基于反学习和正交交叉算子的元胞差分进化算法

doi:10.13190/j.jbupt.2014.03.002

北京邮电大学学报 ›› 2014, Vol. 37 ›› Issue (3): 7-12.doi: 10.13190/j.jbupt.2014.03.002

基于反学习和正交交叉算子的元胞差分进化算法

丁青锋^1,2, 郑国莘¹, 杨柳¹

1. 上海大学特种光纤与光接入重点实验室, 上海 200072;
2. 华东交通大学电气与电子工程学院, 南昌 330013

收稿日期:2013-07-25 出版日期:2014-06-28 发布日期:2014-06-08
作者简介:丁青锋（1980-），男，博士生，讲师，E-mail：brandy724@sina.com；郑国莘（1950-），男，教授，博士生导师.
基金资助:
国家自然科学基金项目（61132003，61171086，51267005）；上海市重点学科项目（S30108）

Cellular Differential Evolution Combined Opposition-Based Learning Initialization with Orthogonal Crossover

DING Qing-feng^1,2, ZHENG Guo-xin¹, YANG Liu¹

1. Key Laboratory of Specialty Fiber Optics and Optical Access Networks, Shanghai University, Shanghai 200072, China;
2. School of Electrical and Electronic Engineering, East China Jiaotong University, Nanchang 330013, China

Received:2013-07-25 Online:2014-06-28 Published:2014-06-08

摘要/Abstract

摘要：

提出了一种基于正交交叉算子的元胞差分进化算法. 进化初期采用反学习初始化方法获得初始候选种群，利用元胞结构的局部搜索方法替代控制参数调节差分进化算法的选择压力，从而平衡差分进化算法的探索能力和开发能力，利用元胞自动机的并行演化机制保持种群的多样性，从而避免陷入局部最优. 该算法利用无交叉因子的正交交叉算子，通过多元素重复试验加速种群收敛速度. 对多个典型测试函数的仿真实验结果表明，所提出的算法相较于多个差分进化改进算法具有更快的收敛速度和更好的计算精度.

关键词: 差分进化算法, 元胞自动机, 反学习, 正交交叉算子

Abstract:

A cellular differential evolution (cDE)algorithm based on orthogonal crossover is presented. The opposition-based learning initialization is used to search better solution in the initial stage, the local search within cellular neighbourhood structure is presented to tune the selection pressure instead of the control parameters. And the parallel evolution mechanism of cellular automata is given to ensure the diversity of the evolution population. In addition, the orthogonal crossover is adopted to accelerate the convergence speed with multi-element repeated trials. The performance of the cDE algorithm is evaluated on a suite of classic benchmark functions and compared favorably with the canonical DE and several DE variants. Simulation shows the proposed algorithm has better convergence performance and higher calculation accuracy.

Key words: differential evolution, cellular automata, opposition-based learning, orthogonal crossover

中图分类号:

TN929.53

丁青锋, 郑国莘, 杨柳. 基于反学习和正交交叉算子的元胞差分进化算法[J]. 北京邮电大学学报, 2014, 37(3): 7-12.

DING Qing-feng, ZHENG Guo-xin, YANG Liu. Cellular Differential Evolution Combined Opposition-Based Learning Initialization with Orthogonal Crossover[J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2014, 37(3): 7-12.

参考文献

[1] Das S, Suganthan P N. Differential evolution: a survey of the state-of-the-art[J]. IEEE Transactions on Evolutionary Computation, 2011, 15(1): 4-31.

[2] 毕晓君, 刘国安. 基于云差分进化算法的约束多目标优化实现[J]. 哈尔滨工程大学学报, 2012, 33(8): 1022-1031. Bi Xiaojun, Liu Guoan. A cloud differential evolutionary algorithm for constrained multi-objective optimization[J]. Journal of Harbin Engineering University, 2012, 33(8): 1022-1031.

[3] Jia Dongli, Zheng Guoxin, Khan M K. An effective memetic differential evolution algorithm based on chaotic local search[J]. Information Sciences, 2011, 181(15): 3175-3187.

[4] Alba E, Tomassini M. Parallelism and evolutionary algorithms[J]. IEEE Transactions on Evolutionary Computation, 2002, 6(5): 443-462.

[5] Alba E, Dorronsoro B. The exploration/exploitation tradeoff in dynamic cellular genetic algorithms[J]. IEEE Transactions on Evolutionary Computation, 2005, 9(2): 126-142.

[6] 鲁宇明, 黎明, 李凌. 一种具有演化规则的元胞遗传算法[J]. 电子学报, 2010, 38(7): 1603-1607. Lu Yuming, Li Ming, Li Ling. The cellular genetic algorithm with evolutionary rule[J]. Acta Electronica Sinica, 2010, 38(7): 1603-1607.

[7] Dorronsoro B, Bouvry P. Cellular genetic algorithms without additional parameters[J]. Journal of Supercomputing, 2013, 63(3): 816-835.

[8] Noman N, Vatanutanon J, Iba H. Tuning selection pressure in differential evolution using local selection[C]//2010 2^nd World Congress on Nature and Biologically Inspired Computing (NaBIC2010). Kitakyushu: IEEE Computer Society, 2010: 66-71.

[9] Rahnamayan S, Tizhoosh H R, Salama M M A. Opposition-based differential evolution[J]. IEEE Transactions on Evolutionary Computation, 2008, 12(1): 64-79.

[10] Leung Y, Wang Yuping. An orthogonal genetic algorithm with quantization for global numberical optimization[J]. IEEE Transactions on Evolutionary Computation, 2001, 5(1): 41-53.

基于反学习和正交交叉算子的元胞差分进化算法

Cellular Differential Evolution Combined Opposition-Based Learning Initialization with Orthogonal Crossover

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价