Sliding Mode Control For Nonlinear Networked Systems Under Event-Triggered Mechanism

Event-triggered control(ETC)is a novel control mechanism.Compared with traditional periodic-triggered control(PTC).it can effectively improve the utilization of limited network resources in the networked systems with limited bandwidth,while ensuring the system stability and performance.The reason is that in the event-triggered mechanism(ETM),the sensor signal or controller signal will not be transmitted unless a certain triggering condition is met.On the other hand,sliding mode control(SMC)is widely used in the fields of industrial process control,networked control and robot because of its strong robustness to system parameter uncertainty,disturbance and input nonlinearity.In recent years,combined with the advantages of ETC and SMC,event-triggered SMC has has become the focus of many scholars.However,the research results of event-triggered sliding mode control for networked systems still have some limitations,which need further discussion.For example,at present,most of the researches on the methods of event-triggered SMC are focused on the state-feedback continuous-time linear systems,and relatively few researches on the more practical output feedback discrete-time nonlinear systems.In addition,there is still a lack of research on the event triggered sliding-mode control for net-worked systems under the influence of factors such as quantization,actuator fault,saturation and dead-zone.In order to make up for the inadequacy of existing achievements,this dissertation studies system modeling,stability analysis and event-triggered mechanism,observer and sliding mode controller co-design problems for a class of discrete-time Lipschitz nonlinear networked systems and T-S fuzzy networked systems with some system constraints,such as saturation,dead-zone,quantization,actuator fault and network delay.The main research contents are summarized as follows:1.The problem of event-triggered observer-based SMC is investigated for a class of un-certain discrete-time Lipschitz nonlinear NCSs with quantizations occurring in both input and output channels.Considering the effects of quantizations,network delay and event-triggered mechanism,the sliding mode dynamics and error dynamics are converted into a linear parameter varying(LPV)time-delay system by using time-delay analysis method and a reformulated Lip-schitz property.Based on this model,with the aid of Lvapunov-Krasovskii functional and linear matrix inequality(LMI)method,a sufficient condition is derived to guarantee the asymptotic sta-bility of the resulting system with apredefined performance level.Then,the event-triggering parameter,observer gain,and sliding mode parameter are co-designed.Considering the influ-ence of input quantization,a new observed-based sliding mode controller is synthesized and the reachability of sliding mode is analyzed.2.The event-triggered fault estimation(FE)and sliding mode fault-tolerant control(FTC)problem for a class of discrete-time Lipschtiz nonlinear NCSs subject to actuator fault is inves-tigated.First,an event-triggered fault/state observer is designed to estimate the actuator fault and system state,simultaneously.By using a reformulated Lipschitz property and time-delay analysis method,the sliding mode dynamics and state/fault error dynamics are converted into a unified LPV time-delay system.Based on Lyapunov-Krasovskii functional method,a delay-dependent sufficient condition is derived to guarantee the asymptotic stability of the resulting system with prescribed performance.And the co-design method is also given.Further-more,an observed-based sliding mode fault-tolerant controller is synthesized to make sure the reachability of the sliding surface.3.The problem of non-fragile observer-based SMC is considered for a class of discrete-time Lipschitz nonlinear NCSs with sensor saturation and dead-zone input via dynamic event-triggering.In order to further reduce unnecessary network data transmission a new dynamic event-triggered mechanism based on saturated output information is proposed.Then,a non-fragile observer is designed to estimate the unmeasured system state,which facilitates the con-struction of the discrete sliding surface.Then,a LPV time-delay system model is established by taking into account the dynamic event-triggered mechanism,output saturation,network delay and parametric uncertainty.Sufficient conditions are derived to guarantee the resulting system to be asymptotically stable with adaptive performance.And the dynamic event-triggering parameter,observer parameter and sliding mode parameter are co-designed.Furthermore,a nov-el event-triggered sliding mode controller is designed to ensure that the system can reach a sliding region near equilibrium point in finite time in the presence of dead-zone input.4.Considering the effects of input quantization,incomplete measurements and two channel event-triggering,the FE and FTC problem for a class of discrete-time Takagi-Sugeno(T-S)fuzzy systems is studied.The incomplete information under consideration includes randomly occurring sensor saturation and randomly occurring quantization.In order to reduce the number of data transmission on the sensor-controller and controller-actuator.a novel dynamic event-triggered mechanism on the sensor side and a static one on the controller side are proposed.Then,an event-triggered fuzzy fault/state observer is designed to estimate actuator fault and system state,simultaneously.Moreover,an augmented T-S fuzzy time-delay system model including error dynamics and sliding mode dynamics is established under a unified framework.A sufficient condition for random asymptotic stability and adaptive performance is derived,and the co-design method is also given to obtain the event-triggering parameter,observer gain and sliding mode pairameter,simultaneuously.Based on this condition,a new fuzzy sliding mode fault-tolerant controller is designed to drive the system trajectory onto a bounded sliding mode region in the presence of input quantization,actuator fault and input ETM.5.In the case of multi-channel saturation,quantization and non-ideal network at the sensor side,the output decentralized dynamic event-triggered SMC problem for a class of discrete-time T-S fuzzy systems is investigated.First,a novel decentralized dynamic event-triggered mechanism is proposed to further reduce the number of the sensor multi-channel network data transmission.Then,an augmented T-S fuzzy time-delay system model including quantization,saturation,decentralized dynamic event-triggered mechanism and sliding mode dynamics is con-structed.The stability of the system model is studied and the parameter co-design method is also given.Furthermore,an observer-based sliding mode controller is synthesized to guarantee the reachability of the sliding surface.