宏观标定方法在Wiedemann驾驶行为阈值研究中的应用  

Application of Macroscopic Calibration in Study of Wiedemann Driving Behaviour Threshold

作　　者：卢守峰[1] 王丽园[1] 

机构地区：[1]长沙理工大学交通运输工程学院,湖南长沙410114

出　　处：《公路交通科技》2014年第9期115-119,共5页Journal of Highway and Transportation Research and Development

基　　金：国家自然科学基金项目(71071024);湖南省自然科学基金项目(12JJ2025);长沙市科技局重点项目(K1106004-11)

摘　　要：基于流量速度图对Wiedemann74模型参数标定的宏观方法进行了研究,建立了集成VISSIM、Matlab、ExcelVBA的参数标定平台,以最小化流量速度图的实测值与仿真值的差异为优化目标,应用图像识别方法判别图像的差异性,利用遗传算法优化参数值,实现了参数自动寻优的迭代过程。建立的参数标定平台能够利用流量、速度等宏观运行数据标定驾驶行为阈值参数,为利用检测器数据实现自动化标定提供了有效手段,为分析驾驶行为特点提供了方法,解决了VISSIM软件中默认参数不适合我国交通状况导致仿真精度不高的问题。利用路侧激光检测器采集长沙市南二环路断面交通数据,根据标定后的参数、实测数据对Wiedemann模型的驾驶行为阈值曲线进行了拟合,根据驾驶行为分区对长沙市南二环路的驾驶行为进行了分析。Speed-volume curve is used to macroscopically calibrate Wiedemann74 model parameters. The parameter calibration platform integrating VISSIM,Matlab and ExcelVBA is built to minimize the differences between the measured values and the simulated values of speed-volume curve. The proposed platform applies image recognition method to determine the image difference,and applies genetic algorithm to optimize the corresponding parameters. This platform realized automatic parameter optimization by iterations. The proposed parameter calibration platform calibrated driving behaviour threshold based on macroscopic operating data such as volume and speed,and provided an automatic way for parameter calibration using detector data,provided an effective method for studying driving behaviour,and solved the low simulation precision problem of default parameters which is unsuitable for Chinese traffic condition. The traffic data from the section of south second ring road of Changsha are collected using roadside laser detector. The driving behaviour threshold curve of Wiedemann model is fitted according to the calibrated parameters and measured data,and the driving behaviours of partition on south second ring road of Changsha City are analyzed.

关 键 词：交通工程 驾驶行为阈值 图像识别 遗传算法 宏观标定平台 

分 类 号：U491[交通运输工程—交通运输规划与管理]