Passenger flow monitoring and early warning system for railway passenger station
-
摘要: 由于缺乏准确的实时客流数据，铁路客运车站的客运组织一直主要依靠经验来调配设备、人员等所需资源。为此，开发了铁路客运车站客流监测与预警系统，通过有效利用铁路客票预售数据、车站历史客流数据、旅客进出站实时数据、列车正晚点数据等相关数据，建立基于K均值聚类的支持向量回归机客流预测模型，实现车站每日进站客流、分时段进站客流、候车室客流的监测、预测及超限预警，方便车站工作人员随时掌握客流动态，及时根据客流变化进行设备和人员动态调配，更加精准、高效、安全、有序地开展车站客运组织作业，有助于改善车站客运服务水平，提升旅客出行体验。Abstract: Due to the lack of accurate real-time passenger flow data, passenger traffic organization of railway passenger station has been mainly relying on experience to allocate the equipment, personnel and other required resources. Therefore, the passenger flow monitoring and early warning system for railway passenger station is developed. Through effective use of railway passenger ticket reservation data, historical station passenger flow data, real-time passenger flow data in and out of the station, train delay data and other relevant data, the monitoring, forecasting and over-limit warning of daily passenger flow, time-phased passenger flow of the station and passenger flow in waiting rooms can be realized using a passenger flow prediction support vector regression model based on K-mean clustering, which can facilitate the station staff to grasp the dynamic changes in passenger flow at any time so as to timely deploy the equipment and personnel, thus carrying out the passenger traffic organization operations of the station more accurately, efficiently, safely and orderly, and helping improve the level of passenger service at railway passenger stations and enhancing passenger travel experience.
-
Key words:
- railway passenger traffic /
- passenger flow monitoring /
- prediction and early warning /
- K-means /
- SVR
-
表 1 济南站客流预警级别划分标准
预警级别 预警颜色 进站口客流预测量E
（人/台）候车室客流预测量W
（m2/人）Ⅳ（一般） 蓝色 300$ \leqslant E $<500 1.1$ \leqslant W $ Ⅲ（较重） 黄色 500$ \leqslant E $<1000 0.9$ \leqslant W < 1 $.1 Ⅱ（严重） 橙色 1000$ \leqslant E $<1500 0.7$ \leqslant W < $0.9 Ⅰ（特别严重） 红色 1500$ \leqslant E $ 0.7>$ W $ 注：依据《铁路旅客运输管理规则》规定，候车厅（室）旅客占用面积标准按1.1～1. 2 m2/人计算。 
下载: 导出CSV
-
[1] 朱建生. 铁路新一代客票系统大数据应用创新研究 [J]. 铁路计算机应用,2019,28(4):1-7. doi: 10.3969/j.issn.1005-8451.2019.04.002 [2] 单杏花,张志强,宁 斐,等. 中国铁路电子客票关键技术应用与系统实现 [J]. 中国铁道科学,2021,42(5):162-173. doi: 10.3969/j.issn.1001-4632.2021.05.19 [3] 郝晓培. 基于大数据的铁路客运用户画像系统研究及应用[D]. 北京: 中国铁道科学研究院, 2018. [4] 郭 淼. 基于支持向量回归的大型客运站客流量预测应用研究 [J]. 铁路计算机应用,2021,30(5):15-18. [5] 李 瑞,吴 恒,杨文韬. 基于售票数据的铁路客运站人流量监测预警系统设计 [J]. 铁道运输与经济,2016,38(12):54-58,73. [6] 阎志远,翁湦元,戴琳琳,等. 铁路客运大数据平台的数据采集技术研究 [J]. 铁路计算机应用,2016,25(9):17-21. doi: 10.3969/j.issn.1005-8451.2016.09.005 [7] 贾成强,刘婷婷,傅 敏,等. 铁路实名制验证检票系统设计与实现 [J]. 铁路技术创新,2012(4):60-62. [8] 徐承禹,韩旭芳. 旅客列车正晚点实时分析系统的设计与实现 [J]. 铁路计算机应用,2012,21(11):18-21. doi: 10.3969/j.issn.1005-8451.2012.11.006 [9] Li S Q, Liu X M, Lin A J. Fractional frequency hybrid model based on EEMD for financial time series forecasting [J]. Communications in Nonlinear Science and Numerical Simulation, 2020(89): 105281. doi: 10.1016/j.cnsns.2020.105281 [10] Feng R L, Xu Z H, Wang J J, et al. Research on the closing prediction of shanghai composite index based on grey prediction model [J]. International Core Journal of Engineering, 2019, 5(8): 116-118. [11] Abad L A, Sarabia S M, Yuzon J M, et al. A short-term load forecasting algorithm using support vector regression & artificial neural network method (SVR-ANN)[C]//2020 11th IEEE Control and System Graduate Research Colloquium (ICSGRC), 8 August 2020, Shah Alam, Malaysia. New York, USA: IEEE, 2020. 138-143. [12] Hu R, Chiu Y C, Hsieh C W, et al. Mass rapid transit system passenger traffic forecast using a Re-sample recurrent neural network [J]. Journal of Advanced Transportation, 2019(2019): 8943291. [13] 荀 超,陈伯建,吴翔宇,等. 基于改进K-means算法的电力短期负荷预测方法研究 [J]. 电力科学与技术学报,2022,37(1):90-95.
点击查看大图
图(4) / 表(1)
