| 基于并行计算模型的并行微粒群算法的性能分析 |
| |
| 引用本文： | 王元元,曾建潮,谭瑛. 基于并行计算模型的并行微粒群算法的性能分析[J]. 微电子学与计算机, 2007, 24(9): 25-28 |
| |
| 作者姓名： | 王元元 曾建潮 谭瑛 |
| |
| 作者单位： | 太原科技大学,系统仿真与计算机应用研究所,山西,太原,030024 |
| |
| 基金项目： | 教育部科学技术研究项目 |
| |
| 摘 要： | 将微粒群算法与并行计算模型相结合,基于三种不同的并行计算模型(带中央控制器的并行计算模型、环形结构带缓存区的并行计算模型、BSP并行计算模型),设计出相应的并行微粒群算法,并对并行算法性能进行详细分析。大量实验表明:子种群之间的通讯周期是个重要的可变参数,当选取合适时,能提高解的质量以及算法的收敛性和最优性。
|
| 关 键 词： | 并行计算模型 并行算法 微粒群算法 |
| 文章编号： | 1000-7180（2007）09-0025-04 |
| 修稿时间： | 2006-06-02 |
Performance Analysis of the Parallel Particle Swarm Optimization Based on the Parallel Computation Models |
| |
| WANG Yuan-yuan,ZENG Jian-chao,TAN Ying. Performance Analysis of the Parallel Particle Swarm Optimization Based on the Parallel Computation Models[J]. Microelectronics & Computer, 2007, 24(9): 25-28 |
| |
| Authors: | WANG Yuan-yuan ZENG Jian-chao TAN Ying |
| |
| Abstract: | In the paper, the parallel particle swarm optimization (PSO) are designed based on three parallel computation models which include parallel computation model with central controller, ring-structure model with buffers, and BSP parallel computation model. The results of experiment show that the period of communication is key parameter. If an appropriate period of communication is chosen, the quality of the result and the performance of algorithm can be improved. |
| |
| Keywords: | parallel computation model parallel algorithm PSO |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
