Short-time traffic flow prediction based on a combined GAT-VMD-LightGBM prediction model

Chuanxiang Ren; Pengfei Jia

doi:10.1117/12.2668496

16 February 2023 Short-time traffic flow prediction based on a combined GAT-VMD-LightGBM prediction model

Chuanxiang Ren, Pengfei Jia

Proceedings Volume 12591, Sixth International Conference on Traffic Engineering and Transportation System (ICTETS 2022); 125913L (2023) https://doi.org/10.1117/12.2668496
Event: 6th International Conference on Traffic Engineering and Transportation System (ICTETS 2022), 2022, Guangzhou, China

Abstract

As the core part of intelligent transportation system, traffic flow prediction is vulnerable to the influence of original signal noise, modal aliasing and other factors, and most of the current traffic flow prediction models cannot take into account the various characteristics of traffic flow, so their stability is poor. In order to improve the accuracy and stability of short-term traffic flow prediction, a combined prediction model based on graph attention network, variational modal decomposition and lightweight gradient elevator is proposed. First, the model uses the mutual information entropy algorithm to process the spatial correlation of road segments and filters out the historical data of the road segments with the highest spatial correlation with the target road segments. The spatial structure between the filtered data and the road segments is used as the input of the graph attention network to obtain the spatial features contained in the data. At the same time, the filtered data is decomposed by variational mode, the arrangement entropy of each component is calculated, the validity of the component is verified, the invalid component is de noised and decomposed again, and the spatial features obtained before are combined with the effective component to form a new time series. Finally, each new time series is input into the LightGBM prediction model as an input, and the predicted values of all new series are superimposed to form the final short-term traffic flow prediction result. The pems04 data set was used as experimental data for verification. The results show that compared with the comparison model, the MAE of GAT-VMD-LightGBM prediction model is reduced by 47.07% and RMSE is reduced by 52.5%, which can prove that the prediction performance of this model is better.

Citation Download Citation

Chuanxiang Ren and Pengfei Jia "Short-time traffic flow prediction based on a combined GAT-VMD-LightGBM prediction model", Proc. SPIE 12591, Sixth International Conference on Traffic Engineering and Transportation System (ICTETS 2022), 125913L (16 February 2023); https://doi.org/10.1117/12.2668496

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
5 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Data modeling

Modal decomposition

Performance modeling

Autoregressive models

Data acquisition

Education and training

Statistical modeling

Show All Keywords

Keywords/Phrases

Search In:

Publication Years