PM2.5 concentration forecasting through a novel multi-scale ensemble learning approach considering intercity synergy

Abstract

Accurate PM 2.5 concentration prediction can provide reliable air pollution warning information to the public. However, previous studies have often focused on the data of the target city itself, ignoring the interaction among cities in the same region. In this paper, we develop a multi-scale ensemble learning approach to forecast daily PM 2.5 concentrations of the target city by modeling its air and climate indicators, and PM 2.5 value of its neighboring cities. First, the proposed approach smooths the multivariate data by singular spectrum analysis and performs multi-feature selection based on distance factor and predictive power of data. Second, the inherent association among the obtained multiple features is captured by multivariate empirical modal decomposition. Third, the Hurst exponent is applied to match each time scale with the corresponding predictor for multi-step prediction. Finally, the forecasting values of all time scales are summed to obtain the PM 2.5 concentration forecasting results of the target city. Four experiments involving Beijing, Wuhan, and Shenzhen are carried out to verify the accuracy and robustness of the proposed approach. The experimental results show that our approach outperforms all benchmark models, and introducing city synergy strategy can improve the forecasting performance significantly. • The synergistic effect of surrounding cities improves the forecasting accuracy. • Double-level feature selection is established to extract the effective features. • Multivariate decomposition algorithm implements multi-source data fusion well. • A forecasting model matching strategy is developed based on the average Hurst exponent. • Average MAPE improvements of the approach against benchmark models are above 8.9%.