Research On Network Systems Analysis And Collaborative Design Based On Event Triggering

One of the most challenging topics in the networked control systems(NCSs)is the codesign problem,either for those which are from the wire NCSs to the wireless NCSs or for those which are from the traditional periodic sampling control to the event-triggered control.Co-design is an analysis and design method of the system which considers resource constraints such as the control strategy and networked communications together,such that the system performance and network resources are optimized.The goal for this thesis is to study the codesign method of corresponding event trigger and state feedback controller in the nonlinear NCSs.The limitation of bandwidth in network communication is an inevitable problem in NCSs.Hence,the aims of this thesis is focus on improving the utilization of the network resources without changing its own system structure.At the meantime,the node driven of NCS will be replaced by the event-triggered mechanism.In this paper,we mainly focus on the eventtriggered transmission strategy for saving network resources as well as the co-design problem with the controller.Meanwhile,the impact of packet loss of system performance in NCSs is also taken into consideration.The main contributions are as follows:Firstly,the stability and H? performance of the nonlinear NCSs based on event-triggered mechanism is analyzed.Based on the system model established from the T-S fuzzy model,the event-triggered mechanism(ETM)strategy is extended from the idea of relative error threshold.Then,the state feedback controller is introduced which is originated from parallel distributed compensation(PDC)technique.Further with the aid of Lyapunov-Krasovskii stability theory and linear matrix inequality(LMI)technique,we obtain the sufficient conditions of which the system can reach the mean-square stability and H? performance is partially fulfilled.Secondly,based on the research output of the previous chapter as well as the effects of the packet loss,we propose the multi-channel event-triggered mechanism(ETM)strategy based on relative error.It is follows by a study of a class of nonlinear NCSs with packet loss based on event-triggered mechanism and H? performance of the state feedback controller,in which the data packet loss is described as an independent and homogeneous Bernoulli progress.In addition,the nonlinear NCSs are established from the T-S fuzzy model,and a state feedback controller is now derived from parallel distributed compensation technique.Based on this model aided with the Lyapunov-Krasovskii stability theory and linear matrix inequality(LMI)technique,we obtain the sufficient conditions of which the system can reach the mean-square stability and H? performance is significantly improved.As a result,the effective event-triggered controller can contribute to the decline in data transmission rate at the node.Thirdly,a co-design method of ETM and controller is studied.The new model of NCSs is derived from the T-S fuzzy method in which the absolute error event trigger is considered.The nonlinear system is initially transformed into a discretized model by the discretization method,while the nonlinear system is modeled by T-S fuzzy model and PDC technique.On the other hand,using Lyapunov-Krasovskii stability theory and LMI technique,a sufficient condition is provided to ensure that the system is uniformly bounded.The proposed method can further reduce the bandwidth utilization of the network,allowing the system state to be adjusted by modifying the ETM parameters.