| 基于RBF神经网络的城市互通立交短时交通量预测 |
| |
| 引用本文： | 臧晓冬,周伟.基于RBF神经网络的城市互通立交短时交通量预测[J].长沙交通学院学报,2009,25(3). |
| |
| 作者姓名： | 臧晓冬 周伟 |
| |
| 作者单位： | 1. 北京工业大学,交通工程北京市重点实验室,北京,100022;广州大学,土木学院,广东,广州,510006
2. 交通部公路科学研究院,北京,100088 |
| |
| 基金项目： | 国家科技支撑计划项目 |
| |
| 摘 要： | 根据实测数据,分析了城市互通立交交通流运行特性,研究了城市互通立交主线总交通量、交织流量比和第2车道大型车比例3种因素对合流区端部交通量的影响,应用神经网络原理建立了输入层为3个神经元,输出层为1个神经元的RBF神经网络,研究了该RBF神经网络的训练算法,应用实测数据进行了网络训练,对训练后的网络进行了仿真.仿真输出结果与实测数据之间的误差很小,证明应用RBF神经网络对互通立交合流区的短时交通量进行预测是一种可行的方法.此研究为分析互通立交内各功能区之间的相互影响,为研究互通立交桥的预先控制和运行状态分析提供了理论参考.
|
| 关 键 词： | 交通工程 预测 神经网络 交通量 |
Short-term traffic volume forecast for urban interchange based on RBF |
| |
| ZANG Xiao-dong,ZHOU Wei.Short-term traffic volume forecast for urban interchange based on RBF[J].Journal of Changsha Communications University,2009,25(3). |
| |
| Authors: | ZANG Xiao-dong ZHOU Wei |
| |
| Affiliation: | ZANG Xiao-dong1,3,ZHOU Wei2(1.Beijing Key Lab of Traffic Engineering,Beijing University of Technology,Beijing 100022,China,2.Research Institute of Highway,Ministry of Communications,Beijing 100088,3.School of Civil Engineering,Guangzhou University,Guangzhou 510006,China) |
| |
| Abstract: | The influence of the total traffic volume of main-line,the ratio of weaving traffic volume to total traffic volume,and the heavy vehicle proportion in the second lane of interchange on the traffic volume of merging area are studied.RBF artificial neural networks with three input nerve cells and one output nerve cell are built.The training algorithm of the RBF neural network is studied,and the RBF neural network is trained according to the field data,then the RBF neural network is simulated and verified by t... |
| |
| Keywords: | traffic engineering forecast neural network traffic volume |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
