• DocumentCode
    3105235
  • Title

    A computer vision early-warning ice detection system for the Smart Grid

  • Author

    Wachal, Randy ; Stoezel, Jean-Sebastien ; Peckover, Monty ; Godkin, Daryl

  • fYear
    2012
  • fDate
    7-10 May 2012
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    A state of the art early ice warning system has been developed and implemented for the Manitoba Hydro Ice Storm Management Program. The vision based ice detection system measures ice accumulation using digital images directly from the overhead line conductors. The ice detection system provides early warning ice alarms, ice accumulation rate information and accurate visual information of ice profiles to the appropriate staff using the corporate WAN infrastructure. Additional research has been undertaken in the development of an algorithm to aid in the detection of hoar frost transformation to ice. Hoar frost on the transmission conductors presents little risk, however under certain conditions hoar frost can transform into ice very quickly, creating a serious condition. New digital image recognition methods are under development and are presented in this paper. It is envisioned that these methods can predict the rate of change of the transformation of hoar frost into ice and thus provide an early warning indication.
  • Keywords
    alarm systems; computer vision; image recognition; overhead line conductors; smart power grids; wide area networks; Manitoba hydro ice storm management program; WAN infrastructure; computer vision early-warning ice detection system; digital image recognition methods; hoar frost transformation; ice accumulation rate information; overhead line conductors; smart grid; transmission conductors; visual information; Cameras; Ice; Machine vision; Monitoring; Servers; Storms; Hoar frost; Ice Detection; Real time data collection; Smart Grid; Vision Based Monitoring System; Weather monitoring;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Transmission and Distribution Conference and Exposition (T&D), 2012 IEEE PES
  • Conference_Location
    Orlando, FL
  • ISSN
    2160-8555
  • Print_ISBN
    978-1-4673-1934-8
  • Electronic_ISBN
    2160-8555
  • Type

    conf

  • DOI
    10.1109/TDC.2012.6281621
  • Filename
    6281621