• DocumentCode
    3168189
  • Title

    Gain-adaptive model-separated predictive control

  • Author

    Chen, Jianmin ; Rink, R.E.

  • Author_Institution
    Dept. of Electr. Eng., Alberta Univ., Edmonton, Alta., Canada
  • Volume
    5
  • fYear
    1995
  • fDate
    22-25 Oct 1995
  • Firstpage
    4149
  • Abstract
    The model algorithmic control (MAC) scheme and gain adaptive deadtime compensator intrinsically can be converted into the internal model controller. One of the attractive features of MAC is that it can be used with only one tuning parameter. However, one-step MAC can not be used for systems with pure time-delay, and can not be used directly for non-minimum phase systems. Combining the advantages of MAC and gain adaptive deadtime compensator, this paper presents a gain-adaptive model-separated predictive control scheme. It can be directly used for time-delay systems and non-minimum phase systems. BIBO stability is obtained under certain conditions, and robustness can be further enhanced through a low-pass filter of control output, namely fc . Both parametric models and non-parametric models can be used by this predictive control scheme
  • Keywords
    compensation; delay systems; predictive control; robust control; BIBO stability; gain adaptive deadtime compensator; gain-adaptive model-separated predictive control; internal model controller; low-pass filter; model algorithmic control; nonminimum phase systems; nonparametric models; parametric models; robustness; time-delay systems; Adaptive control; Delay effects; Digital control; Low pass filters; Parametric statistics; Predictive control; Predictive models; Programmable control; Robust control; Robust stability;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Systems, Man and Cybernetics, 1995. Intelligent Systems for the 21st Century., IEEE International Conference on
  • Conference_Location
    Vancouver, BC
  • Print_ISBN
    0-7803-2559-1
  • Type

    conf

  • DOI
    10.1109/ICSMC.1995.538441
  • Filename
    538441