计算机工程与应用 ›› 2018, Vol. 54 ›› Issue (1): 179-185.DOI: 10.3778/j.issn.1002-8331.1607-0192

• 模式识别与人工智能 • 上一篇    下一篇

反向自适应高斯变异的人工鱼群算法

姚凌波,戴月明,王  艳   

  1. 江南大学 物联网工程学院,江苏 无锡 214122
  • 出版日期:2018-01-01 发布日期:2018-01-15

Opposite adaptive and Gauss mutation artificial fish swarm algorithm

YAO Lingbo, DAI Yueming, WANG Yan   

  1. School of IoT Engineering, Jiangnan University, Wuxi, Jiangsu 214122, China
  • Online:2018-01-01 Published:2018-01-15

摘要: 针对人工鱼群算法存在易陷入局部最优、鲁棒性差以及寻优精度低的问题,提出了反向自适应高斯变异的人工鱼群算法。改进后的算法引入了反向解,根据反向解调整人工鱼的移动方向以及位置,从而提供更多的机会发掘潜在的较优空间,使人工鱼群快速跳出局部最优,从全局角度提升算法的搜索性能。同时提出了一种非线性自适应视野步长策略,更好地平衡了全局搜索与局部搜索之间的关系。为了增加鱼群的多样性,降低人工鱼陷入早熟的可能性,提出了一种最优解引导的高斯变异机制。仿真实验结果表明,该算法能有效地提高人工鱼群的寻优精度、寻优质量及鲁棒性,并且避免了人工鱼群过早收敛。

关键词: 人工鱼群算法, 自适应, 高斯变异, 反向解

Abstract: The Artificial Fish Swarm Algorithm(CAFSA) has some disadvantages such as falling into local optimum, poor robustness and low search accuracy. To solve these problems, this paper proposes an opposite adaptive and Gauss mutation artificial fish swarm algorithm. To provide more opportunities to explore potential better area, the algorithm applies opposite point to adjust direction and location of artificial fish. Thereby, the algorithm can jump out of local optimum fast and improve better global searching ability. In addition, this algorithm balances the global and local searching ability by using a non-linear function to adjust artificial fish’s visual and step. Otherwise, in order to solve early-maturing of artificial fish, using Gauss mutation mechanism based on optimal solution increases the diversity of every artificial fish. The simulation results show that improved artificial fish swarm algorithm has good searching quality, better accuracy and robustness. Meanwhile, the algorithm avoids early-maturing compared with other AFSAs.

Key words: Artificial Fish Swarm Algorithm(AFSA), adaptive, Gauss Mutation(GM), opposite point