计算机工程与应用 ›› 2022, Vol. 58 ›› Issue (10): 68-75.DOI: 10.3778/j.issn.1002-8331.2106-0063

• 理论与研发 • 上一篇    下一篇

融合振荡禁忌搜索的自适应均衡优化算法

刘成汉,何庆   

  1. 1.贵州大学 大数据与信息工程学院,贵阳 550025
    2.贵州大学 贵州省公共大数据重点实验室,贵阳 550025
  • 出版日期:2022-05-15 发布日期:2022-05-15

Adaptive Equilibrium Optimizer Algorithm Combining Oscillating Tabu Search

LIU Chenghan, HE Qing   

  1. 1.College of Big Data & Information Engineering, Guizhou University, Guiyang 550025, China
    2.Guizhou Big Data Academy, Guizhou University, Guiyang 550025, China
  • Online:2022-05-15 Published:2022-05-15

摘要: 为了改善均衡优化(equilibrium optimizer,EO)算法寻优过程中存在的收敛速度慢、易受局部极小值影响的问题,提出一种融合振荡禁忌搜索的自适应均衡优化算法CfOEO。针对EO算法初始化随机性过高导致的收敛速度慢的问题,引入精英反向学习初始化种群,增加算法搜索能力;通过自适应调整收敛因子来平衡算法的局部和全局搜索能力;在禁忌搜索策略中引入振荡算子,提高算法跳出局部极小值的能力。仿真实验采用10个基准测试函数和部分CEC2014测试函数以及基准测试函数的Wilcoxon秩和检测,对CfOEO算法进行寻优性能测试,测试结果验证了CfOEO算法的鲁棒性。

关键词: 均衡优化算法, 精英反向学习, 振荡算子, 禁忌搜索, 自适应收敛因子

Abstract: In order to solve the problems of slow convergence speed and easy to be affected by local minimum in the optimization process of equilibrium optimizer(EO) algorithm, an adaptive equilibrium optimization algorithm(CfOEO) combining oscillating tabu search is proposed. Aiming at the problem of slow convergence speed caused by high initial randomness of EO algorithm, elite reverse learning is introduced to initialize the population to increase the searching ability of the algorithm. The local and global search ability of the algorithm is balanced by adjusting the convergence factor adaptively. Finally, an oscillation operator is introduced into the tabu search strategy to improve the ability of the algorithm to leap out of the local minimum. In the simulation experiment, 10 benchmark test functions, some CEC2014 test functions and the Wilcoxon rank sum test of the standard test functions are used to test the optimization performance of the CfOEO algorithm. The test results verify the robustness of the CfOEO algorithm.

Key words: equilibrium optimizer algorithm, elite reverse learning, oscillation operator, tabu search, adaptive convergence factor