• DocumentCode
    2191170
  • Title

    Strategic power infrastructure defense (SPID)

  • Author

    Chen-Ching Liu

  • Author_Institution
    Adv. Power Technol. Center, Washington Univ., Seattle, WA, USA
  • fYear
    2004
  • fDate
    6-10 June 2004
  • Abstract
    Summary form only given. An advanced system called, "strategic power infrastructure defense (SPID) system," was developed by the Advanced Power Technologies (APT) Consortium consisting of the University of Washington, Arizona State University, Iowa State University and Virginia Tech. By incorporating multi-agent system technologies, the SPID system is able to assess power system vulnerability, monitor hidden failures of protective devices, and provide adaptive control actions to prevent catastrophic failures and cascading sequences of events. The SPID program was sponsored by EPRI and the U.S. Department of Defense. In this session, the panelist will summarize the SPID methodology and the multi-agent system technologies that are critical for the implementation of the SPID system. The software agents in the SPID system are organized in a multi-layer structure to facilitate collaboration among the agents. The agents communicate through a protocol called FIPA. SPID has the ability to adapt to changes in the power infrastructure environment through the embedded machine learning capability. Simulation examples of the multi-agent system is provided.
  • Keywords
    learning (artificial intelligence); multi-agent systems; power engineering computing; power system protection; power system security; adaptive control; machine learning capability; multiagent system; power infrastructure; protective device; strategic power infrastructure defense; Adaptive control; Collaborative software; Condition monitoring; Multiagent systems; Power system control; Power system faults; Power system protection; Power systems; Protocols; Software agents;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Power Engineering Society General Meeting, 2004. IEEE
  • Conference_Location
    Denver, CO
  • Print_ISBN
    0-7803-8465-2
  • Type

    conf

  • DOI
    10.1109/PES.2004.1372735
  • Filename
    1372735