Fault detection of networked dynamical systems: a survey of trends and techniques

Abstract

Fault detection of networked dynamical systems (NDSs) has attracted ever-increasing attention since it can maintain high-quality products as well as operational safety. Considering the utilisation of communication networks, it is desirable to develop engineering-oriented approaches to NDSs subject to the incompleteness from network-induced phenomena (NIP) and the sparsity from communication scheduling. As such, this paper presents a survey of trends and techniques of fault detection in NDSs. First, some typical fault detection methods are summarised based on the various residual assessment functions. Then, some interesting developments are systematically reviewed from two aspects, that is, fault detection with NIP and fault detection under various communication scheduling schemes. In addition, some frontier topics are extensively discussed on fault detection with communication protocols, cyber-attacks, and its applications. Finally, several future works of fault detection problems are investigated to motivate future research.