Short-Term Highway Traffic Flow Prediction via Wavelet–Liquid Neural Network Model

Accurate, efficient, and reliable traffic flow prediction is pivotal for highway operation and management. However, traffic flow series present nonlinear, heterogeneous, and stochastic characteristics, posing significant challenges to precise prediction. To address this issue, this paper proposes a...

Full description

Saved in:
Bibliographic Details
Main Authors: Yongjun Wu, Hongyun Kang, Weipin Wang, Shuli Zhao, Xuening He, Jingyao Chen
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Modelling
Subjects:
Online Access:https://www.mdpi.com/2673-3951/6/2/39
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849705940182368256
author Yongjun Wu
Hongyun Kang
Weipin Wang
Shuli Zhao
Xuening He
Jingyao Chen
author_facet Yongjun Wu
Hongyun Kang
Weipin Wang
Shuli Zhao
Xuening He
Jingyao Chen
author_sort Yongjun Wu
collection DOAJ
description Accurate, efficient, and reliable traffic flow prediction is pivotal for highway operation and management. However, traffic flow series present nonlinear, heterogeneous, and stochastic characteristics, posing significant challenges to precise prediction. To address this issue, this paper proposes a novel wavelet-LNN model, integrating the strengths of wavelet decomposition and liquid neural networks (LNNs). Initially, multi-scale wavelet decomposition is applied to the original traffic flow data to yield approximation components and detailed components. Subsequently, each component is trained using the LNN. Ultimately, the predicted results of all components of the LNN models are aggregated to derive the final traffic flow prediction. The experiments conducted on four highway datasets demonstrate that the proposed wavelet-LNN model surpasses SVR, LSSVM, LSTM, TCN, and transformer models in prediction performance across R2, MSE, and MAE metrics. Notably, the wavelet-LNN model features the fewest parameters (<2% of typical deep learning models).
format Article
id doaj-art-ec2ce84f882d41e89adce6d86e062c07
institution DOAJ
issn 2673-3951
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Modelling
spelling doaj-art-ec2ce84f882d41e89adce6d86e062c072025-08-20T03:16:21ZengMDPI AGModelling2673-39512025-05-01623910.3390/modelling6020039Short-Term Highway Traffic Flow Prediction via Wavelet–Liquid Neural Network ModelYongjun Wu0Hongyun Kang1Weipin Wang2Shuli Zhao3Xuening He4Jingyao Chen5School of Traffic & Transportation, Chonqqing Jiaotong University, Chongqing 400074, ChinaSchool of Traffic & Transportation, Chonqqing Jiaotong University, Chongqing 400074, ChinaChongqing Expressway Network Management Co., Ltd., Chongqing 401120, ChinaChongqing Expressway Network Management Co., Ltd., Chongqing 401120, ChinaChongqing Expressway Network Management Co., Ltd., Chongqing 401120, ChinaChongqing Expressway Network Management Co., Ltd., Chongqing 401120, ChinaAccurate, efficient, and reliable traffic flow prediction is pivotal for highway operation and management. However, traffic flow series present nonlinear, heterogeneous, and stochastic characteristics, posing significant challenges to precise prediction. To address this issue, this paper proposes a novel wavelet-LNN model, integrating the strengths of wavelet decomposition and liquid neural networks (LNNs). Initially, multi-scale wavelet decomposition is applied to the original traffic flow data to yield approximation components and detailed components. Subsequently, each component is trained using the LNN. Ultimately, the predicted results of all components of the LNN models are aggregated to derive the final traffic flow prediction. The experiments conducted on four highway datasets demonstrate that the proposed wavelet-LNN model surpasses SVR, LSSVM, LSTM, TCN, and transformer models in prediction performance across R2, MSE, and MAE metrics. Notably, the wavelet-LNN model features the fewest parameters (<2% of typical deep learning models).https://www.mdpi.com/2673-3951/6/2/39traffic flow predictionwavelet decompositionliquid neural networkintelligent transportation systems
spellingShingle Yongjun Wu
Hongyun Kang
Weipin Wang
Shuli Zhao
Xuening He
Jingyao Chen
Short-Term Highway Traffic Flow Prediction via Wavelet–Liquid Neural Network Model
Modelling
traffic flow prediction
wavelet decomposition
liquid neural network
intelligent transportation systems
title Short-Term Highway Traffic Flow Prediction via Wavelet–Liquid Neural Network Model
title_full Short-Term Highway Traffic Flow Prediction via Wavelet–Liquid Neural Network Model
title_fullStr Short-Term Highway Traffic Flow Prediction via Wavelet–Liquid Neural Network Model
title_full_unstemmed Short-Term Highway Traffic Flow Prediction via Wavelet–Liquid Neural Network Model
title_short Short-Term Highway Traffic Flow Prediction via Wavelet–Liquid Neural Network Model
title_sort short term highway traffic flow prediction via wavelet liquid neural network model
topic traffic flow prediction
wavelet decomposition
liquid neural network
intelligent transportation systems
url https://www.mdpi.com/2673-3951/6/2/39
work_keys_str_mv AT yongjunwu shorttermhighwaytrafficflowpredictionviawaveletliquidneuralnetworkmodel
AT hongyunkang shorttermhighwaytrafficflowpredictionviawaveletliquidneuralnetworkmodel
AT weipinwang shorttermhighwaytrafficflowpredictionviawaveletliquidneuralnetworkmodel
AT shulizhao shorttermhighwaytrafficflowpredictionviawaveletliquidneuralnetworkmodel
AT xueninghe shorttermhighwaytrafficflowpredictionviawaveletliquidneuralnetworkmodel
AT jingyaochen shorttermhighwaytrafficflowpredictionviawaveletliquidneuralnetworkmodel