Machine learning in PEM water electrolysis: A study of hydrogen production and operating parameters

•Advanced machine learning models, such as RF, SVM, and XGBoost, demonstrate high accuracy in predicting hydrogen production rates in PEM electrolyzers.•RF models consistently outperformed the other models, showing exceptional predictive accuracy evidenced from the R² score of 0.9898, RMSE of 19.99...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Computers & chemical engineering Ročník 194; s. 108954
Hlavní autori: Shomope, Ibrahim, Al-Othman, Amani, Tawalbeh, Muhammad, Alshraideh, Hussam, Almomani, Fares
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Elsevier Ltd 01.03.2025
Predmet:
eXtreme gradient boosting algorithm
Hydrogen production
Machine learning
Random forest
Support vector machine
Water electrolysis
Water electrolysis
Random forest
Support vector machine
eXtreme gradient boosting algorithm
Hydrogen production
Machine learning
ISSN:0098-1354
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:•Advanced machine learning models, such as RF, SVM, and XGBoost, demonstrate high accuracy in predicting hydrogen production rates in PEM electrolyzers.•RF models consistently outperformed the other models, showing exceptional predictive accuracy evidenced from the R² score of 0.9898, RMSE of 19.99 mL/min, and MAE value of 10.41 mL/min.•The random forest model outperformed other ML algorithms, providing the most accurate predictions of hydrogen production rates. These results underscore the RF model's robustness in capturing complex relationships within PEM operation, guiding more effective system optimization.•This study establishes a data-driven framework for optimizing PEM electrolyzer operation, paving the way for improved hydrogen production efficiency.•Extended analysis of hydrogen production rates illustrates the effect of varying cell voltage of 0.5 to 3.5 v and cell current of 0 to 100 a on hydrogen flow rate. Proton exchange membrane water electrolysis (PEMWE) powered by renewable energy stands out as a promising technology for the sustainable production of high-purity hydrogen. This study employed three machine learning (ML) algorithms, random forest (RF), support vector machine (SVM), and eXtreme gradient boosting (XGBoost), to predict hydrogen production in PEMWE. Model performance was evaluated using root mean squared error (RMSE), coefficient of determination (R²), and mean absolute error (MAE) metrics. The top-performing models, RF and XGBoost, were further refined through hyperparameter tuning. The final models demonstrated high reliability in predicting hydrogen production rates, with RF consistently outperforming XGBoost. The RF model achieved a predictive accuracy of R² = 0.9898, RMSE = 19.99 mL/min, and MAE = 10.41 mL/min, while the XGBoost model achieved R² = 0.9894, RMSE = 20.43 mL/min, and MAE = 11.50 mL/min. Partial dependency plots (PDPs) emphasized the critical role of optimizing both cell voltage and current to maximize hydrogen production in PEMWE. These insights provide valuable guidance for operational adjustments, ensuring optimal system performance for high efficiency and productivity. The study suggests further research on the impact of parameters like temperature and power density on hydrogen production, incorporating them for better optimization.
ISSN:0098-1354
DOI:10.1016/j.compchemeng.2024.108954