The classical BPR impedance function model was improved in order to more accurately calculate the impedance value of road traffic. The long short-term memory (LSTM) neural network was established to predict the positive and negative of the undetermined coefficient value in the improved function. The traffic data collected from the Shangtang Elevated to Zhonghe Elevated sections of Hangzhou City were used to verify the model. The results were compared with the traditional BPR impedance function method, the classic EMME/2 cone delay function, the BP neural network prediction method and the LSTM neural network prediction method. Results show that the improved model has higher accuracy and reliability under the premise that the data accuracy meets the requirements, indicating that the road impedance calculated by using the improved model can more realistically reflect the traffic operation condition of the road.
Fei WANG,Wei-xiang XU. Improved model of road impedance function based on LSTM neural network. Journal of ZheJiang University (Engineering Science), 2021, 55(6): 1065-1071.
Fig.2Scatter distribution of value $\;\beta $ after introducing value $x$
Fig.3Fitted curve of value $\;\beta $
Fig.4Internal structure of singal LSTM uint
Fig.5Structure of LSTM neural network
Fig.6Comparison chart of value $T/{T_{\rm{f}}}$ calculation results of improved model and classic road resistance model
Fig.7Comparison chart of value $T/{T_{\rm{f}}}$ calculation results of improved model and neural network
Fig.8Comparison of travel time calculation results between improved model and classic road resistance model
Fig.9Comparison of travel time calculation results between improved model and neural network
模型
MAE
MAPE
RMSE
经典BPR
6.34
0.22
11.06
经典EMME/2 锥形延误函数
7.78
0.29
12.31
BP神经网络
18.29
0.69
136.44
LSTM神经网络
5.87
0.21
10.86
本文方法
2.54
0.09
4.91
Tab.1comparison of calculation results between improved model and classic model
[1]
ZHAO Z, CHEN W, WU X LSTM network: a deep learning approach for short-term traffic forecast[J]. IET Intelligent Transport Systems, 2017, 11 (2): 68- 75
doi: 10.1049/iet-its.2016.0208
[2]
DUAN Y J, LV Y S, WANG F Y. Travel time prediction with LSTM neural network[C]// 2016 IEEE 19th International Conference on Intelligent Transportation Systems. Brazil: IEEE, 2016: 1053-1058.
[3]
SIRIPANPORNCHANA C, PANICHPAPIBOON S, CHAOVALIT P. Travel-time prediction with deep learning[C]// Region 10 Conference. Singapore: IEEE, 2017: 1859-1862.
[4]
LI Y F, CHEN M N, ZHAO W Z Investigating long-term vehicle speed prediction based on BP-LSTM algorithms[J]. IET Intelligent Transport Systems, 2019, 13 (8): 1281- 1290
doi: 10.1049/iet-its.2018.5593
[5]
MA Y, ZHANG Z, IHLER A, et al Multi-lane short-term traffic forecasting with convolutional LSTM network[J]. IEEE Access, 2020, 2020 (8): 34629- 34643
[6]
XU W X, ZHAO J M. Research on traffic flow time series model and shortest path algorithm of urban traffic based on travel plans[C]// 2019 International Conference on Intelligent Computing, Automation and Systems. Chongqing: IEEE, 2019: 369-373.
DAVIDSON K B The theoretical basis of a flow-travel time relationship for use in transportation planning[J]. Australian Road Research, 1978, 8 (1): 32- 35
傅白白, 刘法胜, 冯恩民 交通网络费用函数的标定与分析[J]. 交通运输系统工程与信息, 2003, 3 (4): 53- 57 FU Bai-bai, LIU Fa-sheng, FENG En-min Traffic network costs analysis and validation[J]. Journal of Transportation Systems Engineering and Information Technology, 2003, 3 (4): 53- 57
doi: 10.3969/j.issn.1009-6744.2003.04.010
[11]
王树盛, 黄卫, 陆振波 路阻函数关系式推导及其拟合分析研究[J]. 公路交通科技, 2006, 23 (4): 107- 110 WANG Shu-sheng, HUANG Wei, LU Zhen-bo Deduction of link performance function and its regression analysis[J]. Journal of Highway and Transportation Research and Development, 2006, 23 (4): 107- 110
doi: 10.3969/j.issn.1002-0268.2006.04.026
[12]
四兵锋, 钟鸣, 高自友 城市混合交通条件下路段阻抗函数的研究[J]. 交通运输系统工程与信息, 2008, 2 (1): 68- 73 SI Bing-feng, ZHONG Ming, GAO Zi-you A link resistance function of urban mixed traffic network[J]. Journal of Transportation Systems Engineering and Information Technology, 2008, 2 (1): 68- 73
doi: 10.3969/j.issn.1009-6744.2008.01.011
[13]
王素欣, 王雷震, 高利, 等 BPR路阻函数的改进研究[J]. 武汉理工大学学报: 交通科学与工程版, 2009, 33 (3): 446- 449 WANG Su-xin, WANG Lei-zhen, GAO Li, et al Improvement study on BPR link performance function[J]. Journal of Wuhan University of Technology: Transportation Science and Engineering, 2009, 33 (3): 446- 449
doi: 10.3963/j.issn.1006-2823.2009.03.011
[14]
刘宁, 赵胜川, 何南 基于BPR函数的路阻函数研究[J]. 武汉理工大学学报: 交通科学与工程版, 2013, 37 (3): 545- 548 LIU Ning, ZHAO Sheng-chuan, HE Nan Further study of impedance function based on BPR function[J]. Journal of Wuhan University of Technology: Transportation Science and Engineering, 2013, 37 (3): 545- 548
doi: 10.3963/j.issn.2095-3844.2013.03.023
[15]
李昂, 李硕, 李玲 城市道路路段行程时间计算模型研究[J]. 公路工程, 2016, 41 (3): 193- 197 LI Ang, LI Shuo, LI Ling Research on the calculation models of vehicle travel time on urban road segments[J]. Highway Engineering, 2016, 41 (3): 193- 197
doi: 10.3969/j.issn.1674-0610.2016.03.040
[16]
潘义勇, 余婷, 马健霄 基于路段与节点的城市道路阻抗函数改进[J]. 重庆交通大学学报: 自然科学版, 2017, 36 (8): 76- 81 PAN Yi-yong, YU Ting, MA Jian-xiao Improvement of urban road impedance function based on section impedance and node impedance[J]. Journal of Chongqing Jiaotong University: Natural Science, 2017, 36 (8): 76- 81
[17]
李彦瑾, 罗霞 基于模糊神经网络的混合交通流路阻测算模型[J]. 吉林大学学报: 工学版, 2019, 49 (1): 53- 59 LI Yan-jin, LUO Xia Calculation model of road resistance in mixed traffic flow based on fuzzy neural network[J]. Journal of Jilin University: Engineering and Technology Edition, 2019, 49 (1): 53- 59