arrow
Volume 16, Issue 1
A Short-Time Traffic Flow Prediction Model Based on TCN-LSTM with Causal Convolutional Layer

Chenmin Ma

J. Info. Comput. Sci. , 16 (2021), pp. 016-023.

Export citation
  • Abstract

Short term traffic flow prediction is of great significance for traffic management, early guidance and dispersion, avoiding congestion and improving safety. Traffic flow is affected by multiple factors and exists strong time dependency. Therefore, it is very meaningful to establish a prediction model with multiple features, such as time-day-week-biweekly, holiday and weather situation, etc. In this paper, a TCN- LSTM model with causal convolution block is proposed, which is composed of two subnets: LSTM subnetwork is used to extract feature from original traffic flow data sequence, three TCN+LSTM subnetworks are used to extract features from traffic flow data with day-week-biweekly, holiday and weather. TCN is embedded to maintain causation of the input traffic flow data. Finally, features extracted from the two sub networks are merged and imported into top-level full connection network. The prediction sequence of the future short-term traffic flow is obtained at the output layer of FCN. Experimental results show that the proposed TCN-LSTM model has high accuracy and stability in short-term traffic flow prediction.

  • AMS Subject Headings

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{JICS-16-016, author = {Ma , Chenmin}, title = {A Short-Time Traffic Flow Prediction Model Based on TCN-LSTM with Causal Convolutional Layer}, journal = {Journal of Information and Computing Science}, year = {2024}, volume = {16}, number = {1}, pages = {016--023}, abstract = {

Short term traffic flow prediction is of great significance for traffic management, early guidance and dispersion, avoiding congestion and improving safety. Traffic flow is affected by multiple factors and exists strong time dependency. Therefore, it is very meaningful to establish a prediction model with multiple features, such as time-day-week-biweekly, holiday and weather situation, etc. In this paper, a TCN- LSTM model with causal convolution block is proposed, which is composed of two subnets: LSTM subnetwork is used to extract feature from original traffic flow data sequence, three TCN+LSTM subnetworks are used to extract features from traffic flow data with day-week-biweekly, holiday and weather. TCN is embedded to maintain causation of the input traffic flow data. Finally, features extracted from the two sub networks are merged and imported into top-level full connection network. The prediction sequence of the future short-term traffic flow is obtained at the output layer of FCN. Experimental results show that the proposed TCN-LSTM model has high accuracy and stability in short-term traffic flow prediction.

}, issn = {1746-7659}, doi = {https://doi.org/}, url = {http://global-sci.org/intro/article_detail/jics/22375.html} }
TY - JOUR T1 - A Short-Time Traffic Flow Prediction Model Based on TCN-LSTM with Causal Convolutional Layer AU - Ma , Chenmin JO - Journal of Information and Computing Science VL - 1 SP - 016 EP - 023 PY - 2024 DA - 2024/01 SN - 16 DO - http://doi.org/ UR - https://global-sci.org/intro/article_detail/jics/22375.html KW - Short-term traffic flow forecast, Convolutional neural network, Long and short term memory neural network, Causal convolutional layer. AB -

Short term traffic flow prediction is of great significance for traffic management, early guidance and dispersion, avoiding congestion and improving safety. Traffic flow is affected by multiple factors and exists strong time dependency. Therefore, it is very meaningful to establish a prediction model with multiple features, such as time-day-week-biweekly, holiday and weather situation, etc. In this paper, a TCN- LSTM model with causal convolution block is proposed, which is composed of two subnets: LSTM subnetwork is used to extract feature from original traffic flow data sequence, three TCN+LSTM subnetworks are used to extract features from traffic flow data with day-week-biweekly, holiday and weather. TCN is embedded to maintain causation of the input traffic flow data. Finally, features extracted from the two sub networks are merged and imported into top-level full connection network. The prediction sequence of the future short-term traffic flow is obtained at the output layer of FCN. Experimental results show that the proposed TCN-LSTM model has high accuracy and stability in short-term traffic flow prediction.

Ma , Chenmin. (2024). A Short-Time Traffic Flow Prediction Model Based on TCN-LSTM with Causal Convolutional Layer. Journal of Information and Computing Science. 16 (1). 016-023. doi:
Copy to clipboard
The citation has been copied to your clipboard