基于主成分分析和组合神经网络的短时交通流预测方法

提出一种将主成分分析和组合神经网络相结合的方法来预测短时交通流量.将预测路段历史流量及其相关路段的历史流量进行主成分分析,分析结果作为组合神经网络的输入数据,这样不仅减少了输入变量个数,减小网络规模,而且保留了原始变量的主要信息,消除变量之间的相关性.给出了一个说明该方法有效性的例子,用同一组数据比较该方法与典型BP网络的预测效果,结果表明该方法明显优于BP网络.

The Forecasting Approach for Short-term Traffic Flow based on Principal Component Analysis and Combined NN

A combination approach based on Principal Component Analysis(PCA) and combined NN is presented for short-term traffic flow forecasting.The historical data of the traffic flows of the forecasted road and interrelated roads have been processed by Principal Component Analysis.The results of PCA are input data for combined NN.It not only reduces the number of input variables and the size of NN,but also reserves the main information of original variables and irrelevance among variables.An example for explanation of validity is given.The forecast results show that this approach is better than the typical BP NN with the same data.