【摘要】 精确描述高速铁路跨线列车晚点时长分布是准确计算跨线列车晚点引起的后效晚点的基础,也是精细化管理能力紧张的高速铁路的依据,具有重要的理论和实践意义。通过对高速铁路跨线列车晚点实绩数据的分析,运用超统计理论对列车晚点机理进行研究,构建跨线列车晚点时长q-指数分布模型。以2018年9月京沪高速铁路跨线列车晚点数据为样本,通过非线性回归拟合和Kolmogorov-Smirnov检验比较常见的正偏态厚尾分布模型及用于建模普速铁路列车晚点的指数分布模型对跨线列车晚点时长分布曲线的拟合效果。结果表明:q-指数分布模型相比于常见的正偏态厚尾分布模型,有更优的拟合效果,能更精确地描述高速铁路跨线列车晚点时长分布。更多还原

【Abstract】 Accurate description of the delay distribution of high-speed railway off-line trains, as the basis for precisely calculating the train daughter delay time caused by the off-line trains, as well as the basis for the delicacy management of high-speed railways with limited capacity, has important theoretical and practical significance. Based on the analysis of operational data of high-speed railway off-line trains, this paper studied the train delay mechanism by using the theory of the superstatistics, and established the q-exponential distribution model of off-line train delay. The nonlinear regression fitting and Kolmogorov-Smirnov test were used for the comparison of the fitness of curves between the common positive skewness and fat-tailed distribution models and the exponential distribution model used to model the normal-speed railway trains delay based on the delay data of Beijing-Shanghai high-speed railway off-line trains in September 2018. The results confirm that the q-exponential distribution model has better fitting effect and can describe the delay distribution of high-speed railway off-line trains more accurately than the common positive skewness and fat-tailed distribution model.更多还原