基于神经网络的北京环路交通流短期预测研究

ShortTerm Prediction of Traffic Flow in Beijing Ring Road Based on Neural Network

在总结交通流短期预测方法发展趋势的基础上,分别介绍了基于常规的BP神经网络和基于RBF神经网络的交通流量短期预测模型,并重点研究RBF网络模型的预测性能,确定了关键参数sc的最优值.最后应用两种模型对北京环路实测交通流数据进行了预测分析,实验结果表明,两种模型都可以满足实际交通流诱导的需要,BP模型在预测精度上稍优于RBF模型,但后者在学习速度和学习稳定性等方面明显优于前者.

Based on the summarization of the stateoftheart of the shortterm traffic flow prediction methods, two short-term prediction models of traffic flow that include BP Neural Network and RBF Neural Network are discussed in this paper. The function of RBF model is studied chiefly and the optimum value of sc which is the model’s key parameter, is determined. Actual traffic flow data of Beijing ring road is predicted and analyzed by means of these two models. Experiment results show that both two models can satisfy the need of traffic flow guide. Contrastively, BP model is better than RBF model in precision, but RBF model has advantages obviously in learning rate and learning stability.