摘要
数据丢失是动态交通数据采集中的常见现象,丢失数据的补齐质量对交通信息的准确性具有重要影响。本文根据交通数据随时间变化的特点,利用交通检测器采集到的动态数据,对训练数据进行了4种插值方法的插值效果比较,进行了少量数据丢失和大量数据丢失的补齐方法研究,建立了拟合多项式函数,根据保形拟合确定了合理的补齐数值,补齐数据和原始数据的平均相对误差5.0%,数据补齐精度高。
Data missing is a common phenomenon in dynamic traffic data collection. But its completing quality has important influence on traffic information accuracy. According to dynamic data collected by traffic detector and the characteristics of time variation of them,comparison of four kinds of interpolation methods was conducted by using training data. Then,completing method of small amount of missing data and large amount of missing data is carried on research,and fitting polynomial function is built. At last,reasonable data is calculated by shape preserving fitting method with only 5 percent average relative error. The study shows the data completing method has high accuracy.
作者
王小霞
傅惠
杨马进
卢建锋
WANG Xiaoxia FU Hui YANG Majin LU Jianfeng(Faculty of Civil and Transportation Engineering, Guangdong University of Technology, Guangzhou 510006, China Faculty of Electromechanical Engineering, Guangdong University of Technology, Guangzhou 510006, China)
出处
《内蒙古农业大学学报（自然科学版）》
CAS
2016年第3期93-96,共4页
Journal of Inner Mongolia Agricultural University(Natural Science Edition)
基金
国家自然科学基金项目(61573110)
广东工业大学博士基金(12ZK0400)
广东省大学生创新创业训练项目(201411845118)
关键词
动态交通
数据丢失
数据补齐
数据插值
Dynamic traffic
data missing
data completing
data interpolation
