计算机工程 ›› 2009, Vol. 35 ›› Issue (13): 179-182.doi: 10.3969/j.issn.1000-3428.2009.13.062

• 人工智能及识别技术 • 上一篇    下一篇

基于单纯形算子的混合差分进化算法

刘 洁1,吴亮红2,刘建勋3   

  1. (1. 湖南工程学院设计艺术学院,湘潭 411104;2. 湖南科技大学信息与电气工程学院,湘潭 411201; 3. 湖南科技大学计算机科学与工程学院,湘潭 411201)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-07-05 发布日期:2009-07-05

Hybrid Differential Evolution Algorithm Based on Simplex Operator

LIU Jie1, WU Liang-hong2, LIU Jian-xun3   

  1. (1. School of Design Art, Hunan Institute of Engineering, Xiangtan 411104; 2. School of Information and Electric Engineering, Hunan University of Science and Technology, Xiangtan 411201; 3. School of Computer Science and Engineering, Hunan University of Science and Technology, Xiangtan 411201)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-07-05 Published:2009-07-05

摘要: 针对DE/rand/1/bin方案收敛速度慢的缺点,提出一种将单纯形确定性算法和差分进化随机搜索算法相结合的混合优化算法。利用差分进化算法搜索范围广、全局搜索能力强和单纯形算法局部搜索能力强、收敛速度快的特性,较大地提高了差分进化算法的收敛速度和搜索精度。典型Benchmarks复杂函数优化实验表明,该算法优化效率高、优化性能好、对初值具有较强的鲁棒性,性能优于单一的优化方法。

关键词: 复杂非线性函数, 差分进化算法, 单纯形法, 混合优化算法

Abstract: Aiming at the slow convergence of DE/rand/1/bin strategy, this paper presents a hybrid optimization algorithm named SMDE incorporated Simplex Method(SM) into Differential Evolution(DE) algorithm. It takes use of good global searching ability of DE and good local searching ability and fast convergence of SM, so that the convergence speed and solution precision of DE are improved. Experimental results on several classical Benchmarks complex functions show that the hybrid optimization algorithm is effective, efficient and fairly robust to initial conditions, and its performances excels those single optimization methods.

Key words: complex nonlinear function, Differential Evolution(DE) algorithm, Simplex Method(SM), hybrid optimization algorithm

中图分类号: