The design of fault detectors for fault detection and isolation (FDI) in dynamic systems is considered in this paper from a norm based point of view. An analysis of norm based threshold selection is given based on di erent formulations of FDI problems. Both the nominal FDI problem as well as the uncertain FDI problem will be considered. With reference to this analysis, a performance index based on norms of the involved transfer functions is given. A method for designing FDI ® lters which will minimize the performance index is also given.

No abstract

The design of fault detectors for fault detection and isolation (FDI) in dynamic systems is considered in this paper from a norm based point of view. An analysis of norm based threshold selection is given based on different formulations of FDI problems. Both the nominal FDI problem as well as the uncertain FDI problem will be considered. Based on this analysis, a performance index based on norms of the involved transfer functions is given. The performance index allow also to optimize the structure of the fault detection filter directly.

No abstract

Robust 3c2 performance conditions under structured uncertainty, analogous to well known methods for X, performance, have recently emerged in both discrete and continuous-time. This paper considers the extension into uncertain sampled-data (SD) systems, taking into account inter-sample behavior. Convex conditions for robust 312 performance are derived for different uncertainty sets.