吉林大学学报(工学版) ›› 2023, Vol. 53 ›› Issue (12): 3458-3464.doi: 10.13229/j.cnki.jdxbgxb.20221192

• 交通运输工程·土木工程 • 上一篇    

基于深度异步残差网络的路网短时交通流预测算法

高海龙1(),徐一博2,侯德藻2(),王雪松3   

  1. 1.交通运输部公路科学研究院 智能交通技术交通运输行业重点实验室,北京 100088
    2.交通运输部公路科学研究院 智能交通技术交通运输行业重点实验室,北京 100088
    3.同济大学 交通运输工程学院,上海 201804
  • 收稿日期:2022-09-15 出版日期:2023-12-01 发布日期:2024-01-12
  • 通讯作者: 侯德藻 E-mail:ghl19650815@163.com;hdz0328@163.com
  • 作者简介:高海龙(1965-),男,研究员,博士.研究方向:交通安全与智能交通.E-mail:ghl19650815@163.com
  • 基金资助:
    国家自然科学基金项目(U21B2089)

Shortterm traffic flow prediction algorithm for road network based on deep asynchronous residual network

Hai-long GAO1(),Yi-bo XU2,De-zao HOU2(),Xue-song WANG3   

  1. 1.Key Laboratory of Technology on Intelligent Transportation Systems,Research Institute of Highway,Ministry of Transport,Beijing 100088,China
    2.Key Laboratory of Technology on Intelligent Transportation Systems,Research Institute of Highway,Ministry of Transport,Beijing 100088,China
    3.School of Transportation Engineering,Tongji University,Shanghai 201804,China
  • Received:2022-09-15 Online:2023-12-01 Published:2024-01-12
  • Contact: De-zao HOU E-mail:ghl19650815@163.com;hdz0328@163.com

摘要:

将整个路网划分成符合路网短时交通预测要求的子路网,从路网短时交通量中提取时间序列性、周期相似性以及空间序列性3个时空特性。将残差单元加入卷积神经网络中,构建了深度异步网络模型,实现路网短时交通流预测。实验结果证明,本文算法的均方根误差值和平均绝对误差值较小,流量预测值趋势与流量真实值趋势几乎一致,交通流量预测性能较好。

关键词: 智能交通, 深度异步残差网络, 交通流预测性, 时空特性, 卷积神经网络, 残差单元

Abstract:

The whole road network was divided into sub-networks that meet the requirements of short-term traffic forecast. Three temporal and spatial characteristics are extracted from the short-term traffic volume of the road network, namely time sequence, periodic similarity and spatial sequence, and residual units were added to convolutional neural networks to construct a deep asynchronous network model to realize the short-term traffic flow forecast of the road network. The experimental results show that the root mean square error and average absolute error of the proposed algorithm are small, the trend of traffic forecast value is almost consistent with the trend of real traffic value, and the traffic flow forecast performance is good.

Key words: intelligent transportation, deep asynchronous residual network, predictability of traffic flow, space time characteristics, convolutional neural network, residual unit

中图分类号: 

  • TP399

图1

广义空间距离与其影响因素关系结构图"

图2

残差单元图"

表1

交通流量预测的数据划分"

数据集训练数据测试数据
南京Taxi4952456
杭州Bike3469269

图3

各类算法的均方根误差值对比"

图4

各类算法的平均绝对误差值对比"

图5

TaxiNJ流量实际值与各类算法的预测值对比"

图6

BikeHZ流量实际值与各类算法的预测值对比"

1 周永强, 王波, 钱欢,等. 智能交通系统中CMIMO-SM-ARQ协议性能建模分析[J]. 计算机应用研究, 2021, 38(8): 2449-2453.
Zhou Yong-qiang, Wang Bo, Qian Huan, et al. Performance modeling and analysis of CMIMO-SM-ARQ protocol in ITS[J]. Application Research of Computers, 2021, 38(8): 2449-2453.
2 陈喜群, 曹震, 沈楼涛, 等. 融合路段传输模型和深度学习的城市路网短时交通流状态预测[J]. 中国公路学报, 2021, 34(12): 203-216.
Chen Xi-qun, Cao Zhen, Shen Lou-tao, et al. Short-term traffic-state prediction of urban road networks based on the fusion of a link-transmission model and deep learning[J]. China Journal of Highway and Transport, 2021, 34(12): 203-216.
3 蒲悦逸, 王文涵, 朱强, 等. 基于CNN-ResNet-LSTM模型的城市短时交通流量预测算法[J]. 北京邮电大学学报, 2020, 43(5): 9-14.
Pu Yue-yi, Wang Wen-han, Zhu Qiang, et al. Urban short-term traffic flow prediction algorithm based on CNN-ResNet-LSTM model[J]. Journal of Beijing University of Posts and Telecommunications, 2020, 43(5): 9-14.
4 刘铭, 鱼昕. 基于改进LSTM算法的短时交通流量预测[J]. 桂林理工大学学报, 2021, 41(2): 409-414.
Liu Ming, Yu Xin. Short-term traffic flow prediction based on MVF-LSTM algorithm[J]. Journal of Guilin University of Technology, 2021, 41(2): 409-414.
5 代亮, 梅洋, 钱超, 等. 基于生成对抗网络的大规模路网交通流预测算法[J]. 控制与决策, 2021, 36(12): 2937-2945.
Dai Liang, Mei Yang, Qian Chao, et al. Traffic flow forecasting algorithm for large-scale road network based on GAN[J]. Control and Decision, 2021, 36(12): 2937-2945.
6 钟异莹, 陈坚, 邵毅明, 等. 强制性换道的空间特征对分流区交通流的影响[J]. 交通运输系统工程与信息, 2020, 20(5): 114-120.
Zhong Yi-ying, Chen Jian, Shao Yi-ming, et al. Spatial feature of mandatory lane changing and its impact on traffic flow at diverging area[J]. Journal of Transportation Systems Engineering and Information Technology, 2020, 20(5): 114-120.
7 温长鹏, 周磊, 潘兵宏. 基于交通流稳定距离的匝道连续入口最小间距研究[J]. 中外公路, 2021, 41(3): 392-397.
Wen Chang-peng, Zhou Lei, Pan Bing-hong. Research on minimum spacing of ramp continuous entrance based on traffic flow steady distance[J]. Journal of China & Foreign Highwa, 2021, 41(3): 392-397.
8 雷毅, 张善关, 谢云驰, 等. 基于流形距离的高速公路短时交通流预测模型[J]. 科学技术与工程, 2020, 20(18): 7312-7317.
Lei Yi, Zhang Shan-guan, Xie Yun-chi, et al. Short-term traffic flow prediction model of the highway based on manifold distance[J]. Science Technology and Engineering, 2020, 20(18): 7312-7317.
9 程浩, 王起琮, 杨奕华, 等. R-Q因子分析在海相石灰岩岩相识别中的应用——以塔里木盆地巴楚-塔中地区奥陶系石灰岩为例[J]. 西安石油大学学报: 自然科学版, 2020, 35(4): 1-10, 18.
Cheng Hao, Wang Qi-cong, Yang Yi-hua, et al. Application of R-Q factor analysis in lithofacies identification of marine limestone: taking the ordovician limestone in Bachu-Tazhong area of Tarim basin as an example[J]. Journal of Xi'an Shiyou University(Natural Science), 2020, 35(4): 1-10, 18.
10 吴晋武, 张海峰, 冉旭东. 基于数据约减和支持向量机的非参数回归短时交通流预测算法[J]. 公路交通科技, 2020, 37(7): 129-134.
Wu Jin-wu, Zhang Hai-feng, Ran Xu-dong. Nonparametric regressive short-term traffic flow forecast algorithm based on data reduction and SVM[J]. Journal of Highway and Transportation Research and Development, 2020, 37(7): 129-134.
11 黄振盛, 汪玉美, 韩江洪, 等. 基于MLS-SVM和时空特性的短时交通流量预测方法[J]. 合肥工业大学学报: 自然科学版, 2020, 43(1):57-63.
Huang Zhen-sheng, Wang Yu-mei, Han Jiang-hong, et al. Short-term urban traffic flow prediction based on MLS-SVM and spatiotemporal features[J]. Journal of Hefei University of Technology (Natural Science), 2020, 43(1): 57-63.
12 向乔怡, 祁辉. 有向图上高维时间序列模型及其在交通网络中的应用[J]. 数学杂志, 2022, 42(4): 345-358.
Xiang Qiao-yi, Qi Hui. High-dimensional time series model on directed graph and its application in traffic network[J]. Journal of Mathematics, 2022, 42(4): 345-358.
13 熊炜, 管来福, 王传胜,等. 基于预测和残差细化网络的道路提取算法研究[J]. 计算机工程与科学, 2020, 42(4): 683-690.
Xiong Wei, Guan Lai-fu, Wang Chuan-sheng, et al. Road extraction algorithm based on prediction and residual refinement networks[J]. Computer Engineering and Science, 2020, 42(4): 683-690.
14 张永凯, 武志昊, 林友芳, 等. 面向交通流量预测的时空超关系图卷积网络[J]. 计算机应用, 2021, 41(12): 3578-3584.
Zhang Yong-kai, Wu Zhi-hao, Lin You-fang, et al. Spatio-temporal hyper-relationship graph convolutional network for traffic flow forecasting[J]. Journal of Computer Applications, 2021, 41(12): 3578-3584.
15 李洪均, 丁宇鹏, 李超波, 等. 基于特征融合时序分割网络的行为识别研究[J]. 计算机研究与发展, 2020, 57(1): 145-158.
Li Hong-jun, Ding Yu-peng, Li Chao-bo, et al. Action recognition of temporal segment network based on feature fusion[J]. Journal of Computer Research and Development, 2020, 57(1): 145-158.
[1] 车翔玖,徐欢,潘明阳,刘全乐. 生物医学命名实体识别的两阶段学习算法[J]. 吉林大学学报(工学版), 2023, 53(8): 2380-2387.
[2] 张振海,季坤,党建武. 基于桥梁裂缝识别模型的桥梁裂缝病害识别方法[J]. 吉林大学学报(工学版), 2023, 53(5): 1418-1426.
[3] 刘培勇,董洁,谢罗峰,朱杨洋,殷国富. 基于多支路卷积神经网络的磁瓦表面缺陷检测算法[J]. 吉林大学学报(工学版), 2023, 53(5): 1449-1457.
[4] 朱冰,李紫薇,李奇. 基于改进SegNet的遥感图像建筑物分割方法[J]. 吉林大学学报(工学版), 2023, 53(1): 248-254.
[5] 申铉京,张雪峰,王玉,金玉波. 像素级卷积神经网络多聚焦图像融合算法[J]. 吉林大学学报(工学版), 2022, 52(8): 1857-1864.
[6] 高明华,杨璨. 基于改进卷积神经网络的交通目标检测方法[J]. 吉林大学学报(工学版), 2022, 52(6): 1353-1361.
[7] 杨怀江,王二帅,隋永新,闫丰,周跃. 简化型残差结构和快速深度残差网络[J]. 吉林大学学报(工学版), 2022, 52(6): 1413-1421.
[8] 王学智,李清亮,李文辉. 融合迁移学习的土壤湿度预测时空模型[J]. 吉林大学学报(工学版), 2022, 52(3): 675-683.
[9] 李向军,涂洁莹,赵志宾. 基于多尺度融合卷积神经网络的熔解曲线有效性分类[J]. 吉林大学学报(工学版), 2022, 52(3): 633-639.
[10] 李先通,全威,王华,孙鹏程,安鹏进,满永兴. 基于时空特征深度学习模型的路径行程时间预测[J]. 吉林大学学报(工学版), 2022, 52(3): 557-563.
[11] 高文志,王彦军,王欣伟,张攀,李勇,董阳. 基于卷积神经网络的柴油机失火故障实时诊断[J]. 吉林大学学报(工学版), 2022, 52(2): 417-424.
[12] 张龙,徐天鹏,王朝兵,易剑昱,甄灿壮. 基于卷积门控循环网络的齿轮箱故障诊断[J]. 吉林大学学报(工学版), 2022, 52(2): 368-376.
[13] 张友俊,程顺延. 基于卷积神经网络的微带滤波器结构参数估计[J]. 吉林大学学报(工学版), 2022, 52(12): 3022-3028.
[14] 杨军,高志明. 基于无监督变形网络的三维模型稠密对应关系计算[J]. 吉林大学学报(工学版), 2022, 52(12): 2971-2983.
[15] 申铉京,刘同壮,王玉,刘嘉伟. 基于卷积网络结构重参数化的车位状态检测算法[J]. 吉林大学学报(工学版), 2022, 52(12): 2898-2905.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!