Long-Term Forecasting of Air Pollution Particulate Matter (PM2.5) and Analysis of Influencing Factors

Abstract

Long-term forecasting and analysis of PM2.5, a significant air pollution source, is vital for environmental governance and sustainable development. We evaluated 10 machine learning and deep learning models using PM2.5 concentration data along with environmental variables. Employing explainable AI (XAI) technology facilitated explainability and formed the basis for factor analysis. At a 30-day forecasting horizon, ensemble learning surpassed deep learning in performance, with CatBoost emerging as the top-performing model. For forecasting horizons of 90 and 180 days, Bi-SLTM and Bi-GRU, respectively, exhibited the highest performance. Through an analysis of influencing factors by SHAP, it was observed that PM10 exerted the greatest impact on PM2.5 forecasting. However, this effect was particularly pronounced at higher concentrations of CO. Conversely, at lower CO concentrations, the impact of increased PM10 concentrations on PM2.5 was limited. Hence, it can be inferred that CO plays a pivotal role in driving these effects. Following CO, factors such as “dew point” and “temperature” were identified as influential. These factors exhibited varying levels of linear correlation with PM2.5, with temperature showing a negative correlation, while PM10, CO, and dew point generally demonstrated positive correlations with PM2.5.

Insightful Analysis and Prediction of SCOD Component Variation in Low-Carbon/Nitrogen-Ratio Domestic Wastewater via Machine Learning

“…From Figure 8, an increase in the predictive value of Fmax is associated with an increase in C4 and C1, which is consistent with previous studies. The high Fmax values of C3 and C4 positively correlate with SCOD, especially the high Fmax values of component C4, which shows long right tails in the summary plots [78]. However, the four models present different influence trends between C2 and Fmax.…”

PD-LL-Transformer: An Hourly PM2.5 Forecasting Method over the Yangtze River Delta Urban Agglomeration, China

“…RNNs add a memory boost to neural networks, making them way smarter when dealing with sequences and patterns. In addition, to avoid the vanishing gradient predicament inherent in ordinary RNNs, long short-term memory (LSTM) [32,33] and gated recurrent unit (GRU) [34] were invented consecutively and became the most popular RNN architectures. RNNs are intrinsically effective for time-series forecasting tasks, including natural language processing (NLP), recommendation system, and stock prediction.…”