计算机与现代化

• 算法设计与分析 • 上一篇    下一篇

基于SDZ-GRU的多特征短时交通流预测方法

  

  1. (华南师范大学,广东广州510631)
  • 收稿日期:2019-03-13 出版日期:2019-10-28 发布日期:2019-10-29
  • 作者简介:吕田(1993-),男,湖南永州人,硕士研究生,研究方向:交通模式识别,时空数据挖掘,E-mail: tianl9528@m.scnu.edu.cn。

A Multi-features Short-term Traffic Flow Prediction Method Based on SDZ-GRU

  1. (South China Normal University, Guangzhou 510631, China)
  • Received:2019-03-13 Online:2019-10-28 Published:2019-10-29

摘要: 针对当前短期交通流量预测方法误差较大,且仅依靠时间序列数据进行预测的问题,提出一种基于SDZ-GRU的多特征短时交通流预测方法(简称SGMTFP)。该方法在现有的时序数据的基础上加入时间信息等一系列辅助数据,并将SDZ(Surprisal-Driven Zoneout)应用于门控循环单元(Gated Recurrent Unit, GRU)构成新的RNN单元SDZ-GRU。通过滚动式嵌套交叉验证实验,本文方法在均方根误差与平均绝对误差上比常规的GRU分别下降了7.68%和14.55%;另外由于SGMTFP方法加入了辅助特征,相比较不使用辅助特征的情况下,均方根误差与平均绝对误差分别下降了10.9%和15.1%,实验结果表明,本文方法能有效减小误差。

关键词: 短时交通流预测, SDZ, GRU, 辅助特征

Abstract: This paper proposes a multi-features short-term traffic flow prediction method based on SDZ-GRU(SGMTFP) to solve the problems that the current short-term traffic flow prediction methods have large errors and only rely on time-series data. This method adds a series of auxiliary data, such as time information, to the existing time-series data. Also, Surprisal-Driven Zoneout (SDZ) is applied to Gated Recurrent Unit (GRU) to make it a new RNN unit called SDZ-GRU. Through the rolling nested cross-validation experiment, the SGMTFP method proposed in this paper is 7.68% and 14.55% lower than the conventional GRU in terms of root-mean-square error and mean absolute error respectively. In addition, since the SGMTFP method adds auxiliary features, compared with the case without auxiliary features, the root-mean-square error and mean absolute error decrease by 10.9% and 15.1%, respectively. The experimental results show that this method can effectively reduce the error.

Key words: short-term traffic flow prediction, SDZ, GRU, auxiliary features

中图分类号: