Data driven one day ahead air quality forecast system for SO2 and PM10 pollutants

• Main Author: Lu, Jingjun
• Other Authors: Wong Kin Shun, Terence
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/167158
• Institution: Nanyang Technological University

Air quality is an extremely important topic today and has been for some time. The quality of the air we breathe has a significant impact on our health, as well as the health of the environment. In recent years, there has been growing concern about air pollution and its effects on human health and the environment. Many countries have implemented regulations and policies to address air pollution and improve air quality. Therefore, an air quality forecast system for air pollutants will be extremely useful. However, unlike air temperature and wind direction, there is no simple prediction method for outdoor air quality based on current and recent concentrations of air pollutants. Some recent studies carried out on this still has only limited accuracy. Therefore, a more accurate air quality forecast system based on historical air quality data is necessary. This project explores the use of a data driven machine learning (ML) approach to develop models to predict the next day air quality based on the concentration of pollutants and other climate data in the previous few days. The trained models generate acceptable results in predicting the air quality of the next day, comparing to the actual results. This air quality forecast system, with high accuracy, will help people a lot in future decision making and policy formulation for environmental issues.