混合交通环境下降雨对车辆跟驰行为的影响分析与模型构建  

Analysis of vehicle following behavior and construction of model under mixed traffic flow in rainfall environment

作　　者：邢国政 高超 XING Guo-zheng;GAO Chao(School of Highway,Chang'an University,Xi'an 710064,Shaanxi,China;Key Laboratory of lntelligent Construction and Maintenance of CAAC,Xi'an 710064,Shaanxi,China)

机构地区：[1]长安大学公路学院,陕西西安710064 [2]长安大学民航机场智慧建造与维养重点实验室,陕西西安710064

出　　处：《长安大学学报（自然科学版）》2024年第3期138-148,共11页Journal of Chang’an University(Natural Science Edition)

基　　金：交通运输行业重点科技项目(2018-ZD1-001);中央高校基本科研业务费专项资金项目(300102212107)。

摘　　要：针对混合交通环境下车辆跟驰分析时缺少对降雨环境考虑的问题,通过研究驾驶人在降雨环境下驾驶行为的改变,构建在不同降雨强度下适用的跟驰模型。首先,通过Python程序从云控平台数据库中提取车辆跟驰数据和降雨强度数据,分析了车辆在不同降雨强度和交通流状态下的跟驰速度、加速度和车头间距等变化特征以及其相互作用关系,进一步探索了降雨与正常天气下车辆跟驰行为的差异。其次,从优化速度跟驰模型出发,使用加速度衡量跟驰行为特性,综合考虑降雨对前后车相对距离和相对速度的影响,提出了适用于不同降雨强度下的安全间距和驾驶人期望速度确定方法,构建了不同降雨强度下的跟驰模型,并采用遗传算法对新模型进行了参数标定。最后,通过交叉验证方法对标定结果进行了评估,计算了模型输出值与实际值之间的误差。研究结果表明:降雨环境下车辆速度和加速度随着降雨强度的增加有不同幅度的降低,但是车辆加速度变化程度比车辆速度变化程度明显,这说明加速度对于降雨反应更加灵敏;降雨环境下,车头间距受前车速度影响,进而控制后车跟驰速度,不同降雨强度和不同前车速度区间都影响车头间距大小,后车跟驰速度随着车头间距的增大而增大;模型计算结果与实测值的均方根百分比误差不超过0.5,说明所构建的考虑降雨强度的跟驰模型能够很好地描述在不同降雨强度下高速公路驾驶人的跟驰行为。Aiming at the problem that lacks accurate car-following models in rainfall environments,a car-following model under various rainfall intensities was proposed.Firstly,the date of car-following and rainfall intensity were extracted from the cloud control platform database using Python program.The changing characteristics of car-following speed,acceleration,and headway distance,as well as their inter-relationships,were investigated under various rainfall intensities and traffic flow.The difference of car-following behavior between rainfall and normal weather was further explored.Subsequently,a method based on optimal velocity model was proposed to determine the safe distance and driver's desired speed under various rainfall intensities by considering the impacts of rainfall on the relative distance and relative speed between the leading and following vehicles.Then car-following model was created for various rainfall intensities,and the parameters of the new model were calibrated using genetic algorithms.Finally,the calibrated results were evaluated using cross-validation methods,and the error between the model output and actual values was calculated.The results show that speed and acceleration decrease to varying extents with increasing rainfall intensity.However,the variations in acceleration are more significant than those in speed,indicating a more accurate sensitivity of acceleration to rainfall.The headway distance is influenced by the leading vehicle's speed,which in turn controls the following vehicle's speed.Both the rainfall intensity and the speed range of the leading vehicle affect the headway distance,and the following speed of the following vehicle increases with the increase of headway.Model validation results indicate that the root mean square percentage error between the model outputs and collected values is no more than 0.5,demonstrating that the car-following behavior is accurately described by new car-following model considering rainfall intensity on highway.5 tabs,6 figs,26 refs.

关 键 词：交通工程 高速公路 车辆跟驰模型 优化速度模型 降雨环境 

分 类 号：U491.112[交通运输工程—交通运输规划与管理]