城市道路短期交通流预测VHSSA模型

VHSSA Model for Predicting Short-term Traffic Flow of Urban Road

针对出行者出行时对交通信息预报以及动态路径规划的要求,对路段的历史交通流时间序列数据进行了研究,利用城市路段交通流的周期相似性特征提出了基于纵横序列相似性的短期交通流预测VHSSA模型,该模型克服了以往预测模型只考虑纵向时间序列周期性相似的缺陷,将全时间序列数据进行小波变换后分解为反映基本变化规律的基序列和反映波动变化情况的波动序列,既可只进行基序列预测,也可通过置信区间对波动序列进行修正,再与基序列叠加进行全序列预测。经试例验证,VHSSA模型和基于纵向序列相似性的VSSA模型分别与实测序列的基序列和全序列进行比对,VHSSA模型的预测效果总体优于VSSA模型,误差可满足实际要求。

In order to meet the need of accurate traffic prediction and dynamic route planning, we proposed a short-term traffic predication model VHSSA based on the similarity of vertical and horizontal sequence by analyzing the historical traffic time series data and the cyclic similarity characteristics of traffic volume of urban road. The model can overcome the deficiency of the traditional models which only focus on the vertical cyclic sequence similarity. The full-time data are transformed into basis sequence which can reflect the basis characteristics and fluctuant sequence which can reflect the variation characteristics by using wavelet transformation function. By this transformation, both basis sequence prediction and complete sequence prediction can be achieved. When doing complete sequence prediction, we corrected the fluctuant sequence based on the confidence interval, overlapped it with the basis sequence. The verification experiments are carried out to compare the practical basis sequence and complete sequence of the VHSSA model and the VSSA model which consider only the vertical sequence similarity. The results show that the prediction result of the former is better than that of the latter, the error probability of the former is lower than that of the latter. The prediction error can meet actual requirement.