Scholarlink

<p>This paper concerns the decentralized event-based <i>H</i><sub>∞</sub> filter design problem for networked dynamic system (NDS). A more practical situation is studied, in which the communication between subsystems is affected by uncertainties and only local sampled measurement output is available for each filter in the developed filter scheme. Firstly, an event-triggered mechanism is introduced for each subsystem to process the sampled output in order to reduce the communication load. Secondly, on the basis of the well-posedness, the augmented filtering error system composed of the original NDS and the filter is modeled as a time-delay system of high dimension. After that, by employing the Lyapunov functional approach and space construction method, novel computationally attractive sufficient conditions are derived to check the well-posedness, asymptotic stability and <i>H</i><sub>∞</sub> performance of the augmented filtering error system. Thirdly, a co-design method of the filter and event-trigger matrices is obtained by using Finsler lemma and slack matrix approach. Finally, a numerical example is provided to demonstrate the effectiveness of the derived design approach.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Decentralized Event-Triggered H∞ Filter Design for Nonideally Interconnected Networked Dynamic System

  • 期刊论文

  • 发表日期：

    2025年7月3日

查看全文
  • Dan Zhang,
  • Hui Wang,
  • Zhengtao Ding,
  • Cuihua Zhang,
  • Xiaojuan Xue

摘要

This paper concerns the decentralized event-based H∞ filter design problem for networked dynamic system (NDS). A more practical situation is studied, in which the communication between subsystems is affected by uncertainties and only local sampled measurement output is available for each filter in the developed filter scheme. Firstly, an event-triggered mechanism is introduced for each subsystem to process the sampled output in order to reduce the communication load. Secondly, on the basis of the well-posedness, the augmented filtering error system composed of the original NDS and the filter is modeled as a time-delay system of high dimension. After that, by employing the Lyapunov functional approach and space construction method, novel computationally attractive sufficient conditions are derived to check the well-posedness, asymptotic stability and H∞ performance of the augmented filtering error system. Thirdly, a co-design method of the filter and event-trigger matrices is obtained by using Finsler lemma and slack matrix approach. Finally, a numerical example is provided to demonstrate the effectiveness of the derived design approach.