摘要In order to further fully extract the spatio-temporal and periodic characteristics of traffic flow, this paper adopts a combined model (DLSTM-AE) of Autoencoder architecture and Deep Long Short-Term Memory neural networks, and introduces rectified adaptive moment estimation (RAdam) algorithm for model training. Firstly, the characteristics of traffic flow sequence information are collected by using the deep long short-term memory network model, and the collected information is compressed into a fixed dimension representation vector with the help of the automatic encoder structure. Then, the vector is reconstructed by the decoder to realize further information mining. Finally, in the process of model training, the RAdam algorithm is used to optimize and update the momentum parameters in batches, so as to shorten the time to find the optimal solution. This can improve the timeliness and accuracy of model prediction. Simulate on real data sets of highway traffic flow and compare with other model methods, DLSTM-AE combined model not only has obvious advantages in prediction results, but also has good curve fitting ability in traffic flow periodicity. The root mean square error (RMSE) value of DLSTM-AE combined model on the test set decreased by about 0.445 to 1.826, the average absolute error (MAE) value decreased by about 0.282 to 0.984, and the coefficient of determination (R2) value increased by about 0.005 to 0.023. In terms of periodicity, the prediction accuracy of working days corresponding to adjacent weeks is much higher than that of the control group. Simulation results show that the model can capture more potential spatio-temporal and periodic information in traffic flow sequence. It can better meet the needs of highway traffic flow prediction.
Abstract：In order to further fully extract the spatio-temporal and periodic characteristics of traffic flow, this paper adopts a combined model (DLSTM-AE) of Autoencoder architecture and Deep Long Short-Term Memory neural networks, and introduces rectified adaptive moment estimation (RAdam) algorithm for model training. Firstly, the characteristics of traffic flow sequence information are collected by using the deep long short-term memory network model, and the collected information is compressed into a fixed dimension representation vector with the help of the automatic encoder structure. Then, the vector is reconstructed by the decoder to realize further information mining. Finally, in the process of model training, the RAdam algorithm is used to optimize and update the momentum parameters in batches, so as to shorten the time to find the optimal solution. This can improve the timeliness and accuracy of model prediction. Simulate on real data sets of highway traffic flow and compare with other model methods, DLSTM-AE combined model not only has obvious advantages in prediction results, but also has good curve fitting ability in traffic flow periodicity. The root mean square error (RMSE) value of DLSTM-AE combined model on the test set decreased by about 0.445 to 1.826, the average absolute error (MAE) value decreased by about 0.282 to 0.984, and the coefficient of determination (R2) value increased by about 0.005 to 0.023. In terms of periodicity, the prediction accuracy of working days corresponding to adjacent weeks is much higher than that of the control group. Simulation results show that the model can capture more potential spatio-temporal and periodic information in traffic flow sequence. It can better meet the needs of highway traffic flow prediction.
黄艳国, 周陈聪, 左可飞. Traffic Flow Prediction of DLSTM-AE Model Based on RAdam Optimization[J]. Journal of Highway and Transportation Research and Development, 2023, 17(1): 92-100.
HUANG Yan-guo, ZHOU Chen-cong, ZUO Ke-fei. Traffic Flow Prediction of DLSTM-AE Model Based on RAdam Optimization. Journal of Highway and Transportation Research and Development, 2023, 17(1): 92-100.
[1] Office of Safety Commission of Ministry of Transport. Opinions on Further Strengthening Construction of Transport Safety Production System (Draft for Comments)[EB/OL]. (2021-10-27) [2021-12-11]. (in Chinese) [2] LIN Y X, WANG P, MA M. Intelligent Transportation System (ITS): Concept, Challenge and Opportunity[C]// 2017 IEEE 3rd International Conference on Big Data Security on Cloud (Big Data Security). Beijing: IEEE, 2017. [3] LIU Jing, GUAN Wei. A Summary of Traffic Flow Forecasting Methods[J]. Journal of Highway and Transportation Research and Development, 2014, 21(3): 82- 85. (in Chinese) [4] HAN Chao, SONG Su, WANG Cheng-hong. A Real-time Short-term Traffic Flow Adaptive Forecasting Method Based on ARIMA Model [J]. Journal of System Simulation,2004,16(7):1530-1532,1535. (in Chinese) [5] WILLIAMS B M, HOEL L A. Modeling and Forecasting Vehicular Traffic Flow as a Seasonal ARIMA Process: Theoretical Basis and Empirical Results[J]. Journal of Transportation Engineering, 2003, 129(6): 664-672. [6] SMOLA A J, SCHÖLKOPF B. A Tutorial on Support Vector Regression[J]. Statistics and Computing, 2004, 14(3): 199-222. [7] CASTRO-NETO M, JEONG Y S, JEONG M K, et al. Online-SVR for Short-term Traffic Flow Prediction under Typical and Atypical Traffic Conditions[J]. Expert Systems with Applications, 2009, 36(3P2):6164-6173. [8] LIU Y Z, SHAO X J, LI X H. Short-term Traffic Flow Prediction Model Based on Lagrange Support Vector Regression[J]. Journal of Transport Information and Safety, 2007, 25(5): 46-49. [9] WANG Xiang-xue, XU Lun-hui. Short-term Traffic Flow Prediction Based on Deep Learning[J]. Journal of Transportation Systems Engineering and Information Technology, 2018,18(1):81-88. (in Chinese) [10] FU R, ZHANG Z, LI L. Using LSTM and GRU Neural Network Methods for Traffic Flow Prediction[C]//2016 31st Youth Academic Annual Conference of Chinese Association of Automation (YAC). Wuhan: IEEE, 2016: 324-328. [11] LV Y, DUAN Y, KANG W, et al. Traffic Flow Prediction with Big Data: A Deep Learning Approach[J]. IEEE Transactions on Intelligent Transportation Systems, 2015, 16(2): 865-873. [12] LECUN Y, BENGIO Y, HINTON G. Deep Learning[J]. Nature, 2015, 521: 436-444. [13] SAGHEER A, KOTB M. Time Series Forecasting of Petroleum Production Using Deep LSTM Recurrent Networks[J]. Neurocomputing, 2018, 323: 203-213. [14] SAGHEER A, KOTB M. Unsupervised Pre-training of A Deep LSTM-Based Stacked Autoencoder for Multivariate Time Series Forecasting Problems[J]. Scientific Reports, 2019, 9(1): 1-16. [15] LIU L, JIANG H, HE P, et al. On the Variance of the Adaptive Learning Rate and Beyond[EB/OL]. (2020-02-25) [2021-11-11]. https://arxiv.org/pdf/1908.03265.pdf. [16] LI Lei, ZHANG Qing-miao, ZHAO Jun-hui, et al. Short-Term Traffic Flow Prediction Method of Different Periods Based on Improved CNN-LSTM[J]. Journal of Applied Sciences,2021,39(2):185-198. (in Chinese) [17] LI Tao-ying, WANG Ting, ZHANG Yu-qi. Highway Traffic Flow Prediction Model with Multi-features[J]. Journal of Transportation Systems Engineering and Information Technology,2021,21(3):101-111. (in Chinese) [18] CHEN Meng, GAN Ke, LI Kai, et al. A Spatial-temporal Graph Convolutional Network Model for Accurate and Robust Traffic Flow Prediction Based on Real-time Multimodal Spatial-temporal Data[J]. Journal of Highway and Transportation Research and Development, 2021, 38(8):134-139,158. (in Chinese) [19] POLSON N G, SOKOLOV V O. Deep Learning for Short-term Traffic Flow Prediction[J]. Transportation Research Part C: Emerging Technologies, 2017, 79: 1-17. [20] TSCHANNEN M, BACHEM O, LUCIC M. Recent Advances in Autoencoder-based Representation Learning[EB/OL]. (2018-12-12)[ 2021-12-11]. https://arxiv.org/abs/1812.05069. [21] WANG, X L, GUAN, X F, CAO J, et al. Forecast Network-wide Traffic States for Multiple Steps Ahead: A Deep Learning Approach Considering Dynamic Non-local Spatial Correlation and Non-stationary Temporal Dependency[J]. Transportation Research Part C:Emerging Technologies,2020,119:102763. [22] MA L, QU S. A Sequence to Sequence Learning Based Car-following Model for Multi-step Predictions Considering Reaction Delay[J]. Transportation Research Part C:Emerging Technologies,2020,120:102785. [23] MAYA S, UENO K, NISHIKAWA T. DLSTM: A New Approach for Anomaly Detection Using Deep Learning with Delayed Prediction[J]. International Journal of Data Science and Analytics, 2019, 8(2): 137-164.
2023, Vol. 17 
