吉林大学学报(工学版) ›› 2022, Vol. 52 ›› Issue (11): 2618-2625.doi: 10.13229/j.cnki.jdxbgxb20210346

• 交通运输工程·土木工程 • 上一篇    

电动自行车出行特性及骑行决策行为建模

董春娇1(),董黛悦1,2,诸葛承祥3,甄理1   

  1. 1.北京交通大学 综合交通运输大数据应用技术交通运输行业重点实验室，北京 100044
    2.浙江省交通运输科学研究院 交通发展研究中心，杭州 310023
    3.香港理工大学 土地测量及地理资讯学系，香港 999077
  • 收稿日期:2021-04-20 出版日期:2022-11-01 发布日期:2022-11-16
  • 作者简介:董春娇（1982-），女，教授，博士. 研究方向：交通规划，交通安全，交通流理论，出行行为分析. E-mail： cjdong@bjtu.edu.cn
  • 基金资助:
    国家重点研发计划项目(2019YFF0301400)

Trip characteristics and decision⁃making behaviors modeling of electric bicycles riding

Chun-jiao DONG1(),Dai-yue DONG1,2,Cheng-xiang ZHU-GE3,Li ZHEN1   

  1. 1.Key Laboratory of Transport Industry of Big Data Application Technologies for Comprehensive Transport，Ministry of Transport，Beijing Jiaotong University，Beijing 100044，China
    2.Transportation Development Research Center，Zhejiang Scientific Research Institute of Transport，Hangzhou 310023，China
    3.The Department of Land Surveying and Geo-Informatics，The Hong Kong Polytechnic University，Hongkong 999077，China
  • Received:2021-04-20 Online:2022-11-01 Published:2022-11-16

RICH HTML

 1   

PDF (PC)

119

摘要：

通过开展问卷调查和断面交通流参数调查，分析了电动自行车出行特性和交通流量、速度特征。针对电动自行车正常骑行、占用机动车道骑行、在机动车之间穿插骑行和逆向骑行4种电动自行车骑行决策行为，建立了基于Elman神经网络和多元logistic回归的电动自行车骑行决策行为模型。以深圳市福田区为例，进行了实证研究，结果表明：基于Elman神经网络的电动自行骑行决策行为模型的预测效果明显优于基于多元logistic回归的电动自行车骑行决策行为模型，平均预测精度为91.62%，相比多元logistic回归模型提高了9.93%；构建的多元logistic回归模型能揭示影响因素和骑行决策行为之间的关联关系，可为制定减少电动自行车不安全骑行决策行为的策略和政策提供理论支撑。

关键词: 交通工程, 城市交通, 电动自行车, 骑行决策行为, Elman神经网络, 多元logistic回归

Abstract:

Through questionnaire survey and cross-section traffic flow parameter survey， the characteristics of electric bicycle travel， traffic flow and speed are analyzed. Aiming at four kinds of decision-making behaviors of electric bike， such as normal riding， occupying motorway riding， interleaving between vehicles and reverse riding， a decision-making behavior model of electric bike riding based on Elman neural network and multiple logistic regression was established. Finally， taking Futian District of Shenzhen as an example， this paper makes an empirical study. The results show that： the prediction effect of electric bicycle riding decision-making behavior model based on Elman neural network is significantly better than that based on multiple logistic regression， with an average prediction accuracy of 91.62%， which is 9.93% higher than that of multiple logistic regression model； the established multiple logistic regression model can reveal the relationship between influencing factors and riding decision-making behavior. The relationship can provide theoretical support for the formulation of strategies and policies to reduce the unsafe riding behavior of electric bikes.

Key words: traffic engineering, urban traffic, electric bike, riding decision behavior, Elman neural network, multivariate logistic regression

中图分类号: 

  • U491.1

图1

电动自行车出行特性"

图2

电动自行车流流量和速度调查地点分布图"

表1

电动自行车小时流量调查结果"

道路等级平均总流量/（辆?h?1）正常骑行占用机动车道骑行在机动车之间穿插骑行逆向骑行
数量占比/%数量占比/%数量占比/%数量占比/%
主干路298.0035.0011.15166.0055.203.001.3594.0032.30
次干路464.33157.1736.28216.3348.6554.007.4324.677.60
支路554.5083.5047.95407.0045.5564.006.500.000.00
平均值449.10118.0033.59244.4049.3445.806.0333.6011.02

图3

电动自行车骑行速度调查结果"

图4

基于Elman神经网络的电动自行车骑行决策行为模型"

图5

预测结果对比分析"

表2

基于多元logistic回归的电动自行车骑行决策行为模型参数估计结果"

决策行为变量β显著性
占道骑行常量-0.5750.456
收入5万以下-1.2260.007
收入5~10万-1.1380.006
收入10~20万-1.2840.003
收入20~50万-1.3080.005
回家时间18∶00~19∶000.7300.030
快递/外卖用途1.8760.000
车道过窄0.7840.000
机动车较少1.0370.000
公交车占道-0.5780.046
穿插骑行常量-1.8480.001
回家时间17∶00之前-1.0490.012
回家时间19∶00~22∶00-0.7390.045
出行时长0.4060.001
快递/外卖用途1.6820.001
路边停车1.0870.001
公交车停靠-0.5660.009
伺机转弯0.8270.002
逆向骑行常量-2.3470.001
回家时间17∶00~18∶000.9040.017
出行时长0.2570.005
快递/外卖用途1.1870.001
机动车流量大1.3210.001
左侧流量少0.5790.006
目的地在左侧0.5950.001
1 魏丽英, 崔裕枫, 魏家蓉. 基于局部最大熵换道规则的电动自行车流元胞自动机仿真模型[J]. 吉林大学学报: 工学版, 2017, 47(5): 1436-1445.
Wei Li-ying, Cui Yu-feng, Wei Jia-rong. Cellular automata model based on local maximum entropy lane-changing rules for electric bicycle flow[J]. Journal of Jilin University(Engineering and Technology Edition), 2017, 47(5): 1436-1445.
2 高雨虹, 曲昭伟, 宋现敏. 信号交叉口电动自行车提前启亮时间的确定[J]. 吉林大学学报: 工学版, 2020, 50(4): 1355-1369.
Gao Yu-hong, Qu Zhao-wei, Song Xian-min. Determination of early green time for e-bikes at signalized intersections[J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(4): 1355-1369.
3 Fyhri A, Fearnley N. Effects of E-bikes on bicycle use and mode share[J]. Transportation Research Part D: Transport and Environment, 2015, 36: 45-52.
4 Schepers P, Fishman E, Hertog P D, et al. The safety of electrically assisted bicycles compared to classic bicycles[J]. Accident Analysis & Prevention, 2014, 73(73): 174-180.
5 Wolf A, Seebauer S. Technology adoption of electric bicycles: a survey among early adopters[J]. Transportation Research Part A, 2014, 69: 196-211.
6 王涛, 黎文皓, 陈峻. 电动自行车风险驾驶行为因子分析及相关因素研究[J]. 重庆交通大学学报: 自然科学版, 2019, 38(11): 118-126.
Wang Tao, Li Wen-hao, Chen Jun. Factor analysis and related factors study on electric bicycle risk driving behavior[J]. Journal of chongqin jiaotong university(Natural Science), 2019, 38(11): 118-126.
7 毛霖, 杨新苗, 常玉林. 基于Logistic回归模型的电动自行车出行特征分析[J]. 交通运输工程与信息学报, 2007(1): 114-117.
Mao Lin, Yang Xin-miao, Chang Yu-lin. Analysis of the trip characteristics of E-bike based on logistic regression model[J]. Journal of Transportation Engineering and Information, 2007(1): 114-117.
8 Wu C X, Yao L, Zhang K. The red-light running behavior of electric bike riders and cyclists at urban intersections in china: an observational study[J]. Accident Analysis & Prevention, 2012, 49: 186-192.
9 Yang X B, Huan M, Abdel A, et al. A hazard-based duration model for analyzing crossing behavior of cyclists and electric bike riders at signalized intersections[J]. Accident Analysis & Prevention, 2014, 74: 33-41.
10 Tang K S, Dong S, Wang F, et al. Behavior of riders of electric bicycles at onset of green and yellow at signalized intersections in China[J]. Transportation Research Record: Journal of the Transportation Research Board, 2012(1): 85-96.
11 Nematchoua M, Deuse C, Cools M, et al. Evaluation of the potential of classic and electric bicycle commuting as an impetus for the transition towards environmentally sustainable cities: a case study of the university campuses in liege, belgium[J]. Transportation Research Part A, 2014, 69: 196-211.
[1] 张鑫,张卫华. 快速路合流区主线不同交通状态下的安全性分析[J]. 吉林大学学报(工学版), 2022, 52(6): 1308-1314.
[2] 邝先验,陈自如. 基于CA的无信号灯控制路段行人过街横道处动态博弈礼让行为[J]. 吉林大学学报(工学版), 2022, 52(4): 837-846.
[3] 贾洪飞,邵子函,杨丽丽. 终点不确定条件下网约车合乘匹配模型及算法[J]. 吉林大学学报(工学版), 2022, 52(3): 564-571.
[4] 曲大义,赵梓旭,贾彦峰,王韬,刘琼辉. 基于Lennard-Jones势的车辆跟驰动力学特性及模型[J]. 吉林大学学报(工学版), 2022, 52(11): 2549-2557.
[5] 张文会,伊静,刘委,于秋影,王连震. 基于MADYMO的大客车追尾碰撞事故乘员损伤机理[J]. 吉林大学学报(工学版), 2022, 52(1): 118-126.
[6] 曲大义,黑凯先,郭海兵,贾彦峰,王韬. 车联网环境下车辆换道博弈行为及模型[J]. 吉林大学学报(工学版), 2022, 52(1): 101-109.
[7] 杨世军,裴玉龙,潘恒彦,程国柱,张文会. 城市公交车辆驻站时间特征分析及预测[J]. 吉林大学学报(工学版), 2021, 51(6): 2031-2039.
[8] 刘志伟,刘建荣,邓卫. 基于潜在类别的无人驾驶汽车选择行为[J]. 吉林大学学报(工学版), 2021, 51(4): 1261-1268.
[9] 徐进,潘存书,符经厚,刘俊,王郸祁. 典型道路场景以及场景切换时的速度行为特性[J]. 吉林大学学报(工学版), 2021, 51(4): 1331-1341.
[10] 查伟雄,蔡其燕,李剑,严利鑫. 边路车辆出入条件下城市干线信号协调相位差优化[J]. 吉林大学学报(工学版), 2021, 51(2): 565-574.
[11] 孙宝凤,姜源,郑黎黎,崔万坤,任欣欣. 变覆盖半径下城市轨道交通维护保障网络设计模型[J]. 吉林大学学报(工学版), 2020, 50(2): 526-534.
[12] 陈磊,王江锋,谷远利,闫学东. 基于思维进化优化的多源交通数据融合算法[J]. 吉林大学学报(工学版), 2019, 49(3): 705-713.
[13] 魏丽英, 崔裕枫, 魏家蓉. 基于局部最大熵换道规则的电动自行车流元胞自动机仿真模型[J]. 吉林大学学报(工学版), 2017, 47(5): 1436-1445.
[14] 高继东, 曾必强, 彭伟. 电动自行车与轿车碰撞中骑车人的伤害特征[J]. 吉林大学学报(工学版), 2016, 46(6): 1786-1791.
[15] 徐程, 曲昭伟, 陶鹏飞. 混合自行车交通流下的自行车道通行能力估计[J]. 吉林大学学报(工学版), 2016, 46(1): 63-69.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 《吉林大学学报(工学版)》编辑部
地址：长春市人民大街5988号　电话：+86-431-85095297　传真：+86-431-85094074　E-mail: xbgxb@jlu.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发