Intelligent Fault Diagnosis and Prognosis for Engineering Systems

by Vachtsevanos, George; Lewis, Frank L.; Roemer, Michael; Hess, Andrew; Wu, Biqing

• ISBN13: 9780471729990
• ISBN10: 047172999X
• Edition: 1st
• Format: Hardcover
• Copyright: 2006-09-29
• Publisher: Wiley

Summary

Expert guidance on theory and practice in condition-based intelligent machine fault diagnosis and failure prognosis Intelligent Fault Diagnosis and Prognosis for Engineering Systems gives a complete presentation of basic essentials of fault diagnosis and failure prognosis, and takes a look at the cutting-edge discipline of intelligent fault diagnosis and failure prognosis technologies for condition-based maintenance. It thoroughly details the interdisciplinary methods required to understand the physics of failure mechanisms in materials, structures, and rotating equipment, and also presents strategies to detect faults or incipient failures and predict the remaining useful life of failing components. Case studies are used throughout the book to illustrate enabling technologies. Intelligent Fault Diagnosis and Prognosis for Engineering Systems offers material in a holistic and integrated approach that addresses the various interdisciplinary components of the field--from electrical, mechanical, industrial, and computer engineering to business management. This invaluably helpful book: * Includes state-of-the-art algorithms, methodologies, and contributions from leading experts, including cost-benefit analysis tools and performance assessment techniques * Covers theory and practice in a way that is rooted in industry research and experience * Presents the only systematic, holistic approach to a strongly interdisciplinary topic

Author Biography

George Vachtsevanos, Phd, is Director of the Intelligent Control Systems Laboratory in the School of Electrical and Computer Engineering at Georgia Institute of Technology, in Atlanta, Georgia.

Frank L. Lewis, Phd, is Head of the Advanced Controls, Sensors, and MEMS Group in the Automation and Robotics Research Institute at The University of Texas at Arlington, in Fort Worth, Texas.

Michael Roemer, Phd, is Director of Engineering at Impact Technologies, LLC, in Rochester, New York.

Andrew Hess is Air System PHM Lead and Development Manager in the Joint Strike Fighter Program Office at Naval Air Systems Command, in Patuxent River, Maryland.

Biqing Wu, Phd, works on various topics of active disturbance control and CBM/PHM. She is currently serving as a research engineer at the Georgia Institute of Technology, in Atlanta, Georgia.

Table of Contents

PREFACE	xi
ACKNOWLEDGMENTS	xiii
PROLOGUE	xv
1 INTRODUCTION	1	(12)
1.1 Historical Perspective	1	(2)
1.2 Diagnostic and Prognostic System Requirements	3	(1)
1.3 Designing in Fault Diagnostic and Prognostic Systems	4	(1)
1.4 Diagnostic and Prognostic Functional Layers	5	(2)
1.5 Preface to Book Chapters	7	(5)
1.6 References	12	(1)
2 SYSTEMS APPROACH TO CBM/PHM	13	(43)
2.1 Introduction	13	(3)
2.2 Trade Studies	16	(2)
2.3 Failure Modes and Effects Criticality Analysis (FMECA)	18	(8)
2.4 System CBM Test-Plan Design	26	(2)
2.5 Performance Assessment	28	(9)
2.6 CBM/PHM Impact on Maintenance and Operations: Case Studies	37	(12)
2.7 CBM/PHM in Control and Contingency Management	49	(5)
2.8 References	54	(2)
3 SENSORS AND SENSING STRATEGIES	56	(39)
3.1 Introduction	56	(1)
3.2 Sensors	57	(20)
3.3 Sensor Placement	77	(3)
3.4 Wireless Sensor Networks	80	(9)
3.5 Smart Sensors	89	(2)
3.6 References	91	(4)
4 SIGNAL PROCESSING AND DATABASE MANAGEMENT SYSTEMS	95	(77)
4.1 Introduction	95	(1)
4.2 Signal Processing in CBM/PHM	96	(1)
4.3 Signal Preprocessing	97	(7)
4.4 Signal Processing	104	(24)
4.5 Vibration Monitoring and Data Analysis	128	(18)
4.6 Real-Time Image Feature Extraction and Defect/Fault Classification	146	(4)
4.7 The Virtual Sensor	150	(2)
4.8 Fusion or Integration Technologies	152	(10)
4.9 Usage-Pattern Tracking	162	(1)
4.10 Database Management Methods	163	(2)
4.11 References	165	(7)
5 FAULT DIAGNOSIS	172	(112)
5.1 Introduction	172	(3)
5.2 The Diagnostic Framework	175	(6)
5.3 Historical Data Diagnostic Methods	181	(10)
5.4 Data-Driven Fault Classification and Decision Making	191	(20)
5.5 Dynamic Systems Modeling	211	(16)
5.6 Physical Model–Based Methods	227	(8)
5.7 Model-Based Reasoning	235	(8)
5.8 Case-Based Reasoning (CBR)	243	(15)
5.9 Other Methods for Fault Diagnosis	258	(9)
5.10 A Diagnostic Framework for Electrical/Electronic Systems	267	(4)
5.11 Case Study: Vibration-Based Fault Detection and Diagnosis for Engine Bearings	271	(8)
5.12 References	279	(5)
6 FAULT PROGNOSIS	284	(71)
6.1 Introduction	284	(5)
6.2 Model-Based Prognosis Techniques	289	(12)
6.3 Probability-Based Prognosis Techniques	301	(11)
6.4 Data-Driven Prediction Techniques	312	(21)
6.5 Case Studies	333	(18)
6.6 References	351	(4)
7 FAULT DIAGNOSIS AND PROGNOSIS PERFORMANCE METRICS	355	(45)
7.1 Introduction	355	(1)
7.2 CBM/PHM Requirements Definition	356	(2)
7.3 Feature-Evaluation Metrics	358	(4)
7.4 Fault Diagnosis Performance Metrics	362	(23)
7.5 Prognosis Performance Metrics	385	(10)
7.6 Diagnosis and Prognosis Effectiveness Metrics	395	(1)
7.7 Complexity/Cost-Benefit Analysis of CBM/PHM Systems	396	(2)
7.8 References	398	(2)
8 LOGISTICS: SUPPORT OF THE SYSTEM IN OPERATION	400	(17)
8.1 Introduction	400	(1)
8.2 Product-Support Architecture, Knowledge Base, and Methods for CBM	401	(3)
8.3 Product Support without CBM	404	(1)
8.4 Product Support with CBM	405	(4)
8.5 Maintenance Scheduling Strategies	409	(1)
8.6 A Simple Example	410	(6)
8.7 References	416	(1)
APPENDIX	417	(7)
INDEX	424