| Symbols and Abbreviations | p. xv |
| Introduction | p. 1 |
| Nomenclature | p. 3 |
| Fault Detection and Identification Methods based on Analytical Redundancy | p. 5 |
| Model-based Fault Detection Methods | p. 7 |
| Model Uncertainty and Fault Detection | p. 8 |
| The Robustness Problem in Fault Detection | p. 9 |
| System Identification for Robust FDI | p. 11 |
| Fault Identification Methods | p. 12 |
| Report on FDI Applications | p. 13 |
| Outline of the Book | p. 16 |
| Summary | p. 18 |
| Model-based Fault Diagnosis Techniques | p. 19 |
| Introduction | p. 19 |
| Model-based FDI Techniques | p. 20 |
| Modelling of Faulty Systems | p. 21 |
| Residual Generator General Structure | p. 28 |
| Residual Generation Techniques | p. 31 |
| Residual Generation via Parameter Estimation | p. 32 |
| Observer-based Approaches | p. 35 |
| Fault Detection with Parity Equations | p. 40 |
| Change Detection and Symptom Evaluation | p. 44 |
| The Residual Generation Problem | p. 45 |
| Fault Diagnosis Technique Integration | p. 51 |
| Fuzzy Logic for Residual Generation | p. 51 |
| Neural Networks in Fault Diagnosis | p. 53 |
| Neuro-fuzzy Approaches to FDI | p. 54 |
| Structure Identification of NF Models | p. 56 |
| NF Residual Generation Scheme for FDI | p. 57 |
| Summary | p. 59 |
| System Identification for Fault Diagnosis | p. 61 |
| Introduction | p. 61 |
| Models for Linear Systems | p. 62 |
| Parameter Estimation Methods | p. 64 |
| System Identification in Noiseless Environment | p. 65 |
| System Identification in Noisy Environment | p. 68 |
| The Frisch Scheme in the MIMO Case | p. 73 |
| Models for Non-linear Dynamic Systems | p. 75 |
| Piecewise Affine Model | p. 75 |
| Model Continuity and Domain Partitioning | p. 79 |
| Local Affine Model Identification | p. 82 |
| Multiple-Model Identification | p. 85 |
| Fuzzy Modelling and Identification | p. 89 |
| Fuzzy Multiple Inference Identification | p. 90 |
| Takagi-Sugeno Multiple-Model Paradigm | p. 92 |
| Fuzzy Clustering for Fuzzy Identification | p. 95 |
| Product Space Clustering and Fuzzy Model Identification | p. 100 |
| Non-linear Regression Problem and Black-Box Models | p. 103 |
| Fuzzy Model Identification From Clusters | p. 107 |
| Conclusion | p. 112 |
| Residual Generation, Fault Diagnosis and Identification | p. 115 |
| Introduction | p. 115 |
| Output Observers for Robust Residual Generation | p. 116 |
| Unknown Input Observer | p. 119 |
| UIO Mathematical Description | p. 120 |
| UIO Design Procedure | p. 122 |
| FDI Schemes Based on UIO and Output Observers | p. 122 |
| Sliding Mode Observers for FDI | p. 127 |
| Sliding Mode Observers | p. 128 |
| Kalman Filtering and FDI from Noisy Measurements | p. 130 |
| Residual Robustness to Disturbances | p. 131 |
| Disturbance Distribution Matrix Estimation | p. 132 |
| Additive Non-linear Disturbance and Noise | p. 133 |
| Model Complexity Reduction | p. 133 |
| Parameter Uncertainty | p. 134 |
| Distribution Matrix Low Rank Approximation | p. 135 |
| Model Estimation with Bounded Uncertainty | p. 135 |
| Disturbance Vector and Disturbance Matrix Estimation | p. 136 |
| Distribution Matrix Optimisation | p. 139 |
| Disturbance Distribution Matrix Identification | p. 139 |
| Residual Generation via Parameter Estimation | p. 141 |
| Residual Generation via Fuzzy Models | p. 142 |
| FDI Using Neural Networks | p. 143 |
| Neural Network Basics | p. 145 |
| Fault Diagnosis of an Industrial Plant at Different Operating Points Using Neural Networks | p. 147 |
| Operating Point Detection and Fault Diagnosis | p. 147 |
| FDI Method Development | p. 149 |
| Neuro-fuzzy in FDI | p. 150 |
| Methods of Neuro-fuzzy Integration | p. 151 |
| Neuro-fuzzy Networks | p. 152 |
| Residual Generation Using Neuro-fuzzy Models | p. 154 |
| Neuro-fuzzy-based Residual Evaluation | p. 155 |
| Summary | p. 156 |
| Fault Diagnosis Application Studies | p. 157 |
| Introduction | p. 157 |
| Physical Background and Modelling Aspects of an Industrial Gas Turbine | p. 158 |
| Gas Turbine Model Description | p. 158 |
| Identification and FDI of a Single Shaft Industrial Gas Turbinel68 | |
| System Identification | p. 169 |
| FDI Using Dynamic Observers | p. 176 |
| FDI Using Kalman Filters | p. 183 |
| Fuzzy System Identification and FDI | p. 189 |
| Sensor Fault Identification Using Neural Networks | p. 191 |
| Multiple Working Conditions FDI Using Neural Networks | p. 196 |
| FDI Method Development | p. 196 |
| Multiple Operating Point Simulation Results | p. 197 |
| Identification and FDI of Double Shaft Industrial Gas Turbine | p. 199 |
| Process Description | p. 199 |
| System Identification | p. 201 |
| FDI Using Unknown Input Observers | p. 203 |
| FDI Using Kalman Filters | p. 208 |
| Disturbance Decoupled Observers for Sensor FDI | p. 209 |
| Fuzzy Models for Fault Diagnosis | p. 210 |
| Modelling and FDI of a Turbine Prototype | p. 214 |
| System Modelling and Identification | p. 215 |
| Turbine FDI Using Output Observers | p. 220 |
| Case 1: Compressor Failure (Component Fault) | p. 221 |
| Case 2: Fault Diagnosis of the Output Sensor | p. 223 |
| Case 3: Turbine Damage (Turbine Component Fault) | p. 227 |
| Case 4: Actuator Fault (Controller Malfunctioning) | p. 228 |
| FDI in Noisy Environment Using Kalman Filters | p. 233 |
| Fault Isolation | p. 235 |
| Minimal Detectable Faults | p. 239 |
| FDI with Eigenstructure Assignment | p. 242 |
| Robust Fault Diagnosis of the Industrial Process | p. 243 |
| Robust Residual Generation Problem | p. 247 |
| Summary | p. 249 |
| Concluding Remarks | p. 251 |
| Suggestions for Future Work | p. 253 |
| Frequency Domain Residual Generation | p. 253 |
| Adaptive Residual Generators | p. 255 |
| Integration of Identification, FDI and Control | p. 256 |
| Fault Identification | p. 256 |
| Fault Diagnosis of Non-Linear Dynamic Systems | p. 258 |
| References | p. 261 |
| Index | p. 279 |
| Table of Contents provided by Publisher. All Rights Reserved. |