Jing ZUOThis email address is being protected from spambots. You need JavaScript enabled to view it., Zhao YU, Ming HE, and Guoyan LIU
School of Automation and Electrical Engineering, Lanzhou Jiaotong University, Lanzhou Gansu 730070, China
Received: November 8, 2023 Accepted: March 21, 2024 Publication Date: April 16, 2024
Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.
The imbalance of temporal and spatial distribution of passenger flow under network conditions occurs from time to time, which brings great challenges to the operation of urban rail transit system, and the effective prediction of short-time passenger flow status of trains is of great significance to optimize the transportation strategy of line network and meet the real-time passenger flow demand. To this end, a short-time passenger flow prediction model for urban rail trains based on GCN-AMBiLSTM model is proposed. Firstly, a rail traffic timing map is established to capture the global spatial correlation features of different levels of neighbouring stations using graph convolutional neural network; secondly, an attention mechanism is introduced into the bidirectional long and short-term memory neural network, and the AMBiLSTM module is constructed to extract and assign the importance of train passenger flow timing features from the forward and backward dimensions. Finally, based on the feature fusion network, the temporal and spatial features are integrated and the prediction results are output. Based on the Chengdu Metro, the model is tested, and the results show that compared with several baseline models, the model in this paper achieves the optimal values of root mean square error, average absolute error and average absolute percentage error, and the predicted values have a high degree of fit with the actual passenger flow values, and the prediction efficiency can fully satisfy the timeliness requirements of the field, which has a good application prospect.
[1] D. D. Chuwang, W. Chen, and M. Zhong, (2023) “Short-term urban rail transit passenger flow forecasting based on fusion model methods using univariate time series" Applied Soft Computing 147: DOI: 10.1016/j.asoc.2023.110740.
[2] Y. Xin, Q. Xue, M. Ding, J. Wu, and Z. Gao, (2020) “Short-term prediction of passenger volume for urban rail systems: A deep learning approach based on smartcard data" International Journal of Production Economics: DOI: 10.1016/j.ijpe.2020.107920.
[3] Z. Zhu, Y. Zhang, S. Qiu, Y. Zhao, J. Ma, and Z. He, (2023) “Ridership Prediction of Urban Rail Transit Stations Based on AFC and POI Data" Journal of Transportation Engineering, Part A: Systems 149: DOI: 10.1061/JTEPBS.TEENG-7808.
[4] J. Sun, J. Yao, M. Wang, D. Zhang, M. Sabah, and C. Alessandro, (2021) “Subway passenger flow analysis and management optimization model based on AFC data" Journal of Intelligent & Fuzzy Systems 41: 4773–4783. DOI: 10.3233/JIFS-189963.
[5] Q. Yu, N. He, W. Ye, and L. Pang, (2021) “Research on the frequency prediction of barrier-free resources in airport terminal based on ARIMA model" IEEE: DOI: 10.1109/ICPECA51329.2021.9362603.
[6] Z. Yang, P. Jiao, X. Yun, and W. Hong, (2021) “alman Filtering Short-Term Traffic Flow Prediction Model Based on Phase Space Reconstruction" Journal of Beijing University of Civil Engineeringand Architecture 37: 43–50. DOI: 10.19740/j.2096-9872.2021.04.06.
[7] Y. Wang, J. Ma, and J. Zhang, (2019) “Metro Passenger Flow Forecast with a Novel Markov-Grey Model" Periodica Polytechnica Transportation Engineering 48(1): 70–75. DOI: 10.3311/PPtr.11131.
[8] Z. Shi, N. Zhang, S. P. M, and J. Zhang, (2020) “Shortterm metro passenger flow forecasting using ensemblechaos support vector regression" Transportmetrica A: Transport Science 16: 194–212. DOI: 10.1080/23249935.2019.1692956.
[9] M. Tang, Z. Li, and G. Tian, (2019) “A Data-DrivenBased Wavelet Support Vector Approach for Passenger Flow Forecasting of the Metropolitan Hub." IEEE Access 7: 7176–7183. DOI: 10.1109/ACCESS.2019.2890819.
[10] W. Xing, J. Wang, K. Zhou, H. Li, Y. Li, and Z. Yang, (2023) “A hierarchical methodology for vessel traffic flow prediction using Bayesian tensor decomposition and similarity grouping" Ocean Engineering 286: DOI: 10.1016/J.OCEANENG.2023.115687.
[11] L. Li, L. Qin, X. Qu, J. Zhang, Y. Wang, and B. Ran, (2019) “Day-ahead traffic flow forecasting based on a deep belief network optimized by the multi-objective particle swarm algorithm" Knowledge-Based Systems 172: 1– 14. DOI: 10.1016/j.knosys.2019.01.015.
[12] J. Ma, X. Zeng, X. Xue, and R. Deng, (2022) “Metro Emergency Passenger Flow Prediction on Transfer Learning and LSTM Model" Applied Sciences 12: 1644–1644. DOI: 10.3390/APP12031644.
[13] P. Ren, Z. Zuo, and H. Chen, (2022) “Short-term Passenger Flow Prediction of Urban Rail Transit Based on SSA-LSTM" Journal of Wuhan University of Technology 44: 44–52.
[14] Z. Huang, Q. Li, F. Li, and J. Xia, (2019) “A Novel BusDispatching Model Based on Passenger Flow and Arrival Time Prediction" IEEE Access 7: 106453–106465. DOI: 10.1109/access.2019.2932801.
[15] J. Wang, R. Wang, and X. Zeng, (2022) “Short-term passenger flow forecasting using CEEMDAN meshed CNNLSTM-attention model under wireless sensor network" IET Communications 16: 1253–1263. DOI: 10.1049/CMU2.12350.
[16] J. Zhao, J. Shi, c. Sun, L. Ren, and C. Liu, (2020) “Shorttime inflow and outflow prediction of metro stations based on hybrid deep learning" Journal of Transportation Systems Engineering and Information Technology 20: 128–134. DOI: 10.16097/j.cnki.1009-6744.2020.05.019.
[17] J. Bruna, W. Zaremba, A. Szlam, and Y. LeCun, (2013) “Spectral Networks and Locally Connected Networks on Graphs." CoRR abs/1312.6203: DOI: arXiv.1312.6203.
[18] Z. Tao, J. Tang, and K. Hou, (2021) “Online estimation model for passenger flow state in urban rail transit using multi-source data" Computer-Aided Civil and Infrastructure Engineering 36: 762–780. DOI: 10.1111/MICE.12671.
[19] J. Zuo and Z. Yu, (2023) “Real-time Train Passenger Flow Detection Algorithm Based on Convolutional Neural Network" Journal of Railway Science and Engineering 20: 836–845. DOI: 10.19713/j.cnki.43-1423/u.t20220662.
[20] J. Zhang, J. Liu, and Z. Wang, (2021) “Convolutional Neural Network for Crowd Counting on Metro Platforms" Symmetry 13: 703–703. DOI: 10.3390/SYM13040703.
[21] Q. Xiao, Y. Xiao, and F. Chen. “The Passenger Flow Counting Research of Subway Video based on Image Processing”. In: Proceedings of the 29th China Control and Decision-Making Conference (4). Institute of Information Engineering, Shenyang University;] 2017, 755–758. DOI: 10.1109/CCDC.2017.7979418.
[22] X. Dawen, Y. Nan, J. Shunying, H. Yang, and L. Huaqing, (2022) “SW-BiLSTM: a Spark-based weighted BiLSTM model for traffic flow forecasting" Multimedia Tools and Applications 81: 23589–23614. DOI: 10.1007/s11042-022-12039-3.
[23] P. Chen, X. Fu, and X. Wang, (2021) “A Graph Convolutional Stacked Bidirectional Unidirectional-LSTM Neural Network for Metro Ridership Prediction" IEEE Transactions on Intelligent Transportation Systems 23(7): 6950–6972. DOI: 10.1109/TITS.2021.3065404
We use cookies on this website to personalize content to improve your user experience and analyze our traffic. By using this site you agree to its use of cookies.
