Title: Unified spatiotemporal modelling for accurate gas concentration prediction

Authors: Yu Zhang; Lei Cheng

Addresses: School of Information Science and Engineering, Wuhan University of Science and Technology, Wuhan, Hubei, China ' School of Information Science and Engineering, Wuhan University of Science and Technology, Wuhan, Hubei, China

Abstract: Accurate gas concentration prediction is crucial for air quality management and environmental protection. However, the complexity of spatiotemporal dependencies and data gaps within air quality datasets pose significant challenges to this task. In this paper, we propose a spatiotemporal unified attention-based gas prediction model (STUAGPM). The model utilises an innovative data processing approach to effectively integrate temporal and spatial factors, thereby improving data utilisation efficiency. Additionally, the attention mechanism is employed to uniformly model spatiotemporal dependencies, while incorporating meteorological variables and other gas concentrations. Experimental results on two real-world datasets demonstrate that STUAGPM outperforms traditional models, including LSTM, GRU, TCN, and transformer, in terms of mean absolute error (MAE) and mean squared error (MSE). The findings indicate that STUAGPM improves the accuracy and robustness of data processing in complex air quality datasets.

Keywords: spatiotemporal unified modelling; data processing; attention mechanism; gas concentration prediction.

DOI: 10.1504/IJSPM.2024.143854

International Journal of Simulation and Process Modelling, 2024 Vol.21 No.3, pp.192 - 203

Received: 19 Jul 2024
Accepted: 04 Oct 2024
Published online: 10 Jan 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article