基于精英协同的混洗差分进化算法及其应用

运筹与管理 ›› 2013, Vol. 22 ›› Issue (5): 17-23.

基于精英协同的混洗差分进化算法及其应用

张大斌^1,2, 杨添柔¹, 潘玉辰¹, 周茜¹, 张文生²

1.华中师范大学信息管理学院,湖北武汉,430079;
2.中国科学院自动化研究所,北京 100190

收稿日期:2012-08-20 出版日期:2013-10-25
作者简介:张大斌(1969-),男,教授、博士,研究方向:经济与社会预测预警,商务智能,信息系统等;杨添柔,硕士研究生。
基金资助:
国家自然科学基金资助项目(70971052);中国博士后基金资助项目(2012M510607)

Shuffled Differential Evolution Algorithm Based on Elite Synergy and Its Application

ZHANG Da-bin^1,2, YANG Tian-rou¹, PAN Yu-chen¹, ZHOU Xi¹, ZHANG Wen-sheng²

1. School of Information Management, Central China Normal University, Wuhan 430079, China;
2. Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China

Received:2012-08-20 Online:2013-10-25

摘要/Abstract

摘要： 提出了基于精英协同的混洗差分进化算法(Shuffled Differential Evolution,SDE)。该算法引入反向学习的初始化机制,并对设置的普通群和虚拟精英群采用不同的差分策略,进而将精英个体作为信息通道实现种群间的信息交流;同时,借助定期混洗机制实现种群间的文化交流,从而达到协同进化的目的;此外,对长期停滞的个体进行跳变操作,以充分挖掘种群的搜索潜能,增强搜索的有效性。通过函数仿真,并与PSO及其它差分进化算法比较,结果表明该算法具有较好的寻优能力。

关键词: 最优化理论, 差分进化, 反向学习机制, 协同机制, 混洗思想, 多种群, 连续域问题

Abstract: This paper presents a novel Shuffled Differential Evolution algorithm(SDE)based on elite synergy. The algorithm introduces the initialization mechanism of opposition-based learning, employs different differential strategies fof several ordinary groups and a virtual elite group so as to take the elite individuals as the information channel for achieving information exchange among different groups. Meanwhile, it realizes the inter-cultural exchange among different groups by using a regularly shuffled mechanism which regroups the small groups via hash function, so as to achieve the population co-evolution. In addition, hopping operation on the individuals which are in the long-term stagnation can fully tap the potential of population search and enhance the effectiveness of the search. By the benchmark function experiments, the SDE performs better optimization capability in comparison with the Particle Swarm Optimization and other Differential Evolution algorithms.

Key words: optimization theory, differential evolution, opposition-based learning, collaborative mechanism, shuffled idea, multi-population, continuous problem

中图分类号:

TP301.6

张大斌, 杨添柔, 潘玉辰, 周茜, 张文生. 基于精英协同的混洗差分进化算法及其应用[J]. 运筹与管理, 2013, 22(5): 17-23.

ZHANG Da-bin, YANG Tian-rou, PAN Yu-chen, ZHOU Xi, ZHANG Wen-sheng. Shuffled Differential Evolution Algorithm Based on Elite Synergy and Its Application[J]. Operations Research and Management Science, 2013, 22(5): 17-23.

参考文献

[1] Storn R, Price K. Differential evolution-a simple and efficient adaptive scheme for global optimization over continuous spaces[R]. International Computer Science Institute, 1995.
[2] Storn R, Price K. Differential evolution-a simple and efficient heuristic for global optimization over continuous spaces[J]. Journal of Global Optimization, 1997, 11(4): 341-359.
[3] 刘波,王凌,金以慧.差分进化算法研究进展[J].控制与决策,2007,22(7):721-729.
[4] 吴亮红,王耀南,袁小芳,周少武.自适应二次变异差分进化算法[J].控制与决策,2006,21(8):899-902.
[5] Teo J. Exploring dynamic self-adaptive populations in differential evolution[J]. Soft Computing, 2006, 10(8): 673-686.
[6] Rahnamayan S, Tizhoosh H R, Salama M M A. Opposition-based differential evolution[J]. IEEE Transactions on Evolutionary Computation, 2008, 12(1): 64-79.
[7] Wang Y H, Wang X K, Teng H F. Opposition-based cooperative co-evolutionary differential evolution algorithm with gaussian mutation for simplified satellite module optimization[J].Information Technology Journal, 2012, 11(1): 67-75.
[8] 刘罡,李元香,郑昊.保存基因的2-Opt一般反向差分演化算法[J].小型微型计算机系统,2012,33(1):115-120.
[9] Piotrowski A P, Napiorkowski J J, Kiczko A. Differential evolution algorithm with separated groups for multi-dimensional optimization problems[J]. European Journal of Operational Research, 2012, 216: 33-46.
[10] Tasgetiren M F, Suganthan P N. A multi-populated differential evolution algorithm for solving constrained optimization problem[C].//2006 IEEE Congress on Evolutionary Computation. Vancouver BC, Canada. IEEE, 2006. 33-40.
[11] 葛延峰,金文静,高立群,冯达.多种群并行的自适应差分进化算法[J].东北大学学报(自然科学版),2011,32(4):481-484.
[12] 胡中波,熊盛武,苏清华.基于小生境的混合差分演化模拟退化算法[J].计算机工程与应用,2007,43(2):105-107.
[13] Mohammed E A. Performance assessment of foraging algorithm vs. evolutionary algorithms[J]. Information Sciences, 2012, (182): 243-263.
[14] Eusuff M, Lansery K, Pasha F. Shuffled frog-leaping algorithm: a memetic meta-heuristic for discrete optimization[J]. Engineering Optimization, 2006, 38(2): 129-154.