TABLE OF CONTENTS There is an increasing demand for manmade dynamical systems to become safer, more reliable, more cost-effective and less polluting to the environment. These requirements extend beyond normally accepted safety-critical systems of nuclear reactors, chemical plants or aircraft, to new systems such as autonomous vehicles or fast rail systems. An early detection of faults can help avoid the system shut- down, breakdown and even catastrophes involving human fatalities and material damage. A system that includes the capacity of detecting, isolating, identifying or classifying faults is called a fault diagnosis system and the procedure of fault detection and isolation is referred to as an FDI (fault detection and isolation) system. During the 1980s and 1990s many investigations have been made using analytical approaches, based on quantitative models, especially in the control engineering literature [1] [3]. The idea is to generate signals that reflect inconsistencies between the nominal and faulty system operation. Such signals, termed residuals, are usually generated using analytical modelling methods, providing analytical (or functional) redundancy.
Considerable attention has been given to both research and application studies of real processes using analytical redundancy as this is a powerful alternative to the use of repeated hardware (hardware or software redundancy). During the last 20 years, researchers have directed their attention to the use of data-driven FDI techniques and a variety of new improved designs have emerged. The most common approach for model construction for FDI is...
