Please wait a minute...
Chin. Phys. B, 2024, Vol. 33(7): 078901    DOI: 10.1088/1674-1056/ad3349
INTERDISCIPLINARY PHYSICS AND RELATED AREAS OF SCIENCE AND TECHNOLOGY Prev  

WT-FCTGN: A wavelet-enhanced fully connected time-gated neural network for complex noisy traffic flow modeling

Zhifang Liao(廖志芳)1, Ke Sun(孙轲)1, Wenlong Liu(刘文龙)1, Zhiwu Yu(余志武)2,3, Chengguang Liu(刘承光)4,†, and Yucheng Song(宋禹成)1,‡
1 School of Computer Science and Engineering, Central South University, Changsha 410083, China;
2 National Engineering Research Center of High-speed Railway Construction Technology, Changsha 410075, China;
3 School of Civil Engineering, Central South University, Changsha 410075, China;
4 School of Traffic & Transportation Engineering, Central South University, Changsha 410083, China
Abstract  Accurate forecasting of traffic flow provides a powerful traffic decision-making basis for an intelligent transportation system. However, the traffic data's complexity and fluctuation, as well as the noise produced during collecting information and summarizing original data of traffic flow, cause large errors in the traffic flow forecasting results. This article suggests a solution to the above mentioned issues and proposes a fully connected time-gated neural network based on wavelet reconstruction (WT-FCTGN). To eliminate the potential noise and strengthen the potential traffic trend in the data, we adopt the methods of wavelet reconstruction and periodic data introduction to preprocess the data. The model introduces fully connected time-series blocks to model all the information including time sequence information and fluctuation information in the flow of traffic, and establishes the time gate block to comprehend the periodic characteristics of the flow of traffic and predict its flow. The performance of the WT-FCTGN model is validated on the public PeMS data set. The experimental results show that the WT-FCTGN model has higher accuracy, and its mean absolute error (MAE), mean absolute percentage error (MAPE) and root mean square error (RMSE) are obviously lower than those of the other algorithms. The robust experimental results prove that the WT-FCTGN model has good anti-noise ability.
Keywords:  traffic flow modeling      time-series      wavelet reconstruction  
Received:  08 January 2024      Revised:  26 February 2024      Accepted manuscript online:  13 March 2024
PACS:  89.75.Fb (Structures and organization in complex systems)  
  45.70.Vn (Granular models of complex systems; traffic flow)  
  05.45.Tp (Time series analysis)  
  95.75.Wx (Time series analysis, time variability)  
Fund: The Science and Technology Research and Development Program Project of China Railway Group Ltd provided funding for this study (Project Nos. 2020-Special-02 and 2021-Special-08).
Corresponding Authors:  Chengguang Liu, Yucheng Song     E-mail:  chengguangliu@csu.edu.cn;234703024@csu.edu.cn

Cite this article: 

Zhifang Liao(廖志芳), Ke Sun(孙轲), Wenlong Liu(刘文龙), Zhiwu Yu(余志武), Chengguang Liu(刘承光), and Yucheng Song(宋禹成) WT-FCTGN: A wavelet-enhanced fully connected time-gated neural network for complex noisy traffic flow modeling 2024 Chin. Phys. B 33 078901

[1] Guo K, Hu Y, Qian Z, Sun Y, Gao J and Yin B 2020 IEEE Trans. Intell. Transp. Syst. 23 1009
[2] Wei P, Cao Y and Sun D 2013 Transp. Res. Part B Methodol. 53 1
[3] Jiang W and Luo J 2022 Expert Syst. Appl. 207 117921
[4] Zheng C, Fan X, Wang C and Qi J 2020 Proceedings of the AAAI conference on artificial intelligence 34 1234
[5] Yin X, Wu G, Wei J, Shen Y, Qi H and Yin B 2021 IEEE Trans. Intell. Transp. Syst. 23 4927
[6] Liu Y, Liu Z and Jia R 2019 Transp. Res. Part C Emerg. Technol. 101 18
[7] Nai W, Liu L, Wang S and Dong D 2017 Algorithms 10 139
[8] Mousavizadeh Kashi S O and Akbarzadeh M 2019 J. Intell. Transp. Syst. 23 60
[9] Liu Y, Song Y, Zhang Y and Liao Z 2022 Phys. Stat. Mech. Its Appl. 603 127817
[10] Wang Y, Zhang D, Liu Y, Dai B and Lee L H 2019 Transp. Res. Part C Emerg. Technol. 99 144
[11] Wu Y, Tan H, Qin L, Ran B and Jiang Z 2018 Transp. Res. Part C Emerg. Technol. 90 166
[12] Peng Y and Xiang W 2020 Phys. Stat. Mech. Its Appl. 549 123913
[13] Dai G, Ma C and Xu X 2019 IEEE Access 7 143025
[14] Zhang H, Wang X, Cao J, Tang M and Guo Y 2018 Appl. Intell. 48 3827
[15] Vlahogianni E I, Golias J C and Karlaftis M G 2004 Transp. Rev. 24 533
[16] Van Lint J W C and Van Hinsbergen C 2012 Artif. Intell. Appl. Crit. Transp. Issues 22 22
[17] Dudek G 2016 Electr. Power Syst. Res. 130 139
[18] Van Der Voort M, Dougherty M and Watson S 1996 Transp. Res. Part C Emerg. Technol. 4 307
[19] Sun H, Zhang C and Ran B 2004 The 7th International IEEE Conference on Intelligent Transportation Systems (IEEE Cat. No. 04TH8749) (IEEE) p. 410
[20] Okutani I and Stephanedes Y J 1984 Transp. Res. Part B Methodol. 18 1
[21] Ojeda L L, Kibangou A Y and De Wit C C 2013 2013 American Control Conference (IEEE) p. 4724
[22] Silver D, Huang A, Maddison C J, Guez A, Sifre L, Van Den Driessche G, Schrittwieser J, Antonoglou I, Panneershelvam V and Lanctot M 2016 Nature 529 484
[23] Park D and Rilett L R 1999 Comput. Civ. Infrastruct. Eng. 14 357
[24] Huang W, Song G, Hong H and Xie K 2014 IEEE Trans. Intell. Transp. Syst. 15 2191
[25] Siłka J, Wieczorek M and Woźniak M 2022 Neural Comput. Appl. 34 13305
[26] Dey R and Salem F M 2017 2017 IEEE 60th international midwest symposium on circuits and systems (MWSCAS) (IEEE) p. 1597
[27] Cho K, Van Merriënboer B, Gulçehre Ç, Bahdanau D, Bougares F, Schwenk H and Bengio Y 2014 Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) p. 1724
[28] Sherstinsky A 2020 Phys. Nonlinear Phenom. 404 132306
[29] Hora S K, Poongodan R, de Prado R P, Wozniak M and Divakarachari P B 2021 Appl. Sci. 11 11263
[30] Graves A and Graves A 2012 Supervised Seq. Label. Recurr. Neural Netw. pp. 37-45
[31] Fu R, Zhang Z and Li L 2016 2016 31st Youth Academic Annual Conference of Chinese Association of Automation (YAC) (IEEE) p. 324
[32] Jin Y, Xu W, Wang P and Yan J 2018 2018 5th International Conference on Information, Cybernetics, and Computational Social Systems (ICCSS) (IEEE) p. 241
[33] Zhang J, Zheng Y and Qi D 2017 AAAI Conf. Artif. Intell. 31 10735
[34] Liu Y, Zheng H, Feng X and Chen Z 2017 2017 9th International Conference on Wireless Communications and Signal Processing (WCSP) (IEEE) pp. 1-6
[35] Ke J, Zheng H, Yang H and Chen X M 2017 Transp. Res. Part C Emerg. Technol. 85 591
[36] Yu H, Wu Z, Wang S, Wang Y and Ma X 2017 Sensors 17 1501
[37] Li Y, Yu R, Shahabi C and Liu Y 2017 ArXiv Prepr. ArXiv170701926
[38] Diao Z, Wang X, Zhang D, Liu Y, Xie K and He S 2019 Proceedings of the AAAI conference on artificial intelligence 33 890
[39] Oreshkin B N, Carpov D, Chapados N and Bengio Y 2019 ArXiv Prepr. ArXiv190510437
[40] Tang J, Chen X, Hu Z, Zong F, Han C and Li L 2019 Phys. Stat. Mech. Its Appl. 534 120642
[41] Kim T, Kim J, Tae Y, Park C, Choi J H and Choo J 2021 International Conference on Learning Representations
[1] Deep learning framework for time series classification based on multiple imaging and hybrid quantum neural networks
Jianshe Xie(谢建设) and Yumin Dong(董玉民). Chin. Phys. B, 2023, 32(12): 120302.
[2] GEKF, GUKF and GGPF based prediction of chaotic time-series with additive and multiplicative noises
Wu Xue-Dong(伍雪冬) and Song Zhi-Huan(宋执环). Chin. Phys. B, 2008, 17(9): 3241-3246.
No Suggested Reading articles found!