ABSTRACT
Automatic control systems are becoming more and more sophisticated and the control algorithms more and more complex. To improve the reliability of control systems, much attention has been paid to the problem of fault detection and diagnosis (FDD) in dynamic systems over the past two decades. Fruitful results can be found in references [8, 76, 109]. Effective methodologies mainly include observer (or filter)-based schemes, identification-based approaches, and statistic methods. Among these methods, observer-based approaches have been effectively applied to dynamic systems, where observer signals are used to infer information about the fault. Many significant approaches have been presented and applied to practical processes successfully (see [34, 71, 90, 115, 168, 229], and references therein).
