《计算机应用》唯一官方网站 ›› 2025, Vol. 45 ›› Issue (7): 2253-2261.DOI: 10.11772/j.issn.1001-9081.2024070956

• 数据科学与技术 • 上一篇    下一篇

基于聚类多变量时间序列模型的交通状态实时预测

郭书君(), 任卫军, 陈倩倩, 游广飞   

  1. 长安大学 信息工程学院,西安 710064
  • 收稿日期:2024-07-09 修回日期:2024-10-04 接受日期:2024-10-09 发布日期:2025-07-10 出版日期:2025-07-10
  • 通讯作者: 郭书君
  • 作者简介:郭书君(2000—),女,四川达州人,硕士研究生,CCF会员,主要研究方向:智能交通、交通状态识别与预测 sjguo@chd.edu.cn
    任卫军(1973—),男,陕西西安人,副教授,博士,主要研究方向:计算机视觉、数据挖掘、智能交通
    陈倩倩(2000—),女,江苏扬州人,硕士研究生,主要研究方向:智能交通、高速公路维护养护策略
    游广飞(2000—),男,陕西安康人,硕士研究生,主要研究方向:智能交通、交通流预测。
  • 基金资助:
    陕西省重点研发计划项目(2021GY-033);西安公路研究院技术研究项目(220224230915)

Real-time prediction of traffic status based on clustering multivariate time series model

Shujun GUO(), Weijun REN, Qianqian CHEN, Guangfei YOU   

  1. School of Information Engineering,Chang’an University,Xi’an Shaanxi 710064,China
  • Received:2024-07-09 Revised:2024-10-04 Accepted:2024-10-09 Online:2025-07-10 Published:2025-07-10
  • Contact: Shujun GUO
  • About author:GUO Shujun, born in 2000, M. S. candidate. Her research interests include intelligent transportation, traffic status recognition and prediction.
    REN Weijun, born in 1973, Ph. D., associate professor. His research interests include computer vision, data mining, intelligent transportation.
    CHEN Qianqian, born in 2000, M. S. candidate. Her research interests include intelligent transportation, highway maintenance strategy.
    YOU Guangfei, born in 2000, M. S. candidate. His research interests include intelligent transportation, traffic flow prediction.
  • Supported by:
    Shaanxi Province Key Research and Development Program(2021GY-033);Technical Research Project of Xi’an Highway Research Institute(220224230915)

摘要:

针对现有的交通状态预测模型不能有效应对高速公路交通状态的模糊性以及模型训练后不能有效使用实时数据流的问题,提出基于聚类的多变量时间序列交通状态实时预测模型。首先,在分析交通流参数后,构建基于改进的模糊C均值(FCM)算法与熵权法的分类模型对交通状态进行模式定义并设定分类标准,并采用状态指数(SI)指标解决分类边界模糊问题;其次,在分类模型的基础上构建多变量时间序列预测模型,该模型结合卷积网络和注意力机制,能有效地捕捉时间序列数据的长短期依赖关系;然后,利用反向传播更新机制进行在线学习,从而实现预测过程的实时化;最后,将模型在加州交通管理中心性能监控系统(PeMS)数据集上进行测试,把数据集按时间顺序分为训练集、验证集和测试集,并模拟实时数据流进行在线学习和预测。实验结果表明,预测步长为6时,与经典的LightTS(Light Sampling-oriented MLP Structures)模型相比,所提模型的均方误差(MSE)和平均绝对误差(MAE)分别降低了22.81%和14.64%。可见,所提模型能够有效区分交通状态等级,并实现交通状态的实时预测。

关键词: 高速公路, 交通状态分类, 交通状态预测, 模糊聚类, 神经网络, 在线学习

Abstract:

To address the issues that the existing traffic status prediction models cannot handle the fuzziness of highway traffic status effectively and fail to utilize real-time data streams after model training, a real-time prediction method for traffic status based on clustering multivariate time series model was proposed. Firstly, traffic flow parameters were analyzed, and a classification model based on the improved Fuzzy C-Means (FCM) algorithm and entropy weight method was developed to define traffic status and establish classification standards, and a Status Index (SI) indicator was employed to address the issue of classification boundary fuzziness. Secondly, a multivariate time series prediction model was constructed on the basis of the classification model. By combining convolutional networks and attention mechanisms, this model was able to capture both long-term and short-term dependencies in time series data effectively. Thirdly, a back-propagation update mechanism was applied for online learning to realize real-time prediction. Finally, the model was tested on the California Traffic Management Center Performance Measurement System (PeMS) dataset, the dataset was divided into training, validation, and test sets in time order, and real-time data stream simulations were conducted for online learning and prediction. Experimental results show that with the prediction step of 6, compared to the classic LightTS (Light Sampling-oriented MLP Structures) model, the proposed model reduces the Mean Squared Error (MSE) by 22.81% and the Mean Absolute Error (MAE) by 14.64%. It can be seen that the proposed model distinguishes traffic status levels effectively and achieves real-time traffic status prediction.

Key words: highway, traffic status classification, traffic status prediction, fuzzy clustering, neural network, online learning

中图分类号: