Event-Triggered Fault Detection for Networked Control Systems Subject to Packet Dropout

Abstract

This paper investigates the problem of event-triggered fault detection for discrete-time networked systems subject to packet dropout. The main aim of the proposed approach is to efficiently use the communication network and reduce the signal transmission over the network. Toward this goal, a dynamic parity space event-triggered fault detection approach is developed to generate a robust residual signal. Moreover, an adaptive threshold is utilized to overcome the limitations of static thresholds. The efficiency of proposed approach is experimentally demonstrated and validated for a laboratory three-tank system.