基于CNN-Bi-LSTM网络的城轨短时客流预测

Short-term Passenger Flow Prediction for Urban Railway Based on CNN-Bi-LSTM Network

城市轨道交通短时客流预测的精确性对城轨站点在工作日早晚高峰的客流管控以及旅客出行方案决策等具有重要意义,因此提出基于深度学习技术构建卷积神经网络(CNN)结合双向长短时记忆神经网络(Bi-LSTM)的城轨短时客流预测模型,结合Bi-LSTM对时序数据特征双向提取能力的同时,利用CNN实现数据局部特征提取,增强模型的泛化能力。选取上海市地铁1号线徐家汇站的工作日客流数据为基础,对CNN层的层数和卷积核长度寻优,并分别对模型进行消融实验与对比实验。实验结果表明,CNN-Bi-LSTM的预测精度最优,RMSE与MAPE分别为16.699 9以及13.52%,证明了该模型在城轨短时客流预测的有效性。

The accuracy of short-term passenger flow prediction for urban rail transit is of great significance for the control of passenger flow at urban rail stations during the morning and evening peaks on weekdays,as well as for the decision-making of passenger travel programs.Therefore,a short-time passenger flow prediction model of urban rail transit using convolutional neural network(CNN)combined with bidirectional long-short-term memory neural network(Bi-LSTM) is proposed based on deep learning,which combines the bidirectional extraction capability of Bi-LSTM for time series data features,and at the same time,CNN is employed in realizing the local feature extraction of data,so as to enhance the generalization capability of the model.The weekday passenger flow data from Xujiahui Station of Shanghai Metro Line 1 is selected as the basis for optimization of the number of layers and convolution kernel length of the CNN layers,and the ablation and comparison experiments are conducted for the model,respectively.The experimental results show that the prediction accuracy of CNN-Bi-LSTM is optimal,and the RMSE and MAPE are 16.699 9 and 13.52%,respectively,which proves the effectiveness of the model in predicting the short-term passenger flow of urban railways.