D-MGDCN-CLSTM: A Traffic Prediction Model Based on Multi-Graph Gated Convolution and Convolutional Long–Short-Term Memory
Real-time and accurate traffic forecasting aids in traffic planning and design and helps to alleviate congestion. Addressing the negative impacts of partial data loss in traffic forecasting, and the challenge of simultaneously capturing short-term fluctuations and long-term trends, this paper presen...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/2/561 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587494921601024 |
|---|---|
| author | Linliang Zhang Shuyun Xu Shuo Li Lihu Pan Su Gong |
| author_facet | Linliang Zhang Shuyun Xu Shuo Li Lihu Pan Su Gong |
| author_sort | Linliang Zhang |
| collection | DOAJ |
| description | Real-time and accurate traffic forecasting aids in traffic planning and design and helps to alleviate congestion. Addressing the negative impacts of partial data loss in traffic forecasting, and the challenge of simultaneously capturing short-term fluctuations and long-term trends, this paper presents a traffic forecasting model, D-MGDCN-CLSTM, based on Multi-Graph Gated Dilated Convolution and Conv-LSTM. The model uses the DTWN algorithm to fill in missing data. To better capture the dual characteristics of short-term fluctuations and long-term trends in traffic, the model employs the DWT for multi-scale decomposition to obtain approximation and detail coefficients. The Conv-LSTM processes the approximation coefficients to capture the long-term characteristics of the time series, while the multiple layers of the MGDCN process the detail coefficients to capture short-term fluctuations. The outputs of the two branches are then merged to produce the forecast results. The model is tested against 10 algorithms using the PeMSD7(M) and PeMSD7(L) datasets, improving MAE, RMSE, and ACC by an average of 1.38% and 13.89%, 1% and 1.24%, and 5.92% and 1%, respectively. Ablation experiments, parameter impact analysis, and visual analysis all demonstrate the superiority of our decompositional approach in handling the dual characteristics of traffic data. |
| format | Article |
| id | doaj-art-53e66640bef943d8802e415fc0ff69f6 |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-53e66640bef943d8802e415fc0ff69f62025-01-24T13:49:21ZengMDPI AGSensors1424-82202025-01-0125256110.3390/s25020561D-MGDCN-CLSTM: A Traffic Prediction Model Based on Multi-Graph Gated Convolution and Convolutional Long–Short-Term MemoryLinliang Zhang0Shuyun Xu1Shuo Li2Lihu Pan3Su Gong4Shanxi Intelligent Transportation Laboratory Co., Ltd., Taiyuan 030036, ChinaSchool of Computer Science and Technology, Taiyuan University of Science and Technology, Taiyuan 030024, ChinaSchool of Information Science and Technology, Southwest Jiaotong University, Chengdu 611756, ChinaSchool of Computer Science and Technology, Taiyuan University of Science and Technology, Taiyuan 030024, ChinaSchool of Information Science and Technology, Southwest Jiaotong University, Chengdu 611756, ChinaReal-time and accurate traffic forecasting aids in traffic planning and design and helps to alleviate congestion. Addressing the negative impacts of partial data loss in traffic forecasting, and the challenge of simultaneously capturing short-term fluctuations and long-term trends, this paper presents a traffic forecasting model, D-MGDCN-CLSTM, based on Multi-Graph Gated Dilated Convolution and Conv-LSTM. The model uses the DTWN algorithm to fill in missing data. To better capture the dual characteristics of short-term fluctuations and long-term trends in traffic, the model employs the DWT for multi-scale decomposition to obtain approximation and detail coefficients. The Conv-LSTM processes the approximation coefficients to capture the long-term characteristics of the time series, while the multiple layers of the MGDCN process the detail coefficients to capture short-term fluctuations. The outputs of the two branches are then merged to produce the forecast results. The model is tested against 10 algorithms using the PeMSD7(M) and PeMSD7(L) datasets, improving MAE, RMSE, and ACC by an average of 1.38% and 13.89%, 1% and 1.24%, and 5.92% and 1%, respectively. Ablation experiments, parameter impact analysis, and visual analysis all demonstrate the superiority of our decompositional approach in handling the dual characteristics of traffic data.https://www.mdpi.com/1424-8220/25/2/561intelligent transportation systemstraffic predictiondeep learninggraph convolutional networksspatio-temporal feature |
| spellingShingle | Linliang Zhang Shuyun Xu Shuo Li Lihu Pan Su Gong D-MGDCN-CLSTM: A Traffic Prediction Model Based on Multi-Graph Gated Convolution and Convolutional Long–Short-Term Memory Sensors intelligent transportation systems traffic prediction deep learning graph convolutional networks spatio-temporal feature |
| title | D-MGDCN-CLSTM: A Traffic Prediction Model Based on Multi-Graph Gated Convolution and Convolutional Long–Short-Term Memory |
| title_full | D-MGDCN-CLSTM: A Traffic Prediction Model Based on Multi-Graph Gated Convolution and Convolutional Long–Short-Term Memory |
| title_fullStr | D-MGDCN-CLSTM: A Traffic Prediction Model Based on Multi-Graph Gated Convolution and Convolutional Long–Short-Term Memory |
| title_full_unstemmed | D-MGDCN-CLSTM: A Traffic Prediction Model Based on Multi-Graph Gated Convolution and Convolutional Long–Short-Term Memory |
| title_short | D-MGDCN-CLSTM: A Traffic Prediction Model Based on Multi-Graph Gated Convolution and Convolutional Long–Short-Term Memory |
| title_sort | d mgdcn clstm a traffic prediction model based on multi graph gated convolution and convolutional long short term memory |
| topic | intelligent transportation systems traffic prediction deep learning graph convolutional networks spatio-temporal feature |
| url | https://www.mdpi.com/1424-8220/25/2/561 |
| work_keys_str_mv | AT linliangzhang dmgdcnclstmatrafficpredictionmodelbasedonmultigraphgatedconvolutionandconvolutionallongshorttermmemory AT shuyunxu dmgdcnclstmatrafficpredictionmodelbasedonmultigraphgatedconvolutionandconvolutionallongshorttermmemory AT shuoli dmgdcnclstmatrafficpredictionmodelbasedonmultigraphgatedconvolutionandconvolutionallongshorttermmemory AT lihupan dmgdcnclstmatrafficpredictionmodelbasedonmultigraphgatedconvolutionandconvolutionallongshorttermmemory AT sugong dmgdcnclstmatrafficpredictionmodelbasedonmultigraphgatedconvolutionandconvolutionallongshorttermmemory |