• DocumentCode
    498874
  • Title

    Genetic algorithm reseach and application in parameter identification inverse problem

  • Author

    Peng, Ya-mian ; Liu, Chun-Feng ; Yang, Ai-min ; Chang, Jin-cai ; Ji, Nan

  • Author_Institution
    Coll. of Sci., Hebei Polytech. Univ., Tangshan, China
  • Volume
    3
  • fYear
    2009
  • fDate
    12-15 July 2009
  • Firstpage
    1749
  • Lastpage
    1753
  • Abstract
    The parameter identification inverse problem is a kind of important practical problem in many science fields. The two-dimensional convection-diffusion equation was introduced in this paper and we try to solve its parameter identification inverse problem by the genetic algorithm. The finite element method was illustrated to solve the steady problem of two-dimensional convection-diffusion equation before we compute parameter identification inverse problem each time. By using genetic algorithm, we can search the best approximate solution from many initial points and obtained the global optimum solution by means of crossover operator and mutation operator. The results of numerical simulation shows that the genetic algorithm have the higher accuracy and the quicker convergent speed. And it is easy to program and calculate and of great applicable in practice.
  • Keywords
    finite element analysis; genetic algorithms; inverse problems; parameter estimation; crossover operator; finite element method; genetic algorithm; mutation operator; parameter identification inverse problem; two-dimensional convection-diffusion equation; Cybernetics; Educational institutions; Equations; Finite element methods; Genetic algorithms; Genetic mutations; Inverse problems; Machine learning; Numerical simulation; Parameter estimation; Convection-diffusion equation; Genetic algorithm; Inverse problem; Parameter identification;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Machine Learning and Cybernetics, 2009 International Conference on
  • Conference_Location
    Baoding
  • Print_ISBN
    978-1-4244-3702-3
  • Electronic_ISBN
    978-1-4244-3703-0
  • Type

    conf

  • DOI
    10.1109/ICMLC.2009.5212248
  • Filename
    5212248