• DocumentCode
    771004
  • Title

    Short term localized load prediction

  • Author

    Basu, Satyendra N.

  • Author_Institution
    Detroit Edison Co., MI, USA
  • Volume
    7
  • Issue
    1
  • fYear
    1992
  • fDate
    2/1/1992 12:00:00 AM
  • Firstpage
    389
  • Lastpage
    397
  • Abstract
    The development of a three hour load prediction model at Detroit Edison is described. All major electric utilities are required to maintain a certain level of integrity, especially during heat storm days. The predicted temperature and the anticipated load may require several operations including: circuit cutover, jumpering, and load transfer; installation of emergency equipment, like portable transformers; additional cooling to some major pieces of equipment; the shutoff of nonessential loads, e.g. air conditioning load, water heating load, etc.; and the adjustment of substation transformer regulators or line voltage regulators as necessary. Many of these operations require prior-knowledge of the system condition, especially loads at various local substations. An accurate short-term load prediction model can provide advance load information, which will help to assign priorities and sequence such operations. In order to perform the necessary remote and local operations including dispatching of the field crews, a three-hour advance notice is considered adequate. The author presents a method of predicting local load at a substation three hours in advance
  • Keywords
    load forecasting; Detroit Edison; circuit cutover; cooling; emergency equipment installation; heat storm days; jumpering; line voltage regulators; load transfer; portable transformers; short-term load prediction model; substation transformer regulators; three hour load prediction model; Load modeling; Power industry; Predictive models; Regulators; Resistance heating; Storms; Substations; Temperature; Transformers; Water heating;
  • fLanguage
    English
  • Journal_Title
    Power Systems, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0885-8950
  • Type

    jour

  • DOI
    10.1109/59.141733
  • Filename
    141733