电子学报 ›› 2021, Vol. 49 ›› Issue (8): 1577-1585.DOI: 10.12263/DZXB.20200143

• 学术论文 • 上一篇    下一篇

基于信息熵的改进狮群算法及其在组合优化中的应用

李彦苍, 巩翔宇   

  1. 河北工程大学土木工程学院,河北 邯郸 056038
  • 收稿日期:2020-02-03 修回日期:2020-12-12 出版日期:2021-08-25
    • 作者简介:
    • 李彦苍 男,1974年生,河北饶阳人.博士、教授、博士生导师,主要从事计算智能理论及其工程应用方面的研究工作.E-mail:liyancang@hebeu.edu.cn
      巩翔宇 男,1990年生,内蒙古呼和浩特人.硕士研究生,主要从事计算智能理论及其工程应用方面的研究工作.E-mail:gongxiangyu321@126.com
    • 基金资助:
    • 河北省高等学校科学技术研究 (ZD2019114); 河北省自然科学基金 (E2020402079)

An Improved Lion Swarm Algorithm Based on Information Entropy and Its Application in Combinatorial Optimization

LI Yan-cang, GONG Xiang-yu   

  1. College of Civil Engineering, Hebei University of Engineering, Handan, Hebei 056038, China
  • Received:2020-02-03 Revised:2020-12-12 Online:2021-08-25 Published:2021-08-25
    • Supported by:
    • Science and Technology Research Project of Colleges and Universities in Hebei Province (ZD2019114); Natural Science Foundation of Heibei Province, China (E2020402079)

摘要:

狮群算法是一种具有较强寻优能力的群智能算法.为了克服基本狮群算法中因狮王替换的长周期性导致收敛速度较慢,幼狮选择策略较盲目导致的前期遍历性不足,幼狮步长扰动因子受解空间影响过大和算法后期局部收敛速度慢等缺陷;本文在原始狮群算法的基础上改良了狮王的替换策略和幼狮选择的概率,引入信息熵分别控制不同幼狮的步长,引入狮王稳定因子解决幼狮后期选择的盲目性,并适当调整狮群整体构成方式.由信息熵的值来度量狮群算法中幼狮选择的不确定性,通过设置不同的扰动因子达到控制算法中不同幼狮的移动范围,实现算法的自适应调节并增大算法的鲁棒性.仿真实验、桁架优化算例和TSP问题求解对比验证了改进算法的有效性.该研究为组合优化问题的求解提供了一种新的思路和方法.

关键词: 群智能, 狮群算法, 信息熵, 组合优化

Abstract:

Lion swarm algorithm is a kind of group intelligent algorithm with strong optimization ability. In order to overcome the slow convergence speed caused by the long periodicity of Lion King replacement in the basic lion group algorithm, the insufficient earlier ergodicity due to the blind selection strategy of young lion, and the slow local convergence speed in the later stage of the algorithm, the replacement strategy of Lion King and the selection probability of lion cubs were improved based on the original lion swarm algorithm. The information entropy was introduced to control the step length of different lion cubs, the Lion King Stabilizer factor was introduced to solve the blindness of lion cubs' later selection, and the overall composition of lion group was adjusted appropriately. The value of information entropy was used to measure the uncertainty of young lion selection in the lion group algorithm. Different disturbance factors were set to achieve the moving range of different young lions in the control algorithm, so as to realize the adaptive adjustment of the algorithm and increase the robustness of the algorithm. The effectiveness of the improved algorithm was verified by simulation, TSP and truss optimization. This study provides a new idea and method for solving structural optimization problems.

Key words: swarm intelligence, lion swarm optimization, information entropy, combinational optimization

中图分类号: