计算机工程与应用 ›› 2022, Vol. 58 ›› Issue (4): 91-99.DOI: 10.3778/j.issn.1002-8331.2101-0125

• 理论与研发 • 上一篇    下一篇

融合改进天牛须搜索的教与学优化算法

欧阳城添,周凯   

  1. 江西理工大学 信息工程学院,江西 赣州 341000
  • 出版日期:2022-02-15 发布日期:2022-02-15

Teaching-Learning Based Optimization Algorithm with Improved Beetle Antennae Search

OUYANG Chengtian, ZHOU Kai   

  1. School of Information Engineering, Jiangxi University of Science and Technology, Ganzhou, Jiangxi 341000, China
  • Online:2022-02-15 Published:2022-02-15

摘要: 针对教与学优化算法易早熟,解精度低,甚至收敛于局部最优的问题,提出一种新的融合改进天牛须搜索的教与学优化算法。该算法利用Tent映射反向学习策略初始化种群,提升初始解质量。在 “教”阶段,对教师个体执行天牛须搜索算法,增强教师教学水平,提高最优解的精确性。在“学”阶段,对学生个体进行混合变异,从而跳出局部最优,平衡算法的全局搜索与局部开发。通过benchmark测试函数和部分CEC2013函数在不同维度对算法进行仿真实验,并进行Wilcoxon秩和检验统计,证明了改进教与学优化算法的优越性。使用压力容器设计优化问题对算法进一步验证,结果表明改进后的算法在求解约束优化问题时也具有更好的寻优性能,不仅收敛速度快,精度也提高了9个数量级。

关键词: 教与学优化, Tent映射, 反向学习, 天牛须搜索, 混合变异

Abstract: For the problems such as premature, low precision and easy to converge to local best in teaching-learning based optimization algorithm, a new teaching-learning based optimization algorithm with improved beetle antennae search is proposed. Tent mapping and opposition-based learning strategy is used to initialize the population and enhance the quality of initial solution. In the “teaching” phase, teacher individual carries out beetle antennae search algorithm to enhance the teaching level and increase the accuracy of the optimal solution. In the “learning” phase, in order to jump out of the local optimum and balance global search with local development, the student population is hybrid mutated. The algorithms are simulated in different dimensions by benchmark test functions and some CEC2013 functions, and Wilcoxon rank sum test statistics are carried out, the results prove the superiority of improved teaching-learning based optimization algorithm. Further verification by pressure vessel design optimization problem shows that the improved algorithm also has better performance in solving constrained optimization problems, not only the convergence speed is fast, but also the accuracy is improved by 9 orders of magnitude.

Key words: teaching-learning based optimization, Tent mapping, opposition-based learning, beetle antennae search, hybrid?mutation