Fault diagnosis in multirate sampled data systems

Three novel schemes for sensor and actuator fault diagnosis for linear time-invariant systems are presented: multirate parity-space techniques, multirate observers, and multirate detection filters. It is shown that these approaches offer distinct advantages over conventional fault diagnosis schemes in terms of ease of failure detection, the speed, and the class of systems for which detection is possible. In particular, the use of multirate sampling in each of these systems provides a more elegant failure detection system and alleviates the problems inherent in application to systems containing nonminimum phase zeros. It is shown that the existence of certain types of fault diagnosis schemes is a generic system property.<
>