Machine learning analysis of photocatalytic glycerol reforming for hydrogen production
International Journal of Hydrogen Energy, cilt.142, ss.1014-1025, 2025 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 142
- Basım Tarihi: 2025
- Doi Numarası: 10.1016/j.ijhydene.2025.04.027
- Dergi Adı: International Journal of Hydrogen Energy
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Artic & Antarctic Regions, Chemical Abstracts Core, Chimica, Compendex, Environment Index, INSPEC
- Sayfa Sayıları: ss.1014-1025
- Anahtar Kelimeler: Glycerol photo reforming, Hydrogen production prediction, Photocatalysis, Machine learning, Random forest, Gradient boosting
- Boğaziçi Üniversitesi Adresli: Evet
Özet
This study investigates the photocatalytic glycerol reforming process through a comprehensive analysis of 888 data points sourced from 126 published articles between 2005 and 2024; it focused on identifying key parameters influencing the hydrogen production rate such as photocatalysts' properties, including semiconductor type, bandgap, preparation methods, and reaction conditions such as glycerol concentration and light intensity. Regression models were developed using random forest and gradient boosting techniques to predict both the bandgap and hydrogen production rate. The random forest model successfully predicted bandgap with an RMSE of 0.07 for the test set, while the gradient boosting model demonstrated reasonable accuracy for hydrogen production, achieving RMSE values of 9367 for the test set. Notably, the type of co-catalyst emerged as the most significant predictor for hydrogen production, along with the critical role of preparation methods and operational conditions. Decision tree classification of hydrogen production also showed that co-catalyst, semiconductor and preparation method plays an important role for achieving high hydrogen production rates. Our findings underline the importance of optimizing both catalyst properties and reaction environments to enhance hydrogen production efficiency.