Researches On The Problems Of Event-driven Control For Two Types Of Linear Switched Systems

Switched systems are an important class of hybrid systems,which consist of several subsystems and switching rules.Since many systems in industrial fields,such as power systems,chemical engineering systems,and aerospace systems,can be described as switched systems,researches on switched systems have received extensive attention.With the rapid increase of the network level,the traditional periodic sampling can cause waste of communication resources and large network load.However,event-driven mechanism overcomes the problems above.An event-driven sampling occurs whenever the event-driven condition is violated.Event-driven sampling mechanism greatly reduces the sampling frequency while ensuring systems performance.Meanwhile,there are almost a large number of constraints in practical systems,and actuator saturation is a very common nonlinear constraint.The existence of saturation has a great impact on system performance,which may make the systems unstable.In addition,the saturation nonlinearity makes the analysis and design of systems more complicated.Therefore,it is a key and significant issue for how to deal with saturation nonlinearity.At present,there are few results on the problem of event-driven control of linear switched systems and event-driven control of linear switched systems with actuator saturation.Therefore,this thesis studies the stabilization problems of these two types of systems.Firstly,considering the stabilization problem of linear switched systems in the framework of event-driven sampling mechanism,based on the common Lyapunov function method,the event-driven controllers are designed and a sufficient condition for the asymptotic stability of the closed-loop switched systems under arbitrary switching is given.Secondly,by combining the single Lyapunov function method and convex combination technique,a co-design of event-driven controllers and switching law is proposed to achieve asymptotic stabilization of the closed loop switched systems.Thirdly,considering the linear switched systems with actuator saturation,where the system states are measurable,the convex combination method is used to handle the saturation nonlinearity.Based on the common Lyapunov function method,a co-design of event-driven state feedback controllers and event-driven condition is proposed to ensure the asymptotic stability of the closed-loop systems,and the corresponding optimization algorithm is given to estimate the attraction domain.Then,for the linear switched systems with actuator saturation,where the system states are unmeasurable,the convex combination method is also used to deal with the saturation nonlinearity.A co-design of event-driven condition and the event-driven controllers is proposed based on the dwell time method.In addition,some sufficient conditions for the asymptotically stability of systems and the estimated attraction domain are given.For the problems above,it is proved that the lower bound of the interval between events is positive,and the Zeno phenomenon in sampling is excluded.Finally,the work done in this thesis and the next research directions are presented.