当前位置: X-MOL 学术arXiv.cs.SY › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Distributed $\mathscr{H}_2$ Control for Interconnected Discrete-Time Systems: A Dissipativity-based Approach
arXiv - CS - Systems and Control Pub Date : 2020-01-14 , DOI: arxiv-2001.04875
Tom R. V. Steentjes, Mircea Lazar, Paul M. J. Van den Hof

This work is concerned with the analysis, existence and synthesis of distributed output-feedback controllers, that achieve stability and $\mathscr{H}_2$ performance for discrete-time linear interconnected systems. We consider an interconnection structure of local controllers that resembles the plant's interconnection structure, which may correspond to an arbitrary graph. The dissipativity-based approach to distributed discrete-time $\mathscr{H}_2$ control presented in this paper complements other dissipativity-based approaches in the literature to distributed continuous-time $\mathscr{H}_2$ control and distributed $\mathscr{H}_\infty$ control. Moreover, the developed method yields a convex alternative to state-of-the-art methods for distributed discrete-time $\mathscr{H}_2$ control, which are typically not convex or consider unstructured problems. We provide an overview of related results and show the relation between sufficient conditions for $\mathscr{H}_2$ and $\mathscr{H}_\infty$ performance, for both discrete- and continuous-time interconnected linear systems. Sufficient conditions are stated for the existence of a distributed controller achieving a pre-specified $\mathscr{H}_2$ performance. A method for subsequent controller reconstruction is provided by an algebraic procedure. We illustrate the controller synthesis for a large-scale oscillator network, for which the central $\mathscr{H}_2$ control problem can be computationally intractable on a modern PC.

中文翻译:

互连离散时间系统的分布式 $\mathscr{H}_2$ 控制:一种基于耗散的方法

这项工作涉及分布式输出反馈控制器的分析、存在和综合,这些控制器可实现离散时间线性互连系统的稳定性和 $\mathscr{H}_2$ 性能。我们考虑类似于工厂互连结构的本地控制器的互连结构,它可能对应于任意图。本文提出的基于耗散的分布式离散时间 $\mathscr{H}_2$ 控制方法补充了文献中其他基于耗散的方法,用于分布式连续时间 $\mathscr{H}_2$ 控制和分布式 $\ mathscr{H}_\infty$ 控制。此外,所开发的方法为分布式离散时间 $\mathscr{H}_2$ 控制的最先进方法产生了凸替代,通常不是凸的或考虑非结构化问题。我们提供了相关结果的概述,并展示了离散和连续时间互连线性系统的 $\mathscr{H}_2$ 和 $\mathscr{H}_\infty$ 性能的充分条件之间的关系。为实现预先指定的 $\mathscr{H}_2$ 性能的分布式控制器的存在陈述了充分的条件。代数过程提供了一种用于后续控制器重构的方法。我们说明了大规模振荡器网络的控制器综合,在现代 PC 上,中心 $\mathscr{H}_2$ 控制问题在计算上是难以处理的。对于离散和连续时间互连线性系统。为实现预先指定的 $\mathscr{H}_2$ 性能的分布式控制器的存在陈述了充分的条件。代数过程提供了一种用于后续控制器重构的方法。我们说明了大规模振荡器网络的控制器综合,在现代 PC 上,中心 $\mathscr{H}_2$ 控制问题在计算上是难以处理的。对于离散和连续时间互连线性系统。为实现预先指定的 $\mathscr{H}_2$ 性能的分布式控制器的存在陈述了充分的条件。代数过程提供了一种用于后续控制器重构的方法。我们说明了大规模振荡器网络的控制器综合,在现代 PC 上,中心 $\mathscr{H}_2$ 控制问题在计算上是难以处理的。
更新日期:2020-01-15
down
wechat
bug