建成环境影响下的城市轨道交通客流多步短时预测

doi:10.16097/j.cnki.1009-6744.2025.01.016

交通运输系统工程与信息 ›› 2025, Vol. 25 ›› Issue (1): 160-172.DOI: 10.16097/j.cnki.1009-6744.2025.01.016

建成环境影响下的城市轨道交通客流多步短时预测

李之红^*1a，郄堃^1a，王健宇^1a，许晗²，陈金政^1b

1. 北京建筑大学，a.通用航空技术北京实验室，b.建筑与城市规划学院，北京100044；2.北京交通发展研究院，北京101160

收稿日期:2024-06-13 修回日期:2024-12-04 接受日期:2024-12-06 出版日期:2025-02-25 发布日期:2025-02-24
作者简介:李之红(1981—)，男，河北秦皇岛人，教授，博士。
基金资助:
国家自然科学基金(52402377)；北京市自然科学基金(9234025)；北京市属高等学校高水平科研创新团队建设支持计划项目(BPHR20220109)。

Multistep Short-term Prediction of Urban Rail Transit Passenger Flow Under Influence of Built Environment

LI Zhihong^*1a, QIE Kun^1a, WANG Jianyu^1a, XU Han², CHEN Jinzheng^1b

1a. Beijing Laboratory for General Aviation Technology, 1b. School of Architecture and Urban Planning, Beijing University of Civil Engineering and Architecture, Beijing 100044, China; 2. Beijing Transport Institute, Beijing 101160, China

Received:2024-06-13 Revised:2024-12-04 Accepted:2024-12-06 Online:2025-02-25 Published:2025-02-24
Supported by:
National Nature Science Foundation of China(52402377)；Natural Science Foundation of Beijing, China (9234025)；Project of Construction and Support for High-level Innovative Teams of Beijing Municipal Institutions (BPHR20220109)。

摘要/Abstract

摘要： 为挖掘客流的复杂时空耦合关系，解析建成环境影响下的轨道交通客流出行规律，本文提出一种考虑城市建成环境的时空双层超图神经网络模型(SpatialTemporal-Double Hypergraph Neural Network, ST-DHGNN)。模型分为双层超图神经网络和时间序列模块，双层超图神经网络模块用于挖掘轨道交通线路站点间的高阶连通关系和相邻同类建成区域站点的集群关系，时间序列模块用于表征历史客流数据的时间依赖关系。同时，以建成环境和线路作为变量构造新的损失函数，旨在剖析建成环境的影响，提高模型的预测性能。最后，以武汉轨道交通数据为例开展实证研究。研究结果显示：考虑建成环境和轨道站点高阶连通关系对客流预测精度的提升效果显著，本模型均方根误差(RMSE)和平均绝对误差(MAE)值分别为52.04和29.32，比基线模型降低了22%以上，性能显著优于基线模型；通过消融实验验证了融合轨道高阶联通关系和建成环境对模型性能的贡献，其中，单步预测任务中，考虑这两种因素使模型性能分别提升了6%和9%，多步预测任务中，分别提升了4%和12%；构造的融合建成环境因素的可解释损失函数，提高了模型的预测性能，同时，使模型具备更好的科学性和可解释性。研究成果为城市轨道交通的客流管理和列车调度提供了技术支持。

关键词: 智能交通, 客流多步预测, 超图时空网络, 城市轨道交通, 建成环境影响, 可解释损失函数

Abstract: To analyze the spatiotemporal coupling relationships of passenger flow and the passengers travel patterns in urban rail transit, this paper proposes a spatial-temporal double-hypergraph neural network model considering the urban built environment, which is called ST-DHGNN (Spatial Temporal-Double Hypergraph Neural Network). The model is consisted of a double hypergraph neural network module and a time series module. The double-hypergraph neural network module is designed to uncover high-order connectivity among rail transit line stations and clustering relationships among neighboring stations within similar built-up areas. The time series module is utilized to represent the temporal dependencies in historical passenger flow data. Additionally, a new loss function is developed with the built environment and lines as variables, aiming to dissect the impact of the built environment and enhance the model's predictive performance. An empirical study is conducted using Wuhan Metro data as an example. The results indicate that: (1) Considering the built environment and high-order connectivity among rail stations significantly improves passenger flow prediction accuracy. The proposed model achieves Root Mean Square Error (RMSE) of 52.04 and Mean Absolute Error (MAE) values of 29.32. The performance was improved by over 22% compared to baseline models. (2) The ablation experiments have verified the contribution of integrating high-order connectivity relationships of movement trajectories and the built environment to the model's performance. Specifically, in the single-step prediction task,considering these two factors improves the model's performance by 6% and 9%, respectively. In the multi-step prediction task, the improvements are 4% and 12%, respectively. (3) The constructed interpretable loss function incorporating built environment factors enhances the model's predictive capability while imparting better scientific rigor and interpretability. The study can provide technical support for passenger flow management and train scheduling in urban rail transit.

Key words: intelligent transportation, multi-step passenger flow forecasting, hypergraph spatiotemporal networks, urban rail transit, built environment impact, interpretable loss function

中图分类号:

U495

李之红, 郄堃, 王健宇, 许晗, 陈金政. 建成环境影响下的城市轨道交通客流多步短时预测[J]. 交通运输系统工程与信息, 2025, 25(1): 160-172.

LI Zhihong, QIE Kun, WANG Jianyu, XU Han, CHEN Jinzheng. Multistep Short-term Prediction of Urban Rail Transit Passenger Flow Under Influence of Built Environment[J]. Journal of Transportation Systems Engineering and Information Technology, 2025, 25(1): 160-172.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/10.16097/j.cnki.1009-6744.2025.01.016

http://www.tseit.org.cn/CN/Y2025/V25/I1/160

参考文献

[1]CHUWANG D D, CHEN W, ZHONG M. Short-term urban rail transit passenger flow forecasting based on fusion model methods using univariate time series[J]. Applied Soft Computing, 2023, 147: 110740.

[2]GAO C, LIU H, HUANG J, et al. Regularized spatial temporal graph convolutional networks for metro passenger flow prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2024, 25(9): 11241 11255.

[3]LI Z, WANG X, CAI H, et al. Novel hybrid spatiotemporal convolution neural network model for short-term passenger flow prediction in a large-scale metro system[J]. Journal of Transportation Engineering, Part A: Systems, 2024, 150(5): 04024016.

[4]WANG J, ZHANG Y, WEI Y, et al. Metro passenger flow prediction via dynamic hypergraph convolution networks[J]. IEEE Transactions on Intelligent Transportation Systems, 2021, 22(12): 7891-7903.

[5]WANG M, ZHANG Y, ZHAO X, et al. Traffic origin destination demand prediction via multi-channel hypergraph convolutional networks[J]. IEEE Transactions on Computational Society Systems, 2024, 11 (4): 5496-5509.

[6]HU L, SUN T, WANG L. Evolving urban spatial structure and commuting patterns: A case study of Beijing, China [J]. Transportation Research Part D: Transport and Environment, 2018, 59: 11-22.

[7]许心越,孔庆雪,李建民,等.建成环境对轨道交通客流的时空异质性影响分析[J].交通运输系统工程与信息, 2023, 23(4): 194-202, 281. [XU X Y, KONG Q X, LI J M, et al. Analysis of spatio-temporal heterogeneity impact of built environment on rail transit passenger flow [J]. Journal of Transportation Systems Engineering and Information Technology, 2023, 23(4): 194-202, 281.]

[8]王菁,万峰,董春娇,等.城市轨道交通站点吸引范围及强度建模[J]. 吉林大学学报(工学版), 2023,53(2): 439-447. [WANG J, WAN F, DONG C J, et al. Modelling on catchment area and attraction intensity of urban rail transit stations[J]. Journal of Jilin University (Engineering and Technology Edition), 2023, 53(2): 439 447.]

[9]SHEN P, OUYANG L, WANG C, et al. Cluster and characteristic analysis of Shanghai metro stations based on metro card and land-use data[J]. Geo-spatial Information Science, 2020, 23(4): 352-361.

[10]ZHAO L, SONG Y, ZHANG C, et al. T-GCN: A temporal graph convolutional network for traffic prediction[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 21(9): 3848-3858.

[11]HUA Y, ZHAO Z, LI R, et al. Deep learning with long short-term memory for time series prediction[J]. IEEE Communications Magazine, 2019, 57(6): 114-119.

[12]张金雷, 陈奕洁,PANCHAMYK, 等.基于注意力机制的城市轨道交通网络级多步短时客流时空综合预测模型[J]. 地球信息科学学报, 2023, 25(4): 698-713. [ZHANG J L, CHEN Y J, PANCHAMY K, et al. Attention-based multi-step short-term passenger flow spatial-temporal integrated prediction model in URT systems[J]. Journal of Geo-information Science, 2023, 25 (4): 698-713.]

[13]JIN M, GUO J, XIE S, et al. Understanding the effect of built environment on travel behavior at multiple geographic scales[J]. Transportation Research Record, 2023, 2677(5): 1062-1076.

[14]YANG H, LU Y, WANG J, et al. Understanding post pandemic metro commuting ridership by considering the built environment: A quasi-natural experiment in Wuhan, China[J]. Sustainable Cities and Society, 2023, 96: 104626. [15]BUI K H N, CHO J, YI H. Spatial-temporal graph neural network for traffic forecasting: An overview and open research issues[J]. Applied Intelligence, 2022, 52(3): 2763-2774.

建成环境影响下的城市轨道交通客流多步短时预测

Multistep Short-term Prediction of Urban Rail Transit Passenger Flow Under Influence of Built Environment

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	孙煦, 马天行, 王天实, 王健宇, 陆化普. 考虑提升队列规模的智能网联车辆编队策略[J]. 交通运输系统工程与信息, 2025, 25(2): 48-57.
[2]	张建旭, 吴春箱. 自动驾驶车辆渗透环境下多车道瓶颈路段行驶策略仿真[J]. 交通运输系统工程与信息, 2025, 25(2): 69-81.
[3]	南斯睿, 于谦, 李铁柱, 尚赞娣, 陈海波. 考虑进站策略的网联电动公交车节能驾驶优化研究[J]. 交通运输系统工程与信息, 2025, 25(2): 82-94.
[4]	魏丽英, 冯梅, 吴润泽. 智能网联公交信号优先的博弈机理和引导策略[J]. 交通运输系统工程与信息, 2025, 25(2): 95-107.
[5]	代亮, 黄自彬, 张中昊, 李臣富. 考虑车道剩余容量的区域交通信号控制方法[J]. 交通运输系统工程与信息, 2025, 25(2): 108-118.
[6]	袁振洲, 周博涵, 陈沫, 杨洋. 期望状态约束下多车协同编队与分布式优化方法[J]. 交通运输系统工程与信息, 2025, 25(2): 119-127.
[7]	辛嵩, 刘晗, 王可, 曲奕润, 宋新宇. 基于概率融合的车辆意图识别与轨迹预测方法[J]. 交通运输系统工程与信息, 2025, 25(2): 128-137.
[8]	林培群, 陈泽沐, 周楚昊. 基于扩散模型的节假日高速公路交通流预测方法[J]. 交通运输系统工程与信息, 2025, 25(2): 169-179.
[9]	姜晓红, 李家伟, 华晶雯, 仲韫豪, 邢吉平. 考虑乘客自主选择的区段交替响应式城乡公交调度[J]. 交通运输系统工程与信息, 2025, 25(2): 201-213.
[10]	蒋沛, 马新露, 李一博, 陈坚. 网联车辆复用公交专用道建模与仿真研究[J]. 交通运输系统工程与信息, 2025, 25(1): 67-75.
[11]	韩烽凡, 刘爽, 朱亚迪, 杨烨, 王曦. 数据驱动的城轨车站行人仿真方法研究[J]. 交通运输系统工程与信息, 2025, 25(1): 86-91.
[12]	许得杰, 钟苗苗, 巩亮, 惠昌武, 曾俊伟. 基于动态安全间隔的城轨大小交路列车运行仿真研究[J]. 交通运输系统工程与信息, 2025, 25(1): 146-159.
[13]	赖树坤, 许宏科, 林杉, 罗永煜, 邹复民, 廖律超. 基于车辆轨迹信息的高速公路路网拓扑结构动态生成方法研究[J]. 交通运输系统工程与信息, 2025, 25(1): 212-220.
[14]	缪鸿志, 李嘉威, 李江晨, 贾洪飞, 李振福, 李歆蔚. 复杂度学习驱动的无人电卡集疏运路径优化[J]. 交通运输系统工程与信息, 2025, 25(1): 270-288.
[15]	蒋贤才, 郭子豪, 宋成举. 网联车辆环境下城市道路交通流分段协同控制方法[J]. 交通运输系统工程与信息, 2024, 24(6): 47-62.