Event-triggered Control For Networked Nonlinear Systems

With the development of a digital control platform,event-triggered control as a control s-trategy that can save network costs has attracted much attention.For nonlinear networked systems,this paper discusses some hot spots and unsolved problems in event-triggered con-trol.In this paper,event-triggered mechanisms under different network environments are proposed for the nonlinear positive system,strict feedback system,and large-scale inter-connected system.Using Lyapunov stability theory,the control synthesis problems are ad-dressed to guarantee system stability performance,reduce network communication cost and save limited network resources.Compared with the current results,the main contributions of this dissertation are as follows:1.The Takagi-Sugeno fuzzy model is first established to approximate the nonlinear systems for the event-triggered control problem of nonlinear positive systems.Considering the cost of network communication and data transmission delay in a network environment,an adap-tive discrete event-triggered mechanism based on a positive system is designed.Based on this data transmission mechanism,a dynamic output feedback controller is designed consid-ering the existence of deceptive network attacks under shared network conditions.A new Lyapunov functional is constructed to analyze the system’s stability,and it is proved that the closed-loop system is randomly stable under L₁gain indexγ.The conditions of stabilization and positive-preserving are developed in the form of Linear Programming,which is different from the non-positive system.2.For the finite time event-triggered impulsive control problem of a nonlinear positive sys-tem,Takagi-Sugeno fuzzy model is designed for a nonlinear positive system.A discrete event-triggered mechanism is designed to release impulsive signals for the system.An in-stant of impulsive control is generated based on an event-triggered mechanism of a positive system.The combination of discrete event-triggered mechanisms and impulsive control can achieve low-cost,fast,and efficient control tasks of the system and save communication cost-s.The impulse controller is designed to discuss system behaves in finite time when given a bound on the initial conditions.Finally,Linear Programming form conditions are obtained to guarantee the positivity of the nonlinear positive system and make the closed-loop system finite time stability and finite time contractive stability.3.For the output regulation problem of strict feedback systems with uncertain mismatched parameters,based on a plant-controller network,through the comprehensive design of event-triggered strategy,adaptive algorithm,and controller,an event-triggered adaptive algorithm for output regulation is presented.The updating of the adaptive parameters only needs the discrete data given by the event-triggered detector.The measurement error caused by the event-triggered mechanism is compensated by introducing compensated dynamics.With Lyapunov stability theory,the stability of compensated error,filtering error and compen-sated dynamics is analyzed.The obtained event-triggered condition parameters are less conservative than the current results,and better control performance is achieved under more minor communication requirements.4.For the output tracking problem of nonlinear interconnected systems,a new event trig-gered mechanism is designed over the plant-controller communication network and the fuzzy logic system is used to deal with the unknown interconnections.Considering the case of actuator failure,a decentralized adaptive fault-tolerant algorithm using discrete data is designed based on the event triggered mechanism.The output tracking problem of with actuator failure is effectively solved and the Zeno phenomenon is excluded carefully.Com-pared with the current results,the proposed adaptive algorithm and fault-tolerant mechanism have fewer communication costs and can save network resources.5.A finite-time stability guaranteed event-triggered mechanism and corresponding decen-tralized adaptive scheme are designed for the finite-time control problem of nonlinear in-terconnected systems.The filtering and measurement errors caused by the event-triggered mechanism are compensated by introducing the compensated signal dynamics.The finite-time stability of output of the closed-loop system signal is verified using the Lyapunov sta-bility theory,and the Zeno phenomenon is eliminated.