Title :
Niching genetic algorithm adopting restricted competition selection combined with pattern search method
Author :
Kim, Jae-Kwang ; Cho, Dong-Hyeok ; Jung, Hyun-Kyo ; Lee, Cheol-Gyun
Author_Institution :
Sch. of Electr. Eng., Seoul Nat. Univ., South Korea
fDate :
3/1/2002 12:00:00 AM
Abstract :
Among several niching methods, the restricted competition selection (RCS) method searches almost all the location of multiple solutions in the search space. This paper proposes a neo-niching method using the RCS method combined with pattern search method to reduce the calculation time. Through application to the numerical examples and optimal design of interior permanent magnet synchronous motor, it is verified that the proposed method searches the peaks fast and is appropriate for the design of electric machines
Keywords :
electric machine CAD; genetic algorithms; permanent magnet motors; synchronous motors; RCS method; calculation time; electric machine design; interior permanent magnet synchronous motor; multiple solutions; neo-niching method; niching genetic algorithm; optimal design; pattern search method; restricted competition selection; search space; Ecosystems; Electric machines; Electromagnetic devices; Genetic algorithms; Permanent magnet motors; Search methods; Shape; Stochastic processes;
Journal_Title :
Magnetics, IEEE Transactions on