A novel hybrid prediction model of air quality index based on variational modal decomposition and CEEMDAN-SE-GRU

Abstract

As the economy rapidly develops, the phenomenon of air pollution is occurring more frequently, impacting on human health and environmental ecosystems. The Air Quality Index (AQI) is an essential index for describing air pollution levels. Accurate AQI prediction provides reliable information for air pollution control and protect human health. Therefore, in this paper, the AQI data of Beijing-Tianjin-Hebei region from 20 May 2014–18 November 2023 are used. We propose two basic prediction models based on Complete Ensemble Empirical Modal Decomposition with Adaptive Noise (CEEMDAN) and Gated Recursive Unit (GRU) as well as an improved prediction framework combining Variational Modal Decomposition (VMD). The best mean coefficient of determination (R²) is 0.984, the mean absolute error (MAE) is 2.476, and the R² is more than 0.980 for the prediction of AQI in different regions. These results validate the reliability of this hybrid prediction model.