基于直方图优选的车速分布模型研究

An Analysis of Curve Fitting Functions for vehick Speed Distribution Based on Histogram Improving Method

通过对国内外已有的研究分析,发现车速的分布规律变化不一,很难用单一的某一种分布模型进行拟合.因此,通过对呼和浩特市昭乌达路的速度数据采集,利用直方图优选的方法对速度分布直方图进行改进,并利用正态分布、威布尔分布、伽玛分布和Logistic分布模型对改进后的直方图进行了拟合分析.通过对拟合分析得出的相关系数的比较,给出了这四种分布在拟合速度分布时的适用范围.

According to the findings of domestic and foreign studies,it is found that the range of vehicle speed distribution is enormous,and it's very difficult to use a single model to describe the speed distribution.So,the related data was collected in the Zhaowuda Road of Hohhot by the project team,then a histogram improving method is proposed to deal with bar charts of speed distributing and four kinds of curve fitting functions(Normal,Weibull,Gamma,Logistic)are adopted to describe the distribution.After the comparison of the correlation coefficient,the applicability of each function is presented.