CN114880767B - Aero-engine residual service life prediction method based on attention mechanism Dense-GRU network - Google Patents
Aero-engine residual service life prediction method based on attention mechanism Dense-GRU network Download PDFInfo
- Publication number
- CN114880767B CN114880767B CN202210434877.6A CN202210434877A CN114880767B CN 114880767 B CN114880767 B CN 114880767B CN 202210434877 A CN202210434877 A CN 202210434877A CN 114880767 B CN114880767 B CN 114880767B
- Authority
- CN
- China
- Prior art keywords
- layer
- model
- output
- sensor
- attention mechanism
- 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
- 230000007246 mechanism Effects 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 238000006731 degradation reaction Methods 0.000 claims description 36
- 230000015556 catabolic process Effects 0.000 claims description 27
- 230000004913 activation Effects 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000008034 disappearance Effects 0.000 abstract description 3
- 238000012545 processing Methods 0.000 abstract description 3
- 239000012141 concentrate Substances 0.000 abstract 2
- 230000006870 function Effects 0.000 description 17
- 230000008569 process Effects 0.000 description 6
- 238000012360 testing method Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001568 sexual effect Effects 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]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- 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
- 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]
-
- 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
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an aeroengine residual service life prediction method based on an attention mechanism Dense-GRU network, which comprises the following steps: step one, data preprocessing; step two, constructing an attention mechanism layer; step three, constructing an integral model and training; step four, predicting the residual service life by using a model; the method provides a concentrate-GRU network based on a concentrate mechanism, and the proposed model adds the concentrate mechanism in processing multidimensional sensor data of an engine, so that the network model pays more attention to a sensor which is more effective in prediction. The network main body part is a Dense-GRU network, so that the propagation and reuse of the multidimensional time series characteristics are enhanced, the problem of gradient disappearance of a time series deep network is effectively avoided, and the accuracy of residual service life prediction is effectively improved.
Description
Technical Field
The invention belongs to the technical field of prediction of residual life of an aeroengine, and particularly relates to a method for predicting residual life of an aeroengine based on an attention mechanism Dense-GRU network.
Background
The residual service life prediction is to estimate the residual service life of the engine by analyzing the trend of the historical performance decline of the engine, the accurate residual service life prediction can improve the reliability and the safety of equipment or a system, prevent fatal faults and reduce maintenance cost, and support the establishment of a long-term maintenance plan of an engine fleet for an airline company.
Currently, the remaining life predictions fall into two directions: based on a physical degradation model and based on data driving. The residual life prediction is mostly described using mathematical processes based on a physical degradation model, for example, using algebraic equations and differential equations to describe the degradation process. This approach has too high a priori knowledge dependence on the plant degradation process, and is difficult to accurately describe and model by mathematical models due to the complexity and randomness of the system degradation process. Based on the data driving method, a model is built to explore the potential relation between the sensor monitoring data and the system degradation state based on the monitoring data of the system state. With the development of computer and sensor technologies, data-driven based methods are becoming the mainstream in the field of residual life prediction research, which can effectively reduce research costs and mine complex relationships between degradation data and residual life.
With great success of deep learning in the field of computer vision, the research method of residual service life prediction is gradually changed from the traditional machine learning method to the deep learning method, and the neural network method is gradually the main stream of the residual service life prediction method based on data driving. At present, the neural network predicts the remaining service life and has the following problems: (1) The monitoring data of the aeroengine are time series data of a plurality of sensors, and when the multi-element time series is processed, the importance degree among the sensors is often ignored, so that the accuracy of prediction is reduced. (2) Deep neural networks mostly adopt a deep structure with stacked layers, and the structure can only predict the residual service life by using the extracted advanced features of the last layer, and the features of other layers are ignored, so that information is lost. In order to solve the problems, an aeroengine residual service life prediction method based on an attention mechanism Dense-GRU network is provided.
Disclosure of Invention
The invention aims to solve the technical problem of providing an aeroengine residual service life prediction method based on an attention mechanism Dense-GRU network aiming at the defects in the prior art. Firstly, the proposed model adds an attention mechanism in processing multidimensional sensor data of an engine, so that the weight of the sensor data with great influence on the prediction of the residual service life is improved, and the network model is more aware of the sensor with more effective prediction. And secondly, the network main part is a Dense-GRU network, so that the propagation and reuse of the multidimensional time series characteristics are enhanced, and meanwhile, the gradient disappearance problem of a time series deep network is effectively avoided. The method fully considers the influence of the difference of the sensors on the prediction of the residual service life, fully excavates and utilizes the degradation information of the multi-dimensional sensor sequence, and effectively improves the accuracy of the prediction of the residual service life.
In order to solve the technical problems, the invention adopts the following technical scheme: an aeroengine residual service life prediction method based on an attention mechanism Dense-GRU network is characterized by comprising the following steps:
step one, data preprocessing:
step 101: the degradation process of the engine is provided with N times, and m sensor degradation sequence data X in the complete degradation process are input o1 ＝[X 1 ,X 2 ,…,X m ]The degradation sequence of the jth sensor is expressed asEliminating sensor data with small degree of correlation with performance degradation process or basically unchanged whole degradation period parameters, and finally reserving k pieces of sensor degradation data X o2 ＝[X 1 ,X 2 ,…,X k ]K is smaller than m;
step 102: normalizing the input k sensor degradation data according to the formulaCalculating normalized data X of each sensor nor ＝[X 1 ,X 2 ,…,X k ]Wherein the j-th sensor normalized data is +.>
Step 103: for the normalized k sensor data X nor ＝[X 1 ,X 2 ,…,X k ]Training sample division is performed. Dividing samples by adopting a sliding window method, wherein the width of the samples is the dimension k of the sensor, the length of the samples is n, and the samples are expressed as X= [ X ] 1 ,X 2 ,…,X k ]WhereinThe residual life value corresponding to each sample is the residual life value Y at the last moment of the sample RUL ；
Step two, constructing an attention mechanism layer:
step 201: according to formula h 1 ＝ReLU(W 1 ·X+b 1 ) Establishing a first linear layer of the attention mechanism layer, wherein x= [ X ] 1 ,X 2 ,…,X k ]To input samples, h 1 For the output of the first linear layer, W 1 And b 1 The weight and bias of the first linear layer are represented, and the ReLU function is adopted as the activation function;
step 202: according to formula h 2 ＝ReLU(W 2 ·h 1 +b 2 ) Establishing a second linear layer of the attention mechanism layer, wherein h 1 For the output of the first linear layer, h 2 For the output of the second linear layer, W 2 And b 2 The weight and bias of the second linear layer are represented, and the ReLU function is adopted as the activation function;
step 203: according to formula W x ＝softmax(h 2 ) Establishing a weight generation layer of the attention mechanism layer, wherein h 2 For the output of the second linear layer, W x Representing the generated weight, wherein a softmax function is adopted as a calculation function; according to formula X att ＝W x X weighting the initial input samples with the weights generated, ultimately generating the output X of the attention mechanism layer att ；
Step three, building an overall model and training:
step 301: the integral model comprises L layers of GRU unit layers, and the L layers of GRU units are connected in a dense connection mode; according to the formulaCalculating the output of the d-layer GRU unit, wherein the input of the current GRU layer is the output of the previous d-1 layer, d=1, 2, … L, and finally calculating to obtain the output of the last layer GRU unit>
Step 302: according to the formulaCalculating a final residual life prediction value, whereinFor the output of the attention mechanism Dense-GRU model, FC represents the fully connected layer,/->Representing a final residual life prediction value;
step 303: according to the formulaTraining loss of the calculation model, wherein ∈>Representing the predicted output of the model, Y t Representing the real output of the samples, S representing the number of samples; the model activation function selects a ReLU function, and the optimizer is selected as a RMSProp optimizer;
step four, predicting the residual service life by using a model:
step 401: let the sample to be predicted be X pre ＝[X 1 ,X 2 ,…,X k ]The degradation sequence of the jth sensor is expressed asWherein T represents the length of the sample; normalizing the sample data according to step 102, matching the sample size to the model input size according to step 103, generating an input sample X' pre ＝[X 1 ',X' 2 ,…X' k ]The degradation sequence of the jth sensor is denoted +.>Where n represents the length of the model input sample;
step 402: sample X 'to be tested' pre ＝[X 1 ',X' 2 ,…X' k ]Inputting the model into the trained model in the third step, and calculating to obtainPredicting residual service life value of aero-engine at time T+1
Compared with the prior art, the invention has the following advantages:
1. the model provided by the invention adds an attention mechanism in processing multidimensional sensor data, so that the weight of the sensor data with great influence on the prediction of the residual service life is improved, and the network model is more aware of the sensor with more effective prediction.
2. The network model main body part is based on GRU units, and combines the densely connected Dense structure and the GRU units to form a Dense-GRU network, so that the degradation information of the multidimensional sensor sequence is fully mined and utilized, and the propagation and reuse of the characteristics are enhanced.
In summary, the attention mechanism-based Dense-GRU network model fully considers the influence of the difference of the sensors on the prediction of the residual service life, and meanwhile, the network model effectively digs and utilizes the degradation information of the multidimensional sensor sequence, enhances the propagation and reuse of the characteristics, effectively avoids the gradient disappearance problem of the time sequence deep network and provides a new scheme for the prediction of the residual service life of the aeroengine based on data driving.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
FIG. 1 is a block flow diagram of the present invention
FIG. 2 is a layer diagram of an attention mechanism
FIG. 3 is a diagram showing the overall structure of the model
FIG. 4 is a box plot of experimental MAE
FIG. 5 is a sample set generated from the FD001 data set
FIG. 6 shows the experimental results of different methods
Detailed Description
The process of the present invention is described in further detail below with reference to examples.
As shown in fig. 1, the present invention includes the steps of:
step one, data preprocessing:
to verify the validity of the proposed method, the verification was performed using the C-MAPSS dataset disclosed by NASA. The FD001 dataset in the C-MAPSS dataset was used for training models and testing.
Step 101: for the normalized k sensor data X nor ＝[X 1 ,X 2 ,…,X k ]Training sample division is performed. Dividing samples by adopting a sliding window method, wherein the width of the samples is the dimension k of the sensor, the length of the samples is n, and the samples are expressed as X= [ X ] 1 ,X 2 ,…,X k ]WhereinThe residual life value corresponding to each sample is the residual life value Y at the last moment of the sample RUL ；
In the actual use process, the window length selected by the sliding window method is the minimum length of incomplete degradation data of the engines in the test set, namely, the length with the shortest degradation period is selected from the incomplete degradation data of 100 aeroengines in the test set to be used as the sliding window length of the generated sample. In the FD001 dataset, the sensors were finally screened as 16-dimensional sensors, and the shortest length selected for the test set was 31 flight cycles, so the length of the resulting samples was 31, the dimension was 16, and the sample set was as shown in fig. 5.
Step two, constructing an attention mechanism layer:
step 201: according to formula h 1 ＝ReLU(W 1 ·X+b 1 ) Establishing a first linear layer of the attention mechanism layer, wherein x= [ X ] 1 ,X 2 ,…,X k ]To input samples, h 1 For the output of the first linear layer, W 1 And b 1 The weight and bias of the first linear layer are represented, and the ReLU function is adopted as the activation function;
step 202: according to formula h 2 ＝ReLU(W 2 ·h 1 +b 2 ) Establishing a second linear layer of the attention mechanism layer, wherein h 1 For the first lineOutput of sexual layer, h 2 For the output of the second linear layer, W 2 And b 2 The weight and bias of the second linear layer are represented, and the ReLU function is adopted as the activation function;
step 203: according to formula W x ＝softmax(h 2 ) Establishing a weight generation layer of the attention mechanism layer, wherein h 2 For the output of the second linear layer, W x Representing the generated weight, wherein a softmax function is adopted as a calculation function; according to formula X att ＝W x X weighting the initial input samples with the weights generated, ultimately generating the output X of the attention mechanism layer att ；
In the actual use process, as shown in fig. 2, the initial input passes through two linear layers with the activation function of ReLU, a series of weights are generated through softmax, and the finally generated weights and the initial input weight are used as the output of the attention mechanism layer.
Step three, calculating the state distance of the multisource sensor to represent the degradation state:
step 301: the integral model comprises L layers of GRU unit layers, and the L layers of GRU units are connected in a dense connection mode; according to the formulaCalculating the output of the d-layer GRU unit, wherein the input of the current GRU layer is the output of the previous d-1 layer, d=1, 2, … L, and finally calculating to obtain the output of the last layer GRU unit>
Step 302: according to the formulaCalculating a final residual life prediction value, whereinFor the output of the attention mechanism Dense-GRU model, FC represents the fully connected layer,/->Representing a final residual life prediction value;
step 303: according to the formulaTraining loss of the calculation model, wherein ∈>Representing the predicted output of the model, Y t Representing the real output of the samples, S representing the number of samples; the model activation function selects a ReLU function, and the optimizer is selected as a RMSProp optimizer;
in the actual use process, the whole model is shown in fig. 3, the network model adds an attention mechanism layer at the input end, GRU network is used as a basic unit in the network, dense connection structures are adopted among different GRU unit blocks to form a Dense-GRU, and the final output is used for obtaining a final residual service life value through a full connection layer.
Step four, predicting the residual service life by using a model:
step 401: let the sample to be predicted be X pre ＝[X 1 ,X 2 ,…,X k ]The degradation sequence of the jth sensor is expressed asWherein T represents the length of the sample; normalizing the sample data according to step 102, matching the sample size to the model input size according to step 103, generating an input sample X' pre ＝[X 1 ',X' 2 ,…X' k ]The degradation sequence of the jth sensor is denoted +.>Where n represents the length of the model input sample;
step 402: sample X 'to be tested' pre ＝[X 1 ',X' 2 ,…X' k ]Inputting the model into the trained model in the step three, and calculating to obtain the predicted aero-engine at the moment T+1Remaining service life value of machine
In the actual use process, an FD001 data set is utilized for experiments, classical methods such as a long-short time memory network (LSTM), a gate control loop unit network (GRU), a bidirectional long-short time memory network (Bi-LSTM) and a residual error network (Resnet) are selected as comparison experiments for verifying the effectiveness of the proposed method, an average absolute error MAE, a root mean square error RMSE and a Score index are selected as evaluation criteria, the Score index is defined for a model predicted by the residual life, and the smaller Score value indicates the higher accuracy of the model. Table 2 shows the experimental results of the different methods on the test set in the FD001 data set, with the method MAE of 18.4415, RMSE of 22.9736 and Score of 13.9694, which performed best. FIG. 4 depicts a box plot of MAE metrics for different models under the FD001 data set, where the box plot may represent the magnitude and dispersion of errors in the residual life predictions for the different models, the rectangular box and dashed line segments represent the degree of variation in MAE, the longer the rectangular box and dashed line segments indicate, the more dispersed the predictions for the model, i.e., the worse the prediction stability for the model, where the line in the middle of the rectangular box represents the median of the MAE, and the smaller the line values indicate that the prediction errors for the model are smaller. From the results, the validity of the invention is also demonstrated.
The foregoing is merely an embodiment of the present invention, and the present invention is not limited thereto, and any simple modification, variation and equivalent structural changes made to the foregoing embodiment according to the technical matter of the present invention still fall within the scope of the technical solution of the present invention.
Claims (1)
1. An aeroengine residual service life prediction method based on an attention mechanism Dense-GRU network is characterized by comprising the following steps:
step one, data preprocessing:
step 101: the degradation process of the engine is provided with N times, and the degradation sequence data of m sensors in the complete degradation process are inputX o1 ＝[X 1 ,X 2 ,…,X m ]The degradation sequence of the jth sensor is expressed asEliminating sensor data with small degree of correlation with performance degradation process or basically unchanged whole degradation period parameters, and finally reserving k pieces of sensor degradation data X o2 ＝[X 1 ,X 2 ,…,X k ]K is smaller than m;
step 102: normalizing the input k sensor degradation data according to the formulaCalculating normalized data X of each sensor nor ＝[X 1 ,X 2 ,…,X k ]Wherein the normalized data of the jth sensor is
Step 103: for the normalized k sensor data X nor ＝[X 1 ,X 2 ,…,X k ]Dividing training samples; dividing samples by adopting a sliding window method, wherein the width of the samples is the dimension k of the sensor, the length of the samples is n, and the samples are expressed as X= [ X ] 1 ,X 2 ,…,X k ]WhereinThe residual life value corresponding to each sample is the residual life value Y at the last moment of the sample RUL ；
Step two, constructing an attention mechanism layer:
step 201: according to formula h 1 ＝ReLU(W 1 ·X+b 1 ) Establishing a first linear layer of the attention mechanism layer, wherein x= [ X ] 1 ,X 2 ,…,X k ]To input samples, h 1 For the output of the first linear layer, W 1 And b 1 The weight and bias of the first linear layer are represented, and the ReLU function is adopted as the activation function;
step 202: according to formula h 2 ＝ReLU(W 2 ·h 1 +b 2 ) Establishing a second linear layer of the attention mechanism layer, wherein h 1 For the output of the first linear layer, h 2 For the output of the second linear layer, W 2 And b 2 The weight and bias of the second linear layer are represented, and the ReLU function is adopted as the activation function;
step 203: according to formula W x ＝softmax(h 2 ) Establishing a weight generation layer of the attention mechanism layer, wherein h 2 For the output of the second linear layer, W x Representing the generated weight, wherein a softmax function is adopted as a calculation function; according to formula X att ＝W x X weighting the initial input samples with the weights generated, ultimately generating the output X of the attention mechanism layer att ；
Step three, building an overall model and training:
step 301: the integral model comprises L layers of GRU unit layers, and the L layers of GRU units are connected in a dense connection mode; according to the formulaCalculating the output of the d-layer GRU unit, wherein the input of the current GRU layer is the output of the previous d-1 layer, d=1, 2, … L, and finally calculating to obtain the output of the last layer GRU unit>
Step 302: according to the formulaCalculating a final residual life prediction value, whereinFor the output of the attention mechanism Dense-GRU model, FC represents the fully connected layer,/->Representing a final residual life prediction value;
step 303: according to the formulaCalculating training loss of model, wherein Y t pre Representing the predicted output of the model, Y t Representing the real output of the samples, S representing the number of samples; the model activation function selects a ReLU function, and the optimizer is selected as a RMSProp optimizer;
step four, predicting the residual service life by using a model:
step 401: let the sample to be predicted be X pre ＝[X 1 ,X 2 ,…,X k ]The degradation sequence of the jth sensor is expressed asWherein T represents the length of the sample; normalizing the sample data according to step 102, matching the sample size to the model input size according to step 103, generating an input sample X' pre ＝[X' 1 ,X' 2 ,…X' k ]The degradation sequence of the jth sensor is denoted +.>Where n represents the length of the model input sample;
step 402: sample X 'to be tested' pre ＝[X' 1 ,X' 2 ,…X' k ]Inputting the residual service life value into the trained model in the step three, and calculating to obtain the residual service life value of the predicted aero-engine at the moment T+1。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210434877.6A CN114880767B (en) | 2022-04-24 | 2022-04-24 | Aero-engine residual service life prediction method based on attention mechanism Dense-GRU network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210434877.6A CN114880767B (en) | 2022-04-24 | 2022-04-24 | Aero-engine residual service life prediction method based on attention mechanism Dense-GRU network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114880767A CN114880767A (en) | 2022-08-09 |
| CN114880767B true CN114880767B (en) | 2024-03-08 |
Family
ID=82670736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210434877.6A Active CN114880767B (en) | 2022-04-24 | 2022-04-24 | Aero-engine residual service life prediction method based on attention mechanism Dense-GRU network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114880767B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116050665B (en) * | 2023-03-14 | 2024-04-02 | 淄博热力有限公司 | Heat supply equipment fault prediction method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807257A (en) * | 2019-11-04 | 2020-02-18 | 中国人民解放军国防科技大学 | Method for predicting residual life of aircraft engine |
| CN113051689A (en) * | 2021-04-25 | 2021-06-29 | 石家庄铁道大学 | Bearing residual service life prediction method based on convolution gating circulation network |
| CN114186475A (en) * | 2021-10-28 | 2022-03-15 | 南京工业大学 | Pivoting support service life prediction method based on Attention-MGRU |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3958185A1 (en) * | 2020-06-26 | 2022-02-23 | Tata Consultancy Services Limited | Neural networks for handling variable-dimensional time series data |
-
2022
- 2022-04-24 CN CN202210434877.6A patent/CN114880767B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807257A (en) * | 2019-11-04 | 2020-02-18 | 中国人民解放军国防科技大学 | Method for predicting residual life of aircraft engine |
| CN113051689A (en) * | 2021-04-25 | 2021-06-29 | 石家庄铁道大学 | Bearing residual service life prediction method based on convolution gating circulation network |
| CN114186475A (en) * | 2021-10-28 | 2022-03-15 | 南京工业大学 | Pivoting support service life prediction method based on Attention-MGRU |
Non-Patent Citations (2)
| Title |
|---|
| 基于改进GRU的航空发动机剩余寿命预测;车畅畅;王华伟;倪晓梅;付强;;航空计算技术;20200125(01) * |
| 基于注意力机制BiLSTM的设备智能故障诊断方法;王太勇;王廷虎;王鹏;乔卉卉;徐明达;;天津大学学报(自然科学与工程技术版);20200427(06) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114880767A (en) | 2022-08-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112131760B (en) | CBAM model-based prediction method for residual life of aircraft engine | |
| CN109611087B (en) | Volcanic oil reservoir parameter intelligent prediction method and system | |
| CN115758212B (en) | Mechanical equipment fault diagnosis method based on parallel network and transfer learning | |
| CN111190429B (en) | Unmanned aerial vehicle active fault-tolerant control method based on reinforcement learning | |
| CN114266278B (en) | Dual-attention network-based equipment residual service life prediction method | |
| CN114218872B (en) | DBN-LSTM semi-supervised joint model-based residual service life prediction method | |
| CN112147432A (en) | BiLSTM module based on attention mechanism, transformer state diagnosis method and system | |
| CN111680875A (en) | Unmanned aerial vehicle state risk fuzzy comprehensive evaluation method based on probability baseline model | |
| CN108805195A (en) | A kind of motor group method for diagnosing faults based on two-value deep-neural-network | |
| CN114880767B (en) | Aero-engine residual service life prediction method based on attention mechanism Dense-GRU network | |
| CN110874616A (en) | Transformer operation prediction method based on LSTM network and Markov chain correction error | |
| CN114239935A (en) | Prediction method for non-uniform track sequence | |
| Xu et al. | New RUL prediction method for rotating machinery via data feature distribution and spatial attention residual network | |
| CN114169091A (en) | Method for establishing prediction model of residual life of engineering mechanical part and prediction method | |
| CN115660198B (en) | Method for predicting residual service life of rolling bearing | |
| CN114841063A (en) | Aero-engine residual life prediction method based on deep learning | |
| CN112560252B (en) | Method for predicting residual life of aeroengine | |
| CN115048873A (en) | Residual service life prediction system for aircraft engine | |
| Wang et al. | Similarity-based echo state network for remaining useful life prediction | |
| CN114201914A (en) | Solid engine test data analysis method based on BP neural network | |
| Luo et al. | A novel method for remaining useful life prediction of roller bearings involving the discrepancy and similarity of degradation trajectories | |
| Ribeiro Machado da Silva et al. | Convolutional Neural Networks Applied to Flexible Pipes for Fatigue Calculations | |
| US20220188647A1 (en) | Model learning apparatus, data analysis apparatus, model learning method and program | |
| Liu et al. | Structural damage identification based on rough sets and artificial neural network | |
| CN114741948B (en) | Aero-engine degradation trend prediction method based on residual stacking convolution network of sequence reconstruction |
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 |