基于小波分解与重构的交通流短时预测法

Study on the Short-term Forecasting for Traffic Flow Based on Wavelet Analysis

交通流短时预测是交通控制与交通诱导系统的关键问题之一 .随着预测时间跨度的缩短 ,交通流量的变化显示出越来越强的不确定性 ,使得一般的预测方法难以奏效 .本文探讨了小波分析在交通流短时预测中应用的可行性 ,提出了一种基于多分辨率小波分解与重构的交通流短时预测方法 ,介绍了该方法的原理 ,讨论了模型参数的确定 。

The short\|term forecasting for traffic flow is one of the key problems of traffic control and route guidance. With the shortening of the forecasting term, the uncertainty of traffic flow becomes more and more seriously, so that the general approaches is incapable to forecast the short\|term volume effectively. In this paper, the feasibility of the short\|term forecasting for traffic flow based on wavelet analysis is analyzed and an approach based on wavelet decomposition and reconstruction is put forward. Then the determination of the parameters is discussed and the simulation results are given.