摘要
针对交通拥堵问题,设计了基于神经网络模型的交通流量预测系统。交通控制系统是基于避免交通不稳定和使交通流均匀化的思想,使事故风险最小化,交通流最大化。将人工神经网路应用于交通量之短期预测。除了交通量、速度和密度外,模型还将时间和星期几作为输入变量。并利用收集到的实际高速公路流量数据对神经网络结构模型进行验证。实验表明,将每一类车辆的速度都被单独视为输入变量,人工神经网络在这项研究中取得了很好的效果。
Aiming at the problem of traffic congestion,this paper designs a traffic flow prediction system based on neural network model.Traffic control system is based on the idea of avoiding traffic instability and homogenizing traffic flow,minimizing accident risk and maximizing traffic flow.In this study,artificial neural networks are applied to shortterm traffic volume prediction.In addition to traffic volume,speed and density,the model also takes time and weeks as input variables.The structure model of the neural network is validated by using the collected actual traffic data of the expressway.Experiments show that the speed of each type of vehicle is regarded as an input variable separately,and the artificial neural network has achieved good results in this study.
作者
宋伟
张杨
Song Wei;Zhang Yang(Shaanxi Radio Television University,Xi'an 710068,China;International Business Machines Corporation,Xi'an 710000,China)
出处
《国外电子测量技术》
2019年第12期27-31,共5页
Foreign Electronic Measurement Technology
关键词
神经网络
交通流量
智能交通系统
模型
artificial neural netw ork
traffic flow
intelligent transportation system
modelling
