摘要
对建设项目进行交通影响分析越来越受到交通工程专家和规划管理部门的重视,如何科学预测项目建成后对周围路网的交通影响一直是交通影响分析理论和实践中的重要问题.根据人工神经网络的基本原理,利用三层BP网络,建立了输入变量是建设项目产生的交通量、路网中各路段的背景饱和度、离开建设项目的距离,输出变量是项目建成后路段饱和度的交通影响预测模型,并以哈尔滨国际会展体育中心为例,详细介绍模型的建立过程.模型预测值与实际值之间的相对误差<5%,拟合度达到0.99.
The traffic impact analysis of projects is more and more emphasized by the transportation engineering experts, the planning and management departments. It is the important problem in the theory research and practice on current traffic impact analysis that how to predict the impact on the surrounding road network. According to the basic theory of artificial neural network and using the three-layer BP network, the traffic impact predict model are established. The input variables are the traffic production of the project, the background degree of saturation of sections of highway and the distance from the project, and the output variable is the traffic impact predict model of the road degree of saturation after the project is established. The process of establishing model is illustrated by the gymnasium center of Harbin international conference exhibition, the error analyses outcome of the model and the traffic impact of the center on surrounding road network are also analyzed.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2004年第8期1034-1037,共4页
Journal of Harbin Institute of Technology
基金
国家自然科学基金资助项目(50278026).
关键词
BP神经网络
交通影响预测模型
饱和度
误差分析
交通量
Backpropagation
Error analysis
Forecasting
Gymnasiums
Neural networks
Railroad traffic control
