摘要
网络控制系统与传统的点对点控制方式不同,数据传输具有时延和随机丢包,且为了对信息资源进行合理的分配和调度常采用时变采样方式。本文首先给出NCS的离散区间动态模型,该模型以Bernoulli分布序列描述NCS数据丢包的随机性,并以不确定区间矩阵描述NCS的采样周期与传输时延的时变性;在此基础上,基于线性矩阵不等式(LMI)技术,给出具有时变有界采样周期和数据随机丢包NCS的鲁棒H∞控制器存在充分条件,且将控制器的求解问题转化为一个有限维LMI凸优化问题,便于数值计算。所设计的控制器可以保证系统渐近稳定且具有一定的H∞扰动抑制水平γ。数值仿真表明提出控制器设计方法的有效性。
Networked control system is different from traditional point to point control system, data transmission often gets into trouble, such as transmission time-delay and random packet dropouts. To obtain reasonable allocation of information resources and scheduling, time-varying sampling period method is often proposed. Firstly, discrete interval dynamic model of NCS was built, in which the randomness of the data packet loss was described via a stochastic Bernoulli distribution sequence, and the time varying characteristics of sampling period and transmission time-delay were represented by uncertain interval matrices. Moreover, sufficient condition for the existence of robust H∞ controller for NCS with time-varying sampling period and random packet dropouts was established in terms of linear matrix inequalities. To facilitate the calculation, the design of controller was cast into a finite-dimensional LMI convex optimization problem. The admissible controller guarantees robust asymptotic stability and a desired H∞ noise attenuation levelγ. The simulation results of numerical example illustrate the effectiveness of the proposed design scheme.
出处
《计算技术与自动化》
2013年第3期1-5,共5页
Computing Technology and Automation
基金
国家青年基金项目(61004067)
黑龙江省自然科学基金项目(F12511002)
