This book investigates in detail model-based fault diagnosis methods, including observer-based residual generation, residual evaluation based on threshold computation, observer-based fault isolation strategies, observer-based fault estimation, Kalman filter-based fault diagnosis methods, and parity space approach. Studies on model-based fault diagnosis have attracted engineers and scientists from various disciplines, such as electrical, aerospace, mechanical, and chemical engineering. Pursuing a holistic approach, the book establishes a fundamental framework for this topic, while emphasizing the importance of state-space approach. The methods introduced in the book are systemic and easy to follow. The book is intended for undergraduate and graduate students who are interested in fault diagnosis and state estimation, researchers investigating fault diagnosis and fault-tolerant control, and control system design engineers working on safety-critical systems. Preface Contents Acronyms Notation 1 Introduction 1.1 Basic Concepts in Fault Diagnosis 1.2 Model-Based Fault Diagnosis Methods 1.2.1 Observer-Based Methods 1.2.2 Parity Space Approaches 1.2.3 Parameter Estimation Methods 1.3 Outline of the Book References 2 Modelling for Fault Diagnosis 2.1 Modelling of Nominal Systems 2.2 Modelling of Faults 2.3 Description of Uncertainties 2.4 Conclusion References 3 Observer-Based Residual Generation 3.1 Residual Generation Based on Analytical Redundancy 3.2 Observer-Based Residual Generation for Continuous-Time Systems 3.3 Observer-Based Residual Generation for Discrete-Time Systems 3.4 Conclusion References 4 Robust Residual Generation Based on Unknown Input Observer 4.1 UIO-Based Residual Generation for Continuous-Time Systems 4.2 UIO-Based Residual Generation for Discrete-Time Systems 4.3 Conclusion References 5 Optimal Robust Residual Generation 5.1 Optimal Robust Residual Generation for Continuous-Time Systems 5.2 Optimal Robust Residual Generation for Discrete-Time Systems 5.3 Conclusion References 6 Residual Evaluation Based on Threshold Computation 6.1 Residual Evaluation Based on Peak-to-Peak Analysis 6.1.1 Residual Evaluation for Continuous-Time Systems 6.1.2 Residual Evaluation for Discrete-Time Systems 6.2 Residual Evaluation Based on Interval Analysis 6.2.1 Residual Evaluation for Continuous-Time Systems 6.2.2 Residual Evaluation for Discrete-Time Systems 6.3 Conclusion References 7 Fault Isolation Based on Structured Residual Generation 7.1 Sensor Fault Isolation Based on Dedicated Observer Scheme 7.2 Sensor Fault Isolation Based on Simplified Observer Scheme 7.3 Sensor Fault Isolation Based on Generalized Observer Scheme 7.4 Actuator Fault Isolation Based on Structured Residual Generation 7.5 Conclusion References 8 Fault Isolation Based on Directional Residual Generation 8.1 Observer-Based Directional Residual Generator 8.2 Directional Residual Generator Design for Systems with Fully Measurable States 8.3 Directional Residual Generator Design for General LTI Systems 8.4 FDI Based on Directional Residual Generation 8.5 Conclusion References 9 Observer-Based Fault Estimation 9.1 Adaptive Observer-Based Fault Estimation 9.2 Fault Estimation Based on Augmented State Observer 9.3 Conclusion References 10 Kalman Filter-Based Fault Diagnosis 10.1 Kalman Filter-Based Fault Detection 10.2 Fault Isolation Based on Dedicated Kalman Filters 10.3 Fault Estimation Based on Augmented State Kalman Filter 10.4 Conclusion References 11 Parity Space Approaches 11.1 Residual Generation Using Parity Space Approach 11.1.1 Basic Parity Space Approach 11.1.2 Optimal Parity Space Approach 11.2 Residual Evaluation Methods in Parity Space Approach 11.2.1 Norm-Based Analysis Method 11.2.2 Interval-Based Analysis Method 11.2.3 Statistic Test Method 11.3 Fault Isolation Strategies Based on Parity Space Approach 11.3.1 Structured Residual Generation Based on Parity Space Approach 11.3.2 Directional Residual Generation Based on Parity Space Approach 11.4 Conclusion References
