Computer Engineering and Applications ›› 2017, Vol. 53 ›› Issue (19): 55-60.DOI: 10.3778/j.issn.1002-8331.1610-0049

Previous Articles     Next Articles

Baldwin effect-based intuitionistic fuzzy adaptive differential evolution

WEN Tong, HUA Jixue, WANG Yi, MEI Haitao, JIA Qi, YANG Jinshuai   

  1. Air and Missile Defense College, Air-force Engineering University, Xi’an 710051, China
  • Online:2017-10-01 Published:2017-10-13

基于Baldwin效应的直觉模糊自适应差分进化算法

文  童,华继学,王  毅,梅海涛,贾  琪,杨进帅   

  1. 空军工程大学 防空反导学院,西安 710051

Abstract: To overcome the premature convergence of current versions of Differential Evolution(DE) algorithm, an advanced adaptive DE is proposed. Elite individuals, conducted with Baldwin learning, preserve their genotype and try several phenotypes to guide other individuals in uncertain generations; meanwhile, mutation factors are controlled with feedback by intuitionistic fuzzy reasoning. Tested by 19 typical benchmark functions and compared with other state-of-the-art advanced DE, the new method performs better global search capability and convergence speed.

Key words: Differential Evolution(DE), adaptive, intuitionistic fuzzy reasoning, Baldwin effect

摘要: 针对现有改进差分进化算法易陷入局部最优解的不足,提出一种改进的自适应差分进化算法。该算法对精英个体实施Baldwin学习,使其在不确定代数内保持基因型不变并尝试多种表现型以引导种群中其他个体进化;同时用直觉模糊推理的方法对缩放因子进行自适应反馈控制。通过对19个典型benchmark函数进行测试,并与其他知名改进差分进化算法对比,仿真结果表明该改进方法具有较强的跳出局部最优解能力和较快的收敛速度。

关键词: 差分进化, 自适应, 直觉模糊推理, Baldwin效应