文献类型：期刊文献

英文题名：RNN-Based Subway Passenger Flow Rolling Prediction

作者：Sha, Shouwei[1];Li, Jing[1];Zhang, Ke[2,3];Yang, Zifan[2,3];Wei, Zijian[1];Li, Xueyan[4];Zhu, Xin[4]

第一作者：Sha, Shouwei

通讯作者：Wei, ZJ[1]

机构：[1]Beijing Jiaotong Univ, Dept Sch Econ & Management, Beijing 100044, Peoples R China;[2]Beijing Municipal Transportat Operat Coordinat Ct, Beijing 100073, Peoples R China;[3]Beijing Key Lab Integrated Traff Operat Monitorin, Beijing 100073, Peoples R China;[4]Beijing Union Univ, Sch Management, Beijing 100101, Peoples R China

第一机构：Beijing Jiaotong Univ, Dept Sch Econ & Management, Beijing 100044, Peoples R China

通讯机构：[1]corresponding author), Beijing Jiaotong Univ, Dept Sch Econ & Management, Beijing 100044, Peoples R China.

年份：2020

卷号：8

起止页码：15232-15240

外文期刊名：IEEE ACCESS

收录：;EI(收录号：20200908219878);Scopus(收录号：2-s2.0-85079783564);WOS:【SCI-EXPANDED(收录号:WOS:000524739700010)】；

基金：This work was supported in part by the National Nature Science Foundation of China-Young Scientists Fund under Grant 71103014, in part by the Beijing Municipal Philosophy Office under Grant 14JGC095, and in part by the Science and Technology Project of the Beijing Traffic Commission under Grant 201905-ZHJC2.

语种：英文

外文关键词：Data analysis; time series analysis; predictive models; neural networks; LSTM; GRU

摘要：The subway station passenger flow prediction model can forecast passenger volume in the future. This model helps to carry out safety warnings and evacuation of passenger flow in advance. Based on the data of the Shanghai traffic card, the passenger volume in all the time intervals is clustered into three different models for prediction. Taking the Nanjing East Road Station in Shanghai as an example, the time series of passenger volumes was combined with weather data to create several supervised sequences and was converted to supervised sequences according to different values of timestep. To accelerate convergence, two artificial features were added as input. The gated recurrent unit (GRU) network model achieves accurate rolling prediction from 15 minutes to 6 hours. Finally, comparing it with the long short-term memory (LSTM) network and the back-forward propagation network (BPN), it was confirmed that the GRU network with a timestep of 1.5 hours is the best model for the long-term (more than 3 hours) traffic flow rolling prediction, while GRU with a timestep of 45 minutes has the best result for short-term rolling prediction.