Model-Based Fault Diagnosis Techniques Design Schemes, Algorithms and Tools [Hardcover]

Guaranteeing a high system performance over a wide operating range is an important issue surrounding the design of automatic control systems with successively increasing complexity. As a key technology in the search for a solution, advanced fault detection and identification (FDI) is receiving considerable attention. This book introduces basic model-based FDI schemes, advanced analysis and design algorithms, and mathematical and control-theoretic tools.
This second edition of Model-Based Fault Diagnosis Techniques contains:
new material on fault isolation and identification and alarm management;
extended and revised treatment of systematic threshold determination for systems with both deterministic unknown inputs and stochastic noises;
addition of the continuously-stirred tank heater as a representative process-industrial benchmark; and
enhanced discussion of residual evaluation which now deals with stochastic processes.
Model-based Fault Diagnosis Techniques will interest academic researchers working in fault identification and diagnosis and as a text it is suitable for graduate students in a formal university-based course or as a self-study aid for practising engineers working with automatic control or mechatronic systems from backgrounds as diverse as chemical process and power engineering.This book gives readers a framework of model-based FDI techniques, helping them to become familiar with the basic ideas and schemes in a systematic way. Examples and benchmarks provide a means of practising the ideas and judging the methods described.Basic Ideas, Major Issues, and Tools in the Observer-Based FDI Framework.- Modelling of Technical Systems.- Structural Fault Detectability, Isolability and Identifiability.- Basic Residual Generation Methods.- Perfect Unknown Input Decoupling.- Residual Generation with Enhanced Robustness against Unknown Inputs.- Residual Generation with Enhanced Robustness againstlҼ