基于大数据的城市轨道交通出行站外OD位置点识别方法研究

Research on Identification Method of OD Points Outside Urban Rail Transit Stations Based on Big Data

城市轨道交通客流出行特征分析是制定线网规划方案和诊断轨道交通运营组织问题的重要依据,为解决传统城市轨道交通站外OD点识别算法参数设置较主观、识别精度较低、普适性与抗干扰性较弱等问题,研究以时空密度聚类算法为基础,融合遗传算法优化聚类算法参数,构建个体城市轨道交通出行站外OD位置点识别方法。识别过程中,根据志愿者信令数据、出行日志数据与GPS数据,结合遗传算法,标定时空密度聚类算法中聚类半径阈值EPS、聚类时间阈值?T等参数最优值;以此为基础,构建时空密度聚类算法,高效识别轨道出行站外OD。结果表明,通过比较志愿者实际出行日志、GPS等数据,志愿者出行站外OD位置点识别平均误差为633.75 m,算法精度可以满足实际需求。

The analysis of passenger flow travel characteristics in urban rail transit is an important basis for formulating line network planning schemes and diagnosing rail operation organization problems.Traditional algorithms for identifying OD points outside urban rail transit stations have subjective parameter settings,low identification accuracy,weak universality and anti-interference performance.In order to solve these problems,based on the spatio-temporal density clustering algorithm,this paper utilized a genetic algorithm to optimize parameters in the clustering algorithm and constructed a new method for identifying the OD point outside an individual station.Specifically,during the identification process,the paper employed a genetic algorithm and calibrated the optimal values of the clustering radius threshold(EPS),clustering time threshold(?T),and other parameters in the clustering algorithm according to the volunteer signaling data,travel log data,and GPS data.On this basis,the clustering algorithm was constructed to efficiently identify the OD point outside the station.Results show that through the comparison between the volunteers’actual travel logs and GPS data,the average error in identifying OD points outside the stations chosen by volunteers is 633.75 m,and the algorithm can meet the actual needs in terms of accuracy.