摘要
根据典型高速公路隧道群交通运行状况,从时间因素、交通动态因素、道路条件和隧道群环境因素等方面选取输入变量,以运行速度为输出变量,建立基于人工神经网络的高速公路隧道群速度预测模型;然后,通过灵敏度分析方法,研究各个输入变量对输出变量的影响程度,并对各个输入变量的灵敏度分析结果进行比较分析。研究结果表明:该方法能够针对隧道群交通流的实际情况,充分利用与速度密切相关的信息来模拟,克服了传统算法难以建模的缺陷,适合交通流限速控制的在线建模。该方法切实可行、准确度较高,可为制定高速公路隧道群速度限制提供理论基础。
Based on the traffic operation status of the typical freeway tunnel group,the import variable is selected from the time factor,traffic dynamic factor,road condition and tunnel environment factor.The speed forecast model of freeway tunnel group is established based on artificial neural network by taking operation speed as the output variable.The article studies the influence level of each import variable on the output variable by the sensitivity analysis method,and compares and analyzes the sensitivity analysis result of each import variable.The study result shows that this method can fully use the information closely relevant to the speed to simulate the practical traffic flow of tunnel group,overcome the fault of traditional arithmetic hard to establish the model,and is suitable for the on-line model establishment of traffic flow speed limit control.This method is feasible and its accuracy is higher,which can provide the theoretical basis for drafting the speed limit of freeway tunnel group.
出处
《城市道桥与防洪》
2015年第1期138-141,15,共4页
Urban Roads Bridges & Flood Control
关键词
人工神经网络
高速公路隧道群
限速
artificial neural network
freeway tunnel group
speed limit