摘要
为了提高算法的优化性能,在标准生物地理学优化算法基础上,概括了生物地理学理论的基本平衡定理,探索了在该定理下优化算法的各种迁移率模型的特点及行为,同时对这些迁移率模型进行一些典型基准函数的性能测试.通过函数优化实验可知:不同的迁移率模型将对算法的优化性能产生重要影响,同时符合自然规律的复杂迁移率模型的性能要优于简单的线性迁移率模型的性能;另外,变异率的不同也将对算法性能产生影响,对提高解集的适应度起着一定的作用.通过分析表明生物地理学优化算法是一种具有发展潜力的新型优化算法,并得出当前最有效的迁移率模型.
In order to improve the performance of biogeography-based optimization(BBO),this paper generalizes the basic principle of biogeography theory,explores the characteristics and behaviors of various migration rate models in BBO,and investigates performance through representative benchmark functions.The experimental results indicate that different migration rate models in BBO result in significant changes in performance,and complicated models which are closer to natural law outperform simple linear models.In addition,mutation rate does have an influence on optimization performance,and can provide a valuable approach for enhancing solutions.The performance study shows that it is a promising candidate for optimization and the most effect migration rate model for BBO is obtained.
出处
《东南大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2009年第S1期16-21,共6页
Journal of Southeast University:Natural Science Edition
关键词
优化
进化计算
迁移率模型
生物地理学
optimization
evolutionary computation
migration rate model
biogeography