Multi-source variational mode transfer learning for enhanced PM2.5 concentration forecasting at data-limited monitoring stations

Bozhi Yao, Guang Ling, Feng Liu, Ming-Feng Ge

Abstract: Highlights•Explored the effectiveness of air quality data transfer among multiple stations.•Multi-Source Variational Mode Transfer Learning (MSVMTL) is proposed.•MSVMTL integrates data decomposition, deep learning, and transfer learning.•The framework significantly improved prediction accuracy at the target station.•Each stage of the framework demonstrates excellent performance.