| 基于GPU改进的并行人工蜂群算法 |
| |
| 引用本文： | 王智广,王文亮,张同举,鲁强,刘伟峰.基于GPU改进的并行人工蜂群算法[J].中南民族学院学报(自然科学版),2013(4):86-90,114. |
| |
| 作者姓名： | 王智广 王文亮 张同举 鲁强 刘伟峰 |
| |
| 作者单位： | [1]中国石油大学北京地球物理与信息工程学院,北京102249 [2]中国石油化工股份有限公司石油勘探开发研究院,北京100083 [3]中国石油化工集团公司多波地震技术重点实验室,北京100083 |
| |
| 基金项目： | 国家自然科学基金资助项目（60803159）;国家科技重大专项资助项目（2011zx05005-005-006） |
| |
| 摘 要： | 相对于先前的并行人工蜂群算法进行了一些改进，主要采用OpenCL本地内存、并行规约等技术，提出了一种基于图形处理器（GPU）改进的并行人工蜂群算法．该算法将采蜜蜂映射为OpenCL一个工作项，跟随蜂采用右邻域优先的局部选择机制．实验结果表明：文中提出的改进并行人工蜂群算法提高了算法的执行效率，收敛速度得到提升．
|
| 关 键 词： | 人工蜂群算法 并行处理 图形处理器 开放计算语言 |
An Improved Parallel Artificial Bee Colony Algorithm Based on GPU |
| |
| Wang Zhiguang,Wang Wenliang,Zhang TongjuI,Lu QiangI,Liu Weifeng.An Improved Parallel Artificial Bee Colony Algorithm Based on GPU[J].Journal of South-Central University for Nationalities(Natural Sciences),2013(4):86-90,114. |
| |
| Authors: | Wang Zhiguang Wang Wenliang Zhang TongjuI Lu QiangI Liu Weifeng |
| |
| Institution: | 2＇3 (1 College of Geophysics and Information Engineering, China University of Petroleum, Beijing 102249, China; 2 Sinopec Exploration and Production Research Institute, Beijing 100083, China; 3 Sinopec Key Laboratory of Multi Components Seismic Technology, Beijing 100083, China) |
| |
| Abstract: | In this paper, compared with previous parallel artificial bee colony algorithm, we have improved this algorithm in many aspects, such as OpenCL local memory and parallel reduction. An improved parallel artificial bee colony algorithm was proposed based on GPU. The process of the improved parallel artificial bee colony algorithm is that each employee bee mappings to OpenCL work item and onlookers use local selection of right neighborhood. The simulation results show that the improved parallel artificial bee colony algorithm has improved efficiency and the convergence rate is raised. |
| |
| Keywords: | artificial bee colony parallel process GPU OpenCL |
| 本文献已被 维普 等数据库收录！ |
|
