Available:*
Library | Item Barcode | Call Number | Material Type | Item Category 1 | Status |
---|---|---|---|---|---|
Searching... | 30000010029198 | TA169.6 S56 2003 | Open Access Book | Book | Searching... |
On Order
Summary
Summary
Safety in industrial process and production plants is a concern of rising importance but because the control devices which are now exploited to improve the performance of industrial processes include both sophisticated digital system design techniques and complex hardware, there is a higher probability of failure. Control systems must include automatic supervision of closed-loop operation to detect and isolate malfunctions quickly. A promising method for solving this problem is "analytical redundancy", in which residual signals are obtained and an accurate model of the system mimics real process behaviour. If a fault occurs, the residual signal is used to diagnose and isolate the malfunction. This book focuses on model identification oriented to the analytical approach of fault diagnosis and identification covering: choice of model structure; parameter identification; residual generation; and fault diagnosis and isolation. Sample case studies are used to demonstrate the application of these techniques.
Table of Contents
Symbols and Abbreviations | p. xv |
1 Introduction | p. 1 |
1.1 Nomenclature | p. 3 |
1.2 Fault Detection and Identification Methods based on Analytical Redundancy | p. 5 |
1.3 Model-based Fault Detection Methods | p. 7 |
1.4 Model Uncertainty and Fault Detection | p. 8 |
1.5 The Robustness Problem in Fault Detection | p. 9 |
1.6 System Identification for Robust FDI | p. 11 |
1.7 Fault Identification Methods | p. 12 |
1.8 Report on FDI Applications | p. 13 |
1.9 Outline of the Book | p. 16 |
1.10 Summary | p. 18 |
2 Model-based Fault Diagnosis Techniques | p. 19 |
2.1 Introduction | p. 19 |
2.2 Model-based FDI Techniques | p. 20 |
2.3 Modelling of Faulty Systems | p. 21 |
2.4 Residual Generator General Structure | p. 28 |
2.5 Residual Generation Techniques | p. 31 |
2.5.1 Residual Generation via Parameter Estimation | p. 32 |
2.5.2 Observer-based Approaches | p. 35 |
2.5.3 Fault Detection with Parity Equations | p. 40 |
2.6 Change Detection and Symptom Evaluation | p. 44 |
2.7 The Residual Generation Problem | p. 45 |
2.8 Fault Diagnosis Technique Integration | p. 51 |
2.8.1 Fuzzy Logic for Residual Generation | p. 51 |
2.8.2 Neural Networks in Fault Diagnosis | p. 53 |
2.8.3 Neuro-fuzzy Approaches to FDI | p. 54 |
2.8.4 Structure Identification of NF Models | p. 56 |
2.8.5 NF Residual Generation Scheme for FDI | p. 57 |
2.9 Summary | p. 59 |
3 System Identification for Fault Diagnosis | p. 61 |
3.1 Introduction | p. 61 |
3.2 Models for Linear Systems | p. 62 |
3.3 Parameter Estimation Methods | p. 64 |
3.3.1 System Identification in Noiseless Environment | p. 65 |
3.3.2 System Identification in Noisy Environment | p. 68 |
3.3.3 The Frisch Scheme in the MIMO Case | p. 73 |
3.4 Models for Non-linear Dynamic Systems | p. 75 |
3.4.1 Piecewise Affine Model | p. 75 |
3.4.2 Model Continuity and Domain Partitioning | p. 79 |
3.4.3 Local Affine Model Identification | p. 82 |
3.4.4 Multiple-Model Identification | p. 85 |
3.5 Fuzzy Modelling and Identification | p. 89 |
3.5.1 Fuzzy Multiple Inference Identification | p. 90 |
3.5.2 Takagi-Sugeno Multiple-Model Paradigm | p. 92 |
3.5.3 Fuzzy Clustering for Fuzzy Identification | p. 95 |
3.5.4 Product Space Clustering and Fuzzy Model Identification | p. 100 |
3.5.5 Non-linear Regression Problem and Black-Box Models | p. 103 |
3.5.6 Fuzzy Model Identification From Clusters | p. 107 |
3.6 Conclusion | p. 112 |
4 Residual Generation, Fault Diagnosis and Identification | p. 115 |
4.1 Introduction | p. 115 |
4.2 Output Observers for Robust Residual Generation | p. 116 |
4.3 Unknown Input Observer | p. 119 |
4.3.1 UIO Mathematical Description | p. 120 |
4.3.2 UIO Design Procedure | p. 122 |
4.4 FDI Schemes Based on UIO and Output Observers | p. 122 |
4.5 Sliding Mode Observers for FDI | p. 127 |
4.5.1 Sliding Mode Observers | p. 128 |
4.6 Kalman Filtering and FDI from Noisy Measurements | p. 130 |
4.7 Residual Robustness to Disturbances | p. 131 |
4.7.1 Disturbance Distribution Matrix Estimation | p. 132 |
4.7.2 Additive Non-linear Disturbance and Noise | p. 133 |
4.7.3 Model Complexity Reduction | p. 133 |
4.7.4 Parameter Uncertainty | p. 134 |
4.7.5 Distribution Matrix Low Rank Approximation | p. 135 |
4.7.6 Model Estimation with Bounded Uncertainty | p. 135 |
4.7.7 Disturbance Vector and Disturbance Matrix Estimation | p. 136 |
4.7.8 Distribution Matrix Optimisation | p. 139 |
4.7.9 Disturbance Distribution Matrix Identification | p. 139 |
4.8 Residual Generation via Parameter Estimation | p. 141 |
4.9 Residual Generation via Fuzzy Models | p. 142 |
4.10 FDI Using Neural Networks | p. 143 |
4.10.1 Neural Network Basics | p. 145 |
4.11 Fault Diagnosis of an Industrial Plant at Different Operating Points Using Neural Networks | p. 147 |
4.11.1 Operating Point Detection and Fault Diagnosis | p. 147 |
4.11.2 FDI Method Development | p. 149 |
4.12 Neuro-fuzzy in FDI | p. 150 |
4.12.1 Methods of Neuro-fuzzy Integration | p. 151 |
4.12.2 Neuro-fuzzy Networks | p. 152 |
4.12.3 Residual Generation Using Neuro-fuzzy Models | p. 154 |
4.12.4 Neuro-fuzzy-based Residual Evaluation | p. 155 |
4.13 Summary | p. 156 |
5 Fault Diagnosis Application Studies | p. 157 |
5.1 Introduction | p. 157 |
5.2 Physical Background and Modelling Aspects of an Industrial Gas Turbine | p. 158 |
5.2.1 Gas Turbine Model Description | p. 158 |
5.3 Identification and FDI of a Single Shaft Industrial Gas Turbinel68 | |
5.3.1 System Identification | p. 169 |
5.3.2 FDI Using Dynamic Observers | p. 176 |
5.3.3 FDI Using Kalman Filters | p. 183 |
5.3.4 Fuzzy System Identification and FDI | p. 189 |
5.3.5 Sensor Fault Identification Using Neural Networks | p. 191 |
5.3.6 Multiple Working Conditions FDI Using Neural Networks | p. 196 |
5.3.7 FDI Method Development | p. 196 |
5.3.8 Multiple Operating Point Simulation Results | p. 197 |
5.4 Identification and FDI of Double Shaft Industrial Gas Turbine | p. 199 |
5.4.1 Process Description | p. 199 |
5.4.2 System Identification | p. 201 |
5.4.3 FDI Using Unknown Input Observers | p. 203 |
5.4.4 FDI Using Kalman Filters | p. 208 |
5.4.5 Disturbance Decoupled Observers for Sensor FDI | p. 209 |
5.4.6 Fuzzy Models for Fault Diagnosis | p. 210 |
5.5 Modelling and FDI of a Turbine Prototype | p. 214 |
5.5.1 System Modelling and Identification | p. 215 |
5.6 Turbine FDI Using Output Observers | p. 220 |
5.6.1 Case 1: Compressor Failure (Component Fault) | p. 221 |
5.6.2 Case 2: Fault Diagnosis of the Output Sensor | p. 223 |
5.6.3 Case 3: Turbine Damage (Turbine Component Fault) | p. 227 |
5.6.4 Case 4: Actuator Fault (Controller Malfunctioning) | p. 228 |
5.6.5 FDI in Noisy Environment Using Kalman Filters | p. 233 |
5.6.6 Fault Isolation | p. 235 |
5.6.7 Minimal Detectable Faults | p. 239 |
5.7 FDI with Eigenstructure Assignment | p. 242 |
5.7.1 Robust Fault Diagnosis of the Industrial Process | p. 243 |
5.8 Robust Residual Generation Problem | p. 247 |
5.9 Summary | p. 249 |
6 Concluding Remarks | p. 251 |
6.1 Suggestions for Future Work | p. 253 |
6.1.1 Frequency Domain Residual Generation | p. 253 |
6.1.2 Adaptive Residual Generators | p. 255 |
6.1.3 Integration of Identification, FDI and Control | p. 256 |
6.1.4 Fault Identification | p. 256 |
6.1.5 Fault Diagnosis of Non-Linear Dynamic Systems | p. 258 |
References | p. 261 |
Index | p. 279 |