ISSN 0253-2778

CN 34-1054/N

Open AccessOpen Access JUSTC

Improving firefly algorithm by keeping individual activity

Cite this:
https://doi.org/10.3969/j.issn.0253-2778.2016.02.005
  • Received Date: 16 January 2015
  • Rev Recd Date: 01 December 2015
  • Publish Date: 28 February 2016
  • There are some disadvantages in the basic firefly algorithm(FA), such as low solving precision, premature convergence and etc. To overcome these disadvantages, a novel improved FA(IFA) was proposed that keeps individual activity. Firstly, an adaptive control for gamma value was designed by using swarm distance. Secondly, a position calculation for fireflies was updated by using the search process information. Thirdly, a special mutation for the firefly swarm was executed to activate individuals and to make them explore the search space when losing activity. Finally, a perturbation and local search method for the best individual was proposed. Based on ten multi-model test functions, the test results show that the IFA has a better convergence speed and precision than the basic FA, PSO, ABC and other improved FA. The improved FA is a good method for complex function optimization.
    There are some disadvantages in the basic firefly algorithm(FA), such as low solving precision, premature convergence and etc. To overcome these disadvantages, a novel improved FA(IFA) was proposed that keeps individual activity. Firstly, an adaptive control for gamma value was designed by using swarm distance. Secondly, a position calculation for fireflies was updated by using the search process information. Thirdly, a special mutation for the firefly swarm was executed to activate individuals and to make them explore the search space when losing activity. Finally, a perturbation and local search method for the best individual was proposed. Based on ten multi-model test functions, the test results show that the IFA has a better convergence speed and precision than the basic FA, PSO, ABC and other improved FA. The improved FA is a good method for complex function optimization.
  • loading
  • 加载中

Catalog

    Article Metrics

    Article views (187) PDF downloads(124)
    Proportional views

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return