Event-triggered Mechanism Design For Control Systems Subject To Actuator Saturation

The event-triggered(ET)control is a kind of data transmission protocol to exchange information,when the signal error exceeds a given threshold.It can guarantee the stability and some performance of control systems under the premise of a certain aspect.The main purpose of ET control is to reduce the transmission burden of feedback loop,and reduce the quantity of calculation,namely,the utilization of computing resources.In practical engineering,all physical systems are subject to actuator saturation.Such a non-ideal factor degrades the control performance of the systems,even can make the system unstable and cause severe damage.Therefore,this thesis makes an attempt to study the stability of event-triggered control systems subject to actuator saturation.First,we propose the controller design of continuous-time system with actuator saturation.In order to achieve a better quadratic performance,the linear quadratic(LQ)optimal control method is applied to design the controller.We also derive global and local stability conditions for the ET control systems.For event-triggered and self-triggered(ST)transmission mechanism,Zeno behavior is the obstacle for computer implementation.The minimum triggering time interval for ET and ST control is explicitly given to avoid Zeno behavior.Moreover,to obtain a larger stability region,a re-design method is put forward.According to numerical simulation,we know that the re-design method brings a larger stability region.Second,we study the discrete-time case for event-triggered system in the presence of actuator saturation.The ET control mechanism for discrete-time system is discussed,and two different kinds of controller design methods are investigated.These two methods can achieve good control effect in different aspects.We give a sufficient condition to estimate the stability region with Matlab.The feasible solution is,however,a non-convex linear matrix inequality(LMI),we introduce the cone complementary linearization method to get the optimal solution.Finally,we consider the anti-windup compensation to find a larger stability region.The designing method of dynamic output feedback controller is first given at the beginning.Then an iterative optimization algorithm is proposed to get the anti-windup compensator gain.In addition,due to the introduction of output feedback control method,we make some appropriate changes in the triggering condition to avoid Zeno behavior.An exponential convergence term is added into the original trigger conditions to further reduce the triggering number.The effect in expanding stability region is validated by an example.