• DocumentCode
    582004
  • Title

    On fault detection of NCSs subject to limited communication capacity

  • Author

    Li, Tao ; Zheng, Wei Xing

  • Author_Institution
    Dept. of Inf. & Commun., Nanjing Univ. of Inf. Sci. & Technol., Nanjing, China
  • fYear
    2012
  • fDate
    25-27 July 2012
  • Firstpage
    2765
  • Lastpage
    2770
  • Abstract
    This paper addresses the problem of fault detection filter design for networked control systems (NCS) subject to limited communication capacity and a class of sensor stuck faults. Considering the communication limitations (e.g., measurement quantization, signal transmission delays, and data packet dropouts) and all possible sensor stuck faults, a unified mathematical model is first presented. Based on this framework, a full-order fault detection filter is designed such that the residual system is asymptotically stable with the prescribed attenuation level in the generalized H sense. In order to further improve the detection performance, an optimization algorithm is proposed to minimize the threshold. Finally, a spring-mass-damper system is utilized to show the effectiveness of the proposed method.
  • Keywords
    fault diagnosis; filtering theory; networked control systems; optimisation; sensors; shock absorbers; signal processing; vibration control; NCS; asymptotic stability; attenuation level; communication limitations; fault detection filter design; full-order fault detection filter; generalized H sense; limited communication capacity; networked control systems; residual system; sensor stuck faults class; spring-mass-damper system; unified mathematical model; Actuators; Attenuation; Circuit faults; Delay; Educational institutions; Fault detection; Quantization; Communication Capacity; Fault Detection; NCSs;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Control Conference (CCC), 2012 31st Chinese
  • Conference_Location
    Hefei
  • ISSN
    1934-1768
  • Print_ISBN
    978-1-4673-2581-3
  • Type

    conf

  • Filename
    6390393