and damage size information, the fault position information comprises inner ring fault, rolling body fault and outer ring fault, the damage size information comprises 0.178mm, 0.356mm and 0.533mm, the label in the embodiment comprises ten types of label data, wherein nine types of labels simultaneously comprise fault position information and damage size information, and the tenth type of label is a label of the bearing health state.
In this step, extracting the corresponding time domain features is to extract a mean value, a mean square value, a root mean square value, an average amplitude value, a kurtosis value, a peak-to-peak value, a standard deviation, a variance, a skewness value, a pulse factor, a skewness coefficient, a form factor, a kurtosis coefficient, a margin coefficient, a kurtosis factor, and a form entropy from the reconstructed signal data to form a total data set.
And S4, screening the classified signal data according to a set rule, and dividing the screened data into a training data set and a test data set.
In this embodiment, a Maximum Mean Difference (MMD) algorithm is used to screen the total data set to select a data set with a small difference in feature distribution. Given two distributions s and t, the MMD between them is defined as:
where E is the expectation for the allocation,to map raw data to a function of the Regenerated Kernel Hilbert Space (RKHS), when s and t distributions are the same, MMD2(s, t) ═ 0, and the kernel function associated with this map is k (x)s,xt)=<φ(xs),φ(xt)>;
Andthe empirical estimate of MMD, representing distribution s and distribution t, respectively, is as follows:
when the data is consistent with 0 being less than or equal to LM(Ds,Dt) If the condition is less than 0.16, the data is selected as the screened data, and if the condition is not met, the data is removed.
And selecting 80% of the screened data as a training set, 20% of the screened data as a test set, setting 10% of the data in the training set with labels, and the rest 90% of the data in the training set as unlabeled data, and predicting and identifying the labels by using a semi-supervised deep belief network. In the embodiment, because the dependence on the label is greatly reduced by the semi-supervised deep belief network, training can be performed only by ensuring that 10% of data in the training set is labeled, and various costs for making label data are reduced.
And S5, inputting the training data set into the semi-supervised deep belief network, and carrying out deep training on the network.
In this embodiment, a semi-supervised deep belief network (SSDBN) is formed by stacking semi-supervised limited boltzmann machines (SSRBMs), where the SSDBN is composed of an input layer, a plurality of hidden layers, and an output layer, as shown in fig. 2, the SSDBN in this embodiment is configured by stacking 3 SSRBMs in total, and 4 layers in total, each layer has a plurality of neurons, and the neurons only have two states of activation and deactivation, in this embodiment, the number of neurons in the output layer is 10, the neurons in each layer are not connected to each other, the neurons in each layer are all connected to each other, and transmit information, update weight, and bias weight by using an activation function, the SSRBM is composed of a visible layer, a hidden layer, and a supervision layer, and an energy function of the SSRBM is defined as:
where v is a visible layer, h is a hidden layer, u is a supervisory layer, ψ ═ wij,pkj,ai,ck,bj) A is a visual layer bias value, b is a hidden layer bias value, c is a supervised layer bias value, λ is a weight parameter for controlling a ratio of supervised learning to unsupervised learning, w is a weight between the visual layer and the hidden layer, p is a weight between the supervised layers, and ψ is (w ═ isij,pkj,ai,ck,bj) Can be updated as:wherein tau is iteration times, eta is learning rate;
the activation function connecting the SSRBM of each layer is generally a sigmoid function, and is modified into an Isigmoid function because the activation function is easy to cause the problems of gradient disappearance and gradient explosion:
wherein a is a threshold, α is a slope, and satisfies:wherein alpha isminMinimum slope to work for the Isigmoid function.
In the embodiment, the Isigmoid activation function is used to reduce the probability of gradient explosion and gradient disappearance of the model, so that the model is more stable.
And S6, inputting the test data set into a semi-supervised deep belief network, and carrying out fault classification judgment on the data in the test data set through a deeply trained model.
In this embodiment, the semi-supervised deep belief network retains the trained weights and bias matrices, the supervision layer refers to the labeled data to assist in predicting the label of the unlabelled data, and the test data is transmitted from the input layer to the output layer according to the SSDBN after being input, and is finally classified.
In the embodiment, the judged bearing fault can be directly output by inputting the working data, the automation degree is high, the characteristics do not need to be manually selected, expert knowledge does not need to be relied on, and an end-to-end learning mode is completely realized; the method is more suitable for bearing fault diagnosis under multiple working conditions, has stronger compatibility and universality, and improves the bearing fault diagnosis precision; and the semi-supervised deep belief network needs a small amount of label data during model training, so that the training cost is reduced. In bearing fault diagnosis, effective data can be extracted from strong noise; automatic and intelligent automatic bearing fault diagnosis results can be realized without manually extracting features and depending on expert experience knowledge; the use of the labeled data is less, the cost for manufacturing the labeled data is reduced, and the precision of bearing fault diagnosis under the multi-working-condition is improved.
The principles and embodiments of the present application are explained herein using specific examples, which are presented only to assist in understanding the method and its core concepts of the present application. The foregoing is only a preferred embodiment of the present application, and it should be noted that there are objectively infinite specific structures due to the limited character expressions, and it will be apparent to those skilled in the art that a plurality of modifications, decorations or changes may be made without departing from the principle of the present application, and the technical features described above may be combined in a suitable manner; such modifications, variations, combinations, or adaptations of the invention using its spirit and scope, as defined by the claims, may be directed to other uses and embodiments, or may be learned by practice of the invention.
Claims (8)
1. A bearing fault diagnosis method based on a semi-supervised deep belief network is characterized by comprising the following steps:
collecting vibration signal data;
performing wavelet transformation noise reduction on the vibration signal data and completing reconstruction;
classifying the reconstructed signal data according to the labels, and extracting corresponding time domain characteristics;
screening the classified signal data according to a set rule, and dividing the screened data into a training data set and a test data set;
inputting a training data set into a semi-supervised deep belief network, and carrying out deep training on the network;
inputting the test data set into a semi-supervised deep belief network, and carrying out fault classification judgment on the data in the test data set through a deeply trained model.
2. The bearing fault diagnosis method based on the semi-supervised deep belief network as claimed in claim 1, wherein the wavelet transform denoising and reconstructing are performed on the vibration signal data, specifically comprising:
considering the vibration signal data x (t) as a space V in a subspace of the function space0Space V0Can be decomposed into V1And W1I.e. V0=V1+W1,V1Can be decomposed into V2And W2By analogy, Vj-1Can be decomposed into VjAnd WjJ is resolution, j is 0 or 1;
the signal decomposition and noise reduction process of the vibration signal data x (t) is as follows:
whereinIs a linear combination weight with a resolution of 0, P0x (t) is x (t) at V0A smooth approximation of (i.e. a profile of x (t) at a resolution of 0, phi0k(t) is a scale function;
whereinFor linear combining weights at resolution 1, P1x (t) is x (t) at V1A smooth approximation of (i.e. a profile of x (t) at a resolution of 1, phi1k(t) is a scale function;
whereinDiscrete details at resolution 1, D1x (t) is x (t) in W1Projection of1k(t) is a wavelet function;
P0x(t)=P1x(t)+D1x(t);
the decomposed and denoised signals are respectively:
the reconstructed signal is:
3. the bearing fault diagnosis method based on the semi-supervised deep belief network of claim 1, wherein the label comprises fault location information and damage size information of the bearing, the fault location information comprises inner ring faults, rolling body faults and outer ring faults, and the damage size information comprises 0.178mm, 0.356mm and 0.533 mm; the labels comprise ten types, wherein nine types are labels simultaneously containing fault position information and damage size information, and the tenth type is labels of the bearing health state.
4. The bearing fault diagnosis method based on the semi-supervised deep belief network as claimed in claim 1, wherein the extracting of the corresponding time domain features specifically comprises: and extracting a mean value, a mean square value, a root mean square value, an average amplitude value, a kurtosis value, a peak-peak value, a standard deviation, a variance, a skewness value, a pulse factor, a skewness factor, a waveform factor, a kurtosis coefficient, a margin coefficient, a kurtosis factor and a waveform entropy from the reconstructed signal data to form a total data set.
5. The bearing fault diagnosis method based on the semi-supervised deep belief network as claimed in claim 4, wherein the step of screening the classified signal data according to a set rule specifically comprises the steps of:
the total data set was screened using the Maximum Mean Difference (MMD) algorithm, given two distributions s and t, the MMD between them is defined as:
where E is the expectation for the allocation,to map raw data to a function of the Regenerated Kernel Hilbert Space (RKHS), when s and t distributions are the same, MMD2(s, t) ═ 0, and the kernel function associated with this map is k (x)s,xt)=<φ(xs),φ(xt)>;
Andthe empirical estimate of MMD, representing distribution s and distribution t, respectively, is as follows:
when the data is consistent with 0 being less than or equal to LM(Ds,Dt) If the condition is less than 0.16, the data is selected as the screened data, and if the condition is not met, the data is removed.
6. The method for diagnosing the bearing fault based on the semi-supervised deep belief network as recited in claim 5, wherein 80% of screened data are selected as a training set, 20% are selected as a test set, 10% of the data in the training set are provided with labels, and the remaining 90% of the data in the training set are unlabeled data, and the labels of the data are predicted and identified by the semi-supervised deep belief network.
7. The bearing fault diagnosis method based on the semi-supervised deep belief network as claimed in claim 1, wherein the inputting of the training data set into the semi-supervised deep belief network for deep training of the network specifically comprises:
the semi-supervised deep belief network (SSDBN) is formed by stacking semi-supervised restricted Boltzmann machines (SSRBMs), the SSDBN is composed of an input layer, a plurality of hidden layers and an output layer, each layer is provided with a plurality of neurons, the neurons only have two states of activation and non-activation, the neurons of each layer are not connected with each other, the neurons in each layer are fully connected, information is transmitted by using an activation function, weight values are updated, and weight values are weighted, the SSRBMs are composed of a visible layer, a hidden layer and a supervision layer, and the energy function of the SSRBMs is defined as:
where v is a visible layer, h is a hidden layer, u is a supervisory layer, ψ ═ wij,pkj,ai,ck,bj) A is a visual layer bias value, b is a hidden layer bias value, c is a supervised layer bias value, λ is a weight parameter for controlling a ratio of supervised learning to unsupervised learning, w is a weight between the visual layer and the hidden layer, p is a weight between the supervised layers, and ψ is (w ═ isij,pkj,ai,ck,bj) Can be updated as:wherein tau is iteration times, eta is learning rate;
the activation function connecting the layers of the SSRBM is the Isigmoid function:
8. The bearing fault diagnosis method based on the semi-supervised deep belief network as claimed in claim 1, wherein the step of inputting the test data set into the semi-supervised deep belief network and performing fault classification and judgment on the data in the test data set through the model after deep training specifically comprises the steps of: the semi-supervised deep belief network reserves the trained weight and bias matrix, the supervision layer refers to the label data to assist in predicting the label of the unlabelled data, the test data are transmitted from the input layer to the output layer according to the SSDBN after being input, and finally the test data are classified.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010807298.2A CN111982514B (en) | 2020-08-12 | 2020-08-12 | Bearing fault diagnosis method based on semi-supervised deep belief network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010807298.2A CN111982514B (en) | 2020-08-12 | 2020-08-12 | Bearing fault diagnosis method based on semi-supervised deep belief network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111982514A true CN111982514A (en) | 2020-11-24 |
CN111982514B CN111982514B (en) | 2023-03-24 |
Family
ID=73434917
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010807298.2A Active CN111982514B (en) | 2020-08-12 | 2020-08-12 | Bearing fault diagnosis method based on semi-supervised deep belief network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111982514B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113405799A (en) * | 2021-05-20 | 2021-09-17 | 新疆大学 | Bearing early fault detection method based on health state index construction and fault early warning limit self-learning |
CN114354185A (en) * | 2021-12-29 | 2022-04-15 | 重庆邮电大学 | Gear box fault diagnosis method based on decoupling network |
CN114993679A (en) * | 2022-05-25 | 2022-09-02 | 国网重庆市电力公司电力科学研究院 | Multi-fan combined bearing fault self-diagnosis method and system |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160116378A1 (en) * | 2013-08-26 | 2016-04-28 | Mtelligence Corporation | Population-based learning with deep belief networks |
CN108304927A (en) * | 2018-01-25 | 2018-07-20 | 清华大学 | Bearing fault modality diagnostic method and system based on deep learning |
CN108414226A (en) * | 2017-12-25 | 2018-08-17 | 哈尔滨理工大学 | Fault Diagnosis of Roller Bearings under the variable working condition of feature based transfer learning |
JP2019101789A (en) * | 2017-12-04 | 2019-06-24 | 日本電信電話株式会社 | Model learning device, model learning method, and prediction system |
US20190243735A1 (en) * | 2018-02-05 | 2019-08-08 | Wuhan University | Deep belief network feature extraction-based analogue circuit fault diagnosis method |
CN110533070A (en) * | 2019-07-26 | 2019-12-03 | 西安交通大学 | A kind of mechanical breakdown intelligent diagnosing method based on migration original shape network under small sample |
CN110543860A (en) * | 2019-09-05 | 2019-12-06 | 燕山大学 | Mechanical fault diagnosis method and system based on TJM (machine learning model) transfer learning |
CN110542819A (en) * | 2019-09-25 | 2019-12-06 | 贵州电网有限责任公司 | transformer fault type diagnosis method based on semi-supervised DBNC |
-
2020
- 2020-08-12 CN CN202010807298.2A patent/CN111982514B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160116378A1 (en) * | 2013-08-26 | 2016-04-28 | Mtelligence Corporation | Population-based learning with deep belief networks |
JP2019101789A (en) * | 2017-12-04 | 2019-06-24 | 日本電信電話株式会社 | Model learning device, model learning method, and prediction system |
CN108414226A (en) * | 2017-12-25 | 2018-08-17 | 哈尔滨理工大学 | Fault Diagnosis of Roller Bearings under the variable working condition of feature based transfer learning |
CN108304927A (en) * | 2018-01-25 | 2018-07-20 | 清华大学 | Bearing fault modality diagnostic method and system based on deep learning |
US20190243735A1 (en) * | 2018-02-05 | 2019-08-08 | Wuhan University | Deep belief network feature extraction-based analogue circuit fault diagnosis method |
CN110533070A (en) * | 2019-07-26 | 2019-12-03 | 西安交通大学 | A kind of mechanical breakdown intelligent diagnosing method based on migration original shape network under small sample |
CN110543860A (en) * | 2019-09-05 | 2019-12-06 | 燕山大学 | Mechanical fault diagnosis method and system based on TJM (machine learning model) transfer learning |
CN110542819A (en) * | 2019-09-25 | 2019-12-06 | 贵州电网有限责任公司 | transformer fault type diagnosis method based on semi-supervised DBNC |
Non-Patent Citations (4)
Title |
---|
GONGMINGWANG 等: "Improved Classification with Semi-supervised Deep Belief Network", 《IFAC PAPERSONLINE》 * |
MEI WANG 等: "Deep face recognition with clustering based domain adaptation", 《NEUROCOMPUTING》 * |
YI QIN 等: "The optimized deep belief networks with improved logistic Sigmoid units and their application in fault diagnosis for planetary gearboxes of wind turbines", 《IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS》 * |
赵蕾: "基于改进粒子滤波和深度置信网络的故障诊断研究", 《中国优秀博硕士学位论文全文数据库(硕士) 工程科技Ⅱ辑》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113405799A (en) * | 2021-05-20 | 2021-09-17 | 新疆大学 | Bearing early fault detection method based on health state index construction and fault early warning limit self-learning |
CN114354185A (en) * | 2021-12-29 | 2022-04-15 | 重庆邮电大学 | Gear box fault diagnosis method based on decoupling network |
CN114993679A (en) * | 2022-05-25 | 2022-09-02 | 国网重庆市电力公司电力科学研究院 | Multi-fan combined bearing fault self-diagnosis method and system |
Also Published As
Publication number | Publication date |
---|---|
CN111982514B (en) | 2023-03-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111982514B (en) | Bearing fault diagnosis method based on semi-supervised deep belief network | |
WO2021128510A1 (en) | Bearing defect identification method based on sdae and improved gwo-svm | |
CN110866365B (en) | Mechanical equipment intelligent fault diagnosis method based on partial migration convolutional network | |
CN113935406B (en) | Mechanical equipment unsupervised fault diagnosis method based on countermeasure flow model | |
CN106124212B (en) | Fault Diagnosis of Roller Bearings based on sparse coding device and support vector machines | |
CN106895975B (en) | Bearing fault diagnosis method based on Stacked SAE deep neural network | |
CN106682688B (en) | Particle swarm optimization-based stacked noise reduction self-coding network bearing fault diagnosis method | |
CN110163302B (en) | Indicator diagram identification method based on regularization attention convolution neural network | |
CN111964908B (en) | Bearing fault diagnosis method under variable working condition based on multi-scale weight distribution convolutional neural network model | |
CN107657250B (en) | Bearing fault detection and positioning method and detection and positioning model implementation system and method | |
CN112577747A (en) | Rolling bearing fault diagnosis method based on space pooling network | |
CN110991424A (en) | Fault diagnosis method based on minimum entropy deconvolution and stacking sparse self-encoder | |
CN113505655B (en) | Intelligent bearing fault diagnosis method for digital twin system | |
CN109543838B (en) | Image increment learning method based on variational self-encoder | |
CN106570477A (en) | Vehicle model recognition model construction method based on depth learning and vehicle model recognition method based on depth learning | |
CN106769052A (en) | A kind of mechanical system rolling bearing intelligent failure diagnosis method based on cluster analysis | |
CN114358123B (en) | Generalized open set fault diagnosis method based on deep countermeasure migration network | |
CN107016241A (en) | Based on the rotating machinery lifetime stage recognition methods for adding sample enhancing depth own coding learning network of making an uproar | |
CN115508073B (en) | Prototype adaptive mechanical equipment fault diagnosis method based on multi-scale attention | |
CN114169377A (en) | G-MSCNN-based fault diagnosis method for rolling bearing in noisy environment | |
CN110991471A (en) | Fault diagnosis method for high-speed train traction system | |
CN114861719A (en) | High-speed train bearing fault diagnosis method based on ensemble learning | |
CN116026569A (en) | Mechanical equipment unsupervised migration intelligent fault diagnosis method for releasing source data | |
CN115221958A (en) | Mechanical equipment selective migration fault diagnosis method based on composite weight | |
CN112380932B (en) | Vibration signal characteristic value selection method and elevator health state evaluation or fault diagnosis 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 | ||
CB03 | Change of inventor or designer information |
Inventor after: Han Xu Inventor after: Ye Nan Inventor after: Chang Peize Inventor after: Zhang Luyu Inventor after: Wang Jia Inventor after: Li Jiahang Inventor before: Ye Nan Inventor before: Chang Peize Inventor before: Zhang Luyu Inventor before: Wang Jia Inventor before: Li Jiahang |
|
CB03 | Change of inventor or designer information | ||
GR01 | Patent grant | ||
GR01 | Patent grant |