改进膨胀时空图卷积网络的短时交通流预测

doi:10.19678/j.issn.1000-3428.0069765

摘要/Abstract

摘要：

路网交通流预测在智能交通领域起着关键性作用, 交通流不仅具有高度的空间相关性, 同时在时间特征上也存在时间相关性和周期性。现有的时空交通流量预测在时间特征提取方面更多关注交通流的局部时间特征。针对上述问题, 提出一种改进膨胀时空图卷积网络(IDTS-GCN)模型, 以改进的图卷积网络(GCN)为基础提取空间特征, 将膨胀卷积的顺序操作改为并行操作后将膨胀序列嵌入双向长短期记忆(Bi-LSTM)中提取交通流短期局部与宏观长期时间特征, 低膨胀率的序列提取短期局部时间特征, 高膨胀率的序列提取长期宏观时间特征, 在此基础上添加残差连接融合时空特征得到最终预测结果。为了验证IDTS-GCN模型的有效性, 在PeMS04和PeMS08数据集上进行测试, 结果表明, IDTS-GCN模型在两种数据集下相较STSGCN时空联合学习模型, 平均绝对误差(MAE)、平均绝对百分比误差(MAPE)和均方根误差(RMSE)平均下降了4.917%、3.371%、6.079%和6.291%、5.842%、4.395%。

关键词: 智能运输系统, 交通流预测, 时空特征, 图信号处理, 双向长短期记忆

Abstract:

Traffic flow prediction for road networks plays a key role in intelligent transportation. Traffic flow not only exhibits high spatial correlation but also exhibits long-term correlation and periodicity in time characteristics. Existing spatio-temporal traffic flow prediction models focus more on the local time characteristics of traffic flows through time feature extraction. To address this issue, this study proposes an Improved Dilated Temporal-Spatial Graph Convolution Network (IDTS-GCN) model to extract spatial features based on an improved Graph Convolution Network (GCN). The sequential operation of dilated convolution is changed to a parallel operation, and the dilated series is embedded into Bi-directional Long Short-Term Memory (Bi-LSTM) to extract short-term local and macro-long-term time features of traffic flow. Short-term local temporal features are extracted from sequences with low dilated rates, and long-term macro-temporal features are extracted from sequences with high dilated rates. Accordingly, residual connections are added to fuse the spatio-temporal features to obtain the final prediction. In evaluations on the PeMS04 and PeMS08 datasets, IDTS-GCN reduces the Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Root Mean Square Error (RMSE) by 4.917%, 3.371%, and 6.079% and 6.291%, 5.842%, and 4.395%, respectively, compared with that achieved by the STSGCN spatiotemporal joint learning model.

Key words: Intelligent Transportation Systems (ITS), traffic flow prediction, spatial-temporal features, Graph Signal Processing (GSP), Bi-directional Long Short-Term Memory (Bi-LSTM)

罗向龙, 徐忠承, 苏勇东, 何西槟, 刘若辰. 改进膨胀时空图卷积网络的短时交通流预测[J]. 计算机工程, 2025, 51(12): 346-356.

LUO Xianglong, XU Zhongcheng, SU Yongdong, HE Xibin, LIU Ruochen. Short-Term Traffic Flow Prediction Based on Improved Dilated Temporal-Spatio Graph Convolutional Network[J]. Computer Engineering, 2025, 51(12): 346-356.

https://www.ecice06.com/CN/Y2025/V51/I12/346

图/表 16

图1 路网图描述示例

Fig.1 Examples of road network diagram descriptions

图2 IDTS-GCN模型框架

Fig.2 IDTS-GCN model framework

图3 改进空间特征提取模块

Fig.3 Improved spatial feature extraction module

图4 一维膨胀卷积

Fig.4 One-dimensional dilated convolution

图5 动态膨胀时间特征提取模块

Fig.5 Dynamic dilated time feature extraction module

图6 在PeMS04数据集上IDTS-GCN预测结果

Fig.6 Prediction results of IDTS-GCN on the PeMS04 dataset

图7 在PeMS08数据集上IDTS-GCN预测结果

Fig.7 Prediction results of IDTS-GCN on the PeMS08 dataset

图8 在PeMS04数据集上不同模型预测结果

Fig.8 Prediction results of different models on the PeMS04 dataset

图9 在PeMS08数据集上不同模型预测结果

Fig.9 Prediction results of different models on the PeMS08 dataset

图10 多步预测性能比较

Fig.10 Multi-step predictive performance comparison

图11 IDTS-GCN模型训练误差曲线

Fig.11 IDTS-GCN model training error curve

参考文献 36

1	ZHU L , YU F R , WANG Y G , et al. Big data analytics in intelligent transportation systems: a survey. IEEE Transactions on Intelligent Transportation Systems, 2019, 20 (1): 383- 398. doi: 10.1109/TITS.2018.2815678
2	ATLURI G , KARPATNE A , KUMAR V . Spatio-temporal data mining. ACM Computing Surveys, 2019, 51 (4): 1- 41.
3	ZHAO J H , NI S J , YANG L H , et al. Multiband cooperation for 5G HetNets: a promising network paradigm. IEEE Vehicular Technology Magazine, 2019, 14 (4): 85- 93. doi: 10.1109/MVT.2019.2935793
4	YUAN H T , LI G L . A survey of traffic prediction: from spatio-temporal data to intelligent transportation. Data Science and Engineering, 2021, 6 (1): 63- 85. doi: 10.1007/s41019-020-00151-z
5	XIE P , LI T R , LIU J , et al. Urban flow prediction from spatiotemporal data using machine learning: a survey. Information Fusion, 2020, 59, 1- 12. doi: 10.1016/j.inffus.2020.01.002
6	KOESDWIADY A , SOUA R , KARRAY F . Improving traffic flow prediction with weather information in connected cars: a deep learning approach. IEEE Transactions on Vehicular Technology, 2016, 65 (12): 9508- 9517. doi: 10.1109/TVT.2016.2585575
7	周晓, 唐宇舟, 刘强. 基于卡尔曼滤波的道路平均速度预测模型研究. 浙江工业大学学报, 2020, 48 (4): 392-396, 404.
	ZHOU X , TANG Y Z , LIU Q . Research on road average speed prediction model based on Kalman filter. Journal of Zhejiang University of Technology, 2020, 48 (4): 392-396, 404.
8	杨立宁, 李艳婷. 基于SVD和ARIMA的时空序列分解与预测. 计算机工程, 2021, 47 (3): 53- 61. doi: 10.19678/j.issn.1000-3428.0056885
	YANG L N , LI Y T . Spatio-temporal sequence decomposition and prediction based on SVD and ARIMA. Computer Engineering, 2021, 47 (3): 53- 61. doi: 10.19678/j.issn.1000-3428.0056885
9	KUMAR S V . Traffic flow prediction using Kalman filtering technique. Procedia Engineering, 2017, 187, 582- 587. doi: 10.1016/j.proeng.2017.04.417
10	LINT H , HINSBERGEN C V . Short-term traffic and travel time prediction models. Artificial Intelligence Applications to Critical Transportation Issues, 2012, 22 (1): 22- 41.
11	SUMALATHA V, DINGARI M, JAYALAKSHMI C. Short term road traffic flow forecasting using multi layer perceptron neural networks[C]//Proceedings of ICMSA'19. Hyderabad, India: American Institute of Physics Publishing, 2020: 311-325.
12	DEKEL O , SINGER Y . Support vector machines on a budget. Neural Information Processing Systems, 1997, 28 (7): 779- 784.
13	YANG Y J, DUAN Z Y. A novel prediction method of traffic flow: least squares support vector machines based on spatial relation[C]//Proceedings of International Conference of Transportation Professionals. Washington D.C., USA: IEEE Press, 2014: 1807-1818.
14	谷远利, 张源, 芮小平, 等. 基于免疫算法优化LSSVM的短时交通流预测. 吉林大学学报(工学版), 2019, 49 (6): 1852- 1857.
	GU Y L , ZHANG Y , RUI X P , et al. Short-term traffic flow prediction based on LSSVM optimized by immune algorithm. Journal of Jilin University(Engineering and Technology Edition), 2019, 49 (6): 1852- 1857.
15	WU H , GAO L , ZHANG Z . Analysis of crash data using quantile regression for counts. Journal of Transportation Engineering, 2014, 140 (4): 04014026.
16	PARK J, LI D, MURPHEY Y L, et al. Real time vehicle speed prediction using a neural network traffic model[C]//Proceedings of the 2011 International Joint Conference on Neural Networks. Washington D.C., USA: IEEE Press, 2011: 2991-2996.
17	XIA D W , YANG N , JIAN S Y , et al. SW-BiLSTM: a spark-based weighted BiLSTM model for traffic flow forecasting. Multimedia Tools and Applications, 2022, 81 (17): 23589- 23614. doi: 10.1007/s11042-022-12039-3
18	张惠臻, 高正凯, 李建强, 等. 基于循环神经网络的城市轨道交通短时客流预测. 吉林大学学报(工学版), 2023, 53 (2): 430- 438.
	ZHANG H Z , GAO Z K , LI J Q , et al. Short-term passenger flow forecasting of urban rail transit based on recurrent neural network. Journal of Jilin University(Engineering and Technology Edition), 2023, 53 (2): 430- 438.
19	ABDULJABBAR R L , DIA H , TSAI P W . Unidirectional and bidirectional LSTM models for short-term traffic prediction. Journal of Advanced Transportation, 2021, 2021, 5589075.
20	SHI X M , QI H , SHEN Y M , et al. A spatial-temporal attention approach for traffic prediction. IEEE Transactions on Intelligent Transportation Systems, 2021, 22 (8): 4909- 4918. doi: 10.1109/TITS.2020.2983651
21	罗向龙, 郭凰, 廖聪, 等. 时空相关的短时交通流宽度学习预测模型. 计算机工程与应用, 2022, 58 (9): 181- 186.
	LUO X L , GUO H , LIAO C , et al. Spatiotemporal short-term traffic flow prediction based on broad learning system. Computer Engineering and Applications, 2022, 58 (9): 181- 186.
22	LIU Q C , WANG B C , ZHU Y Q . Short-term traffic speed forecasting based on attention convolutional neural network for arterials. Computer-Aided Civil and Infrastructure Engineering, 2018, 33 (11): 999- 1016. doi: 10.1111/mice.12417
23	CHENG Z Y , LU J , ZHOU H J , et al. Short-term traffic flow prediction: an integrated method of econometrics and hybrid deep learning. IEEE Transactions on Intelligent Transportation Systems, 2022, 23 (6): 5231- 5244. doi: 10.1109/TITS.2021.3052796
24	RANJAN N , BHANDARI S , ZHAO H P , et al. City-wide traffic congestion prediction based on CNN, LSTM and transpose CNN. IEEE Access, 2020, 8, 81606- 81620. doi: 10.1109/ACCESS.2020.2991462
25	ISHIDA S , TERAYAMA K , KOJIMA R , et al. Prediction and interpretable visualization of retrosynthetic reactions using graph convolutional networks. Journal of Chemical Information and Modeling, 2019, 59 (12): 5026- 5033. doi: 10.1021/acs.jcim.9b00538
26	LI Z R , GONG J W , LU C , et al. Interactive behavior prediction for heterogeneous traffic participants in the urban road: a graph-neural-network-based multitask learning framework. ASME Transactions on Mechatronics, 2021, 26 (3): 1339- 1349. doi: 10.1109/TMECH.2021.3073736
27	WU Z H , PAN S R , CHEN F W , et al. A comprehensive survey on graph neural networks. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32 (1): 4- 24. doi: 10.1109/TNNLS.2020.2978386
28	LI W , WANG X , ZHANG Y W , et al. Traffic flow prediction over Muti-sensor data correlation with graph convolution network. Neurocomputing, 2021, 427, 50- 63. doi: 10.1016/j.neucom.2020.11.032
29	CHEN C , LI K L , TEO S G , et al. Gated residual recurrent graph neural networks for traffic prediction. Artificial Intelligence, 2019, 33 (1): 485- 492.
30	MA M Y , NA S , WANG H Y . AEGCN: an autoencoder-constrained graph convolutional network. Neurocomputing, 2021, 432, 21- 31. doi: 10.1016/j.neucom.2020.12.061
31	YIN X Y , WU G Z , WEI J Z , et al. Multi-stage attention spatial-temporal graph networks for traffic prediction. Neurocomputing, 2021, 428, 42- 53. doi: 10.1016/j.neucom.2020.11.038
32	YU B, YIN H T, ZHU Z X. Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting[C]//Proceedings of the 27th International Joint Conference on Artificial Intelligence. Washington D.C., USA: IEEE Press, 2018: 3634-3640.
33	GUO S N , LIN Y F , FENG N , et al. Attention based spatial-temporal graph convolutional networks for traffic flow forecasting. Artificial Intelligence, 2019, 33 (1): 922- 929.
34	SONG C , LIN Y F , GUO S N , et al. Spatial-temporal synchronous graph convolutional networks: a new framework for spatial-temporal network data forecasting. Artificial Intelligence, 2020, 34 (1): 914- 921.
35	KINGMA D , BA J . Adam: a method for stochastic optimization. Computer Science, 2015, 32, 112- 121.
36	WU Z H, PAN S R, LONG G D, et al. Graph WaveNet for deep spatial-temporal graph modeling[C]//Proceedings of the 28th International Joint Conference on Artificial Intelligence. Washington D.C., USA: IEEE Press, 2019: 1907-1913.

[1]	翟志鹏, 曹阳, 沈琴琴, 施佺. 基于多时空图融合与动态注意力的交通流预测[J]. 计算机工程, 2025, 51(9): 139-148.
[2]	武东辉, 王金凤, 仇森, 刘国志. 基于EWBiLSTM-ATT的数据手套手语识别[J]. 计算机工程, 2025, 51(8): 107-119.
[3]	武东辉, 王金凤, 仇森, 刘国志. 基于EWBiLSTM-ATT的数据手套手语识别[J]. 计算机工程, 2025, 51(8): 107-119.
[4]	周莎, 车生兵, 考友琛, 张旭, 郭甚驿. 基于特征选择和时空特征的网络入侵检测[J]. 计算机工程, 2025, 51(7): 223-231.
[5]	柳大格, 游进国, 耿齐祁. 融合全局与局部语义的跨领域方面词抽取[J]. 计算机工程, 2025, 51(6): 116-126.
[6]	徐永刚, 孙琦烜, 李凡甲, 程健维, 戴佳俊. 基于扩展时间和时空特征融合图卷积网络的骨架行为识别[J]. 计算机工程, 2025, 51(4): 281-292.
[7]	朱子斌, 李千林, 张小燕, 韩双双. 基于双向时空特征学习的人体深度图像估计[J]. 计算机工程, 2025, 51(10): 258-269.
[8]	耿贞伟, 李申章, 于凤荣. 考虑空间关联的超短期、短期多区域电力负荷单步和多步预测[J]. 计算机工程, 2025, 51(10): 392-400.
[9]	党小超, 刘涧, 董晓辉, 祝忠彦, 李芬芳. 面向不平衡数据的机械设备故障命名实体识别[J]. 计算机工程, 2024, 50(9): 104-112.
[10]	屈潇雅, 李兵, 温立强. 面向行政执法案件文本的事件抽取研究[J]. 计算机工程, 2024, 50(9): 63-71.
[11]	张亚洲, 和玉, 戎璐, 王祥凯. 基于上下文知识增强型Transformer网络的抑郁检测[J]. 计算机工程, 2024, 50(8): 75-85.
[12]	刘建敏, 林晖, 汪晓丁. 基于图注意力机制的无地图场景轨迹预测方法[J]. 计算机工程, 2024, 50(7): 144-153.
[13]	于洋, 孙芳芳, 吕华, 李扬, 王晓民. 基于多尺度时空注意力网络的微表情检测方法[J]. 计算机工程, 2024, 50(6): 228-235.
[14]	周昭辰, 方清茂, 吴晓红, 胡平, 何小海. 基于MacBERT与对抗训练的机器阅读理解模型[J]. 计算机工程, 2024, 50(5): 41-50.
[15]	史昕, 曹凤腾, 纪艺, 马峻岩. 基于多尺度时空特征与软注意力机制的交通流预测方法[J]. 计算机工程, 2024, 50(12): 346-357.

选择文件类型/文献管理软件名称

选择包含的内容