基于数据拟合的城市轨道交通客流量时间序列模型优化分析

Optimization Analysis of Urban Rail Transit Passenger Flow Time Series Model Based on Data Fitting

为确保行车调度、预防处理轨道交通突发事故,通过曲线拟合充分发掘客流量时间序列趋势性,基于客流量时间分布数据,以整体拟合与自动分段拟合两种分析方法,针对B市多个地铁站客流量,优化分析时间序列模型。优化结果表明,相比整体拟合方法,自动分段拟合可较好去除客流量时间序列趋势,有效提高时间序列预测精确度;自动分段拟合不需要人工,可避免人为失误,既保障了客流量时间序列预测精确性,又实现了自动智能优化,在很大程度上为行车调度与应急安全管控奠定了技术基础。

In order to ensure traffic scheduling,prevent and deal with rail transit accidents,the trend of passenger flow time series is fully developed through curve fitting.Based on the passenger flow time distribution data,two analysis methods of overall fitting and automatic subsection fitting are used to optimize the analysis time series model for the passenger flow of multiple subway stations in B city.The optimization results show that,compared with the whole fitting method,the automatic subsection fitting can remove the trend of passenger flow time series better and improve the accuracy of time series prediction effectively;the automatic subsection fitting does not need manual work and can avoid human errors,which not only guarantees the accuracy of passenger flow time series prediction,but also realizes the automatic intelligent optimization,to a large extent,it is the traffic scheduling and emergency safety management Control laid the technical foundation.