计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (20): 31-35.

• 理论研究、研发设计 • 上一篇    下一篇

基于logistic模型的自适应布谷鸟算法

陈  华,张艺丹   

  1. 中国石油大学(华东) 理学院,山东 青岛 266580
  • 出版日期:2015-10-15 发布日期:2015-10-30

Adaptive cuckoo algorithm based on logistic model

CHEN Hua, ZHANG Yidan   

  1. College of Science, China University of Petroleum (East China), Qingdao, Shandong 266580, China
  • Online:2015-10-15 Published:2015-10-30

摘要: 针对标准布谷鸟算法的相关问题,提出了一种基于logistic模型的动态步长控制因子和动态发现概率的改进布谷鸟算法。改进的算法在运行时可以自动调节步长控制因子和发现概率的大小,且在算法初期可以使种群保持多样性,提高了全局最优值的搜索能力;随着局部最优值搜索能力的增强,算法在后期逐渐趋于稳定。通过用几种典型的Benchmarks函数进行模拟试验,试验结果证实了所提出的算法计算精度高、收敛速度快。

关键词: 布谷鸟算法, logistic模型, 动态步长控制因子, 动态发现概率

Abstract: On the basis of related presentations of the standard cuckoo algorithms, an improved cuckoo algorithm of dynamic step-size control factor and dynamic find-statistic based on logistic model is presented. The improved algorithms can automatically adjust step-size control factor and find-statistic during the running time, so it can keep the individuals diversity and improve searching ability of global optimum in the population at the initial generations. However, the algorithm is gradually stabilized with searching ability of local optimum improved at a later time. Several classic Benchmarks functions are tested and the results show that the proposed algorithms have fast convergence and higher calculation accuracy.

Key words: cuckoo algorithm, logistic model, dynamic step-size control factor, dynamic find-statistic