论文部分内容阅读
非复合分布模型可用于分析交通流量达1 800 vph的车辆时间间隔,但并不适用于更高交通流量的情况.为解决此类问题,提出了一些基于复合分布的模型.但这类模型的参数标定过程复杂,在一定程度上限制了其应用.针对流量介于1 900 vph到4 100 vph的车辆时间间隔,本文分别采用5种复合分布模型进行分析,即指数-极值分布(EEV)、对数正态-极值分布(LEV)、威布尔-极值分布(WEV)、威布尔-对数正态分布(WLN)和指数-对数正态分布(ELN).然后采用两种方法进行拟合优度检验——基于累计函数分布检验(CDF)和双样本(Cramer-von Mises)&K样本(Anderson-Darling)检验.结果表明,在分析车辆时间间隔方面,威布尔-极值分布(WEV)是最佳的复合分布模型,在Cramer-von Mises检验和K样本Anderson-Darling检验中均具有良好的一致性.
The non-compound distribution model can be used to analyze the vehicle time-lapse of traffic flow up to 1 800 vph, but it is not suitable for the case of higher traffic flow.In order to solve such problems, some models based on compound distribution are proposed The parameter calibration process is complicated and its application is restricted to a certain extent.For the vehicle time interval from 1 900 vph to 4 100 vph, five composite distribution models are used to analyze the exponential-extreme value distribution (EEV) , Lognormal-Extreme Value Distribution (LEV), Weibull-Extreme Value Distribution (WEV), Weibull-Lognormal Distribution (WLN) and Exponential-Lognormal Distribution (ELN) Method was used to test the goodness-of-fit test based on CDF and Anderson-Darling test.The results showed that in the analysis of vehicle time interval, Weibull- Distribution (WEV) is the best composite distribution model with good agreement between the Cramer-von Mises test and the K-sample Anderson-Darling test.