正在加载图片...
第10卷第3期 智能系统学报 Vol.10 No.3 2015年6月 CAAI Transactions on Intelligent Systems Jun.2015 D0:10.3969/j.issn.1673-4785.201404010 网络出版地址:http://www.cnki.net/kcms/detail/23.1538.tp.20150409.1657.001.html 一种改进的人工鱼群优化算法 吴昌友 (山东工商学院管理科学与工程学院,山东烟台264005) 摘要:对人工鱼群优化算法的觅食行为、群聚行为、追尾行为和公告板设置等基本原理进行分析,指出算法在 复杂优化问题上产生初始人工鱼群难和陷入局部最优解的原因,提出了改进人工鱼群优化算法,给出了初始 人工鱼群产生的方法,在人工鱼群优化算法的觅食行为、群聚行为、追尾行为中引人了自适应移动步长,同时 在算法中引入变异策略,避免算法陷入局部最优,提高全局寻优能力。最后通过对4个测试函数进行实验,对 于函数∫和f来说,虽然改进的人工鱼群算法和标准人工鱼群算法都达到了最优值,但是改进的人工鱼群 算法收敛的速度更快:函数来说,标准人工鱼群算法运行多次都陷入最优解,无法找到全局最优解。因此, 实验说明了改进算法的有效性与精确性。 关键词:人工鱼群优化算法:觅食:群聚;追尾:移动步长:变异策略 中图分类号:TP18文献标志码:A文章编号:1673-4785(2015)03-0465-05 中文引用格式:吴昌友.一种新的改进人工鱼群优化算法[J].智能系统学报,2015,10(3):465-469. 英文引用格式:WU Changyou.An improved artificial fish swarm optimization algorithm[J].CAAI Transactions on Intelligent Sys- tems,2015,10(3):465-469. An improved artificial fish swarm optimization algorithm WU Changyou School of Management Science and Engineering,Shandong Institute of Business And Technology,Yantai 264005,China) Abstract:In this paper,the basic principles of artificial fish's behaviors of prey,swarm,follow and bulletin board set were analyzed.Investigations were conducted to explore the reasons why it is difficult to produce the initial artifi- cial fish swarm,and why it always falls into local optional solution.The proposed solution improves the artificial fish algorithm with the method of the produce of initial artificial fish swarm,in the artificial fish's behaviors of prey, swarm and follow introduced the adaptive mobile step length with mutation strategy into the artificial fish at the same time,avoiding fish caught in local optima,improving the ability of global optimization.Finally,through the experi- ment of the 4 test functions concluded that as for the function of f,f and f,while the improved artificial fish swarm algorithm and artificial fish swarm algorithm have reached the optimal value,but the convergence of the im- proved artificial fish swarm algorithm is faster.As to the function of f,the standard artificial fish swarm algorithm run in to the optimal solution in several times'operation and the global optimal solution cannot be found.Therefore, the experiment shows the effectiveness and accuracy of the improved algorithm. Keywords:artificial fish swarm optimization algorithm;prey;swarm;follow;moving step length;mutation strategy 人工鱼群优化算法是李晓磊提出的一种智能优 化算法,该算法通过模拟鱼群的觅食、聚群、追尾等 收稿日期:2014-04-08.网络出版日期:2015-04-09. 行为,实现集群智能的一种优化方法,具有较强的鲁 基金项目:国家自然科学基金资助项目(71272122,71373148):山东 棒性和并行分布处理能力等优点[)。该算法与遗 省社科规划项目(13DGLJ05):山东能源经济协同创新中 心资助项目(2014sDXT005);山东省软科学项目 传算法和粒子群优化算法一样容易陷入局部最优解 (2014RKB01021). 和收敛速度慢,这些问题引起了广大学者们的注意, 通信作者:吴昌友.E-mail:wuchangyou_81@163.com.第 10 卷第 3 期 智 能 系 统 学 报 Vol.10 №.3 2015 年 6 月 CAAI Transactions on Intelligent Systems Jun. 2015 DOI:10.3969 / j.issn.1673⁃4785.201404010 网络出版地址:http: / / www.cnki.net / kcms/ detail / 23.1538.tp.20150409.1657.001.html 一种改进的人工鱼群优化算法 吴昌友 (山东工商学院 管理科学与工程学院,山东 烟台 264005) 摘 要:对人工鱼群优化算法的觅食行为、群聚行为、追尾行为和公告板设置等基本原理进行分析,指出算法在 复杂优化问题上产生初始人工鱼群难和陷入局部最优解的原因,提出了改进人工鱼群优化算法,给出了初始 人工鱼群产生的方法,在人工鱼群优化算法的觅食行为、群聚行为、追尾行为中引入了自适应移动步长,同时 在算法中引入变异策略,避免算法陷入局部最优,提高全局寻优能力。 最后通过对 4 个测试函数进行实验,对 于函数 f 1 、f 2和 f 4来说,虽然改进的人工鱼群算法和标准人工鱼群算法都达到了最优值,但是改进的人工鱼群 算法收敛的速度更快;函数 f 3来说,标准人工鱼群算法运行多次都陷入最优解,无法找到全局最优解。 因此, 实验说明了改进算法的有效性与精确性。 关键词:人工鱼群优化算法;觅食;群聚;追尾;移动步长;变异策略 中图分类号:TP18 文献标志码:A 文章编号:1673⁃4785(2015)03⁃0465⁃05 中文引用格式:吴昌友. 一种新的改进人工鱼群优化算法[J]. 智能系统学报, 2015, 10(3): 465⁃469. 英文引用格式:WU Changyou. An improved artificial fish swarm optimization algorithm[J]. CAAI Transactions on Intelligent Sys⁃ tems, 2015, 10(3): 465⁃469. An improved artificial fish swarm optimization algorithm WU Changyou (School of Management Science and Engineering, Shandong Institute of Business And Technology, Yantai 264005, China) Abstract:In this paper, the basic principles of artificial fish's behaviors of prey, swarm, follow and bulletin board set were analyzed. Investigations were conducted to explore the reasons why it is difficult to produce the initial artifi⁃ cial fish swarm, and why it always falls into local optional solution. The proposed solution improves the artificial fish algorithm with the method of the produce of initial artificial fish swarm, in the artificial fish's behaviors of prey, swarm and follow introduced the adaptive mobile step length with mutation strategy into the artificial fish at the same time, avoiding fish caught in local optima, improving the ability of global optimization. Finally, through the experi⁃ ment of the 4 test functions concluded that as for the function of f 1 , f 2 and f 4 , while the improved artificial fish swarm algorithm and artificial fish swarm algorithm have reached the optimal value, but the convergence of the im⁃ proved artificial fish swarm algorithm is faster. As to the function of f 3 , the standard artificial fish swarm algorithm run in to the optimal solution in several times' operation and the global optimal solution cannot be found. Therefore, the experiment shows the effectiveness and accuracy of the improved algorithm. Keywords:artificial fish swarm optimization algorithm; prey; swarm; follow; moving step length; mutation strategy 收稿日期:2014⁃04⁃08. 网络出版日期:2015⁃04⁃09. 基金项目:国家自然科学基金资助项目( 71272122,71373148);山东 省社科规划项目( 13DGLJ05);山东能源经济协同创新中 心 资 助 项 目 ( 2014SDXT005 ); 山 东 省 软 科 学 项 目 (2014RKB01021). 通信作者:吴昌友. E⁃mail: wuchangyou_81@ 163.com. 人工鱼群优化算法是李晓磊提出的一种智能优 化算法,该算法通过模拟鱼群的觅食、聚群、追尾等 行为,实现集群智能的一种优化方法,具有较强的鲁 棒性和并行分布处理能力等优点[1-3] 。 该算法与遗 传算法和粒子群优化算法一样容易陷入局部最优解 和收敛速度慢,这些问题引起了广大学者们的注意
向下翻页>>
©2008-现在 cucdc.com 高等教育资讯网 版权所有