Big Data-Driven Intelligent Fault Diagnosis and Prognosis for Mechanical Systems

Big Data-Driven Intelligent Fault Diagnosis and Prognosis for Mechanical Systems
اسم المؤلف
Yaguo Lei , Naipeng Li , Xiang Li
التاريخ
10 يناير 2023
المشاهدات
460
التقييم
(لا توجد تقييمات)
Loading...

Big Data-Driven Intelligent Fault Diagnosis and Prognosis for Mechanical Systems
Yaguo Lei , Naipeng Li , Xiang Li
Contents
1 Introduction and Background . 1
1.1 Introduction . 1
1.1.1 AI Technologies for Data Processing 4
1.1.2 Big Data-Driven Intelligent Predictive Maintenance . 5
1.1.3 Big Data Analytics Platform Practices 6
1.2 Overview of Big Data-Driven PHM . 9
1.2.1 Data Acquisition . 9
1.2.2 Data Processing 11
1.2.3 Diagnosis . 12
1.2.4 Prognosis . 13
1.2.5 Maintenance 15
1.3 Preface to Book Chapters 16
References . 18
2 Conventional Intelligent Fault Diagnosis 21
2.1 Introduction . 21
2.2 Typical Neural Network-Based Methods . 23
2.2.1 Introduction to Neural Networks . 23
2.2.2 Intelligent Diagnosis Using Radial Basis Function
Network 27
2.2.3 Intelligent Diagnosis Using Wavelet Neural Network 31
2.2.4 Epilog 37
2.3 Statistical Learning-Based Methods . 37
2.3.1 Introduction to Statistical Learning . 38
2.3.2 Intelligent Diagnosis Using Support Vector Machine 39
2.3.3 Intelligent Diagnosis Using Relevant Vector Machine . 49
2.3.4 Epilog 57
2.4 Conclusions . 57
References . 58
viiviii Contents
3 Hybrid Intelligent Fault Diagnosis . 61
3.1 Introduction . 61
3.2 Multiple WKNN Fault Diagnosis . 62
3.2.1 Motivation 62
3.2.2 Diagnosis Model Based on Combination of Multiple
WKNN . 63
3.2.3 Intelligent Diagnosis Case Study of Rolling Element
Bearings 67
3.2.4 Epilog 69
3.3 Multiple ANFIS Hybrid Intelligent Fault Diagnosis . 71
3.3.1 Motivation 71
3.3.2 Multiple ANFIS Combination with GA . 72
3.3.3 Fault Diagnosis Method Based on Multiple ANFIS
Combination 73
3.3.4 Intelligent Diagnosis Case of Rolling Element
Bearings 75
3.3.5 Epilog 80
3.4 A Multidimensional Hybrid Intelligent Method . 81
3.4.1 Motivation 81
3.4.2 Multiple Classifier Combination 82
3.4.3 Diagnosis Method Based on Multiple Classifier
Combination 84
3.4.4 Intelligent Diagnosis Case of Gearboxes . 87
3.4.5 Epilog 91
3.5 Conclusions . 91
References . 92
4 Deep Transfer Learning-Based Intelligent Fault Diagnosis . 95
4.1 Introduction . 95
4.2 Deep Belief Network for Few-Shot Fault Diagnosis . 98
4.2.1 Motivation 98
4.2.2 Deep Belief Network-Based Diagnosis Model
with Continual Learning 99
4.2.3 Few-Shot Fault Diagnosis Case of Industrial Robots 106
4.2.4 Epilog 110
4.3 Multi-Layer Adaptation Network for Fault Diagnosis
with Unlabeled Data 111
4.3.1 Motivation 111
4.3.2 Multi-Layer Adaptation Network-Based Diagnosis
Model 113
4.3.3 Fault Diagnosis Case of Locomotive Bearings
with Unlabeled Data 121
4.3.4 Epilog 125
4.4 Deep Partial Adaptation Network for Domain-Asymmetric
Fault Diagnosis 126Contents ix
4.4.1 Motivation 126
4.4.2 Deep Partial Transfer Learning Net-Based Diagnosis
Model 127
4.4.3 Partial Transfer Diagnosis of Gearboxes with Domain
Asymmetry . 136
4.4.4 Epilog 142
4.5 Instance-Level Weighted Adversarial Learning for Open-Set
Fault Diagnosis 144
4.5.1 Motivation 144
4.5.2 Instance-Level Weighted Adversarial Learning-Based
Diagnosis Model . 146
4.5.3 Fault Diagnosis Case of Rolling Bearing Datasets . 151
4.5.4 Epilog 161
4.6 Conclusions . 163
References . 164
5 Data-Driven RUL Prediction 167
5.1 Introduction . 167
5.2 Deep Separable Convolutional Neural Network-Based RUL
Prediction . 169
5.2.1 Motivation 169
5.2.2 Deep Separable Convolutional Network . 169
5.2.3 Architecture of DSCN 170
5.2.4 RUL Prediction Case of Accelerated Degradation
Experiments of Rolling Element Bearings . 173
5.2.5 Epilog 180
5.3 Recurrent Convolutional Neural Network-Based RUL
Prediction . 181
5.3.1 Motivation 181
5.3.2 Recurrent Convolutional Neural Network 181
5.3.3 Architecture of RCNN 182
5.3.4 RUL Prediction Case Study of FEMTO-ST
Accelerated Degradation Tests of Rolling Element
Bearings 188
5.3.5 Epilog 194
5.4 Multi-scale Convolutional Attention Network-Based RUL
Prediction . 195
5.4.1 Motivation 195
5.4.2 Multi-scale Convolutional Attention Network 195
5.4.3 Architecture of MSCAN 196
5.4.4 RUL Prediction Case of a Life Testing of Milling
Cutters . 202
5.4.5 Epilog 207
5.5 Conclusions . 208
References . 209x Contents
6 Data-Model Fusion RUL Prediction 213
6.1 Introduction . 213
6.2 RUL Prediction with Random Fluctuation Variability 215
6.2.1 Motivation 215
6.2.2 RUL Prediction Considering Random Fluctuation
Variability 216
6.2.3 RUL Prediction Case of FEMTO-ST Accelerated
Degradation Tests of Rolling Element Bearings . 222
6.2.4 Epilog 227
6.3 RUL Prediction with Unit-to-Unit Variability . 227
6.3.1 Motivation 227
6.3.2 RUL Prediction Model Considering Unit-to-Unit
Variability 229
6.3.3 RUL Prediction Case of Turbofan Engine Degradation
Dataset . 237
6.3.4 Epilog 239
6.4 RUL Prediction with Time-Varying Operational Conditions 241
6.4.1 Motivation 241
6.4.2 RUL Prediction Model Considering Time-Varying
Operational Conditions . 243
6.4.3 RUL Prediction Case of Accelerated Degradation
Experiments of Thrusting Bearings . 252
6.4.4 Epilog 255
6.5 RUL Prediction with Dependent Competing Failure Processes 256
6.5.1 Motivation 256
6.5.2 RUL Prediction Model Considering Dependent
Competing Failure Processes 258
6.5.3 RUL Prediction Case of Accelerated Degradation
Experiments of Rolling Element Bearings . 270
6.5.4 Epilog 275
6.6 Conclusions . 275
References . 276
Glossary 279 About the Authors
Glossary
AACO Accumulative amplitudes of carrier orders
AC Alternating current
ADT Accelerated degradation test
AE Acoustic emission
AI Artificial intelligence
ANN Artificial neural network
AR Autoregressive
ARE Absolute relative error
ARMA Autoregressive moving average
AUC Area under the receiver operation characteristic curve
BFP Bearing fault in the planet gear
BM Brownian motion
BN Batch normalization
CaAE Capsule auto-encoder
CBM Condition-based maintenance
CD Contrastive divergence
CDET Compensation distance evaluation technique
CDF Cumulative distribution function
CLSTM Convolutional long short-term memory
CNC Computer numerical control
CNN Convolutional neural network
CPG Crack in the planetary gear
CS Crack in the sun gear
CWRU Case Western Reserve University
DAFD Domain adaptation for fault diagnosis
DAN Deep adaptation network
DAQ Data acquisition
DBN Deep belief network
DBNCL Deep belief network with continual learning
DCFP Dependent competing failure process
DCN Deep convolutional network
DDC Deep domain confusion
DDL Dynamic dense layer
DNN Deep neural network
DPS Degradation process simulation
DPTLN Deep partial transfer learning network
DSCN Deep separable convolutional network
EEMD Ensemble empirical mode decomposition
EM Expectation maximization
EMD Empirical mode decomposition
ERDS Energy ratio based on difference spectrum
ERM Empirical risk minimization
FCL Fully-connected layer
FFT Fast Fourier transform
FHT First hitting time
FPT First predicting time
FT Failure threshold
GAN Generative adversarial network
GAP Global average pooling
GA Genetic algorithm
GMP Global max pooling
HI Health indicator
HPP Homogeneous Poisson process
i.i.d. Independent identically distributed
IF Inner race failure
ILWAL Instance-level weighted adversarial learning
IMF Intrinsic mode function
IoT Internet of Things
KNN K nearest neighbor
KKT Karush-Kuhn-Tucker
KL Kullback-Leibler
LOESS Locally weighted scatter smoothing method
mAP Mean average precision
MCNN Multi-scale convolutional neural network
MLAN Multi-layer adaptation network
MLP Multi-layer perceptron
MMD Maximum mean discrepancy
MPE Multi-scale displacement entropy
MSCAN Multi-scale convolutional attention network
MSE Mean square error
MRVM Multiclass relevance vector machine
OAA One-against-all
OAO One-against-one
OF Outer race failure
OSVM Open set support vector machine
PCA Principal component analysisGlossary 281
PDF Probability density function
PE Permutation entropy
PHM Prognostics and health management
RBF Radial basis function
RBM Restricted Boltzmann machine
RCNN Recurrent convolutional neural network
RDN Residual dense network
ReLU Rectified linear unit
ResNet Residual network
RF Rolling element failure
RKHS Reproducing Hilbert space
RMS Root mean square
RMSE Root mean square error
RUL Remaining useful life
RVM Relevance vector machine
SD Standard deviation
SE Squeeze and excitation
SLT Statistical learning theory
SNR Signal-to-noise ratio
SRM Structural risk minimization
SPRO Spectrum peak ratio of bearing outer race
SPRI Spectrum peak ratio of bearing inner race
SPRR Spectrum peak ratio of bearing roller
STFT Short-time Fourier transform
SVM Support vector machine
SW Wear in the sun gear
TCA Transfer component analysis
TD Temporal dimension
TSA Time synchronous average
t-SNE T-distributed stochastic neighbor embedding
UMLE Unit maximum likelihood estimation
UtUV Unit-to-unit variability
VC Vapnik-Chervonenkis
WKNN Weighted K nearest neighbor
WNN Wavelet neural network
WPM Wiener process-based method
WPT Wavelet packet transform
WPTLS Wavelet packet transform with the lifting scheme
WPW Wear in the planetary gear
WS Wear in the sun gear

كلمة سر فك الضغط : books-world.net
The Unzip Password : books-world.net

تحميل

يجب عليك التسجيل في الموقع لكي تتمكن من التحميل
تسجيل | تسجيل الدخول

التعليقات

اترك تعليقاً