Deep Learning-Based State-of-Health Estimation of Proton-Exchange Membrane Fuel Cells under Dynamic Operation Conditions
Abstract
:1. Introduction
- A dynamic load cycle durability test for PEMFCs was developed. The experimental procedure included designing a suitable dynamic load cycle, determining the optimal operating parameters, and performing dynamic load durability testing on two fuel cells. The durability tests lasted for over 550 h, with all data being collected using a G20 test station.
- Four deep learning models were selected to perform the prediction tasks. These models were LSTM, GRU, TCN, and transformer, and were utilized to predict the degradation of the fuel cells. The predictive performance of these different models was thoroughly evaluated.
- The impact of durability data resampling intervals and model training data proportions on prediction accuracy was investigated. This study was validated and compared using two durability datasets.
2. Experimental
2.1. Preparation of Membrane Electrode Assembly
2.2. PEMFC Dynamic Durability Test
3. Materials and Methods
3.1. Singular Spectrum Analysis
3.2. Long Short-Term Memory
3.3. Gated Recurrent Unit
3.4. Temporal Convolutional Network
3.5. Transformer
3.6. Evaluation Criteria of Prediction Performance
4. Results and Discussion
4.1. Data Analysis
4.2. Data Resampling Intervals
4.3. Training Dataset Size
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Le, T.; Sharma, P.; Bora, B.; Tran, V.; Truong, T.; Le, H.; Nguyen, P. Fueling the future: A comprehensive review of hydrogen energy systems and their challenges. Int. J. Hydrogen Energy 2024, 54, 791–816. [Google Scholar] [CrossRef]
- Boretti, A.; Castelletto, S. Hydrogen energy storage requirements for solar and wind energy production to account for long-term variability. Renew. Energy 2024, 221, 119797. [Google Scholar] [CrossRef]
- Hassan, Q.; Sameen, A.; Salman, H.; Jaszczur, M.; Al-Jiboory, A. Hydrogen energy future: Advancements in storage technologies and implications for sustainability. J. Energy Storage 2023, 72, 108404. [Google Scholar] [CrossRef]
- Agyekum, E.; Odoi-Yorke, F.; Abbey, A.; Ayetor, G. A review of the trends, evolution, and future research prospects of hydrogen fuel cells—A focus on vehicles. Int. J. Hydrogen Energy 2024, 72, 918–939. [Google Scholar] [CrossRef]
- Tang, X.; Zhang, Y.; Xu, S. Temperature sensitivity characteristics of PEM fuel cell and output performance improvement based on optimal active temperature control. Int. J. Heat Mass Transf. 2023, 206, 123966. [Google Scholar] [CrossRef]
- Halder, P.; Babaie, M.; Salek, F.; Shah, K.; Stevanovic, S.; Bodisco, T.; Zare, A. Performance, emissions and economic analyses of hydrogen fuel cell vehicles. Renew. Sustain. Energy Rev. 2024, 199, 114543. [Google Scholar] [CrossRef]
- Tang, X.; Yang, M.; Shi, L.; Hou, Z.; Xu, S.; Sun, C. Adaptive state-of-health temperature sensitivity characteristics for durability improvement of PEM fuel cells. Chem. Eng. J. 2024, 491, 151951. [Google Scholar] [CrossRef]
- Liu, Z.; Cai, S.; Tu, Z.; Chan, S. Recent development in degradation mechanisms of proton exchange membrane fuel cells for vehicle applications: Problems, progress, and perspectives. Energy Storage Sav. 2024, 3, 106–152. [Google Scholar] [CrossRef]
- Allal, Z.; Noura, H.; Chahine, K. Efficient health indicators for the prediction of the remaining useful life of proton exchange membrane fuel cells. Energy Convers. Manag. X 2024, 21, 100503. [Google Scholar] [CrossRef]
- Zuo, J.; Steiner, N.; Li, Z.; Hissel, D. Health management review for fuel cells: Focus on action phase. Renew. Sustain. Energy Rev. 2024, 20, 114613. [Google Scholar] [CrossRef]
- Lv, J.; Yu, Z.; Zhang, H.; Sun, G.; Muhl, P.; Liu, J. Transformer Based Long-Term Prognostics for Dynamic Operating PEM Fuel Cells. IEEE Trans. Transp. Electrif. 2023, 10, 1747–1757. [Google Scholar] [CrossRef]
- Zhang, Y.; Ma, R.; Xie, R.; Feng, Z.; Liang, B.; Li, Y. A Degradation Prediction Method for PEM Fuel Cell Based on Deep Temporal Feature Extraction and Transfer Learning. IEEE Trans. Transp. Electrif. 2024, 10, 203–212. [Google Scholar] [CrossRef]
- Wang, C.; Dou, M.; Li, Z.; Outbib, R.; Zhao, D.; Zuo, J.; Wang, Y.; Liang, B.; Wang, P. Data-driven prognostics based on time-frequency analysis and symbolic recurrent neural network for fuel cells under dynamic load. Reliab. Eng. Syst. Saf. 2023, 233, 109123. [Google Scholar] [CrossRef]
- Pan, M.; Hu, P.; Gao, R.; Liang, K. Multistep prediction of remaining useful life of proton exchange membrane fuel cell based on temporal convolutional network. Int. J. Green Energy 2023, 20, 408–422. [Google Scholar] [CrossRef]
- Lv, J.; Shen, X.; Gao, Y.; Liu, J.; Sun, G. The seasonal-trend disentangle based prognostic framework for PEM fuel cells. Renew. Energy 2024, 228, 120648. [Google Scholar] [CrossRef]
- Morizet, N.; Desforges, P.; Geissler, C.; Pahon, E.; Jemei, S.; Hissel, D. An adaptative approach for estimating the remaining useful life of a heavy-duty fuel cell vehicle. J. Power Sources 2024, 597, 234152. [Google Scholar] [CrossRef]
- Chen, D.; Wu, W.; Chang, K.; Li, Y.; Pei, P.; Xu, X. Performance degradation prediction method of PEM fuel cells using bidirectional long short-term memory neural network based on Bayesian optimization. Energy 2023, 285, 129469. [Google Scholar] [CrossRef]
- Wan, P.; Liu, H.; Chen, J.; Qin, X.; Lehnert, W.; Shao, Z.; Li, R. A novel degradation model of proton exchange membrane fuel cells for state of health estimation and prognostics. Int. J. Hydrogen Energy 2021, 46, 31353–31361. [Google Scholar]
- Ou, M.; Zhang, R.; Shao, Z. A novel approach based on semi-empirical model for degradation prediction of fuel cells. J. Power Sources 2021, 488, 229435. [Google Scholar] [CrossRef]
- Pei, P.; Meng, Y.; Chen, D.; Ren, P.; Wang, M.; Wang, X. Lifetime prediction method of proton exchange membrane fuel cells based on current degradation law. Energy 2023, 265, 126341. [Google Scholar] [CrossRef]
- Yang, Y.; Yu, X.; Zhu, W.; Xie, C.; Zhao, B.; Zhang, L.; Shi, C.; Huang, L.; Zhang, R. Degradation prediction of proton exchange membrane fuel cells with model uncertainty quantification. Renew. Energy 2023, 219, 119525. [Google Scholar] [CrossRef]
- Zhao, Y.; Luo, M.; Yang, J.; Chen, B.; Sui, P. Numerical analysis of PEMFC stack performance degradation using an empirical approach. Int. J. Hydrogen Energy 2024, 56, 147–163. [Google Scholar] [CrossRef]
- Tang, X.; Qin, X.; Wei, K.; Xu, S. A novel online degradation model for proton exchange membrane fuel cell based on online transfer learning. Int. J. Hydrogen Energy 2023, 48, 13617–13632. [Google Scholar] [CrossRef]
- Wilberforce, T.; Alaswad, A.; Garcia-Perez, A.; Xu, Y.; Ma, X.; Panchev, C. Remaining useful life prediction for proton exchange membrane fuel cells using combined convolutional neural network and recurrent neural network. Int. J. Hydrogen Energy 2023, 48, 291–303. [Google Scholar] [CrossRef]
- Sun, B.; Liu, X.; Wang, J.; Wei, X.; Yuan, H.; Dai, H. Short-term performance degradation prediction of a commercial vehicle fuel cell system based on CNN and LSTM hybrid neural network. Int. J. Hydrogen Energy 2023, 48, 8613–8628. [Google Scholar] [CrossRef]
- Yue, M.; Zhang, X.; Teng, T.; Meng, J.; Pahon, E. Recursive performance prediction of automotive fuel cell based on conditional time series forecasting with convolutional neural network. Int. J. Hydrogen Energy 2023, 56, 248–258. [Google Scholar] [CrossRef]
- Teng, T.; Zhang, X.; Yue, M.; Lv, Q.; Li, C. Research of proton exchange membrane fuel cell degradation index and prediction method for automotive vehicles. Sustain. Energy Technol. Assess. 2024, 65, 103789. [Google Scholar] [CrossRef]
- He, K.; Liu, Z.; Sun, Y.; Mao, L.; Lu, S. Degradation prediction of proton exchange membrane fuel cell using auto-encoder based health indicator and long short-term memory network. Int. J. Hydrogen Energy 2022, 47, 35055–35067. [Google Scholar] [CrossRef]
- Fu, S.; Zhang, D.; Xiao, Y.; Zheng, C. A non-stationary transformer-based remaining useful life prediction method for proton exchange membrane fuel cells. Int. J. Hydrogen Energy 2024, 60, 1121–1133. [Google Scholar] [CrossRef]
- Zhou, N.; Cui, B.; Zhou, J. Transformer-based prediction of the RUL of PEMFC. In Proceedings of the 2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS), Guangzhou, China, 22–24 July 2022; pp. 345–348. [Google Scholar]
- Hong, J.; Yang, H.; Liang, F.; Li, K.; Zhang, X.; Zhang, H.; Zhang, C.; Yang, Q.; Wang, J. State of health prediction for proton exchange membrane fuel cells combining semi-empirical model and machine learning. Energy 2024, 291, 130364. [Google Scholar] [CrossRef]
- Wang, Y.; Wu, K.; Zhao, H.; Li, J.; Sheng, X.; Yin, Y.; Du, Q.; Zu, B.; Han, L.; Jiao, K. Degradation prediction of proton exchange membrane fuel cell stack using semi-empirical and data-driven methods. Energy AI 2023, 11, 100205. [Google Scholar] [CrossRef]
- Yue, M.; Li, Z.; Roche, R.; Jemei, S.; Zerhoumi, N. Degradation identification and prognostics of proton exchange membrane fuel cell under dynamic load. Control Eng. Pract. 2022, 118, 104959. [Google Scholar] [CrossRef]
- Xiao, F.; Shao, M. Electrocatalyst and electrode design strategies for durable proton exchange membrane fuel cells. Matter 2024, 7, 351–377. [Google Scholar] [CrossRef]
- Zamel, N. The catalyst layer and its dimensionality—A look into its ingredients and how to characterize their effects. J. Power Sources 2016, 309, 141–159. [Google Scholar] [CrossRef]
- Sun, C.; Negro, E.; Nale, A.; Pagot, G.; Vezzu, K.; AZawodzinski TMeda, L.; Gambaro, C.; Noto, V. An efficient barrier toward vanadium crossover in redox flow batteries: The bilayer [Nafion/(WO3)x] hybrid inorganic-organic membrane. Electrochim. Acta 2021, 378, 138133. [Google Scholar] [CrossRef]
- Tsotridis, G.; Pilenga, A.; Marco, G. EU Harmonised Test Protocols for PEMFC MEA Testing in Single Cell Configuration for Automotive Applications; JRC Science for Policy Report; EUR 27632 EN; Publications Office of the European Union: Luxembourg, 2015. [Google Scholar]
- Zuo, J.; Lv, H.; Zhou, D.; Xue, Q.; Jin, L.; Zhou, W.; Yang, D.; Zhang, C. Deep learning based prognostic framework towards proton exchange membrane fuel cell for automotive application. Appl. Energy 2021, 281, 115937. [Google Scholar] [CrossRef]
- Broomhead, D.; King, G. Extracting qualitative dynamics from experimental data. Phys. D Nonlinear Phenom. 1986, 20, 217–236. [Google Scholar] [CrossRef]
- Zhang, Y.; Liu, Z.; Yang, C.; Huang, X.; Lou, S.; Zhang, H.; Yan, D. Unveiling dynamics changes: Singular spectrum analysis-based method for detecting concept drift in industrial data streams. Knowl. Based Syst. 2024, 293, 111640. [Google Scholar] [CrossRef]
- Yu, Y.; Si, X.; Hu, C.; Zhang, J. A Review of Recurrent Neural Networks: LSTM Cells and Network Architectures. Neural Comput. 2019, 31, 1235–1270. [Google Scholar] [CrossRef]
- Chen, Y.; Zhang, J.; Zhai, S.; Hu, Z. Data-driven modeling and fault diagnosis for fuel cell vehicles using deep learning. Energy AI 2024, 16, 100345. [Google Scholar] [CrossRef]
- Yamak, P.; Li, Y.; Gadosey, P. A Comparison between ARIMA, LSTM, and GRU for Time Series Forecasting. In Proceedings of the ACAI ‘19’: Proceedings of the 2019 2nd International Conference on Algorithms, Sanya, China, 20–22 December 2019. [Google Scholar]
- Izadi, M.; Hassani, P.; Raeesi, M. A novel WaveNet-GRU deep learning model for PEM fuel cells degradation prediction based on transfer learning. Energy 2024, 293, 130602. [Google Scholar] [CrossRef]
- Bai, S.; Kolter, Z.; Koltun, V. An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling. arXiv 2018, arXiv:1803.01271. [Google Scholar]
- Sasha, T.; Almeida, D.; Lyman, K. Resnet in resnet: Generalizing residual architectures. arXiv 2016, arXiv:1603.08029. [Google Scholar]
- Vaswani, A.; Shazeer, N.; Parma, N. Attention Is All You Need. In Proceedings of the Advances in Neural Information Processing Systems 30, Long Beach, CA, USA, 4–9 December 2017. [Google Scholar]
- Chitty-Venkata, K.; Mittal, S.; Emani, M.; Vishwanath, V.; Somani, A. A survey of techniques for optimizing transformer inference. J. Syst. Archit. 2023, 144, 102990. [Google Scholar] [CrossRef]
- He, W.; Liu, T.; Ming, W.; Li, Z.; Du, J.; Li, X.; Guo, X.; Sun, P. Progress in prediction of remaining useful life of hydrogen fuel cells based on deep learning. Renew. Sustain. Energy Rev. 2024, 192, 114193. [Google Scholar] [CrossRef]
- Song, K.; Huang, X.; Huang, P.; Sun, H.; Chen, Y.; Huang, D. Data-driven health state estimation and remaining useful life prediction of fuel cells. Renew. Energy 2024, 227, 120491. [Google Scholar] [CrossRef]
Technical Parameters | Value |
---|---|
Activated area | 5 × 5 cm2 |
Thickness of membrane | 12 μm |
Thickness of GDL | 220 μm |
Thickness of CL | 15 μm |
Radius of Pt particle | 5 nm |
Pt loading of the anode | 0.05 mg·cm−2 |
Pt loading of the cathode | 0.3 mg·cm−2 |
RMSE (20 min) | RMSE (40 min) | RMSE (60 min) | ||
---|---|---|---|---|
PEMFC without cracks | LSTM | 0.01602 | 0.02546 | 0.04539 |
GRU | 0.02495 | 0.02574 | 0.05289 | |
TCN | 0.01689 | 0.03968 | 0.04866 | |
Transformer | 0.05242 | 0.05017 | 0.08404 | |
PEMFC with uniform cracks | LSTM | 0.00934 | 0.02923 | 0.04423 |
GRU | 0.0128 | 0.03297 | 0.03911 | |
TCN | 0.0234 | 0.03908 | 0.05153 | |
Transformer | 0.03273 | 0.03424 | 0.05219 |
RMSE (70%) | RMSE (50%) | RMSE (30%) | ||
---|---|---|---|---|
PEMFC without cracks | LSTM | 0.00532 | 0.00607 | 0.01602 |
GRU | 0.00546 | 0.00831 | 0.02495 | |
TCN | 0.00867 | 0.01516 | 0.01689 | |
Transformer | 0.0223 | 0.03349 | 0.05242 | |
PEMFC with uniform cracks | LSTM | 0.00533 | 0.00857 | 0.00934 |
GRU | 0.01073 | 0.01242 | 0.0128 | |
TCN | 0.00922 | 0.01604 | 0.02341 | |
Transformer | 0.02101 | 0.03053 | 0.03273 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zhang, Y.; Tang, X.; Xu, S.; Sun, C. Deep Learning-Based State-of-Health Estimation of Proton-Exchange Membrane Fuel Cells under Dynamic Operation Conditions. Sensors 2024, 24, 4451. https://doi.org/10.3390/s24144451
Zhang Y, Tang X, Xu S, Sun C. Deep Learning-Based State-of-Health Estimation of Proton-Exchange Membrane Fuel Cells under Dynamic Operation Conditions. Sensors. 2024; 24(14):4451. https://doi.org/10.3390/s24144451
Chicago/Turabian StyleZhang, Yujia, Xingwang Tang, Sichuan Xu, and Chuanyu Sun. 2024. "Deep Learning-Based State-of-Health Estimation of Proton-Exchange Membrane Fuel Cells under Dynamic Operation Conditions" Sensors 24, no. 14: 4451. https://doi.org/10.3390/s24144451
APA StyleZhang, Y., Tang, X., Xu, S., & Sun, C. (2024). Deep Learning-Based State-of-Health Estimation of Proton-Exchange Membrane Fuel Cells under Dynamic Operation Conditions. Sensors, 24(14), 4451. https://doi.org/10.3390/s24144451