KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Author: search.filters.author.Mizutani, KenjiSubject: search.filters.subject.Mutation rate

Search Results

Now showing 1 - 2 of 2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Development of search space establishment-less type high speed genetic algorithm considering mutation rate and using parallel computer
    (2007-12-01)
    Takahiro, Ujiie
    ;
    Mizutani, Yosibumi
    ;
    Mizutani, Kenji
    ;
    Leelajindakrairerk, Monthon
    This paper presents a new genetic algorithm "Search Space Establishment-less Type High Speed Genetic Algorithm Considering Mutation Rate and Using Parallel Computer: PMSSELGA" for optimization control problems. Through the PMSSELGA, all the search space of the tuning parameters for the control are set in the range of 1000.0-0.001. And, a mutation rate is gradually to decrease with the generation from 100% to 15%. The PMSSELGA is speeded up by the parallel computation and the avoidance of repetitive calculations. It is applied to the optimum design of Multi-input power system stabilizer using Series Compensation Method in power system, which contains low-frequency oscillations, and the efficiency through the PMSSELGA is superior to conventional one.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Design of MPSS by search space logarithm type GA considering mutation rate
    (2007-09-25)
    Leelajindakrairerk, Monthon
    ;
    Mizutani, Kenji
    ;
    Mizutani, Yoshibumi
    ;
    Okabe, Takatugu
    ;
    Hatano, Shoichi
    This paper presents a new genetic algorithm "Search Space Logarithm Type Genetic Algorithm Considering Mutation Rate: MSSLGA" for optimization control problems of the Multi-input Power System Stabilizer (MPSS) design. Through the MSSLGA, in general, all the search space of tuning parameters for MPSS can be presumed in the range of 0.01 - 100.0. First, this paper proposes developing it into the logarithm to cover the large scale disadvantage. Second, it is presented that mutation rates are decreased together gradually with the generation from the big value. The MSSLGA is possible to obtain the higher-fitness answer easily than the conventional Genetic Algorithm using a constant mutation rate. It is applied to the optimum design of MPSS using Series Compensation Method in power system, which contains low-frequency oscillations, and the efficiency through the MSSLGA is superior to conventional one.