Aiming at the drawbacks of current traffic network flow prediction methods, such as low capability to explore the potential trend of spatial data and ignoring the long-term data correlation, this paper proposes a novel traffic flow prediction model based on MSCNN and attention mechanism. The model designs an RMSCNN module for considering the spatio-temporal dependence of long and short-term traffic flows, in which a residual block is introduced to alleviate the problem of excessive downsampling loss during feature extraction. In addition, a temporal multi-attention module based on the Transformer framework is used to highlight long time series features. To validate the method, traffic flow prediction experiments were conducted using the PEMS04 dataset. The results show that the RMSCNN-Trans prediction model reduces the Root Mean Square Error (RMSE) by at least 0.14, the Mean Absolute Error (MAE) by at least 1.33 and the Mean Absolute Percentage Error (MAPE) by at least 0.02 compared to several commonly used models.
ShenQ.Research on traffic flow prediction based on spatio-temporal fusion grey model. Jiangsu: Soochow University, 2021. (in Chinese)
4.
FengXLingXZhengH, et al. Adaptive multi-kernel SVM with spatial-temporal correlation for short-term traffic flow prediction. IEEE Trans Intell Transp Syst2018; 20(6): 2001–2013.
5.
LiYChaiSMaZ, et al. A hybrid deep learning framework for long-term traffic flow prediction. IEEE Access2021; 9: 11264–11271.
6.
ZhangWYuanSTaoJ, et al. Bi-LSTM highway traffic flow prediction based on multiple factors. Comput Syst Appl2021; 30(6): 184–190. (in Chinese)
7.
YuBYinHZhuZ. Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting. In: Proceedings of the 27th International Joint Conferences on Artificial Intelligence Organization (IJCAI), Stockholm, Sweden, 13–19 July 2018, pp.3634–3640. IJCAI.
8.
YuanHChenZ.A short-time traffic flow prediction algorithm based on temporal convolutional neural network. J South China Univ Technol (Nat Sci Ed)2020; 48(11): 107–113, 122. (in Chinese)
9.
DengHChenHXiaoQ, et al. Traffic flow prediction based on multi-attention dynamic graph convolutional network. J Taiyuan Univ Technol2024; 55(1): 172–183. (in Chinese)
10.
LiuSChenHChenX, et al. A two-branch spatio-temporal graphical convolutional neural network for traffic flow prediction. Inf Control2023; 52(3): 391–404. (in Chinese)
11.
NarmadhaSVijayakumarV.Spatio-temporal vehicle traffic flow prediction using multivariate CNN and LSTM model. Mater Today Proc2023; 81: 826–833.
12.
MaYChengSMaY, et al. A review of convolutional neural networks and their applications in intelligent transport systems. J Transp Eng2021; 21(4): 48–71.
13.
ZhangXJZhangGNZhangXL, et al. Medium and long-term traffic flow prediction based on ACNN-trans model. J China Univ Posts Telecomm2023; 30(4): 21–32.
HuangLZhuFLiZ. Trendformer: trend adaptive transformer for traffic flow prediction. In: 2022 5th international conference on data science and information technology (DSIT), Shanghai, China, 22–24 July 2022, pp.1–6. New York: IEEE.
16.
LiuQLiuCQiuZ, et al. A traffic flow prediction method based on time-aware transformer. Comput Sci2023; 50(11): 88–96. (in Chinese)
17.
HuangBDouHLuoY, et al. Adaptive spatiotemporal transformer graph network for traffic flow forecasting by IoT loop detectors. IEEE Internet Things J2023; 10(2): 1642–1653.
18.
MéndezMMerayoMGNúñezM.Long-term traffic flow forecasting using a hybrid CNN-BiLSTM model. Eng Appl Artif Intell2023; 121: 106041.
19.
ZhangQLiMXiaC, et al. Spatio-temporal residual graph convolutional network for short-term traffic flow prediction. IEEE Access2023; 11: 84187–84199.
20.
YinQDingFZhuY, et al. Urban traffic flow prediction model based on CNN and multi-scale feature fusion. Comput Technol Dev2022; 32(10): 175–181. (in Chinese)
21.
SzegedyCLiuWJiaY, et al. Going deeper with convolutions. In: Proceedings of the IEEE conference on computer vision and pattern recognition, Boston, MA, USA, 7–12 June 2015, pp.1–9. New York: IEEE.
22.
VaswaniAShazeerNParmarN, et al. Attention is all you need. Adv Neural Inf Process Syst2017; 30: 5998–6008.
23.
ZhaoZChenWWuX, et al. LSTM network: a deep learning approach for short-term traffic forecast. IET Intell Transp Syst2017; 11(2): 68–75.
24.
BaiLYaoLKanhereS, et al. STG2Seq: spatial-temporal graph to sequence model for multi-step passenger demand forecasting. In: Proceedings of the 28th International Joint Conferences on Artificial Intelligence Organization, Macao, China, 10–16 August 2019, pp.1981–1987. IJCAI
25.
WuZPanSLongG, et al. Graph wavenet for deep spatial-temporal graph modeling. In: Proceedings of the 28th International Joint Conferences on Artificial Intelligence Organization, Macao, China, 10–16 August 2019, pp.1907–1913. IJCAI
26.
ChoiJChoiHHwangJ, et al. Graph neural controlled differential equations for traffic forecasting. In: Proceedings of 36th AAAI Conference on Artificial Intelligence, 34th Conference on Innovative Applications of Artificial Intelligence, 12th Symposium on Educational Advances in Artificial Intelligence, 2022, pp.6367–6374. AAAI.
27.
WuHXuJWangJ, et al. Autoformer: decomposition transformers with auto-correlation for long-term series forecasting. Adv Neural Inf Process Syst2021; 34: 22419–22430.
28.
ZhouTMaZWenQ, et al. FEDformer: frequency enhanced decomposed transformer for long-term series forecasting. Proc Mach Learn Res2022; 162: 27268–27286.
