• DocumentCode
    1761370
  • Title

    Synthesis of low-coefficient sensitivity controllers with respect to multiplicative controller coefficient variations

  • Author

    Xiang-Gui Guo ; Qiang Gao ; Zhe Xu

  • Author_Institution
    Tianjin Key Lab. for Control Theor. & Applic. in Complicated Syst., Tianjin Univ. of Technol., Tianjin, China
  • Volume
    9
  • Issue
    1
  • fYear
    2015
  • fDate
    1 2 2015
  • Firstpage
    120
  • Lastpage
    128
  • Abstract
    In this study, a set of optimal state-space descriptions with minimum H-sensitivity to multiplicative controller coefficient variations and a desirable disturbance attenuation level is synthesised for discrete-time linear systems. The first aim is to present a new strategy to evaluate the H-sensitivity of the transfer function to multiplicative controller coefficient variations. The H norm of the sum of coefficient sensitivity functions is regarded as an H-sensitivity measure. The second contribution is to present less conservative conditions to design low-sensitivity controllers via minimisation of the H-sensitivity measure. In addition, a linear matrix inequality-based algorithm is developed to obtain the controller gain matrices. Finally, two numerical examples which demonstrate the effectiveness and advantages of the proposed method are presented.
  • Keywords
    H optimisation; control system synthesis; discrete time systems; linear matrix inequalities; linear systems; minimisation; sensitivity analysis; transfer functions; H-sensitivity measure; controller gain matrices; discrete-time linear systems; disturbance attenuation level; less conservative conditions; linear matrix inequality-based algorithm; low-coefficient sensitivity controllers; minimisation; multiplicative controller coefficient variations; optimal state-space descriptions; transfer function;
  • fLanguage
    English
  • Journal_Title
    Control Theory & Applications, IET
  • Publisher
    iet
  • ISSN
    1751-8644
  • Type

    jour

  • DOI
    10.1049/iet-cta.2014.0272
  • Filename
    6987402