摘要
文章提取山西省太长高速公路视频数据,分析货车比例对高速公路交通流参数的影响,进而根据交通流参数分析货车比例对交通安全的影响。首先利用二项式统计分析模型分析货车比例对平均速度、速度标准差和速度变异系数的影响,然后采用负指数分布拟合车头时距,研究货车比例、车流量与平均车头时距的关系。最终以线性支持向量机验证变异系数分类状态,根据变异系数对交通安全的作用选取二元Logistic回归模型判断货车比例对交通安全的影响。结果表明:货车比例范围在0.15~0.5之间时,平均速度随着货车比例的增加而降低,速度标准差和变异系数随着货车比例的增加而增加,平均车头时距随流量的增加而减小。可知自由流条件下,速度变异系数越大交通流处于危险状态。最终建议货车比例为0.15~0.25之间,交通状态较为安全。
Based on the video data of the Taichang Highway in Shanxi Province,the influence of the proportion of the truck on the highway traffic flow parameters was analyzed,and then the influence of the truck proportion on the traffic safety was analyzed according to these traffic surrogate measures.Firstly,the influence of truck proportion on mean speed,standard deviation of speed and speed coefficient of variation were analyzed by binomial statistics,respectively.Secondly,the time headway fitted via negative exponential distribution and the relationship between the truck proportion,traffic volume and mean time headway were explored,respectively.Finally,the linear support vector machine(SVM)algorithm was used to verify the classification of the coefficient of variation and the binary logistic regression model was selected to determine the influence of the truck proportion on the traffic safety according to the relationship between the speed coefficient of variation and traffic safety.The experimental results show that when the truck proportion ranges from0.15 to 0.5,the mean speed decreases as the truck proportion increases;the standard deviation of speed and the speed coefficient of variation increase with the increase of the proportion of trucks;the mean time headway decreases as traffic volume increases.Under free flow conditions,the greater the coefficient of variation is,the more likely that traffic flow is in a dangerous state.The final recommended proportion of truck range is between 0.15 to 0.25,during which the traffic flow is in the relatively safe state.
作者
张诗雯
彭一川
陆键
陈瑶
张红军
ZHANG Shi-wen;PENG Yi-chuan;LU Jian;CHEN Yao;ZHANG Hong-jun(The Key Laboratory of Road and Traffic of Engineering of Ministry of Education,Tongji University,Shanghai 201804,China;School of Transportation Engineering,Tongji University,Shanghai 201804,China)
出处
《武汉理工大学学报》
北大核心
2017年第12期42-48,共7页
Journal of Wuhan University of Technology
基金
国家自然科学基金(71601143
71671127)
山西省交通厅科技项目(16002380370)
关键词
交通安全
货车比例
支持向量机分类
二元LOGISTIC模型
交通流参数
变异系数
traffic safety
truck proportion
support vector machine classification
binary logistic model
traffic flow parameter
coefficient of variation