EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Interpretable Transformer-Based Voltage Degradation Prediction of Proton Exchange Membrane Fuel Cells Under Constant-Current Operation

Fengyan Yi, Xing Shu, Jinming Zhang, Zongjing Huang, Junling Zhang, Hongtao Gong, Xiangya Liu, Shuaihua Wang, Jiaming Zhou

Electronics · 28.07.2026 · 15 (15) · 3334

DOI: 10.3390/electronics15153334 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 28.07.2026. Verfasst von Fengyan Yi und 8 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Weifang City Science and Technology Development Plan Projects.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate voltage degradation prediction is essential for health management and lifetime extension of proton exchange membrane fuel cell (PEMFC) systems. During long-term constant-current operation, stack voltage evolves nonlinearly and is influenced by coupled variations in temperature, pressure, flow rate, and humidity, while many deep learning-based models lack physical interpretability. This study proposes an interpretable Transformer-based framework for PEMFC voltage degradation prediction under constant-current operation. The framework integrates outlier correction, interpolation, Savitzky–Golay filtering, Z-score normalization, sliding-window reconstruction, Transformer-based prediction, and feature-ablation interpretation. Using multivariate sensor measurements and historical voltage as inputs, the Transformer was compared with RNN, LSTM, and GRU baselines under identical preprocessing and evaluation conditions. The models were evaluated chronologically by continuously applying the sliding-window model over the held-out final 20% of the aging sequence. The Transformer achieved the best performance, with MAE of 8.2 × 10−4, RMSE of 1.18 × 10−3, MAPE of 0.0256%, and R2 of 0.9961. Compared with the second-best RNN model, it reduced MAE, RMSE, and MAPE by 9.89%, 7.81%, and 9.86%, respectively. Feature ablation showed that flow- and pressure-related variables contributed 50.08% and 26.78% of the total importance, respectively.

Förderung

  • Weifang City Science and Technology Development Plan Projects

Zitieren

APA

Yi, F., Shu, X., Zhang, J., Huang, Z., Zhang, J., Gong, H., Liu, X., Wang, S., & Zhou, J. (2026). Interpretable Transformer-Based Voltage Degradation Prediction of Proton Exchange Membrane Fuel Cells Under Constant-Current Operation. Electronics, 15(15), 3334. https://doi.org/10.3390/electronics15153334

Vancouver

Yi F, Shu X, Zhang J, Huang Z, Zhang J, Gong H, et al. Interpretable Transformer-Based Voltage Degradation Prediction of Proton Exchange Membrane Fuel Cells Under Constant-Current Operation. Electronics. 2026;15(15):3334. doi:10.3390/electronics15153334

BibTeX
@article{Yi2026,
  author = {Yi, Fengyan and Shu, Xing and Zhang, Jinming and Huang, Zongjing and Zhang, Junling and Gong, Hongtao and Liu, Xiangya and Wang, Shuaihua and Zhou, Jiaming},
  title = {Interpretable Transformer-Based Voltage Degradation Prediction of Proton Exchange Membrane Fuel Cells Under Constant-Current Operation},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {15},
  pages = {3334},
  doi = {10.3390/electronics15153334}
}
Kostenlos starten →