摘要
提出了一种基于灰色神经网络的道路权重确定方法.首先利用灰色预测模型少数据建模和人工神经网络模型非线性逼近的优点,将两种模型有机结合,实现对交通流的模拟预测;其次利用交通流量-行驶速度以及行驶速度-行驶时间的关系确定交通流量与行驶时间的关系模型;最后结合前两部分建立的模型构建基于灰色神经网络的路阻函数模型,从而确定路段的权重.实验结果表明,该方法具有较高的精度,且模型利用少量的数据就可以确定路段的路阻函数,为路段权值的确定提供了一种有效可行的方法,可用于智能交通的路径规划等应用中.
A road weight evaluation method based on grey neural network was proposed.The advantages of grey forecasting model,which is simple and needs less original data,and neural network were synthesized to establish the grey neural network model of traffic flow.The relationships between traffic flow,velocity and density were used to determine the relational model of flow and time.On these bases,the impedance function model and the weight of road sections were achieved.The experiment results show that the model has higher accuracy and only a few measurement data are needed to confirm the impedance function model of road sections.The model is an effective,feasible and convenient method for getting the weight of road sections.It can be applied in the road planning of intelligent transportation system.
出处
《上海理工大学学报》
CAS
北大核心
2013年第6期552-556,562,共6页
Journal of University of Shanghai For Science and Technology
基金
北京市自然科学基金资助项目(4112016)
北京市属高等学校人才强教计划资助项目(PHR201008239)
关键词
交通阻抗
路径规划
灰色模型
神经网络
traffic impedance
road planning
grey model
neural network