计算机工程与应用 ›› 2016, Vol. 52 ›› Issue (17): 146-151.

• 模式识别与人工智能 • 上一篇    下一篇

一种改进的生物地理学优化算法

鲁宇明1,4,王彦超2,刘嘉瑞3,Wu Liu4   

  1. 1.江西省图像处理与模式识别重点实验室,南昌 330063
    2.南昌航空大学 航空制造工程学院,南昌 330063
    3.西北工业大学 理学院,西安 710072
    4.耶鲁大学 放射治疗系,康涅狄格州纽黑文 06511
  • 出版日期:2016-09-01 发布日期:2016-09-14

Improved biogeography-based optimization algorithm

LU Yuming1,4, WANG Yanchao2, LIU Jiarui3, Wu Liu4   

  1. 1.Key Laboratory of Image Processing and Pattern Recognition in Jiangxi Province, Nanchang 330063, China
    2.College of Aeronautical Manufacturing Engineering, Nanchang Hangkong University, Nanchang 330063, China
    3.School of Natural and Applied Sciences, Northwestern Polytechnical University, Xi’an 710072, China
    4.Department of Radiology Therapeutics, Yale University, Connecticut, Newhaven 06511, United States
  • Online:2016-09-01 Published:2016-09-14

摘要: 生物地理学优化算法(BBO)作为一种新型的智能算法,在其提出不到十年的时间内受到学界的广泛关注和研究,并显示出了广阔的应用前景。为了提高算法的优化性能,对BBO算法提出一种改进,该算法在将差分优化算法(DE)中的局部搜索策略同BBO算法中的迁移策略相结合的基础上,针对迁移算子和变异算子分别进行改进,提出了二重迁移算子和二重变异算子,使得栖息地个体在进化过程中得到更高的进化概率,从而使得算法的寻优能力得到进一步提升。通过6个高维函数的测试,结果表明该算法在优化高维优化问题时,较其他几种生物地理学优化算法具有更好的收敛性和稳定性。

关键词: 生物地理学优化算法, 局部搜索策略, 二重迁移算子, 二重变异算子

Abstract: The Biogeography-Based Optimization algorithm(BBO) is a new intelligent algorithm. It has received wide concern and study by the academic community within the ten years since it was proposed, and shows a broad application prospect. In order to improve the performance of the algorithm, the paper proposes an improved BBO algorithm. The improved algorithm based on the combination of the local search strategy in Differential Evolution(DE) algorithm and the migration strategy in BBO algorithm, which raises a kind of double migration operator and double mutation operator, aims to make the operators work better. These improvements make the habitats get a higher evolutionary probability in the process of evolution and the algorithm’s optimization ability get further improvement. Through the test of 6 high dimension basic functions, the result shows that the improved algorithm proposed in this paper has better convergence and stability compared with other optimization algorithm referred in the optimization of high dimensional optimization problem.

Key words: biogeography-based optimization algorithm, local search strategy, double migration operator, double mutation operator