《计算机应用研究》|Application Research of Computers

混合策略改进正弦余弦算法

Mixed strategy to improve sine cosine algorithm

免费全文下载 (已被下载 次)  
获取PDF全文
作者 林杰,何庆
机构 贵州大学 a.大数据与信息工程学院;b.贵州省公共大数据重点实验室,贵阳 550025
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2020)12-019-3612-06
DOI 10.19734/j.issn.1001-3695.2019.09.0542
摘要 为提高正弦余弦算法在求解函数优化问题的性能,提出混合策略改进正弦余弦算法。首先,利用正切函数作为调节因子代替原本线性变化的参数,用于平衡算法的全局探索和局部开发;然后,引入权重系数,用于控制位置更新处个体上一代位置的影响力,有效提高算法开发能力和寻优速度;最后,构建逐维交叉学习策略,克服最优解无更新的缺点,对最优个体进行扰动更新,跳出局部最优,避免早熟收敛。在不同维数的八个基准函数上进行仿真实验。实验表明,该算法相对于其他群智能优化算法具有更高的寻优精度和收敛速度,相比于最新的正弦余弦改进算法,也表现出更好的收敛性能和稳定性。
关键词 正弦余弦算法; 调节因子; 权重系数; 逐维交叉学习
基金项目 贵州省科技计划项目重大专项项目(黔科合重大专项字[2018]3002,黔科合重大专项字[2016]3022)
贵州省公共大数据重点实验室开放课题(2017BDKFJJ004)
贵州省教育厅青年科技人才成长项目(黔科合KY字[2016]124)
贵州大学培育项目(黔科合平台人才[2017]5788)
本文URL http://www.arocmag.com/article/01-2020-12-019.html
英文标题 Mixed strategy to improve sine cosine algorithm
作者英文名 Lin Jie, He Qing
机构英文名 a.College of Big Data & Information Engineering,b.Guizhou Provincial Key Laboratory of Public Big Data,Guizhou University,Guiyang 550025,China
英文摘要 In order to improve the performance of the sine cosine algorithm in solving the function optimization problem, this paper proposed an improved sine cosine algorithm for mixed strategy(MI-SCA). Firstly, in order to balance the global exploration and local exploitation of the algorithm, it used the tangent function as the adjustment factor instead of the original linear va-riation parameter. Then, it introduced the weight coefficient to control the influence of the previous generation position of the location update, and improved the algorithm exploitation and optimization speed effectively. Finally, it constructed a cross-dimensional cross-learning strategy to overcome the shortcomings of the optimal solution without updating, perturb the optimal individual, avoid falling into the local optimum, and avoid premature convergence. This paper carried simulation experiments out on eight benchmark functions with different dimensions. The experimental results show that the proposed algorithm has higher optimization precision and convergence speed than the comparison algorithm, and shows better convergence performance and stability.
英文关键词 sine cosine algorithm(SCA); adjustment factor; weight coefficient; cross-dimensional cross learning curve function
参考文献 查看稿件参考文献
 
收稿日期 2019/9/23
修回日期 2019/11/5
页码 3612-3617
中图分类号 TP301
文献标志码 A