The invention provides an urban rail transit station short-time passenger flow prediction model and prediction method, and the method comprises the steps: combining the operation data of stations, constructing passenger flow data, deeply mining the spatial features of the stations, and constructing a multi-relation network; in order to realize mining of spatial features among nodes, R-GCN is introduced into a spatial convolution layer by using a spatial convolution module to realize multi-graph fusion; in order to realize mining of time sequence characteristics, a time convolution module is utilized, and a TCN network is used for realizing one-dimensional time sequence convolution, so that short-time passenger flow prediction is carried out. In the application of short-time passenger flow prediction of the urban rail transit station, the method has higher accuracy and adaptability, can adapt to short-time prediction requirements of each station under various road networks, has more significant guiding significance for actual station work, and is beneficial to improving the operation management level of the station.
本发明提供一种城市轨道交通车站短时客流预测模型以及预测方法,结合车站的运营数据,构建客流数据,并深度挖掘车站间空间特征,构建多关系网络;为了实现节点间空间特征的挖掘,利用空间卷积模块,在空间卷积层引入R‑GCN实现多图融合;为了实现时序特征的挖掘,利用时间卷积模块,使用TCN网络实现一维时序卷积,以此进行短时客流预测。本发明在城市轨道交通车站短时客流预测问题的应用中,具有更强的准确性与适应性,能适应各种路网下各车站的短时预测需求,对实际车站工作的指导意义更加显著,有利于提高车站的运营管理水平。
Urban rail transit station short-time passenger flow prediction model and prediction method
城市轨道交通车站短时客流预测模型以及预测方法
2024-07-05
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
Integrated Prediction Model for Urban Rail Transit Station Feeder Passenger Flow
DOAJ | 2024
|Urban rail transit passenger flow prediction method under emergency
European Patent Office | 2023
|Learning Spatial-Temporal Dynamics for Short-Term Passenger Flow Prediction in Urban Rail Transit
Transportation Research Record | 2023
|