计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (7): 48-52.DOI: 10.3778/j.issn.1002-8331.1805-0421

• 理论与研发 • 上一篇    下一篇

修正浓度与适应步长的果蝇优化算法

信成涛,邹  海   

  1. 安徽大学 计算机科学与技术学院,合肥 230601
  • 出版日期:2019-04-01 发布日期:2019-04-15

Fruit Fly Optimization Algorithm with Modified Concentration and Adaptive Step

XIN Chengtao, ZOU Hai   

  1. School of Computer Science and Technology, Anhui University, Hefei 230601, China
  • Online:2019-04-01 Published:2019-04-15

摘要: 基本果蝇优化算法在寻优求解时浓度值只能为正,无法对浓度为负时达到最优的问题进行寻优。另外基本果蝇算法在寻优求解时,步长是随机的,这就容易使算法早熟,陷入局部最优解,算法的求解精度也不高。针对基本果蝇算法的这些问题,提出了一种修正浓度与适应步长的果蝇优化算法。该算法对果蝇得到的浓度值进行了修正,使味道浓度分布在整个正负寻优区间。在迭代时,充分利用果蝇群体已经进行的全局影响因素,对果蝇个体的搜寻距离进行适应性改变。为了验证该算法的效果,选用了几个常用的测试函数对该算法进行实验验证,结果表明,该算法不仅可以有效避免陷入局部最优,在寻优精度上也有一定提升。

关键词: 果蝇优化算法, 修正浓度, 适应步长, 局部最优, 寻优精度

Abstract: The basic fruit fly optimization algorithm can only be positive when solving the optimization problem, and it can not optimize the problem when the concentration is negative. In addition, the step size of the basic fruit fly optimization algorithm is random in iterative optimization, which makes the algorithm precocious and fall into the local optimal solution, and the accuracy of the algorithm is not high. Aiming at these problems of the basic fruit fly algorithm, a modified concentration and adaptive step algorithm for fruit fly is proposed. This algorithm modifies the concentration of fruit fly and makes the taste concentration distribution in the whole range of positive and negative optimization. In iteration, the search distance of fruit fly is changed adaptively by making full use of the global factors that have been carried out in fruit fly population. In order to verify the effectiveness of the algorithm, several commonly used test functions are selected to verify the algorithm. The results show that the algorithm can not only effectively avoid falling into local optimum, but also improve the precision of optimization.

Key words: fruit fly optimization algorithm, modified concentration, adaptive step, local optimum, precision of optimization