Research On Security Control Of Cyber-Physical Systems Based On Event-Triggered Scheme

Cyber-physical Systems(CPSs)are a class of multi-dimensional complex systems integrating sensing,computing,network communication and physical processes.The security problems of CPSs mainly include information security and physical security.In recent years,the safety of CPSs has attracted extensive attention.This paper mainly studies the security control of CPSs under Denial-of-Service(DoS)attacks and False Data Injection(FDI)attacks.The main technical contributions of this paper are as follows:(1)Consider the security control problem of a class of CPSs built by linear systems under periodic DoS attacks.First,a novel event-triggered scheme(ETS)without Zeno behaviors is proposed to save the limited network resources and eliminate the occurrence of invalid events during periodic DoS attacks.Then,by constructing a predictor-based event-triggered control framework,the upper bound of the prediction error related to the DoS attacks parameters is given.Furthermore,sufficient conditions on periodic DoS attacks parameters,under which the input-to-state stability of the closed-loop system can be still guaranteed,are provided.Compared with the existing design method,it is shown that the proposed one can achieve the better system performance when the system suffers the same degree of DoS attacks.(2)Consider the security control problem of a class of CPSs constructed by nonlinear systems with unknown parameters under FDI attacks and actuator failures.A class of malicious attackers considered in this paper can obtain system state information,and use this information to construct a time-varying attack model to modify the state of the system,and the attack is bounded.First,a novel adaptive event-triggered scheme(AETS)without Zeno behaviors is designed to save limited network resources,and its threshold can be continuously adjusted according to the change of system state.Then,an adaptive controller and adaptive laws are designed to effectively compensate for FDI attacks and actuator failures.Furthermore,it is proved that the tracking error of the system can be exponentially converged within a compact set with an adjustable radius.Compared with the existing ETS with a fixed threshold,it is shown that the proposed AETS can significantly reduce the number of data packet transmissions and improve utilization of network resource while ensuring the same system performance.In addition,the proposed AETS can degenerate into the existing ETS with the fixed threshold.Finally,some simulation examples are given to illustrate the effectiveness of the proposed methods.