【作者】 杜豫川； 陈赣浙； 周小鹏； 孙立军；

【Author】 Du Yuchuan;Cheng Ganzhe;Zhou Xiaopeng;Sun Lijun;Key Laboratory of Road and Traffic Engineering of the Ministry of Education,Tongji University;Department of Traffic Police in Shanghai;

【机构】 同济大学道路与交通工程教育部重点实验室； 上海市交警总队；

【摘要】 本文以快速路网为研究对象,探讨了利用车牌识别数据预测OD的可行性。提出了基于ADF单位根检验的快速路OD波动性分析方法和基于K-Means算法的快速路OD相似性分析方法,并根据这两个方法分别对研究范围进行了多时间颗粒度下快速路OD的波动性分析和相似性分析。最后,论文以时间颗粒度为30分钟的工作日快速路OD历史数据作为基础,提出了基于贝叶斯估计的快速路OD预测方法,并通过实例验证了该预测方法的适用性和实用性。更多还原

【Abstract】 This paper discusses the feasibility of forecasting of OD through the license plate recognition data.Then propose the expressway OD volatility analysis method based on the ADF unit root test and the expressway OD similarity analysis method based on K-Means algorithm,and analyzes the expressway OD volatility and similarity under multiple time granularity based on these two methods respectively.Finally,use the workday expressway OD history data in time granularity of 30 minutes as the foundation,designs expressway OD forecasting method based on Bayes estimation.Through examples verify the applicability and practicability of the method.更多还原