城市轨道交通客流预测的历史相似日选择研究
光志瑞1,2魏运1,2薛云雷3谢莎婷1,2吴雁军1,2
Study on Historical Similarity Day Selection in Urban Rail Transit Passenger Flow Prediction
GUANG ZhiruiWEI YunXUE YunleiXIE ShatingWU Yanjun
-
作者信息:1.北京市地铁运营有限公司技术创新研究院, 100039, 北京;
2.地铁运营安全保障技术北京市重点实验室, 100039, 北京;
3.京投新岸线技术有限公司, 100089, 北京
-
Affiliation:Technical Innovation Research Institute of Beijing Mass Transit Railway Operation Co., Ltd., 100039, Beijing, China
-
关键词:
-
Key words:
-
DOI:10.16037/j.1007-869x.2022.07.011
-
中图分类号/CLCN:U293.13
-
栏目/Col:研究报告
摘要:
选择合理的历史相似日（以下简称“相似日”），可有效提高城市轨道交通客流预测的精度。提出了适用于不同客流预测需求的相似日选择方法：首先将相似日选择过程转化为客流模式识别过程；其次提出了前向逐日检索和双向定位检索两种相似日选择算法策略；然后采用定量转化、量纲一化、构造函数法等方法分别计算了星期类型、与预测日的间隔时长、工作日属性、日最高温度等影响因素的相似度，构建了含各影响因素的累乘函数；最后，结合北京城市轨道交通客流数据对相似日选择方法进行拟合及验证。结果表明：该方法的预测结果准确、可靠，可为不同的城市轨道交通客流预测场景提供参考。
Abstracts:
Choosing reasonable historical similar days is an effective way to improve the prediction accuracy of urban rail transit passenger flow. A method of choosing similar days to meet different demands of passenger flow prediction is proposed. First, similar day selection process is converted to passenger flow pattern recognition process. Secondly, two strategies of similar day selection are proposed, which are ′forward day by day retrieval′ and ′bidirectional positioning retrieval′. Then, methods of quantitative transformation, normalization, and function construction are adopted to calculate the similarity degree of each influencing factor, such as week attribute, length of time between forecast days, date attribute, daily maximum temperature, a multiplicative function including all factors is constructed. Finally, Beijing urban rail transit passenger flow data is adopted for calibration and validation with similar day selection method. Results show that the prediction result of this method is accurate and reliable, providing reference for passenger flow prediction in various urban rail transit scenarios.
引文 / Ref:
光志瑞，魏运，薛云雷，等.城市轨道交通客流预测的历史相似日选择研究[J].城市轨道交通研究，2022，25（7）：51.
GUANG Zhirui，WEI Yun，XUE Yunlei，et al.Study on Historical Similarity Day Selection in Urban Rail Transit Passenger Flow Prediction[J].Urban mass transit，2022，25（7）：51.
GUANG Zhirui，WEI Yun，XUE Yunlei，et al.Study on Historical Similarity Day Selection in Urban Rail Transit Passenger Flow Prediction[J].Urban mass transit，2022，25（7）：51.
- 上一篇: 高铁物流运输模式及其可行性
- 下一篇: 新型冠状病毒肺炎疫情对上海轨道交通客流的影响及防控措施分析*
过刊浏览
资料下载
友情链接