计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (6): 40-45.DOI: 10.3778/j.issn.1002-8331.1508-0165

• 理论与研发 • 上一篇    下一篇

新型果蝇优化算法的研究

朱志同1,郭  星1,2,李  炜1,2   

  1. 1.安徽大学 计算智能与信号处理重点实验室,合肥 230039
    2.安徽大学 计算机科学与技术学院,合肥 230601
  • 出版日期:2017-03-15 发布日期:2017-05-11

Research on new fruit fly optimization algorithm

ZHU Zhitong1, GUO Xing1,2, LI Wei1,2   

  1. 1.Key Laboratory of Intelligent Computing Signal Processing, Ministry of Education, Anhui University, Hefei 230039, China
    2.School of Computer Science and Technology, Anhui University, Hefei 230601, China
  • Online:2017-03-15 Published:2017-05-11

摘要: 由于传统果蝇优化算法(FOA)具有寻优精度低和易陷入局部极小点的缺点,提出了一种具有不同飞行半径的分群搜索策略,使得在搜索区间内果蝇的种群多样性大大增加;同时在果蝇个体的飞行距离与方向的步长函数上,针对不同的果蝇子群引入了不同的函数,该类函数具有周期震荡性质,可以很好地避免果蝇群陷入局部极小点而无法求得最优解。通过对8个测试函数的仿真实验,验证了这些策略能够有效地提高搜索精度、收敛速度和稳定性。

关键词: 智能计算, 果蝇算法, 分群策略, 非线性函数

Abstract: For the traditional fruit flies optimization algorithm has the disadvantage of low optimization accuracy and easily
falls into local minimum point, this paper proposes a group search strategy which has different radii of flight to greatly
increase the population diversity of fruit flies in the search region. In the fruit flies individual flight distance and step function, it introduces different functions into different fruit fliessubgroups. Those functions have cycle oscillation properties.
It can well avoid flies group into local minimum point to get the optimal solution. Through the simulation experiment for
the eight test functions, it verifies these strategies can effectively improve the search precision, convergence speed and
stability.

Key words: intelligent computing, fruit flies algorithm, group search strategy, nonlinear function