计算机工程与应用 ›› 2018, Vol. 54 ›› Issue (2): 131-136.DOI: 10.3778/j.issn.1002-8331.1607-0309

• 模式识别与人工智能 • 上一篇    下一篇

改进果蝇优化算法在多目标搜索的应用

张  健,郭  星,李  炜   

  1. 安徽大学 计算机科学与技术学院,合肥 230601
  • 出版日期:2018-01-15 发布日期:2018-01-31

Application of improved?fruit fly optimization?algorithm on multi object search

ZHANG Jian, GUO Xing, LI Wei   

  1. School of Computer Science and Technology, Anhui University, Hefei 230601, China
  • Online:2018-01-15 Published:2018-01-31

摘要: 在实际工程优化问题中多数问题是多目标优化问题,多目标优化问题一直以来就是智能算法的研究热点。提出一种改进的果蝇优化算法,将其应用在多目标搜索领域,并成功使用该算法解决了一种多目标背包问题。算法在基本果蝇优化算法的基础上采用分群策略和动态半径,在群A中从种群位置开始以动态半径探索新的可行解,在群B中则通过非支配个体之间的交叉操作进行密集搜索。果蝇种群的位置在每一轮迭代产生的非劣解集中进行选取,提高了算法的收敛速度。通过在多个数据集下进行测试,并和粒子群算法、NSGA-2做了对比实验,最终结果显示使用该算法在特定条件下能取得较好的搜索效果,证明了使用果蝇优化算法解决多目标问题的可行性。

关键词: 果蝇优化算法, 多目标搜索, 背包问题

Abstract: In the practical engineering optimization problems, most of the problems are multi-objective optimization problem, multi-objective optimization problem is a research hotspot for a long time. This paper proposes a new fruit fly optimization algorithm, and successfully uses the algorithm to solve a multi-objective knapsack problem. The improved fruit fly algorithm uses clustering strategy and dynamic radius. In group A, a new feasible solution is explored from the position of population in the dynamic radius. In group B, it uses the crossover operation between the non dominated individuals. The initial location of the fly population is selected by random in the non dominated solutions. The convergence speed of the algorithm is improved greatly. By testing in multiple data sets with PSO and NSGA-2, the final results show that using the proposed algorithm under certain conditions can take the better search result, which proves the feasibility of using fruit fly optimization algorithm to solve the multi-objective problem.

Key words: Fruit Fly Optimization Algorithm(FFOA), multiple objective search, knapsack problem