CN111581892B - Bearing residual life prediction method based on GDAU neural network - Google Patents
Bearing residual life prediction method based on GDAU neural network Download PDFInfo
- Publication number
- CN111581892B CN111581892B CN202010478293.XA CN202010478293A CN111581892B CN 111581892 B CN111581892 B CN 111581892B CN 202010478293 A CN202010478293 A CN 202010478293A CN 111581892 B CN111581892 B CN 111581892B
- Authority
- CN
- China
- Prior art keywords
- attention
- gate
- neural network
- bearing
- output
- 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.)
- Active
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000005096 rolling process Methods 0.000 claims abstract description 16
- 238000004364 calculation method Methods 0.000 claims abstract description 9
- 230000001133 acceleration Effects 0.000 claims abstract description 7
- 230000002159 abnormal effect Effects 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 31
- 238000005070 sampling Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 12
- 238000010606 normalization Methods 0.000 claims description 6
- 230000009467 reduction Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 2
- 230000036541 health Effects 0.000 description 6
- 238000002474 experimental method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 241000935372 Tisias Species 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000011156 evaluation 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
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
-
- G—PHYSICS
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
-
- 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
- G06F2218/04—Denoising
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Acoustics & Sound (AREA)
- Biomedical Technology (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Geometry (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to a bearing residual life prediction method based on a GDAU neural network, and belongs to the technical field of rolling bearing detection. The method comprises the following steps: firstly, collecting vibration signals of a rolling bearing through an acceleration sensor, obtaining a root mean square value of the vibration signals through calculation, removing and replacing abnormal points of the root mean square value, and finally inputting the processed root mean square value into a GDAU network for life prediction. The invention greatly improves the prediction accuracy of the residual life of the bearing.
Description
Technical Field
The invention belongs to the technical field of rolling bearing detection, and relates to a bearing residual life prediction method based on a GDAU neural network.
Background
The rolling bearing is widely applied to mechanical equipment and is one of the most widely applied mechanical parts. Under the complex working environments such as overload, impact, abrasion and the like, the rolling bearing can be damaged to different degrees, so that the whole mechanical equipment is stopped and damaged, and under serious conditions, the production activity is greatly damaged, and the personal safety problem is caused. Therefore, the health of the rolling bearing restricts the reliability, safety and productivity of the whole mechanical equipment. Therefore, the residual life prediction of the rolling bearing can effectively evaluate the health state of the rolling bearing, and the mechanical equipment can work safely and efficiently.
Although a method for predicting the residual life of the rolling bearing by using a neural network is available at present, the prediction result of the neural network structure is not accurate enough, and the monitoring of the rolling bearing is not effective.
Disclosure of Invention
Accordingly, the present invention aims to provide a method for predicting the remaining life of a bearing based on a GDAU neural network, which constructs a gated dual-attention unit network (GDAU), inputs the time domain characteristic root mean square value of a bearing vibration signal as a health index into the GDAU neural network to predict the remaining life of the bearing, and improves the accuracy of the prediction result.
In order to achieve the above purpose, the present invention provides the following technical solutions:
firstly, collecting vibration signals of a rolling bearing through an acceleration sensor arranged on an experiment table, obtaining a root mean square value of the vibration signals through calculation, removing and replacing abnormal points of the root mean square value, and finally inputting the processed root mean square value into the GDAU network for life prediction;
the method comprises the following steps:
s1: collecting vibration signals in the life cycle of the bearing, and performing noise reduction treatment; let the sampling time be T, the interval between adjacent sampling points be T s The number of samples is n;
s2: calculating root mean square value of each vibration signal after noise reduction treatment to obtain an n multiplied by 1 dimension eigenvalue vector X= [ X ] 1 ,x 2 ,…,x n ] T The method comprises the steps of carrying out a first treatment on the surface of the The eigenvalue vector of the first m samples is selected as a training vector s= [ x ] 1 ,x 2 ,…,x m ] T ；
S3: normalizing the training vector S to obtain a normalized training vector Y= [ Y ] 1 ,y 2 ,…，,y m ] T ；
S4: reconstructing a matrix W;
s5: constructing a gating dual-attention unit network (Gated dual attention unit network, GDAU), wherein the number of input layer units is k, and the number of output layer units is 1;
s6: the k rows in front of the matrix W are used as the input of the GDAU neural network, and the last row is used as the output of the GDAU neural network to train the network;
s7: taking k outputs of the reciprocal as network inputs to obtain the output of the next moment;
s8: repeating step S7, when the preset threshold value is exceeded after the inverse normalization is output, the predicted sampling point number minus m is multiplied by the sum T of the vibration signal interval time and the sampling time s And +T is the residual life of the bearing.
Further, in the step S3, the normalized training vector S adopts a linear normalization method.
Further, in the step S4, the reconstruction matrix is
Further, in the step S5, the constructed GDAU neural network is:
r t ＝σ(U r x t +W r h t-1 +b r )
z t ＝σ(U z x t +W z h t-1 +b z )
wherein r is t Indicating reset gate output, z t Representing an update gate output; x is x t Inputting information for time t of GDAU network, h t-1 Hiding state information for the last moment; w (W) r And U r To reset the gate weight matrix, W z And U z To update the gate weight matrix, W h And U h B is a candidate state weight matrix r ,b z ,b h Is a bias matrix;for attention gate output, A t To reset the gate and to update the attention profile of the gate.
Further, in the constructed GDAU neural network, the first attention gate is to process the input data x with an attention mechanism in each time dimension t And recursive data h t-1 The specific calculation formula is as follows:
s(x t ,h t-1 )＝V T tanh(W s x t +U s h t-1 )
wherein V is an additional parameter related to the attention gate, W s And U s For the weight matrix associated with the scoring function,andfor a weight matrix related to the attention gate output,/a>For attention gate output, α t Is a vector of attention distribution.
Furthermore, in the constructed GDAU neural network, the second attention gate processes the reset gate and the update gate output by using sigmoid and Tanh functions in each time dimension, and the specific calculation formula is as follows:
wherein,is a weight matrix>Is a bias matrix; />Calculating through a sigmoid function, and representing the attention ratio of the reset gate and the update gate output; />A candidate attention value is calculated by using a Tanh function, and the contribution degree of information to prediction is represented.
The invention has the beneficial effects that: according to the invention, the time domain characteristic root mean square value of the bearing vibration signal is used as a health index to be input into the GDAU neural network to predict the residual life of the bearing by introducing the gating attention unit network (GDAU), and compared with the existing neural network prediction result, the accuracy of the prediction result is greatly improved.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objects and other advantages of the invention may be realized and obtained by means of the instrumentalities and combinations particularly pointed out in the specification.
Drawings
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in the following preferred detail with reference to the accompanying drawings, in which:
FIG. 1 is a block diagram of a GDAU neural network;
FIG. 2 is a first attention door construction diagram;
FIG. 3 is a second attention door construction diagram;
FIG. 4 is a flow chart of a method for predicting the residual life of a bearing according to the present invention;
FIG. 5 is a simulation graph of the results of life prediction (final hundred points selected) using RMS for different bearings under the same operating conditions;
FIG. 6 is a graph comparing the predicted effect of using a GDAU neural network to the residual life of a bearing using an existing network.
Detailed Description
Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the illustrations provided in the following embodiments merely illustrate the basic idea of the present invention by way of illustration, and the following embodiments and features in the embodiments may be combined with each other without conflict.
Referring to fig. 1 to 6, the time domain characteristics of the rolling bearing vibration signals reflect the degradation trend of the state and health condition of the rolling vibration signals to a certain extent. In the time domain feature, root Mean Square (RMS) is widely used to evaluate vibration signals to effectively reflect the degradation trend of bearings. Therefore, the method selects the time domain characteristic root mean square value of the bearing vibration signal as the health index to predict the residual life of the bearing. Firstly, collecting vibration signals of a rolling bearing through an acceleration sensor arranged on an experiment table, obtaining a root mean square value of the vibration signals through calculation, removing and replacing abnormal points of the root mean square value, and finally inputting the processed root mean square value into a GDAU network for life prediction.
As shown in fig. 4, the prediction method based on the GDAU neural network includes the following steps:
1. vibration signals in the life cycle of the bearing are collected. Sampling time is T, and interval between adjacent sampling points is T s Let n be the number of samples.
2. The root mean square values after the noise reduction of the vibration signals are calculated respectively, so that an n multiplied by 1-dimensional eigenvalue vector X= [ X ] can be obtained 1 ,x 2 ,…,x n ] T . The eigenvalue vector of the first m samples is selected as a training vector s= [ x ] 1 ,x 2 ,…,x m ] T 。
3. Normalizing the training vector S by using a linear normalization method to obtain a normalized training vector Y= [ Y ] 1 ,y 2 ,…,y m ] T 。
4. Reconstruction matrix
5. And constructing the GDAU neural network, wherein the number of input layer units is k, and the number of output layer units is 1. The calculation formula of the GDAU neural network is as follows:
r t ＝σ(U r x t +W r h t-1 +b r )
z t ＝σ(U z x t +W z h t-1 +b z )
s(x t ,h t-1 )＝V T tanh(W s x t +U s h t-1 )
wherein x is t Inputting information for GDAU network, h t-1 The state information is hidden for the last time. W (W) r And U r To reset the gate weight matrix, W z And U z To update the gate weight matrix, W h And U h B is a candidate state weight matrix r ,b z ,b h Is a bias matrix. V is an additional parameter related to the attention gate. W (W) s And U s Is a matrix of weights associated with the scoring function.And->Is a weight matrix related to the attention gate output. />Is the attention gate output. Alpha t Is a vector of attention distribution. />Is a weight matrix;is a bias matrix. />Calculated by a sigmoid function, which represents a reset gate r t Updating door z t Attention ratio of the output. Calculating candidate attention value +.>It represents the degree of contribution of information to the prediction. A is that t To reset gates and updateThe attention profile of the door.
6. The first k rows of the matrix W are taken as inputs to the neural network and the last row is taken as an output of the neural network to train the network.
7. And taking the k outputs of the reciprocal as network inputs to obtain the output of the next moment.
8. Repeating the step 7 for a certain number of times, and inversely normalizing the outputs and then comparing the output with an actual value x' = (x) m+1 ,x m+2 ,…x n ) T Comparison to demonstrate the effectiveness of this method. Meanwhile, when the output inverse normalization exceeds a set threshold value, the predicted sampling point number minus m is multiplied by the sum T of the vibration signal interval time and the sampling time s And +T is the residual life of the bearing.
Verification experiment:
the bearing data used in this experiment were from the pro tisia test bench, which is shown in fig. 5. The PRONOSTIA test bed mainly comprises a rotating part, a loading part and a measuring part. The rotating part mainly comprises a motor, an accelerator, a gear box and a corresponding rotating shaft, and the rotating speed and the rotating direction of the bearing can be set for the motor by people. The loading part is an important component of the test bed, and in order to accelerate the decay process of the bearing and achieve the purpose of shortening the service life of the bearing, the radial load of the rolling bearing is continuously increased until the maximum rated value of the bearing is reached. The measuring section measures two state indexes of vibration signal and temperature of the bearing by using an acceleration sensor and a temperature sensor. The acceleration sensor can measure acceleration in the horizontal direction and the vertical direction, the sampling frequency is 25.6kHz, the sampling interval time is 10s, and the sampling time is 0.1s. The sampling frequency of the temperature sensor is 10Hz.
As shown in fig. 5, the life prediction was performed using RMS for bearing 1 and bearing 2 at 1800rpm and 4000N load, and bearing 1 and bearing 2 contained 2803 and 2375 RMS values, respectively. Since the first 1000 RMS changes of the bearing 1 are smooth, a total of 1725 RMS values from 1001 to 2725 are selected for life prediction, and the prediction result is shown in fig. 5 (a). Similarly, bearing 2 takes a total of 439 to 2238 1800 RMS values for life prediction, as shown in fig. 5 (b). The failure threshold of bearing 1 and bearing 2 was 1.903. The failure threshold value, training value, predicted value and actual value of the last 100 points are predicted, the predicted result is shown in fig. 5, and the predicted result is basically consistent with the actual value.
Comparison experiment:
in order to fully prove the advantages of the invention, three evaluation indexes of MAE (mean absolute error), MAPE (mean absolute percent error) and RMSE (root mean square error) are respectively compared with the existing neural network model. The comparison result is shown in fig. 6, and it can be seen from fig. 6 that the GDAU neural network adopted in the present invention has higher prediction accuracy than the existing neural network.
Finally, it is noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made thereto without departing from the spirit and scope of the present invention, which is intended to be covered by the claims of the present invention.
Claims (2)
1. The method is characterized in that firstly, vibration signals of a rolling bearing are collected through an acceleration sensor, a root mean square value of the vibration signals is obtained through calculation, abnormal points are removed and replaced on the root mean square value, and finally the processed root mean square value is input into the gated double-attention unit neural network for life prediction; the method specifically comprises the following steps:
s1: collecting vibration signals in the life cycle of the bearing, and performing noise reduction treatment; let the sampling time be T, the interval between adjacent sampling points be T s The number of samples is n;
s2: calculating root mean square value of each vibration signal after noise reduction treatment to obtain an n multiplied by 1 dimension eigenvalue vector X= [ X ] 1 ,x 2 ,…,x n ] T The method comprises the steps of carrying out a first treatment on the surface of the The eigenvalue vector of the first m samples is selected as a training vector s= [ x ] 1 ,x 2 ,…,x m ] T ；
S3: normalizing the training vector S to obtain a normalized training vector Y= [ Y ] 1 ,y 2 ,…,y m ] T ；
S4: the reconstruction matrix W is:
s5: constructing a gate-control double-attention unit neural network, wherein the number of input layer units is k, and the number of output layer units is 1;
the constructed gated dual-attention unit neural network is as follows:
r t ＝σ(U r x t +W r h t-1 +b r )
z t ＝σ(U z x t +W z h t-1 +b z )
wherein r is t Indicating reset gate output, z t Representing an update gate output; x is x t Inputting information, h, for gating the neural network of the double-attention unit at the moment t t-1 Hiding state information for the last moment; w (W) r And U r To reset the gate weight matrix, W z And U z To update the gate weight matrix, W h And U h B is a candidate state weight matrix r ,b z ,b h Is a bias matrix;for attention gate output, A t Attention profile for reset gate and update gate;
in the constructed gated dual-attention unit neural network, the first attention gate is at each time dimensionDegree input data x is processed using an attention mechanism t And recursive data h t-1 The specific calculation formula is as follows:
s(x t ,h t-1 )＝V T tanh(W s x t +U s h t-1 )
wherein V is an additional parameter related to the attention gate, W s And U s For the weight matrix associated with the scoring function,and->For a weight matrix related to the attention gate output,/a>For attention gate output, α t Is an attention distribution vector;
in the constructed gated double-attention unit neural network, the second attention gate processes the reset gate and the update gate output by using sigmoid and Tanh functions in each time dimension, and the specific calculation formula is as follows:
wherein,is a weight matrix>Is a bias matrix; />Calculating through a sigmoid function, and representing the attention ratio of the reset gate and the update gate output; />Calculating candidate attention values by using a Tanh function, and representing the contribution degree of information to prediction;
s6: taking k rows in front of the matrix W as the input of the gating dual-attention unit neural network, and taking the last row as the output of the gating dual-attention unit neural network to train the network;
s7: taking k outputs of the reciprocal as network inputs to obtain the output of the next moment;
s8: repeating step S7, when the preset threshold value is exceeded after the inverse normalization is output, the predicted sampling point number minus m is multiplied by the sum T of the vibration signal interval time and the sampling time s And +T is the residual life of the bearing.
2. The method for predicting the residual life of a bearing based on a gated dual-attention cell neural network according to claim 1, wherein in the step S3, the normalized training vector S adopts a linear normalization method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010478293.XA CN111581892B (en) | 2020-05-29 | 2020-05-29 | Bearing residual life prediction method based on GDAU neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010478293.XA CN111581892B (en) | 2020-05-29 | 2020-05-29 | Bearing residual life prediction method based on GDAU neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111581892A CN111581892A (en) | 2020-08-25 |
| CN111581892B true CN111581892B (en) | 2024-02-13 |
Family
ID=72123630
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010478293.XA Active CN111581892B (en) | 2020-05-29 | 2020-05-29 | Bearing residual life prediction method based on GDAU neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111581892B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112347898B (en) * | 2020-11-03 | 2024-04-09 | 重庆大学 | Rolling bearing health index construction method based on DCAE neural network |
| CN112762100B (en) * | 2021-01-14 | 2021-08-10 | 哈尔滨理工大学 | Bearing full-life-cycle monitoring method based on digital twinning |
| CN112818870A (en) * | 2021-02-03 | 2021-05-18 | 浙江大学 | Method for predicting residual life of bearing based on gated neural network framework |
| CN113505535B (en) * | 2021-07-08 | 2023-11-10 | 重庆大学 | Motor life prediction method based on gating self-adaptive hierarchical attention unit network |
| CN114412685B (en) * | 2021-12-23 | 2024-01-26 | 黄河水利水电开发集团有限公司 | Bearing temperature analysis and anomaly monitoring method for hydroelectric generating set |
| CN115017937B (en) * | 2022-04-28 | 2024-05-14 | 重庆大学 | Gear residual service life prediction method based on double-thread gating circulation unit |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109084980A (en) * | 2018-10-10 | 2018-12-25 | 北京交通大学 | Bearing fault prediction technique and device based on equalization segmentation |
| CN110085249A (en) * | 2019-05-09 | 2019-08-02 | 南京工程学院 | The single-channel voice Enhancement Method of Recognition with Recurrent Neural Network based on attention gate |
| CN110175425A (en) * | 2019-05-31 | 2019-08-27 | 重庆大学 | A kind of prediction technique of the gear remaining life based on MMALSTM |
| CN110210126A (en) * | 2019-05-31 | 2019-09-06 | 重庆大学 | A kind of prediction technique of the gear remaining life based on LSTMPP |
| CN110458278A (en) * | 2019-04-19 | 2019-11-15 | 四川大学 | Degradation prediction method based on quantum attention loop coding decoding neural network |
| CN110705181A (en) * | 2019-10-13 | 2020-01-17 | 重庆交通大学 | Rolling bearing residual life prediction method based on convolution length-time memory cyclic neural network |
| CN111595583A (en) * | 2020-05-29 | 2020-08-28 | 重庆大学 | GAU neural network-based bearing residual life prediction method |
-
2020
- 2020-05-29 CN CN202010478293.XA patent/CN111581892B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109084980A (en) * | 2018-10-10 | 2018-12-25 | 北京交通大学 | Bearing fault prediction technique and device based on equalization segmentation |
| CN110458278A (en) * | 2019-04-19 | 2019-11-15 | 四川大学 | Degradation prediction method based on quantum attention loop coding decoding neural network |
| CN110085249A (en) * | 2019-05-09 | 2019-08-02 | 南京工程学院 | The single-channel voice Enhancement Method of Recognition with Recurrent Neural Network based on attention gate |
| CN110175425A (en) * | 2019-05-31 | 2019-08-27 | 重庆大学 | A kind of prediction technique of the gear remaining life based on MMALSTM |
| CN110210126A (en) * | 2019-05-31 | 2019-09-06 | 重庆大学 | A kind of prediction technique of the gear remaining life based on LSTMPP |
| CN110705181A (en) * | 2019-10-13 | 2020-01-17 | 重庆交通大学 | Rolling bearing residual life prediction method based on convolution length-time memory cyclic neural network |
| CN111595583A (en) * | 2020-05-29 | 2020-08-28 | 重庆大学 | GAU neural network-based bearing residual life prediction method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111581892A (en) | 2020-08-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111581892B (en) | Bearing residual life prediction method based on GDAU neural network | |
| CN111595583B (en) | GAU neural network-based bearing residual life prediction method | |
| CN109931678B (en) | Air conditioner fault diagnosis method based on deep learning LSTM | |
| CN112347898B (en) | Rolling bearing health index construction method based on DCAE neural network | |
| Cui et al. | Quantitative fault analysis of roller bearings based on a novel matching pursuit method with a new step-impulse dictionary | |
| CN112488235A (en) | Elevator time sequence data abnormity diagnosis method based on deep learning | |
| CN110132554B (en) | Rotary machine fault diagnosis method based on deep Laplace self-coding | |
| Di et al. | Ensemble deep transfer learning driven by multisensor signals for the fault diagnosis of bevel-gear cross-operation conditions | |
| CN110610035A (en) | Rolling bearing residual life prediction method based on GRU neural network | |
| CN107991097A (en) | A kind of Method for Bearing Fault Diagnosis based on multiple dimensioned symbolic dynamics entropy | |
| CN112651119B (en) | Multi-performance parameter acceleration degradation test evaluation method for space harmonic reducer | |
| CN113642754A (en) | Complex industrial process fault prediction method based on RF noise reduction self-coding information reconstruction and time convolution network | |
| CN115982896B (en) | Bearing retainer service life detection method and device | |
| CN110472349A (en) | A kind of hot-rolled steel performance prediction method based on EEMD and depth convolutional network | |
| CN111783242A (en) | RVM-KF-based rolling bearing residual life prediction method and device | |
| CN113933054B (en) | Rolling bearing fault monitoring and diagnosing method, equipment, medium and processor | |
| Jin et al. | Fault diagnosis of rotating machines based on EEMD-MPE and GA-BP | |
| Cheng et al. | Interactive hybrid model for remaining useful life prediction with uncertainty quantification of bearing in nuclear circulating water pump | |
| CN116304587A (en) | Rolling bearing degradation trend prediction method based on CAE and AGRU | |
| Yi et al. | On a prediction method for remaining useful life of rolling Bearings via VMD-based dispersion entropy and GAN | |
| CN113609685B (en) | Bearing residual life prediction method based on optimized RVM and mixed degradation model | |
| CN113052060B (en) | Bearing residual life prediction method and device based on data enhancement and electronic equipment | |
| Chen et al. | Prediction model based on XGBoost for mechanical properties of steel materials | |
| CN114252266A (en) | Rolling bearing performance degradation evaluation method based on DBN-SVDD model | |
| Jiang et al. | Multi-feature based LSTM rolling bearing remaining useful life prediction |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |