网络化时滞控制系统的有限时间稳定性分析  

Finite-Time Stability of Networked Control Systems with Time Delay

作　　者：刘薇[1] 周振[2] 王跃灵[2] 

机构地区：[1]浙江东方职业技术学院,温州325011 [2]燕山大学工业计算机控制工程河北省重点实验室,秦皇岛066004

出　　处：《计算机系统应用》2015年第2期28-34,共7页Computer Systems & Applications

摘　　要：针对网络控制系统中存在于传感器-控制器-执行器间的双时延问题,提出了一种基于Markov模型的状态反馈控制策略.与传统应用Markov随机过程的方式相比,该策略采用两个Markov链描述每一个时延,通过状态反馈把该随机系统描述为具有四个随机参量的离散Markov跳变系统.利用Lyapunov有限时间稳定性理论分析得到该系统稳定的充分条件,并利用线性矩阵不等式(LMI)得到了可行的反馈矩阵.数值仿真结果进一步证明了该策略的有效性.This paper is concerned about the networked control systems with random time delays, which occur in both sensor-controller and controller-actuator. For solving the problem of its dynamic features and uncertainty, one puts forward a state feedback control strategy with Markov characteristics. Moreover, in contrast with the way modeled through Markov stochastic process, each time-delay is described with two Markov chains, so that the resulting closed-loop system is established as a discrete Markovian jump system with four random parameters. A sufficient condition on the issue of finite-time bounded and stabilization is established through the theory of Lyapunov, and a linear matrix inequality （LMI） approach is employed to calculate the gain matrix of controller. Finally, an illustrative numerical simulation is given to demonstrate the effectiveness of the strategy proposed.

关 键 词：MARKOV跳变系统 有限时间稳定 状态反馈 随机时延 线性矩阵不等式 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]