《计算机应用》唯一官方网站 ›› 2022, Vol. 42 ›› Issue (1): 258-264.DOI: 10.11772/j.issn.1001-9081.2021010080

• 前沿与综合应用 • 上一篇    

基于改进时空残差卷积神经网络的城市路网短时交通流预测

包银鑫1, 曹阳1,2, 施佺1,2()   

  1. 1.南通大学 信息科学技术学院,江苏 南通 226019
    2.南通大学 交通与土木工程学院,江苏 南通 226019
  • 收稿日期:2021-01-15 修回日期:2021-04-20 接受日期:2021-04-29 发布日期:2021-05-12 出版日期:2022-01-10
  • 通讯作者: 施佺
  • 作者简介:包银鑫(1996—),男,江苏淮安人,博士研究生,主要研究方向:智能信息处理
    曹阳(1986—),江苏如皋人,教授,博士生导师,博士,主要研究方向:科学计算、智能信息处理
    施佺(1973—),男,江苏海门人,教授,博士生导师,博士,CCF高级会员,主要研究方向:智能信息处理。
  • 基金资助:
    国家自然科学基金资助项目(61771265);江苏省“333工程”科研项目(BRA2017475);南通市“226”科研项目(131320633045)

Improved spatio-temporal residual convolutional neural network for urban road network short-term traffic flow prediction

Yinxin BAO1, Yang CAO1,2, Quan SHI1,2()   

  1. 1.School of Information Science and Technology,Nantong University,Nantong Jiangsu 226019,China
    2.College of Transportation and Civil Engineering,Nantong University,Nantong Jiangsu 226019,China
  • Received:2021-01-15 Revised:2021-04-20 Accepted:2021-04-29 Online:2021-05-12 Published:2022-01-10
  • Contact: Quan SHI
  • About author:BAO Yinxin, born in 1996, Ph. D. candidate. His research interests include intelligent information processing.
    CAO Yang, born in 1986, Ph. D., professor. His research interests include scientific computing, intelligent information processing.
    SHI Quan, born in 1973, Ph. D., professor. His research interests include intelligent information processing.
  • Supported by:
    National Natural Science Foundation of China(61771265);“Project 333” Scientific Research Project of Jiangsu(BRA2017475);“226” Scientific Research Project of Nantong(131320633045)

摘要:

城市路网交通流预测受到历史交通流和相邻路口交通流的影响,具有复杂的时空关联性。针对传统时空残差模型缺乏对交通流数据进行相关性分析、捕获微小变化而容易忽略长期时间特征等问题,提出一种基于改进时空残差卷积神经网络(CNN)的城市路网短时交通流预测模型。该模型将原始交通流数据转化成交通栅格数据,利用皮尔逊相关系数(PCC)对交通栅格数据进行相关性分析,确定相关性高的周期序列和邻近序列;同时,建立周期序列模型和邻近序列模型,并引入长短时记忆(LSTM)网络作为混合模型提取时间特征以及捕获两种序列的长期时间特征。利用成都市出租车数据集对模型进行验证,结果表明该模型预测结果优于LSTM、CNN和传统残差模型等基准模型,以均方根误差(RMSE)为评价指标时,所提模型将测试集中交通路网的平均预测精度分别提高了25.6%、13.3%和3.2%。

关键词: 短时交通流预测, 时空分析, 残差网络, 皮尔逊相关系数, 长短时记忆网络, 卷积神经网络, 组合模型

Abstract:

Traffic flow prediction for urban road network is influenced by historical traffic flow and traffic flow at adjacent intersections, which has complex spatio-temporal correlation. For the lack of correlation analysis on traffic flow data, capturing small changes but ignoring long-term time characteristics in traditional spatio-temporal residual models, a short-term traffic flow prediction model for urban road network based on improved spatio-temporal residual Convolutional Neural Network (CNN) was proposed. In the proposed model, the original traffic flow data was transformed into traffic grid data, and Pearson Correlation Coefficient (PCC) was used to analyze the correlation of traffic grid data, so as to determine the periodic series and adjacent series with high correlation. At the same time, the periodic series model and the adjacent series model were established, and Long Short-Term Memory (LSTM) network was introduced as the hybrid model to extract the time characteristics and capture the long-term time characteristics of the two series. Experimental results on Chengdu taxi dataset show that the proposed model can predict traffic flow better than benchmark models of LSTM, CNN and the traditional residual model. When the evaluation index is Root Mean Square Error (RMSE), the average prediction accuracy of traffic road network in the test set is improved by 25.6%, 13.3% and 3.2% respectively.

Key words: short-term traffic flow prediction, spatio-temporal analysis, Residual Network (ResNet), Pearson Correlation Coefficient (PCC), Long Short-Term Memory (LSTM) network, Convolutional Neural Network (CNN), combination model

中图分类号: