引用本文:武云丽,林波,曾海波.带有超大型挠性网状天线航天器姿控系统的参数化多目标设计[J].控制理论与应用,2019,36(5):766~773.[点击复制]
WU Yun-li,LIN-Bo,ZENG Hai-bo.Parametric multi-objective design for spacecrafts attitude control system with super flexible netted antennas[J].Control Theory and Technology,2019,36(5):766~773.[点击复制]
带有超大型挠性网状天线航天器姿控系统的参数化多目标设计
Parametric multi-objective design for spacecrafts attitude control system with super flexible netted antennas
摘要点击 2133  全文点击 914  投稿时间:2019-02-24  修订日期:2019-05-21
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2019.90100
  2019,36(5):766-773
中文关键词  控制系统参数化设计, 多目标设计, 大挠性航天器, 参数摄动, 高阶扰动模态
英文关键词  Parameterized design of control system, multi-objective design, super flexible spacecraft, parameter perturbations, high-order flexible mode
基金项目  
作者单位E-mail
武云丽 北京控制工程研究所 zxwh@sohu.com 
林波 北京控制工程研究所  
曾海波 北京控制工程研究所  
中文摘要
      本文研究了带有超大型挠性附件的卫星的姿态控制问题。关于挠性航天器振动抑制与姿态控制,绝大多 数已有的控制方法都是针对某一单一指标提出的。然而工程上要同时兼顾精度、快速性、平稳性、挠性部件的振 动抑制以及各种鲁棒性,因此挠性航天器的姿态控制系统设计实际上是一个典型的多目标设计问题。本文针对具 有超大挠性网状天线卫星的俯仰通道姿态系统,提出了一种基于输出反馈的鲁棒极点配置的参数化多目标设计方 法。首先给出了系统能控与能观的充分必要条件,然后给出了动态补偿器以及特征向量矩阵的参数化表达,并在 此基础上进一步对自由参向量进行了多目标优化,使得控制系统具有:(1)配置到期望区域的极点;(2)较低 的特征值灵敏度;(3)较强的高阶未建模动态抑制能力;(4)尽可能小的控制增益。最后,本文根据卫星工程 参数进行了控制器设计与仿真验证,仿真结果表明本文提出的方法可以在动态响应、高阶未建模动态抑制能力、 控制量峰值等方面优于传统的PID 控制器。
英文摘要
      This paper investigates the attitude control problem of a satellite with super-large flexible appendages. Most of existing approaches on the vibration suppression and attitude control of flexible spacecraft are developed with a singleobjective. However, in control engineering practice the performance indices such as accuracy, convergence speed, stability, vibration suppression in flexible structures and robustness are required to be considered simultaneously, which leads to a typical multi-objective design problem. In this paper, a parametric multi-objective design method based on output feedback for robust pole placement is proposed for pitch channel attitude systems with very large flexible mesh antenna satellites. Firstly, necessary and sufficient conditions for the controllability and observability of the system are derived. Then the parametric expressions of the dynamic compensator and the eigenvector matrix are given. Based on this, the multi-objective optimization of the free parameter vector is carried out, so that the control system has: 1) poles which are assigned to the desired regions; 2) lower eigenvalue sensitivity; 3) stronger ability to suppress high-order unmodeled dynamics; 4) control gain as small as possible. Finally, the controller design and simulation verification are carried out according to the satellite engineering parameters. The simulation results show that the proposed method can outperform the traditional PID controller in terms of dynamic response, high-order unmodeled dynamic suppression capability and peak control.