A Quantitative Modeling and Prediction Method for Sustained Rainfall-PM2.5 Removal Modes on a Micro-Temporal Scale

PM2.5 is unanimously considered to be an important indicator of air quality. Sustained rainfall a kind typical but complex process in southern China with uncertain duration and intervals. During sustained rainfall, the variation concentrations hour-level time series diverse complex. However, existing analytical methods mainly examine overall removals at annual/monthly scale, missing quantitative analysis mode that applies micro-scale data describe removal phenomenon. In order further achieve quality prediction prevention short term, it necessary analyze its micro-temporal effect for atmospheric environment forecasting. This paper proposed modeling method rainfall-PM2.5 modes on scale. Firstly, set scale were constructed. Then, mode-constrained using factorization machines (FM) was predict future effect. Moreover, historical observation Nanjing city hourly from 2016 January 2020 used modeling. Meanwhile, whole year phenomenon prediction. The experiment shows reasonableness effectiveness method.