CN112213103A - Fault diagnosis method, device, system and medium for rail transit rolling stock bearing - Google Patents
Fault diagnosis method, device, system and medium for rail transit rolling stock bearing Download PDFInfo
- Publication number
- CN112213103A CN112213103A CN201910629360.0A CN201910629360A CN112213103A CN 112213103 A CN112213103 A CN 112213103A CN 201910629360 A CN201910629360 A CN 201910629360A CN 112213103 A CN112213103 A CN 112213103A
- Authority
- CN
- China
- Prior art keywords
- fault
- model
- training
- bearing
- fault diagnosis
- 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 66
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000005096 rolling process Methods 0.000 title claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 62
- 239000013598 vector Substances 0.000 claims abstract description 43
- 238000012544 monitoring process Methods 0.000 claims abstract description 11
- 238000012360 testing method Methods 0.000 claims description 30
- 238000004422 calculation algorithm Methods 0.000 claims description 25
- 239000011159 matrix material Substances 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 9
- 238000002790 cross-validation Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000012512 characterization method Methods 0.000 claims description 4
- 230000003137 locomotive effect Effects 0.000 claims description 4
- 238000007477 logistic regression Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000007637 random forest analysis Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 2
- 238000005192 partition Methods 0.000 claims description 2
- 238000012774 diagnostic algorithm Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
-
- 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
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention discloses a fault diagnosis method, a device, a system and a medium for a rail transit rolling stock bearing, belongs to the technical field of bearing fault diagnosis, solves the technical problem of low precision of the current diagnosis method, and adopts the technical scheme that: establishing a fault sample set: dividing the bearing fault into fault modes corresponding to the structure based on the structure of the bearing; collecting vibration signals of the bearings in the fault modes and normal bearings in the fault sample set during working, and extracting characteristic vectors of the vibration signals; training a multilayer stacking model and adjusting and optimizing model parameters according to the feature vector to establish a final multilayer stacking model; and (3) carrying out real-time state monitoring and fault diagnosis on the bearings of the same model by adopting a mature training stack model. The fault diagnosis method, the fault diagnosis device, the fault diagnosis system and the fault diagnosis medium have the advantages of high diagnosis precision and the like.
Description
Technical Field
The invention mainly relates to the technical field of bearing fault diagnosis, in particular to a fault diagnosis method, a device, a system and a medium for a rail transit locomotive bearing.
Background
The bearing is a typical rotating mechanical device, is also a vulnerable universal part, and is widely applied to the field of rail transit. Whether the bearing operates normally or not directly affects the precision, reliability and service life of locomotive equipment, and the fault diagnosis of the bearing is more and more emphasized. The condition monitoring and fault diagnosis of the bearing by using the vibration signal of the bearing are common and effective methods at present. The fault diagnosis of the vehicle equipment bearing is quickly realized, the reliability and the service life level of locomotive equipment can be improved, and the time and the economic cost required by fault positioning are reduced. The existing technology for carrying out online state monitoring and fault diagnosis on the bearing has the advantages of high complexity and low accuracy of the whole diagnosis method.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the invention provides a method, a device, a system and a medium for diagnosing the fault of a rail transit rolling stock bearing with high diagnosis precision, aiming at the technical problem of lower precision in the existing bearing fault diagnosis technology.
In order to solve the technical problems, the technical scheme provided by the invention is as follows:
a fault diagnosis method for a rail transit rolling stock bearing comprises the following steps:
s01, establishing a fault sample set: dividing the bearing fault into fault modes corresponding to the structure based on the structure of the bearing;
s02, collecting vibration signals of the bearings in the fault modes in the fault sample set and the normal bearings during working, and extracting feature vectors of the vibration signals;
s03, training and model parameter optimizing the multilayer stacking model according to the feature vector, and establishing a final multilayer stacking model;
and S04, performing real-time state monitoring and fault diagnosis on bearings of the same model by adopting a training mature stacking model.
As a further improvement of the above technical solution:
in step S03, a two-layer stack model is used, and the specific training steps include:
training a first-layer stacking model: dividing the feature vector set into a training set and a test set, and completing the conversion process from the characterization to the probability of the training feature vector by using K-fold cross validation; simultaneously, K times of tests are carried out on the test set to obtain a probabilistic result, and then the arithmetic mean value of the output results of the K times of test sets is solved; combining the results output by the training set and the test set to obtain an output matrix;
training a second-layer stacking model: and taking the output matrix as the input of the second-layer stacking model training, and performing self-learning by adopting an integrated learning algorithm to establish a stacking model.
The algorithm in the first-layer stacking model training comprises a plurality of vector machines, a neural network, a logistic regression or an integrated learning algorithm, and the integrated learning algorithm comprises one of GBDT, XGboost or random forest.
The process of K-fold cross validation is as follows: dividing the training set into 5 completely disjoint equal parts, using 4 of 5 parts as the training set to perform model training, and using the remaining 1 part as the test set to perform model precision test; this was done 5 times, and 5 aliquots of the training set partition were made one test set at a time.
In step S01, the failure mode includes one or more of an inner race failure, an outer race failure, a rolling element failure, and a cage failure.
The feature vector includes one or more of a maximum value, a minimum value, an effective value, a singular value, or an energy entropy.
In step S02, the vibration signal is filtered during the feature vector extraction.
The invention also discloses a fault diagnosis device for the rail transit rolling stock bearing, which comprises
A first module to establish a fault sample set: dividing the bearing fault into fault modes corresponding to the structure based on the structure of the bearing;
the second module is used for collecting vibration signals of the bearings of all fault modes in the fault sample set and the normal bearings during working and extracting the characteristic vectors of the vibration signals;
the third module is used for training and model parameter tuning of the multilayer stack model according to the feature vector and establishing a final multilayer stack model;
and the fourth module is used for carrying out real-time state monitoring and fault diagnosis on the bearings of the same model by adopting a stack model with mature training.
The invention further discloses a computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, performs the steps of the fault diagnosis method as described above.
The invention also discloses a fault diagnosis system of the rail transit rolling stock bearing, which comprises a computer device, wherein the computer device is programmed or configured to execute the steps of the fault diagnosis method; or a storage medium of the computer device has stored thereon a computer program programmed or configured to perform the fault diagnosis method as described above.
Compared with the prior art, the invention has the advantages that:
according to the fault diagnosis method, device, system and medium for the rail transit rolling stock bearing, the bearing fault is divided into fault modes corresponding to the structure based on the structure of the bearing, and a multi-layer stacking model is established by adopting multi-layer stacking training, so that higher fault diagnosis precision can be obtained, and accurate online monitoring of bearings of the same type can be realized.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
FIG. 2 is a schematic diagram of a stacked model structure according to the present invention.
FIG. 3 is a schematic diagram of the stacked model training of the present invention.
FIG. 4 is a schematic diagram of input and output of a stack model according to the present invention.
Detailed Description
The invention is further described below with reference to the figures and the specific embodiments of the description.
As shown in fig. 1, the method for diagnosing a fault of a rail transit rolling stock bearing of the embodiment includes the following steps:
s01, establishing a fault sample set: dividing bearing faults into fault modes corresponding to the structures based on the structures of the bearings;
s02, obtaining vibration signals of the bearings of the fault modes in the fault sample set and the normal bearings during working, and extracting feature vectors of the vibration signals;
s03, performing multi-layer stacking training and adjustment according to the feature vectors, and establishing a stacking model;
and S04, diagnosing the real-time data of the bearing by adopting the stacking model.
According to the fault diagnosis method for the rail transit rolling stock bearing, the multi-layer stacking model is trained by using the prior bearing fault data, the well-trained stacking model can be used for fault diagnosis of bearings of the same type, higher fault diagnosis precision can be obtained, and online accurate monitoring of the running state of the bearing is realized. Stack model
In this embodiment, in step S03, two-layer stacking model training is adopted, which specifically includes:
training a first-layer stacking model: dividing the feature vector set into a training set and a test set, and completing the conversion from characterization to probability of the feature vectors on the training set by using K-fold cross validation; simultaneously, K times of tests are carried out on the test set to obtain the arithmetic average value of the output result of the test set; because the first layer stack model comprises a plurality of algorithms, the diagnosis result of the first layer can be output in a matrix form and used as the input of the second layer stack model;
training a second-layer stacking model: and taking the output matrix of the first-layer stacking model training as the input of the second-layer stacking model training, and carrying out self-learning by adopting an integrated learning algorithm to establish a stacking model.
In this embodiment, the algorithm in the first layer stacking model includes multiple support vector machines, neural networks, logistic regression, or ensemble learning algorithms, and the ensemble learning algorithm in the first layer stacking model includes GBDT, XGBoost, or random forest.
In the present embodiment, in step S01, the failure mode includes one or more of an inner ring failure, an outer ring failure, a rolling element failure, and a cage failure;
the feature vector includes one or more of a maximum value, a minimum value, a valid value, or an energy entropy.
In this embodiment, in step S02, the vibration signal is filtered in the process of extracting the feature vector; if the dimension of the feature vector is large, feature selection can be carried out on the feature vector to obtain more important feature vectors.
The invention also discloses a fault diagnosis device for the rail transit rolling stock bearing, which comprises
A first module to establish a fault sample set: dividing the bearing fault into fault modes corresponding to the structure based on the structure of the bearing;
the second module is used for acquiring vibration signals of the bearings of all fault modes in the fault sample set and the normal bearings during working and extracting characteristic vectors;
the third module is used for carrying out multi-layer stacking training and adjustment according to the characteristic vector and establishing a stacking model;
and the fourth module is used for diagnosing the real-time data of the bearing by adopting the stacking model.
The invention also discloses a computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, performs the steps of the fault diagnosis method as described above.
The invention further discloses a fault diagnosis system of a rail transit rolling stock bearing, which comprises a computer device programmed or configured to execute the steps of the fault diagnosis method; or a storage medium of the computer device has stored thereon a computer program programmed or configured to perform the fault diagnosis method as described above.
The process of the invention is further illustrated below with reference to a specific embodiment:
the invention extracts the characteristic vector of the vibration signal by collecting the vibration signal generated when a certain type of bearing works, completes the training of the intelligent model by using the characteristic vector, and realizes the online monitoring and fault diagnosis of the health state of the bearing after the training of the intelligent model is completed and the online operation is finished. In the selection of the intelligent algorithm, a stacking algorithm is adopted as a diagnosis method, and the method specifically comprises the following steps:
step (1) carrying out fault mode analysis on the bearing and manufacturing a fault sample
The bearing comprises an outer ring, an inner ring, a rolling body, a retainer and other parts. According to engineering experience, except for a normal working state, the bearing is divided into four fault modes according to the fault occurrence position, wherein the four common fault modes are as follows: inner ring failure, outer ring failure, rolling element failure, and cage failure.
After identifying the typical four bearing failures, a failure sample was made for each failure mode. In the present embodiment, the method is described by taking the above-described four failure modes as an example (this is not limiting, and in other embodiments, one, two, three, five or more failures may be used for analysis). (note: the failure degree in the same failure mode is different, for example, the failure size of the inner ring of the bearing can be set to 0.007 inch, 0.014 inch and 0.021 inch, which belong to three different failure degrees, and the failure position in the same failure mode is different, for example, the failure of the bearing can belong to three different positions in the 3 o ' clock direction, the 6 o ' clock direction and the 12 o ' clock direction
Step (2) obtaining a vibration signal of the fault sample in the step (1) during working
Simulating the working state of the bearing to perform a test, and acquiring a vibration signal generated in the working process of the bearing;
classifying the collected vibration signals, such as: the bearing normal state vibration signal is recorded as type 1 data, the inner ring fault vibration signal is recorded as type 2 data, the outer ring fault vibration signal is recorded as type 3 data, the rolling element fault vibration signal is recorded as type 4 data, and the retainer fault vibration signal is recorded as type 5 data;
extracting the characteristic vector of the vibration signal in the step (3)
And extracting the feature vector of the vibration signals of 1 to 5 types.
The vibration signal can only reflect the change condition of the vibration amplitude of the bearing along with the running time, and the signal characteristics can more deeply reflect the time domain characteristics, the frequency domain characteristics, the statistical characteristics and the energy characteristics of the signal. Several characteristic features of the signal can be extracted as feature vectors. For example, the maximum, minimum, …, energy entropy, etc. of the signal constitute a feature vector, in the form of X ═ e (maximum, minimum, …, energy entropy).
Step (4) framework building of stacking model
And establishing a fault diagnosis model by taking a Stacking algorithm (Stacking) as a fault diagnosis algorithm, and performing fault diagnosis on the bearing. The stacking algorithm is a multilayer intelligent self-learning diagnostic algorithm, can be flexibly adjusted according to the needs of actual conditions, and adopts two-layer stacking in the embodiment.
The basic structure of the stack model is shown in fig. 2. The first layer is to follow the maximum difference principle, and the diagnosis algorithm of the first layer can be several single classification algorithms with weak diagnosis function in machine learning, such as a support vector machine, a neural network and logistic regression, and can also be an integrated learning algorithm with strong diagnosis function. The second layer of the algorithm usually selects an integrated learning algorithm with stronger learning function, such as GBDT, XGboost and the like, so that the output result of the first layer can be better learned, and the final diagnosis accuracy is improved.
And (5) model training and parameter adjustment of the stack model.
Taking the feature vector X extracted in step (3) (maximum, minimum, …, energy entropy) as the input of the stacked model in step (4), performing layered training on the stacked model (i.e. training the first layer in fig. 2 and then training the second layer in fig. 2), and obtaining higher diagnosis precision by adjusting the model parameters.
1. The first layer of training of the stacking model:
and (4) dividing the feature vector set in the step (3) into a training set and a testing set, wherein the training set accounts for 70%, and the testing set accounts for 30%. In the process of training the stacking model of the first layer, 5-fold cross validation is used to complete the process from characterization to probability of the feature vector X (the algorithm will output the probability of what fault type the feature vector belongs to). The specific operation method comprises the following steps: and dividing the training set into 5 completely disjoint equal parts, training the model of the first layer by using 4 of the 5 parts as the training set, and testing the precision of the first layer by using the remaining 1 part as the testing set.
After 5 times of this, 5 equal parts divided by the training set will be used as a test set, and the complete output of the training set will become the input of the second layer. The training set is converted from a vector form of X (maximum value, minimum value, … and energy entropy) into y (p0, p1, … … and pn), wherein p0 represents the normal state probability of the bearing, p1 represents the fault probability of the inner ring, and … … and pn represent the fault probability of the nth class of the bearing.
At the same time, the test set will be tested (note: the test set is not divided into 5 equal parts, and only needs to be input into the model 5 times), and for 5-fold cross validation reasons, the test set will be tested 5 times, and 5 test results are generated, y1 ═ p0, p1, …, pn, … …, y5 ═ p0, p1, …, pn, and finally, the average values of y1 to y5 are calculated.
2. Training of the second layer of the stack model:
since the first layer of the stack model contains multiple algorithms, the result output by the first layer is no longer a vector y1 (p0, p1, …, pn), but is presented in the form of a matrix, as shown in fig. 4. The first row in the matrix Y represents the probability of each type of fault diagnosed by the first-layer diagnostic algorithm 1, the second row represents the probability of each type of fault diagnosed by the first-layer diagnostic algorithm 2, and the third row represents the probability of each type of fault diagnosed by the first-layer diagnostic algorithm 3. And taking the matrix Y as the input of the second layer of the stacked model, and performing self-learning through a diagnosis algorithm of the second layer to perform fault diagnosis.
If the diagnosis result is not satisfactory, the parameters of the basic model can be adjusted to obtain a more ideal diagnosis result.
Step (6) the fault diagnosis model is on line
After the stacking model obtains ideal fault diagnosis precision, the stacking model can be operated on line, and real-time state monitoring and fault diagnosis are carried out on the bearing.
The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.
Claims (10)
1. A fault diagnosis method for a rail transit rolling stock bearing is characterized by comprising the following steps:
s01, establishing a fault sample set: dividing the bearing fault into fault modes corresponding to the structure based on the structure of the bearing;
s02, collecting vibration signals of the bearings in the fault modes in the fault sample set and the normal bearings during working, and extracting feature vectors of the vibration signals;
s03, training and model parameter optimizing the multilayer stacking model according to the feature vector, and establishing a final multilayer stacking model;
and S04, performing real-time state monitoring and fault diagnosis on bearings of the same model by adopting a training mature stacking model.
2. The method for diagnosing the failure of the rail transit rolling stock bearing of claim 1, wherein in the step S03, a two-layer stacking model is adopted, and the specific training step comprises:
training a first-layer stacking model: dividing the feature vector set into a training set and a test set, and completing the conversion process from the characterization to the probability of the training feature vector by using K-fold cross validation; simultaneously, K times of tests are carried out on the test set to obtain a probabilistic result, and then the arithmetic mean value of the output results of the K times of test sets is solved; combining the results output by the training set and the test set to obtain an output matrix;
training a second-layer stacking model: and taking the output matrix as the input of the second-layer stacking model training, and performing self-learning by adopting an integrated learning algorithm to establish a stacking model.
3. The rail transit locomotive vehicle bearing fault diagnosis method according to claim 2, wherein the algorithm in the first layer of stacked model training comprises a plurality of vector machines, neural networks, logistic regression or ensemble learning algorithms, and the ensemble learning algorithms comprise one of GBDT, XGBoost or random forest.
4. The method for diagnosing the fault of the rail transit rolling stock bearing according to claim 2, wherein the K-fold cross validation process comprises: dividing the training set into 5 completely disjoint equal parts, using 4 of 5 parts as the training set to perform model training, and using the remaining 1 part as the test set to perform model precision test; this was done 5 times, and 5 aliquots of the training set partition were made one test set at a time.
5. The rail transit rolling stock bearing failure diagnosis method according to any one of claims 1 to 4, wherein the failure mode includes one or more of an inner ring failure, an outer ring failure, a rolling body failure and a cage failure at step S01.
6. The method for diagnosing the failure of the rail transit rolling stock bearing according to any one of claims 1 to 4, wherein the feature vector includes one or more of a maximum value, a minimum value, an effective value, a singular value or an energy entropy.
7. The method for diagnosing a failure of a rail transit rolling stock bearing according to any one of claims 1 to 4, wherein the vibration signal is filtered during the feature vector extraction at step S02.
8. A fault diagnosis device for rail transit rolling stock bearings is characterized by comprising
A first module to establish a fault sample set: dividing the bearing fault into fault modes corresponding to the structure based on the structure of the bearing;
the second module is used for collecting vibration signals of the bearings of all fault modes in the fault sample set and the normal bearings during working and extracting the characteristic vectors of the vibration signals;
the third module is used for training and model parameter tuning of the multilayer stack model according to the feature vector and establishing a final multilayer stack model;
and the fourth module is used for carrying out real-time state monitoring and fault diagnosis on the bearings of the same model by adopting a stack model with mature training.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the fault diagnosis method according to any one of claims 1 to 7.
10. A fault diagnosis system for rail transit rolling stock bearings comprising computer means, characterized in that the computer means is programmed or configured to perform the steps of the fault diagnosis method according to any one of claims 1 to 7; or a storage medium of the computer device having stored thereon a computer program programmed or configured to perform the fault diagnosis method of any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910629360.0A CN112213103A (en) | 2019-07-12 | 2019-07-12 | Fault diagnosis method, device, system and medium for rail transit rolling stock bearing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910629360.0A CN112213103A (en) | 2019-07-12 | 2019-07-12 | Fault diagnosis method, device, system and medium for rail transit rolling stock bearing |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112213103A true CN112213103A (en) | 2021-01-12 |
Family
ID=74048561
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910629360.0A Pending CN112213103A (en) | 2019-07-12 | 2019-07-12 | Fault diagnosis method, device, system and medium for rail transit rolling stock bearing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112213103A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113361359A (en) * | 2021-05-31 | 2021-09-07 | 电子科技大学 | XGboost algorithm-based nuclear power pipeline fault identification method |
CN113390645A (en) * | 2021-04-25 | 2021-09-14 | 北京航空工程技术研究中心 | Special test method for peeling fault of main bearing of aircraft engine |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106895975A (en) * | 2017-01-17 | 2017-06-27 | 苏州大学 | Bearing fault diagnosis method based on Stacked SAE deep neural network |
CN107505133A (en) * | 2017-08-10 | 2017-12-22 | 滁州学院 | The probability intelligent diagnosing method of rolling bearing fault based on adaptive M RVM |
CN108304884A (en) * | 2018-02-23 | 2018-07-20 | 华东理工大学 | A kind of cost-sensitive stacking integrated study frame of feature based inverse mapping |
-
2019
- 2019-07-12 CN CN201910629360.0A patent/CN112213103A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106895975A (en) * | 2017-01-17 | 2017-06-27 | 苏州大学 | Bearing fault diagnosis method based on Stacked SAE deep neural network |
CN107505133A (en) * | 2017-08-10 | 2017-12-22 | 滁州学院 | The probability intelligent diagnosing method of rolling bearing fault based on adaptive M RVM |
CN108304884A (en) * | 2018-02-23 | 2018-07-20 | 华东理工大学 | A kind of cost-sensitive stacking integrated study frame of feature based inverse mapping |
Non-Patent Citations (1)
Title |
---|
陈学武: "《城市低收入人群出行方式选择机理与交通发送策略[M]》", 31 December 2018 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113390645A (en) * | 2021-04-25 | 2021-09-14 | 北京航空工程技术研究中心 | Special test method for peeling fault of main bearing of aircraft engine |
CN113390645B (en) * | 2021-04-25 | 2022-09-13 | 北京航空工程技术研究中心 | Special test method for peeling fault of main bearing of aircraft engine |
CN113361359A (en) * | 2021-05-31 | 2021-09-07 | 电子科技大学 | XGboost algorithm-based nuclear power pipeline fault identification method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106769052B (en) | A kind of mechanical system rolling bearing intelligent failure diagnosis method based on clustering | |
CN108376298B (en) | Early warning and diagnosing method for temperature faults of engine of wind turbine generator | |
CN110110768B (en) | Rolling bearing fault diagnosis method based on parallel feature learning and multiple classifiers | |
CN111046945B (en) | Fault type and damage degree diagnosis method based on combined convolutional neural network | |
CN112418277B (en) | Method, system, medium and equipment for predicting residual life of rotating machine parts | |
CN110376522B (en) | Motor fault diagnosis method of data fusion deep learning network | |
CN107797537A (en) | A kind of prognostic and health management method applied to automatic production line | |
CN112257530B (en) | Rolling bearing fault diagnosis method based on blind signal separation and support vector machine | |
CN106912067B (en) | WSN wireless communication module fault diagnosis method based on fuzzy neural network | |
US11137322B2 (en) | Diagnosing method of engine condition and diagnostic modeling method thereof | |
CN113375941A (en) | Open set fault diagnosis method for high-speed motor train unit bearing | |
CN106250935A (en) | The rotary machinery fault diagnosis method that genetic programming and weighted evidence theory merge | |
CN112213103A (en) | Fault diagnosis method, device, system and medium for rail transit rolling stock bearing | |
CN114266197A (en) | Method for diagnosing equipment fault of hydropower station | |
KR20210061517A (en) | Apparatus and method for fault diagnosis using fake data generated by machine learning | |
CN113569685A (en) | Method and system for establishing fault diagnosis model and fault diagnosis of machine tool spindle bearing | |
CN116168720A (en) | Motor sound abnormality fault diagnosis method, system and storable medium | |
CN117170304B (en) | PLC remote monitoring control method and system based on industrial Internet of things | |
CN112417771A (en) | Method for establishing fault diagnosis model of machine tool spindle and method for diagnosing fault of machine tool spindle | |
CN111783941A (en) | Mechanical equipment diagnosis and classification method based on probability confidence degree convolutional neural network | |
CN116451121A (en) | Switch fault diagnosis method, system, equipment and medium | |
CN107016440A (en) | The multiresolution deep neural network intelligent diagnosing method of machine driving failure | |
CN113487141B (en) | Bearing state evaluation method based on multisource information clustering fusion | |
CN113177484B (en) | Intelligent mechanical fault diagnosis method based on LOF self-encoding | |
CN113820123A (en) | Gearbox fault diagnosis method based on improved CNN and selective integration |
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 |
Application publication date: 20210112 |
|
RJ01 | Rejection of invention patent application after publication |