Traffic Congestion Prediction Based on Dynamic Adaptive Gated Graph Convolutional Networks

doi:10.12141/j.issn.1000-565X.250003

Abstract

Abstract:

With the continual rise in the number of motor vehicles in urban areas, traffic congestion has become increasingly severe, adversely affecting environmental protection and urban operational efficiency. Consequently, it is of critical importance to accurately predict traffic congestion for traffic management and optimization. However, existing research still faces limitations in modeling the dynamic, time-varying characteristics of traffic flow and the complex interactions among road segments. To address these challenges, a gated spatiotemporal convolutional network model based on graph neural networks was proposed to more effectively capture and predict traffic congestion. Firstly, an improved K-means clustering algorithm was employed to divide the raw data into multiple congestion-state categories, which are then incorporated as auxiliary features to enhance feature representation. Next, a gated temporal convolutional network was introduced to capture the temporal properties and dynamic dependencies in traffic data, and a dynamic adaptive gated graph convolutional network was constructed to achieve feature fusion and dynamic weight allocation through a signal generation module and a dual-modulation mechanism, thereby facilitating effective extraction of spatiotemporal features. Finally, residual connections were incorporated to improve training stability, and skip connections were utilized to integrate multi-level and multi-scale features. Experimental results on real-world PeMS08 and PeMS04 datasets demonstrate that the proposed model achieves superior prediction accuracy compared with other baseline methods.

Key words: traffic congestion prediction, graph neural network, dynamic adaptive gating, clustering algorithm, gated temporal convolutional network

CLC Number:

TP39

WANG Qingrong, GAO Huanyi, ZHU Changfeng, HE Runtian, MU Zhuangzhuang. Traffic Congestion Prediction Based on Dynamic Adaptive Gated Graph Convolutional Networks[J]. Journal of South China University of Technology(Natural Science Edition), 2025, 53(9): 31-47.

Figures/Tables 24

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Fig.10

Fig.11

Fig.12

Fig.13

Fig.14

Fig.15

Fig.16

Fig.17

Fig.18

Fig.19

Table 1

Table 2

Fig.20

Fig.21

Table 3

References 29

[1]	LI L， DAI F， HUANG B，et al ．ST3DRNet：attention-based spatio-temporal 3D residual neural networks for traffic congestion prediction［J］．Sensors，2024，24（4）：1261/1-19.
[2]	SHAHRIARI S， GHASRI M， SISSON S A，et al ．Ensemble of ARIMA：combining parametric and bootstrapping technique for traffic flow prediction［J］．Transportmetrica A：Transport Science，2020，16（3）：1552-1573.
[3]	PRIAMBODO B， AHMAD A， KADIR R A ．Predicting traffic flow propagation based on congestion at neighbouring roads using hidden Markov model［J］．IEEE Access，2021，9： 85933-85946.
[4]	ZHANG Z， HAN Y， PENG T，et al ．A comprehensive spatio-temporal model for subway passenger flow prediction［J］．ISPRS International Journal of Geo-Information，2022，11（6）：341-341.
[5]	ZHANG H， JIANG C， YANG L ．Forecasting traffic congestion status in terminal areas based on support vector machine［J］．Advances in Mechanical Engineering，2016，8（9）：1-11.
[6]	YU H， JI N， REN Y，et al ．A special event-based K-nearest neighbor model for short-term traffic state prediction［J］．IEEE Access，2019，7：81717-81729.
[7]	姚俊峰，何瑞，史童童，等．基于机器学习的交通流预测方法综述［J］．交通运输工程学报，2023，23（3）：44-67.
	YAO Jun-feng， HE Rui， SHI Tong-tong，et al ．Review on machine learning-based traffic flow prediction methods［J］．Journal of Traffic and Transportation Engineering，2023，23（3）：44-67.
[8]	梁秀霞，夏曼曼，何月阳，等．基于时空多头图注意力网络的交通流预测［J］．电子学报，2024，52（2）：500-509.
	LIANG Xiu-xia， XIA Man-man， HE Yue-yang，et al ．Traffic flow prediction based on spatio-temporal multi-head graph attention network［J］．Acta Electronica Sinica，2024，52（2）：500-509.
[9]	YU B， YIN H， ZHU Z ．Spatio-temporal graph convolutional networks：a deep learning framework for traffic forecasting ［C］∥ Proceedings of the 27th International Joint Conference on Artificial Intelligence．Stockholm：International Joint Conferences on Artificial Intelligence Organization，2018：3634-3640.
[10]	LI Y， YU R， SHAHABI C，et al ．Diffusion convolutional recurrent neural network：data-driven traffic forecasting［C］∥ Proceedings of the 6th International Conference on Learning Representations．Vancouver：OpenReview.net，2018：1-10.
[11]	WU Z， PAN S， LONG G，et al ．Graph WaveNet for deep spatial temporal graph modeling［C］∥ Proceedings of the 28th International Joint Conference on Artificial Intelligence．Macao：ACM，2019：1907-1913.
[12]	贺文武，裴博彧，李雅婷，等．基于双向自适应门控图卷积网络的交通流预测［J］．交通运输系统工程与信息，2023，23（1）：187-197.
	HE Wen-wu， PEI Bo-yu， LI Ya-ting，et al ．Traffic flow forecasting based on bi-directional adaptive gating graph convolutional networks［J］．Journal of Transportation Systems Engineering and Information Technology，2023，23（1）：187-197.
[13]	崔建勋，要甲，赵泊媛．基于深度学习的短期交通流预测方法综述［J］．交通运输工程学报，2024，24（2）：50-64.
	CUI Jian-xun， YAO Jia， ZHAO Bo-yuan ．Review on short-term traffic flow prediction methods based on deep learning［J］．Journal of Traffic and Transportation Engineering，2024，24（2）：50-64.
[14]	WANG X， ZENG R， ZOU F，et al ．STTF：an efficient transformer model for traffic congestion prediction［J］．International Journal of Computational Intelligence Systems，2023，16（1）：1-16.
[15]	张玺君，余光杰，崔勇，等．基于聚类算法和图神经网络的短时交通流预测［J］．吉林大学学报（工学版），2024，54（6）：1593-1600.
	ZHANG Xi-jun， YU Guang-jie， CUI Yong，et al ．Short-term traffic flow prediction based on clustering algorithm and graph neural network［J］．Journal of Jilin University（Engineering and Technology Edition），2024，54（6）：1593-1600.
[16]	BAI S， KOLTER J Z， KOLTUN V ．An empirical evaluation of generic convolutional and recurrent networks for sequence modeling［EB/OL］．（2018-03-04）［2024-11-06］．.
[17]	HU Z， SHAO F， SUN R ．A new perspective on traffic flow prediction：a graph spatial-temporal network with complex network information［J］．Electronics，2022，11：2432/1-22.
[18]	GUO X， KONG X， XING W，et al ．Adaptive graph generation based on generalized pagerank graph neural network for traffic flow forecasting［J］．Applied Intelligence，2023，53（24）：30971- 30986.
[19]	Caltrans.Caltrans performance measurement system ［EB/OL］．（2024-02-25）［2025-03-17］．.
[20]	WANG Q， CHEN X， ZHU C，et al ．Short-term traffic flow prediction based on the SGA-KGCN-LSTM model［J］．Engineering Letters，2023，31（3）：1221-1235.
[21]	SMITH B L， WILLIAMS B M， OSWALD R K ．Comparison of parametric and nonparametric models for traffic flow forecasting［J］．Transportation Research Part C：Emerging Technologies，2002，10（4）：303-321.
[22]	LIN G， LIN A， GU D ．Using support vector regression and K-nearest neighbors for short-term traffic flow prediction based on maximal information coefficient［J］．Information Sciences，2022， 608：517-531.
[23]	龚循强，邱万锦，吕开云，等．变分模态分解与自适应图卷积门控循环网络的交通流量组合预测模型［J］．武汉大学学报（信息科学版），2024，49（12）：2329-2341.
	GONG Xunqiang， QIU Wanjin， Kaiyun LÜ，et al ．A combined traffic flow prediction model based on variational mode decomposition and adaptive graph convolutional gated recurrent network［J］．Geomatics and Information Science of Wuhan University，2024，49（12）：2329-2341.
[24]	王庆荣，慕壮壮，朱昌锋，等．基于HDNNF-CAF的短时交通流预测研究［J］．计算机工程与应用，2025，61（15）： 318-328.
	WANG Qingrong， MU Zhuangzhuang， ZHU Changfeng，et al ．Research on short-term traffic flow prediction based on HDNNF-CAF［J］．Computer Engineering and Applications，2025，61（15）：318-328.
[25]	MIRJALILI S， MIRJALILI S M， LEWIS A ．Grey wolf optimizer［J］．Advances in Engineering Software，2014，69：46-61.
[26]	薛建凯．一种新型的群智能优化技术的研究与应用［D］．上海：东华大学，2020.
[27]	ARTHUR D， VASSILVITSKII S ．K-means++：the advantages of careful seeding［C］∥ Proceedings of the Eighteenth Annual ACM-SIAM Symposium on Discrete Algorithms．New Orleans：ACM，2007：1027-1035.
[28]	DAVIES D L， BOULDIN D W ．A cluster separation measure［J］．IEEE Transactions on Pattern Analysis and Machine Intelligence，1979，1（2）：224-227.
[29]	CALINSKI T， HARABASZ J ．A dendrite method for cluster analysis［J］．Communications in Statistics：Theory and Methods，1974，3（1）：1-27.

时长/min	模型	RMSE/10^-2	MAPE/%	MAE/10^-2	R²
15	HA	10.490	4.932	5.981	0.321
	SVR	8.593	4.780	5.399	0.698
	DCRNN	6.303	2.810	3.198	0.730
	STGCN	6.190	2.802	3.014	0.758
	Graph WaveNet	5.838	2.229	2.723	0.774
	STTF	4.452	1.421	1.893	0.895
	VMD-AGCGRN	4.053	1.140	1.372	0.907
	GSTCN	3.739	0.933	1.195	0.914
30	HA	10.586	4.930	5.993	0.320
	SVR	9.601	4.843	5.547	0.638
	DCRNN	6.946	3.343	3.611	0.683
	STGCN	6.874	3.228	3.448	0.692
	Graph WaveNet	6.641	2.572	3.152	0.725
	STTF	4.916	1.782	2.127	0.852
	VMD-AGCGRN	4.751	1.452	1.918	0.866
	GSTCN	4.476	1.175	1.501	0.874
60	HA	10.675	4.926	6.011	0.317
	SVR	10.335	5.761	6.288	0.561
	DCRNN	7.849	3.414	3.804	0.659
	STGCN	8.021	3.542	3.967	0.647
	Graph WaveNet	7.494	3.034	3.709	0.678
	STTF	5.544	2.234	2.489	0.803
	VMD-AGCGRN	5.387	1.975	2.247	0.814
	GSTCN	5.286	1.579	1.994	0.825

时长/min	模型	RMSE/10^-2	MAPE/%	MAE/10^-2	R²
15	HA	11.762	5.701	6.934	0.231
	SVR	9.573	4.780	5.399	0.573
	DCRNN	7.872	3.256	3.821	0.705
	STGCN	7.983	3.438	3.998	0.691
	Graph WaveNet	7.618	3.048	3.787	0.718
	STTF	5.069	1.618	2.147	0.853
	VMD-AGCGRN	4.677	1.426	1.635	0.872
	GSTCN	4.472	1.115	1.448	0.891
30	HA	11.767	5.702	6.936	0.231
	SVR	10.496	4.314	4.966	0.558
	DCRNN	8.668	3.774	4.372	0.616
	STGCN	8.731	3.996	4.416	0.602
	Graph WaveNet	8.201	3.378	4.194	0.638
	STTF	5.793	1.849	2.454	0.819
	VMD-AGCGRN	5.583	1.650	2.091	0.824
	GSTCN	5.224	1.424	1.839	0.849
60	HA	11.774	5.704	6.939	0.232
	SVR	11.340	4.964	5.726	0.486
	DCRNN	9.188	4.019	4.458	0.588
	STGCN	9.364	4.187	4.683	0.579
	Graph WaveNet	8.640	3.734	4.612	0.605
	STTF	6.533	2.079	2.735	0.770
	VMD-AGCGRN	6.364	1.991	2.572	0.787
	GSTCN	6.054	1.818	2.330	0.802

模型	RMSE/10^-2	MAPE/%	MAE/10^-2
GSTCN（基准模型）	3.739	0.933	1.195
移除DAGGCN	4.168	1.301	1.428
移除注意力	3.846	0.985	1.231
移除自适应GCN	3.935	1.003	1.280
替换DAGGCN为GCN	4.054	1.127	1.398
移除辅助特征	3.794	0.967	1.233