系统仿真学报 ›› 2020, Vol. 32 ›› Issue (9): 1717-1723.doi: 10.16182/j.issn1004731x.joss.19-0087

• 仿真建模理论与方法 • 上一篇    下一篇

一种基于GA的新型生物地理学优化算法研究

王宁, 魏利胜   

  1. 安徽工程大学电气工程学院,安徽 芜湖 241000
  • 收稿日期:2019-03-02 修回日期:2019-04-21 出版日期:2020-09-18 发布日期:2020-09-18
  • 作者简介:王宁(1995-),男,安徽宿州,硕士生,研究方向为复杂系统的建模、控制与优化;魏利胜(1978-),男,安徽巢湖,博士,教授,研究方向为多目标优化,机械视图等。
  • 基金资助:
    安徽省自然科学基金(1608085MF146),安徽工程大学青年拔尖人才项目(2016BJRC008)

Research on a Novel Biogeography-Based Optimization Algorithm Based On GA

Wang Ning, Wei Lisheng   

  1. School of Electrical Engineering, Anhui Polytechnic University, Wuhu 241000, China
  • Received:2019-03-02 Revised:2019-04-21 Online:2020-09-18 Published:2020-09-18

摘要: 为了使生物地理学优化算法的优化能力得到进一步提高,提出了一种基于遗传算法的新型生物地理学优化算法。在迁移操作之前增加了选择操作,采用了“轮盘赌”的方法选择出迁移个体,以使适应度较高的个体可以优先得到迁移,并且变异操作结合了遗传高斯变异操作方法,从而更好地提升了算法的优化性能;在此基础上,从理论上详细推导了该方法的收敛性条件。使用了5种测试函数进行实验,结果证明了改进后的算法在优化结果和收敛速度方面要更优。

关键词: 生物地理学优化算法, 遗传算法, 选择操作, 高斯变异

Abstract: In order to further improve the optimization ability of biogeography-based optimization algorithm, a new genetic algorithm is proposed. The selection operation is added before the migration operation, and the migration individual is selected by the method of "roulette", so that the individuals with higher fitness can be preferentially migrated. The mutation operation combines the genetic gaussian mutation method, and the optimization performance of the algorithm is improved. The convergence condition of the method is derived in theory. Five test functions are used in the experiments, and the results prove that the ameliorated algorithm is better at the results of optimization and rate of convergence.

Key words: biogeography-based optimization (BBO), genetic algorithm, selection operation, gauss mutation

中图分类号: