Sign in Subscribe
Electrolysis (Chemistry)

Advancing Water Electrolysis Technologies for Green Hydrogen Production

Strategic Insights

— 1 min read

Research from Mahidol University in Nakorn Panom, Thailand, has shed light on the importance of proton exchange membrane water electrolyzer (PEMWE) in scaling up green hydrogen production. The study emphasizes the need to optimize hydrogen production rates (QH2) while minimizing electrical energy consumption (E) in PEMWE systems. A novel integrated artificial neural network-genetic algorithm (ANN-GA) framework was proposed to predict and optimize these outcomes.

Key Takeaways:

  • The integrated ANN-GA framework significantly outperforms conventional regression models in predicting hydrogen production rates (QH2) and energy consumption (E) in PEMWE systems, with an R2 of 0.99989 and an MAE of 0.46914.
  • The proposed approach enables fine-tuned control over system parameters, supporting the development of efficient, scalable, and sustainable hydrogen production technologies.
  • The ANN-GA framework takes four critical inputs (water temperature, water flow rate, applied voltage, and current density) to forecast two outputs (QH2 and E).
  • Experimental validation using GA-optimized parameters yields a mean prediction error of less than 13.33%, confirming the robustness of the proposed method.
  • The study highlights the effectiveness of the integrated ANN-GA framework in enhancing prediction accuracy and optimization capabilities for PEMWE operations.
  • The proposed method supports the development of efficient, scalable, and sustainable hydrogen production technologies.
  • The research was conducted by Chakrit Suvanjumrat, Arom Boekfah, and Supachai Rumnum from Mahidol University.

Statistics:

  • 0.99989: R2 value indicating the accuracy of the ANN-GA framework in predicting hydrogen production rates (QH2).
  • 0.46914: Mean absolute error (MAE) value indicating the precision of the AN-GA framework in predicting energy consumption (E).
  • 13.33%: Mean prediction error of the proposed approach in experimental validation.
  • 4: Number of critical inputs taken by the ANN-GA framework (water temperature, water flow rate, applied voltage, and current density).
  • 2: Number of outputs forecasted by the ANN-GA framework (QH2 and E).
  • 10: Number of neurons in the hidden layer of the ANN-GA framework.
  • 4: Number of input nodes in the ANN-GA framework.
  • 2: Number of output nodes in the ANN-GA framework.

Sources:

  • NewsRx LLC. Researchers from Mahidol University Provide Details of New Studies and Findings in the Area of Networks (Modeling and Optimization of Pem Water Electrolysis Via an Ann-ga Hybrid Approach). Journal of Engineering. October 20, 2025; p 4147.
  • Suvanjumrat, C., Boekfah, A., & Rumnum, S. (2025). Modeling and Optimization of Pem Water Electrolysis Via an Ann-ga Hybrid Approach. International Journal of Hydrogen Energy, 177.

Read more