计算机工程与应用 ›› 2021, Vol. 57 ›› Issue (2): 77-90.DOI: 10.3778/j.issn.1002-8331.2001-0290

• 理论与研发 • 上一篇    下一篇

高斯差分变异和对数惯性权重优化的鲸群算法

陈雷,尹钧圣   

  1. 1.天津商业大学 信息工程学院,天津 300134
    2.天津商业大学 理学院,天津 300134
  • 出版日期:2021-01-15 发布日期:2021-01-14

Whale Swarm Optimization Algorithm Based on Gaussian Difference Mutation and Logarithmic Inertia Weight

CHEN Lei, YIN Junsheng   

  1. 1.School of Information Engineering, Tianjin University of Commerce, Tianjin 300134, China
    2.School of Science, Tianjin University of Commerce, Tianjin 300134, China
  • Online:2021-01-15 Published:2021-01-14

摘要:

针对鲸群优化算法在处理高维问题时存在收敛速度慢、容易陷入局部最优和收敛精度低等问题,提出一种基于对数惯性权重和高斯差分变异的鲸群优化算法。通过高斯差分变异对鲸鱼位置更新方程进行变异,增加了种群多样性,提高了鲸群算法的全局搜索能力,防止早熟现象发生;将对数惯性权重引入搜寻猎物阶段,平衡全局搜索和局部开发能力,提高了算法寻优精度。通过测试函数优化实验对算法进行测试,实验结果表明,改进算法具有更高的寻优精度和更快的收敛速度。

关键词: 鲸群优化算法, 对数惯性权重, 高斯差分变异, 群智能优化算法, 收敛性能

Abstract:

Aiming at the problem that whale optimization algorithm has slow convergence speed, easy to fall into local optimum and low convergence precision when dealing with high-dimensional problems, a whale optimization algorithm based on logarithmic inertia weight and Gaussian difference mutation is proposed. Firstly, Gaussian difference mutation is used. The whale position update equation is mutated, which increases the population diversity and improves the global search ability of the whale algorithm to prevent premature phenomenon. Then, the logarithmic inertia weight is introduced into the prey hunting stage, which improves the algorithm’s optimization accuracy while balancing the global search and local development capabilities. Finally, the algorithm is tested from experiments. The experimental results show that the improved algorithm has higher optimization precision and faster convergence speed.

Key words: whale optimization algorithm, logarithmic inertia weight, Gaussian difference mutation, group intelligent optimization algorithm, convergence performance