基于数据融合的城市轨道交通大客流预警研究

Early warning research of large passenger flow in urban rail transit based on data fusion

为了降低因节假日或突发时间导致的客流拥堵,促进客运部门快速应对客流波动,提出了一种基于数据融合的城市轨道交通大客流预警方法。通过对客流参数预测信息进行融合,获得符合实际客流瓶颈点的参数信息,以完成轨道交通大客流预警。提出的方法利用动态贝叶斯网络,实现了历史数据与运输服务信息的融合。以城市假期期间五条线路每日客流量为例,通过对比实验验证了提出方法的有效性与准确性。

In order to reduce the passenger flow congestion caused by holidays or emergencies and promote the passenger transport department to quickly respond to the passenger flow fluctuations,this paper proposes a large passenger flow early warning method for urban rail transit based on data fusion.Through the fusion of passenger flow parameter prediction information,the parameter information in line with the actual bottleneck of passenger flow is obtained to complete the early warning of large passenger flow of rail transit.The proposed method uses dynamic Bayesian network to realize the integration of historical data and transportation service information.Taking the daily passenger flow of five lines during the urban holiday as an example,the effectiveness and accuracy of the proposed method are verified through comparative experiments.