| 摘要: |
| 针对基本人工兔优化(Artificial Rabbits Optimization,ARO)算法在解决复杂优化问题时存在收敛慢、精度不高和容易陷入局部最优等缺陷，本文提出一种改进的ARO算法(记为IARO算法)。IARO算法中的基于正弦函数的非线性递减能量因子能够帮助算法实现从探索阶段到开发阶段的良好过渡，从而提高算法的收敛速度和解的质量。此外，为了提高算法跳出局部最优的概率，IARO算法引入了一种动态透镜成像学习策略。为了证明IARO算法的优越性，首先选取了6个基准测试函数进行数值实验，然后用其求解2个工程设计优化问题和1个包括15个数据集的特征选择问题，并与灰狼优化(GWO)算法、鲸鱼优化算法(WOA)、正弦余弦算法(SCA)和基本ARO算法进行对比。结果表明，IARO算法有着比其他对比算法更优越的性能。 |
| 关键词: 人工兔优化算法 动态透镜成像学习策略 工程优化 特征选择 函数优化 |
| DOI：10.13656/j.cnki.gxkx.20230928.013 |
| 投稿时间：2023-02-10修订日期：2023-03-29 |
| 基金项目:国家自然科学基金项目(12361106)，贵州省教育厅创新群体项目(黔教合KY字[2021]015)和贵州省自然科学基金重点项目(黔科合基础-ZK[2023]重点003)资助。 |
|
| Dynamic Lens Imaging Learning Artificial Rabbits Optimization Algorithm and Its Applications |
|
WANG Wei1, LONG Wen2
|
| (1.School of Management Science and Engineering, Guizhou University of Finance and Economics, Guiyang, Guizhou, 550025, China;2.School of Mathematics and Statistics, Guizhou University of Finance and Economics, Guiyang, Guizhou, 550025, China) |
| Abstract: |
| Aiming at the defects of slow convergence,low precision and easy to fall into local optimum in solving complex optimization problems by Artificial Rabbit Optimization (ARO) algorithm,an improved ARO algorithm (IARO algorithm) is proposed.The nonlinear decreasing energy factor based on sine function in IARO algorithm can help the algorithm to achieve a good transition from the exploration stage to the development stage,so as to improve the convergence speed and solution quality of the algorithm.In order to enhance the probability of helping the algorithm jump out of local optimum,a dynamic lens imaging based learning strategy is introduced.In order to prove the superiority of IARO algorithm,six benchmark functions are selected for numerical experiments,and then it is used to solve two engineering design optimization problems and one feature selection problem including 15 data sets,and compared with Grey Wolf Optimizer (GWO) algorithm Whale Optimization Algorithm (WOA),Sine Cosine Algorithm (SCA) and basic ARO algorithm.The results show that the IARO algorithm has better performance than other comparison algorithms. |
| Key words: artificial rabbits optimization algorithm dynamic lens imaging based learning strategy engineering optimization feature selection function optimization |
