基于时空特性和RBF神经网络的短时交通流预测

针对实际交通流变化具有较明显的动态性、周相似性和相关性,提出一种基于交通流的时空变化特性和RBF神经网络的短时交通流预测方法。该方法充分挖掘和利用了交通流时间序列的周相似性和相关性,以及相邻路段上交通流的相互影响因素,结合RBF神经网络自学习、自组织、自适应功能和大范围的数据融合特性对交通流进行短时预测。用实例进行了仿真计算和分析,结果表明该方法能够提高交通流的预测精度。

Short-term Traffic Flow Forecasting Based on Spatiotemporal Characteristics of Traffic Flow and RBF Neural Network

As traffic flow changes dynamically with week-similarity and relevance,this paper presents a short-term traffic flow forecasting method based on spatial and temporal changes in traffic flow characteristics and RBF neural network.The method takes full advantage of the week-similarity and relevance of the traffic flow time series,considering the adjacent section of the traffic flow of interacting factors,with combination of self-learning,self-organizing,and adaptive function of the RBF neural network,plus a wide range of data integration characteristics in short-term traffic flow forecasting.Finally,examples are simulated and analyzed.The results show that the method can improve the traffic flow prediction accuracy.