Journal of Zhejiang University SCIENCE  A

Accepted manuscript available online (unedited version)


Real-time degradation modeling for automotive PEMFC stacks: a multi-scale fusion network validated on an industrial 215-channel system


Author(s):  Zifei WANG1, 2, Xiangxian ZHU3, Congxin LI4, Daidai CHEN3, Zhitao LIU2, Longhua MA1, Jili TAO1, Hongye SU2

Affiliation(s):  1NingboTech University, Ningbo 315100, China 2State Key Laboratory of Industrial Control Technology, Zhejiang University, Hangzhou 310027, China 3Ningbo Joyson Electronic Corp, Ningbo 315040, China 4Ningbo Green Power Hydrogen Technology Research Institute Co., Ltd. Zhejiang 315033, China

Corresponding email(s):  Jili TAO, taojili@nbt.edu.cn

Key Words:  Large-scale PEMFC stacks; Multi-scale bidirectional fusion network; Degradation prediction; Automotive scenarios


Share this article to： More <<< Previous Paper|Next Paper >>>

Zifei WANG1,2, Xiangxian ZHU3, Congxin LI4, Daidai CHEN3, Zhitao LIU2, Longhua MA1, Jili TAO1, Hongye SU2. Real-time degradation modeling for automotive PEMFC stacks: a multi-scale fusion network validated on an industrial 215-channel system[J]. Journal of Zhejiang University Science A,in press.Frontiers of Information Technology & Electronic Engineering,in press.https://doi.org/10.1631/jzus.A2500337

@article{title="Real-time degradation modeling for automotive PEMFC stacks: a multi-scale fusion network validated on an industrial 215-channel system",
author="Zifei WANG1,2, Xiangxian ZHU3, Congxin LI4, Daidai CHEN3, Zhitao LIU2, Longhua MA1, Jili TAO1, Hongye SU2",
journal="Journal of Zhejiang University Science A",
year="in press",
publisher="Zhejiang University Press & Springer",
doi="https://doi.org/10.1631/jzus.A2500337"
}

%0 Journal Article
%T Real-time degradation modeling for automotive PEMFC stacks: a multi-scale fusion network validated on an industrial 215-channel system
%A Zifei WANG1
%A 2
%A Xiangxian ZHU3
%A Congxin LI4
%A Daidai CHEN3
%A Zhitao LIU2
%A Longhua MA1
%A Jili TAO1
%A Hongye SU2
%J Journal of Zhejiang University SCIENCE A
%P
%@ 1673-565X
%D in press
%I Zhejiang University Press & Springer
doi="https://doi.org/10.1631/jzus.A2500337"

TY - JOUR
T1 - Real-time degradation modeling for automotive PEMFC stacks: a multi-scale fusion network validated on an industrial 215-channel system
A1 - Zifei WANG1
A1 - 2
A1 - Xiangxian ZHU3
A1 - Congxin LI4
A1 - Daidai CHEN3
A1 - Zhitao LIU2
A1 - Longhua MA1
A1 - Jili TAO1
A1 - Hongye SU2
J0 - Journal of Zhejiang University Science A
SP -
EP -
%@ 1673-565X
Y1 - in press
PB - Zhejiang University Press & Springer
ER -
doi="https://doi.org/10.1631/jzus.A2500337"


Abstract: Accurately predicting long-term degradation patterns in proton exchange membrane fuel cell stacks (PEMFCs) under automotive operating conditions remains challenging. Prediction methods are largely constrained by laboratory-scale experiments and limited stack sizes, resulting in insufficient accuracy and generalization capability. To address these limitations, in this paper we propose a multi-scale bidirectional fusion network (MBFNet) tailored for an industrial 215-channel PEMFC stack, enabling accurate degradation prediction under accelerated real-world dynamic conditions using gas-heat-electricity (GHE) co-simulation data. A channel-joint adaptive noise correlation threshold (NCT) algorithm is introduced to account for variable correlations across sensors and operating conditions without relying on prior physical modeling. A multi-scale decomposition module captures degradation dynamics at different temporal scales, while a bidirectional fusion module integrates global trends and local details into the final prediction. Experimental results showed that MBFNet achieved 18.5% lower prediction error and 36.8% fewer parameters than the LSTM-Attention benchmark under real operating scenarios. In multi-step prediction tasks, MBFNet reduced error by 9% relative to LSTM-Attention and by 40.5% relative to 1D-CNN, better satisfying automotive application requirements. Moreover, MBFNet exhibits strong physical interpretability, making it efficient to implement and promising for practical deployment.

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

Open peer comments: Debate/Discuss/Question/Opinion

<1>

Please provide your name, email address and a comment





Full Text:  <336>

CLC number: 

On-line Access: 2026-03-23

Received: 2025-07-22

Revision Accepted: 2025-12-16

Crosschecked: 0000-00-00

Cited: 0

Clicked: 424

Citations:  Bibtex RefMan EndNote GB/T7714

Journal of Zhejiang University-SCIENCE, 38 Zheda Road, Hangzhou 310027, China
Tel: +86-571-87952783; E-mail: cjzhang@zju.edu.cn
Copyright © 2000 - 2026 Journal of Zhejiang University-SCIENCE