摘要
常见的基于实测数据的origin-destination(OD)预测方法分为两类:一类基于历史信息,即根据上一天或上一周同一日相同时段的数据进行预测,简称同比预测法;另一类则是根据同一天相邻时间段的数据预测本时段的OD,简称环比预测法.预测所用基础数据的时段长度称为时间颗粒度.时间颗粒度的大小对OD预测结果的稳定性、准确性具有重要影响.针对上海快速路网,采用ADF单位根检验和K—Means聚类分析方法,研究时间颗粒度对预测结果的影响,提出了时间颗粒度选择的建议,同比预测方法相比环比预测法更容易得出稳定、合理的预测结果,30-60min的时间颗粒度预测效果较好.
Two methods, year-on-year prediction method and circularity-ratio prediction method, are often resorted to in the real-time origin-destination (013) prediction. The former depends on historical information, such as the data about the previous day or week. The latter depends on the data of adjacent period to predict the real-time 0D. The size of time granularity, the time period used in the real-time prediction of expressway network has a significant effect on the stability and accuracy of OD prediction results. The Augmented Dickey-fuller Test (ADF) unit root test and K-means algorithm are employed to study the effect of prediction results, and the size of time granularity is recommended. Compared with circularity-ratio prediction method, year-on- year prediction method can lead to the more stable and accurate result and a more satisfied time granularity isbetween 30 minutes and one hour.
出处
《同济大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2016年第10期1553-1558,共6页
Journal of Tongji University:Natural Science
基金
"十二五"国家科技支撑计划(2014BAG03B05)
关键词
时间颗粒度
快速路OD预测
波动性
相似性
time granularity
expressway origin-destination(OD) prediction
volatility
similarity