Machinery Prognostics and Prognosis Oriented Maintenance Management

This book gives a complete presentatin of the basic essentials of machinery prognostics and prognosis oriented maintenance management, and takes a look at the cutting-edge discipline of intelligent failure prognosis technologies for condition-based maintenance.

• Presents an introduction to advanced maintenance systems, and discusses the key technologies for advanced maintenance by providing readers with up-to-date technologies

• Offers practical case studies on performance evaluation and fault diagnosis technology, fault prognosis and remaining useful life prediction and maintenance scheduling, enhancing the understanding of these technologies

• Pulls togeter recent developments and varying methods into one volume, complemented by practical examples to provide a complete reference

Autorentext

*Jihong Yan, Professor and Head of Department of Industrial Engineering, Harbin Institute of Technology, China
Professor Yan has been working in the area of intelligent maintenance for over ten years, starting at the Centre for Intelligent Maintenance Systems (IMS) funded by NSF in the US as a researcher for three years, mainly focused on prognosis algorithm development. He then joined Pennsylvania State University in 2004 to work on personnel cross training related topics. From 2005 to the present he is a Professor at Harbin Institute of Technology, China. Professor Yan's research is focused on advanced maintenance of machinery, such as online condition monitoring, signal data pre-processing, feature extraction, reliability and performance evaluation, fault diagnosis, fault prognosis and remaining useful life prediction, and maintenance scheduling.

Inhalt
About the Author xi

Preface xiii

Acknowledgements xv

1 Introduction 1

1.1 Historical Perspective 1

1.2 Diagnostic and Prognostic System Requirements 2

1.3 Need for Prognostics and Sustainability-Based Maintenance Management 3

1.4 Technical Challenges in Prognosis and Sustainability-Based Maintenance Decision-Making 4

1.5 Data Processing, Prognostics, and Decision-Making 7

1.6 Sustainability-Based Maintenance Management 9

1.7 Future of Prognostics-Based Maintenance 11

References 12

2 Data Processing 13

2.1 Probability Distributions 13

2.1.1 Uniform Distribution 14

2.1.2 Normal Distribution 16

2.1.3 Binomial Distribution 18

2.1.4 Geometric Distribution 19

2.1.5 Hyper-Geometric Distribution 21

2.1.6 Poisson Distribution 22

2.1.7 Chi-Squared Distributions 24

2.2 Statistics on Unordered Data 25

2.2.1 Treelets Analysis 26

2.2.2 Clustering Analysis 28

2.3 Statistics on Ordered Data 32

2.4 Technologies for Incomplete Data 33

References 34

3 Signal Processing 37

3.1 Introduction 37

3.2 Signal Pre-Processing 38

3.2.1 Digital Filtering 38

3.2.2 Outlier Detecting 39

3.2.3 Signal Detrending 41

3.3 Techniques for Signal Processing 42

3.3.1 Time-Domain Analysis 42

3.3.2 Spectrum Analysis 44

3.3.3 Continuous Wavelet Transform 46

3.3.4 Discrete Wavelet Transform 49

3.3.5 Wavelet Packet Transforms 51

3.3.6 Empirical Mode Decomposition 51

3.3.7 Improved Empirical Mode Decomposition 57

3.4 Real-Time Image Feature Extraction 67

3.4.1 Image Capture System 67

3.4.2 Image Feature Extraction 68

3.5 Fusion or Integration Technologies 72

3.5.1 DempsterShafer Inference 72

3.5.2 Fuzzy Integral Fusion 73

3.6 Statistical Pattern Recognition and Data Mining 74

3.6.1 Bayesian Decision Theory 74

3.6.2 Artificial Neural Networks 76

3.6.3 Support Vector Machine 79

3.7 Advanced Technology for Feature Extraction 85

3.7.1 Group Technology 87

3.7.2 Improved Algorithm of Group Technology 88

3.7.3 Numerical Simulation of Improved Group Algorithm 90

3.7.4 Group Technology for Feature Extraction 91

3.7.5 Application 92

References 96

4 Health Monitoring and Prognosis 101

4.1 Health Monitoring as a Concept 101

4.2 Degradation Indices 101

4.3 Real-Time Monitoring 106

4.3.1 Data Acquisition 106

4.3.2 Data Processing Techniques 115

4.3.3 Example 120

4.4 Failure Prognosis 126

4.4.1 Classification and Clustering 129

4.4.2 Mathematical Model of the Classification Method 130

4.4.3 Mathematical Model of the Fuzzy C-Means Method 130

4.4.4 Theory of Ant Colony Clustering Algorithm 133

4.4.5 Improved Ant Colony Clustering Algorithm 134

4.4.6 Intelligent Fault Diagnosis Method 138

4.5 Physics-Based Prognosis Models 141

4.5.1 Model-Based Methods for Systems 142

4.6 Data-Driven Prognosis Models 144

4.7 Hybrid Prognosis Models 147

References 149

5 Prediction of Remaining Useful Life 153

5.1 Formulation of Problem 153

5.2 Methodology of Probabilistic Prediction 154

5.2.1 Theory of Weibull Distribution 155

5.2.2 Bayesian Theorem 157

5.3 Dynamic Life Prediction Using Time Series 160

5.3.1 General Introduction 160

5.3.2 Prediction Models 162

5.3.3 Applications 173

5.4 Remaining Life Prediction by the Crack-Growth Criterion 176

References 181

6 Maintenance Planning and Scheduling 183 6.1 S...