文献类型：会议论文

英文题名：GPU-based coevolutionary particle swarm optimization

作者：Zhao Liang[1];Zhu Yanxing[2];Zhang Jianyu[1];Ye Zhencheng[1]

机构：[1]East China Univ Sci & Technol, Minist Educ, Key Lab Adv Control & Optimizat Chem Proc, Shanghai 200237, Peoples R China;[2]Ethylene Plant Jilin Petrochem Co Ltd, PetroChina, Jilin 132021, Jilin, Peoples R China

会议论文集：36th Chinese Control Conference (CCC)

会议日期：JUL 26-28, 2017

会议地点：Dalian, PEOPLES R CHINA

语种：英文

外文关键词：GPU; CUDA; Particle swarm optimization; Coevolutionary particle swarm optimization

摘要：Coevolutionary particle swarm optimization (CPSO) algorithm has been investigated and applied in the real world widely. When tackling the large-scale and complex real time optimization problems, the running time of CPSO algorithm is a barrier. In this paper, Graphics Processing Unit (GPU) is introduced to provide speedup in order to meet the real time requirements. The CPSO algorithm has been implemented on GPU concurrently using the CUDA architecture. The performance and run time of CPU-based and GPU-based CPSO algorithms are compared in detail. The experiment result shows that the GPU-based parallel computation mode can shorten the run time of CPSO algorithm apparently.