主管：中国科学院
主办：中国优选法统筹法与经济数学研究会
　　　中国科学院科技战略咨询研究院
中国管理科学

中国管理科学 ›› 2011, Vol. 19 ›› Issue (2): 110-115.

• 论文 • 上一篇    下一篇

粒子群优化的模糊聚类方法在车辆行驶工况中的应用

石琴, 王楠楠, 仇多洋   

  1. 合肥工业大学交通运输工程学院, 安徽合肥 230009
  • 收稿日期:2010-08-31 修回日期:2010-12-26 出版日期:2011-04-30 发布日期:2011-04-30
  • 作者简介:石琴(1963- ),女(汉族),安徽蚌埠人,合肥工业大学交通运输工程学院教授,硕士研究生导师,博士,研究方向:交通环境与安全、车辆现代设计理论与方法。
  • 基金资助:

    国家自然科学基金资助项目(70771036);国家自然科学基金资助项目(71071044)

Application of Fuzzy Clustering Based on Particle Swarm Optimization in Vehicle Driving Cycle

SHI Qin, WANG Nan-nan, QIU Duo-yang   

  1. School of transportation Engineering, Hefei University of Technology, Hefei 230009, China
  • Received:2010-08-31 Revised:2010-12-26 Online:2011-04-30 Published:2011-04-30

PDF (PC)

1153

被引次数

22

摘要： 本文研究了粒子群优化的模糊聚类方法在车辆行驶工况中的应用。采用主成分分析方法将众多反映车辆行驶工况特征的运动学片段特征值进行压缩,用粒子群优化的模糊聚类方法对运动学片段的前三个主成分得分进行聚类,通过Matlab编程将上述理论用于合肥市典型道路行驶工况的构建和分析,按时间比例选取合适片段拟合代表性工况,并将代表性工况和采用K均值聚类法及模糊C均值聚类方法拟合的工况进行对比分析。研究结果表明,将粒子群优化的模糊聚类方法应用到工况的构建中可以有效地提高构建精度。

关键词: 粒子群, 模糊聚类, 主成分分析, K均值聚类, 行驶工况

Abstract: Fuzzy clustering based on particle swarm optimization method in vehicle driving cycle is tested in this paper.Principal component analysis is used to reduce the characteristics of the whole kinematic segments,which represents road running characteristics.The scores of the first three principal components of the kinematic segments are classified by using fuzzy clustering based on particle swarm optimization method.Programming with Matlab,the above theory is used to construct and analyze the typical roads in Hefei,and the representative driving cycle is obtained by selecting proper segments according to the ratio of time.The representative driving cycle and driving cycle obtained from k-means clustering and fuzzy cmeans clustering method are compared with the experimental data respectively.The research results shaws that fuzzy clustering based on particle swarm optimization method which is used to construct driving cycle can improve construction precision effectively.

Key words: particle swarm, fuzzy clustering, principal component analysis, k-means clustering, driving cycle

中图分类号: