• DocumentCode
    2079682
  • Title

    The hybrid genetic algorithm based on the niche´s technology

  • Author

    Jiang Feng-Guo

  • Author_Institution
    Sch. of Civil Eng., Heilongjiang Inst. of Sci. & Technol., Harbin, China
  • fYear
    2010
  • fDate
    29-31 July 2010
  • Firstpage
    5276
  • Lastpage
    5279
  • Abstract
    Genetic arithmetic operators in genetic algorithm be improved, and a hybrid genetic algorithm of a gradient algorithm combining with the genetic algorithm be given against to the defects such as the prematurity, slow on the convergence rate, weak in the ability of local search, all those appeared on the progress of the genetic algorithm´ iteration. The niche´s technology be inducted due to the local optimal solution can easily appear on optimization of the multiply peak value. The analysis result indicates that not only the strong on the local search capacity of gradient algorithm be exhibited but also the strong on the general search capacity of genetic algorithm be combined based on the niche hybrid genetic algorithm, which make the capacity of convergence improve greatly. At the same time, the local optimal solution aroused in the optimization be avoided due to the niche technology be utilized, the hybrid genetic algorithm is an effective structural optimization method be proved in the computational example.
  • Keywords
    convergence; genetic algorithms; gradient methods; search problems; convergence rate; genetic algorithm iteration; gradient algorithm; local search; niche hybrid genetic algorithm; structural optimization method; Algorithm design and analysis; Classification algorithms; Convergence; Electronic mail; Genetics; Machine learning algorithms; Optimization; Genetic Algorithm; Gradient Algorithm; Hybrid Genetic Algorithm; Structural Optimization;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Control Conference (CCC), 2010 29th Chinese
  • Conference_Location
    Beijing
  • Print_ISBN
    978-1-4244-6263-6
  • Type

    conf

  • Filename
    5572372