Data-driven technology for engineering systems health management : design approach, feature construction, fault diagnosis, prognosis, fusion and decisions

• Main Author: Niu, Gang (Author)
• Format: Electronic eBook
• Language: English
• Published: Singapore : Beijing, China : Springer ; Science Press, [2016]
• Subjects: Fault location (Engineering); Systems engineering.; elektronické knihy; electronic books
• ISBN: 9789811020322; 9789811020315
• Physical Description: 1 online resource (xiii, 357 pages) : illustrations

This book introduces condition-based maintenance (CBM)/data-driven prognostics and health management (PHM) in detail, first explaining the PHM design approach from a systems engineering perspective, then summarizing and elaborating on the data-driven methodology for feature construction, as well as...

Table of Contents:

• Preface; Acknowledgements; Contents; 1 Background of Systems Health Management; 1.1 Introduction; 1.2 Maintenance Strategy; 1.3 From Maintenance to PHM; 1.4 Definitions and Terms of Systems Health Management; 1.5 Preface to Book Chapters; References; 2 Design Approach for Systems Health Management; 2.1 Introduction; 2.2 Systems Engineering; 2.3 Systems Engineering, Dependability, and Health Management; 2.4 SHM Lifecycle Stages; 2.4.1 Research Stage; 2.4.2 Requirements Development Stage; 2.4.3 System/Functional Analysis; 2.4.4 Design, Synthesis, and Integration.
• 2.4.5 System Test and Evaluation2.4.6 HM System Maturation; 2.5 A Systems-Based Methodology for PHM/CBM Design; 2.6 A Proposed PHM Design Approach for Rotary Machinery Systems; References; 3 Overview of Data-Driven PHM; 3.1 Introduction; 3.2 PHM Technical Approaches; 3.3 Data-Driven PHM/CBM System Architecture; 3.4 Role of Condition Monitoring, Fault Diagnosis, and Prognosis; 3.5 Fault Diagnosis Framework; 3.6 Problems During Implementation; 3.7 Related Techniques; References; 4 Data Acquisition and Preprocessing; 4.1 Introduction; 4.2 Data Acquisition; 4.2.1 Selecting a Proper Measure.
• 4.2.2 Vibration Transducers4.2.3 Transducer Selection; 4.2.4 Transducer Mounting; 4.2.5 Transducer Location; 4.2.6 Frequency Span; 4.2.7 Data Display; 4.3 Data Processing; 4.4 Data Analysis; 4.4.1 Features in Time Domain; 4.4.2 Features in Frequency Domain; 4.4.3 Features in Time-Frequency Domain; References; 5 Statistic Feature Extraction; 5.1 Introduction; 5.2 Basic Concepts; 5.2.1 Pattern and Feature Vector; 5.2.2 Class; 5.3 Parameter Evaluation Technique; 5.4 Principal Component Analysis (PCA); 5.5 Independent Component Analysis (ICA); 5.6 Kernel PCA; 5.7 Kernel ICA.
• 5.8 Fisher Discriminant Analysis (FDA)5.9 Linear Discriminant Analysis (LDA); 5.10 Generalized Discriminant Analysis (GDA); 5.11 Clustering; 5.11.1 k-Centers Clustering; 5.11.2 k-Means Clustering; 5.11.3 Hierarchical Clustering; 5.12 Other Techniques; References; 6 Feature Selection Optimization; 6.1 Introduction; 6.2 Individual Feature Evaluation (IFE); 6.3 Conditional Entropy; 6.4 Backward Feature Selection; 6.5 Forward Feature Selection; 6.6 Branch and Bound Feature Selection; 6.7 Plus l-Take Away r Feature Selection; 6.8 Floating Forward Feature Selection.
• 6.9 Distance-Based Evaluation Technique6.10 Taguchi Method-Based Feature Selection; 6.11 Genetic Algorithm; 6.11.1 General Concept; 6.11.2 Differences from Other Traditional Methods; 6.11.3 Simple Genetic Algorithm (SGA); 6.11.4 Feature Selection Using GA; 6.12 Summary; References; 7 Intelligent Fault Diagnosis Methodology; 7.1 Introduction; 7.2 Linear Classifier; 7.2.1 Linear Separation of Finite Set of Vectors; 7.2.2 Perceptron Algorithm; 7.2.3 Kozinec's Algorithm; 7.2.4 Multi-class Linear Classifier; 7.3 Quadratic Classifier; 7.4 Bayesian Classifier; 7.5 k-Nearest Neighbors (k-NN).