• DocumentCode
    3474638
  • Title

    Application of integrated weight method and support vector machine in the comprehensive evaluation of power quality

  • Author

    Zhang, Xiaohua ; Chen, Xingying ; Liu, Haoming ; Zhao, Bo

  • Author_Institution
    Coll. of Electr. Eng., Hohai Univ., Nanjing
  • fYear
    2008
  • fDate
    6-9 April 2008
  • Firstpage
    2181
  • Lastpage
    2186
  • Abstract
    In this paper, a novel comprehensive strategy, consisting of an integrated weight method and support vector machine (SVM) method, is proposed to evaluate power quality. In integrated weight method, analysis hierarchy process (AHP) is employed to determine the subjective weights, and improved scatter degree (ISD) is employed to determine the objective weights, then the comprehensive weights could be fixed based on addition principle. Since SVM could deal with small samples and nonlinear problems effectively, it is introduced into the comprehensive evaluation of power quality. Test results show that the proposed strategy is feasible to PQ evaluation problems. At the same time it can obtain reasonable evaluation results quickly. The obtained model has so good generalization performance that it is fit for large practical samples.
  • Keywords
    power engineering computing; power supply quality; support vector machines; PQ evaluation problems; SVM; analysis hierarchy process; improved scatter degree; integrated weight method; power quality; support vector machine; Analytical models; Artificial neural networks; Electricity supply industry; Mathematics; Power quality; Radial basis function networks; Scattering; Support vector machine classification; Support vector machines; Testing; analysis hierarchy process; comprehensive evaluation; improved scatter degree; integrated weight method; power quality; support vector machine;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Electric Utility Deregulation and Restructuring and Power Technologies, 2008. DRPT 2008. Third International Conference on
  • Conference_Location
    Nanjuing
  • Print_ISBN
    978-7-900714-13-8
  • Electronic_ISBN
    978-7-900714-13-8
  • Type

    conf

  • DOI
    10.1109/DRPT.2008.4523772
  • Filename
    4523772