《计算机应用研究》|Application Research of Computers

IYYPO:一种改进的阴阳对优化算法

Improved Yin-Yang-pair optimization algorithm

免费全文下载 (已被下载 次)  
获取PDF全文
作者 李大海,艾志刚,王振东
机构 江西理工大学 信息工程学院,江西 赣州 341000
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2021)01-027-0134-06
DOI 10.19734/j.issn.1001-3695.2019.11.0612
摘要 阴阳对优化算法是一种新颖的轻量级随机优化算法,利用两点(全局探索点<i>P</i><sub>2</sub>和局部开发点<i>P</i><sub>1</sub>)的迭代交换来实现优化搜索。用户定义参数直接影响该算法的全局探索和局部开发之间的平衡,并且对算法的性能有着重要的影响。为提高该算法的优化性能,首先分析了原算法的用户定义参数(缩放因子<i>α</i>)对于性能的影响,随后提出用户定义参数线性与非线性递减三种改进的阴阳对优化算法。采用2013年进化计算大会中单目标实参算法竞赛中使用的28个测试函数进行性能评估,结果表明相比于原算法,改进后的算法具有更高的计算精度和更快的收敛速度。最后通过一个工程优化任务来展示改进后算法的性能。
关键词 阴阳对优化; 进化计算; 单目标优化; 参数优化
基金项目 国家自然科学基金资助项目(61563019)
本文URL http://www.arocmag.com/article/01-2021-01-027.html
英文标题 Improved Yin-Yang-pair optimization algorithm
作者英文名 Li Dahai, Ai Zhigang, Wang Zhendong
机构英文名 School of Information Engineering,Jiangxi University of Science & Technology,Ganzhou Jiangxi 341000,China
英文摘要 Yin-Yang-pair optimization algorithm is a novel lightweight optimization algorithm, which uses the iterative exchange of two points(the global exploration point <i>P</i><sub>2</sub> and the local exploitation point <i>P</i><sub>1</sub>) to achieve the search of optimal solutions. The user defined parameters of the YYPO directly affect the balance of global exploration and local exploitation of the algorithm and impact the performance of the algorithm. In order to improve the optimization performance of the algorithm, this paper analyzed effects of the user-defined parameters(expansion/contraction factor <i>α</i>) to the performance of the original algorithm, then proposed the IYYPO algorithm. IYYPO adopted linear and non-linear strategy to gradually reduce user-defined parameters. This paper also used 28 test functions, which was the test benchmark of the competition of single objective real parameter algorithm in 2013 Evolutionary Computing Conference, as the testbench to evaluate the performance of three new algorithms. Compared with the original algorithm, experiment results illustrate that the proposed algorithm can archive higher accuracy and faster convergence speed. At last, this paper tested the improved algorithms in a practical optimization problem to show the performance of the improved algorithms.
英文关键词 Yin-Yang-pair optimization(YYPO); evolutionary computation; single objective optimization; parametric optimization
参考文献 查看稿件参考文献
 
收稿日期 2019/11/27
修回日期 2020/1/15
页码 134-139,144
中图分类号 TP301.6
文献标志码 A