基于小波神经网络的短时交通量预测

The Prediction of Short-term Traffic Flow Based on the Wavelet Neural Network

为了更有效地引导交通运行,针对城市交通流量,以福州市连江县文笔路与敖江北路的交叉路口为例,构建基于小波神经网络的短时交通量预测模型,应用MATLAB软件进行计算,通过网络训练、修正权值和小波基函数参数,得出预测结果。结果显示,小波神经网络预测的相对误差均值为2.65%,而ARMA预测的相对误差均值为39.06%,说明小波神经网络预测方法可更精确地预测短时交通量。

In order to guide the traffic running effectively,taking the intersection of Wenbi Road and Aojiang North Road in Lianjiang County as an example,the prediction model of the short-term traffic flow was constructed based on the wavelet neural network method,and finally the results were obtained with the soft of MATLAB by the network training of the sample data performed to correct the weights and parameters of the wavelet basis function.The results showed that the wavelet neural network predicted a relative error mean was of 2.65%,while the ARMA was 39.06%,so the wavelet neural network method can predict short-term traffic flow more accurately.