摘要
以合肥市典型道路为例,将大量行驶工况的实验数据划分为运动学片段,并选出11个特征参数进行研究。首先用主成分分析法对运动学片段进行降维处理;接着利用K均值聚类技术对其进行分类。分析结果验证了在城市道路行驶工况研究中应用主成分分析法和K均值聚类法的可行性和有效性;最后拟合出合肥市典型道路的代表性行驶工况。与实验数据的对比结果表明,拟合的行驶工况能综合反映合肥市的交通状况。
Taking the typical roads in Hefei City as example,massive test data of driving cycle are divided into a number of microtrips and 11 characteristic parameters are selected.On these bases a research on driving cycle is carried out.Firstly,dimension reduction treatment is performed on microtrips by principal component analysis,and they are then classified by using K means clustering technique.The results of analysis verify the feasibility and effectiveness of applying principal component analysis and K means clustering in studying driving cycle for city road.Finally a representative driving cycle for the typical roads in Hefei City is worked out and compared with test data.The results show that the fitted driving cycle can represent the overall traffic conditions in Hefei City.
出处
《汽车工程》
EI
CSCD
北大核心
2011年第3期256-261,共6页
Automotive Engineering
基金
国家自然科学基金项目(707710367
1071044)资助
关键词
行驶工况
运动学片段
主成分分析
聚类分析
driving cycle
microtrips
principal component analysis
cluster analysis
