Research on spatio-temporal network prediction model of parallel–series traffic flow based on Transformer and GCAT

Changfeng Zhu, Chunxiao Yu and Jiuyuan Huo

Physica A: Statistical Mechanics and its Applications, 2023, vol. 610, issue C

Abstract: Traffic flow forecasting is critical in transportation research. However, the excessive nonlinearity and complexity of spatial and temporal correlations in traffic flow critically restrict the prediction accuracy. To cope with the challenge, a parallel–series combined deep learning prediction model is proposed. Firstly, the traffic flow data is decomposed into unique time spans consistent with positive rules (weekly, daily, and hourly cycles). Then, a deep learning model named Transformer-Graph Convolutional Attention Networks (TRGCAT) is further used to predict the multi-flow in the traffic network. TRGCAT firstly encodes and concatenates the current hourly, daily, and weekly periodic features of all traffic nodes in parallel, which decodes the middle output with spatial features, and ultimately predicts future single-step, short-term, and long-term multi-step traffic flow by using Graph Convolutional Attention Network (GCAT) in series. We conduct numerical experiments on open-source datasets PeMS, the results show that TRGCAT does better in spatial and temporal fusion and may acquire extra aggressive forecasting consequences than baseline methods.

Keywords: Traffic flow prediction; Transformer; GCAT; Parallel–series combined model; Spatial and temporal fusion (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0378437122009724
Full text for ScienceDirect subscribers only. Journal offers the option of making the article available online on Science direct for a fee of $3,000

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:phsmap:v:610:y:2023:i:c:s0378437122009724

DOI: 10.1016/j.physa.2022.128414

Access Statistics for this article

Physica A: Statistical Mechanics and its Applications is currently edited by K. A. Dawson, J. O. Indekeu, H.E. Stanley and C. Tsallis

More articles in Physica A: Statistical Mechanics and its Applications from Elsevier
Bibliographic data for series maintained by Catherine Liu ().