Abstract

Ultrafine particulate matter (PM<inf>0.1</inf>) is a global and significant environmental issue because it can deeply translocate the human body, causing adverse health effects and leading to a high mortality rate. This study investigates the relationship between meteorological factors and PM<inf>0.1</inf> concentrations, providing insights into the formation and distribution of ultrafine particles. However, accurate measurement of the PM<inf>0.1</inf> concentration information is challenging due to sophisticated processes and expensive instruments that make it difficult to access. This study addresses these concerns with a new deeplearning regression model for PM<inf>0.1</inf> concentration prediction based on meteorological factors. The model is designed and developed to explore the optimal model structures (hidden layers and neurons) to achieve standard laboratory-based PM<inf>0.1</inf> measurement. The model structures are verified by root mean squared error (RMSE) and coefficient of determination (R²) based on predictive performance to prove the laboratory-based standard's accomplishment. The results demonstrate that the proposed model can estimate PM0.1 concentration with high performance, R² = 92.52%, and RMSE = 0.26 µg/m³, which is precise and reliable for an intelligent-driven PM0.1 concentration prediction system to support preventive health decisionmaking. This approach contributes to a more comprehensive understanding of atmospheric composition by enabling widespread monitoring of PM<inf>0.1</inf>, a critical but often unmeasured component of air pollution.