Research On Event-Triggered Control For Networded Control Systems Subject To Denial-of-Service Attacks

Owing to the development of automation technology,networked control systems have attracted extensive attention in recent decades,where different components are connected via shared communication networks.The networked control systems have a wider range of applications than wire-connected control systems,ranging from small,centralized control systems to large-scale decentralized systems.However,the implementation of communication networks inevitably accompanies many inherent drawbacks,such as limited network bandwidth and network attacks.For some security-critical systems,how to run safely with limited bandwidth becomes challenging.Therefore,this paper mainly focuses on the security control problem of a class of networked control systems subject to denialof-service attacks,designs a variety of flexible event-triggered control mechanisms,and uses the analysis framework of hybrid systems to model and analyze the stability of the entire system.The main work of this paper is as follows:Firstly,a dynamic event-triggered mechanism is designed for networked control systems subject to denial-of-service attacks.Considering the data interaction between the sensor and the controller through the wireless network,and the network channel suffers from denial-of-service attacks with frequency and duration characteristics.Under the premise of ensuring the stability and minimum inter-event time,a dynamic event-triggered mechanism is proposed to enable the system to tolerate a certain degree of denial-of-service attacks.For the state feedback control systems,three event-triggered schemes are designed,and the stability conditions of systems are given.Finally,the effectiveness of the proposed theoretical results is verified by simulation examples.Furthermore,for networked control systems with data transmission through sensorcontroller-actuator networks,considering the problem of asynchronous denial-of-service attacks and asynchronous data transmission,a hybrid dynamic event-triggered mechanism is proposed to further improve the triggering interval,and a regularization variable is introduced to exclude Zeno behavior.Based on the hybrid system framework,the explicit design of the event-triggered mechanism and the stability conditions of the system are given to ensure the exponential stability and Lp gain of the system.In addition,taking the Lipschitz nonlinear system as an example,the linear matrix inequality technique is utilized to formulate the design conditions.Finally,the validity of the results is verified by the simulation of linear and nonlinear numerical examples.Finally,aiming at the problem that the event generator fails to fully utilize the known model information of the system,a resilient model-based hybrid dynamic event-triggered mechanism is proposed to resist denial-of-service attacks and further reduce the communication burden in the sensor-controller-actuator network.By using model-based computing,the newly proposed event-triggered mechanism is superior to the existing eventtriggered mechanisms in terms of network utilization and robustness.In addition to the centralized model-based hybrid dynamic event-triggered mechanism,it also provides a decentralized design method for large-scale systems composed of multiple interconnected subsystems.Finally,the improvement effect of the proposed model-based hybrid dynamic event-triggered mechanism is verified by two simulation examples.