Optimisation of an ash fouling model for predicting cleanliness levels in coal-fired power plant

This study develops an optimised ash fouling model for coal-fired power plants by integrating feature selection with an advanced regression model. The aim is to accurately predict cleanliness levels, providing a reliable indication for soot-blowing operations to maintain boiler performance. Comprehe...

Full description

Bibliographic Details
Main Authors: Achmad Nazeem, Nabil, Mohd Ibrahim, Maslina, Che Soh, Azura, Ishak, Asnor Juraiza, Raja Ahmad, Raja Mohd Kamil, Norsahperi, Nor Mohd Haziq, Mohd Radzi, Mohd Amran
Format: Conference or Workshop Item
Language:English
Published: IEEE 2025
Online Access:http://psasir.upm.edu.my/id/eprint/119224/
http://psasir.upm.edu.my/id/eprint/119224/1/119224.pdf
Description
Summary:This study develops an optimised ash fouling model for coal-fired power plants by integrating feature selection with an advanced regression model. The aim is to accurately predict cleanliness levels, providing a reliable indication for soot-blowing operations to maintain boiler performance. Comprehensive feature selection techniques are used to identify key variables influencing soot accumulation, while Gaussian Process Regression (GPR) predicts cleanliness levels. Three feature selection methods are implemented and compared to determine the most accurate approach for guiding soot-blowing operations. Using operational data from a power plant station in Malaysia, this study aims to enhance sootblowing practices, improve efficiency, and reduce environmental impact. By optimising soot-blowing strategies, the research supports sustainable energy practices and contributes to improved power plant performance.