Transmission-Dependent Fault Detection and Isolation Strategy for Networked Systems Under Finite Capacity Channels.

Abstract

This paper addresses a novel transmission-dependent fault detection and isolation (FDI) scheme in finite frequency domain for networked control systems with consideration of limited communication capacity, which includes multiple transmission intervals and delays, media accessing constraints as well as packet losses. By focusing on these phenomena, a switched stochastic system with multistochastic parameters is first modeled to represent the network-induced features. Then, with the aid of finite frequency stochastic performance index and geometric analysis method, a novel FDI scheme is proposed in finite frequency domain. The FDI filters are designed corresponding to the different transmission intervals and complete the task of detection and isolation only by partially available measurements. A novel lemma is investigated subsequently to capture the desired finite frequent stochastic performance for derived systems. Based on this lemma, sufficient conditions are developed to characterize the filter gains. Finally, an application to the VTOL aircraft is presented to show the effectiveness of the proposed approach.