ABSTRACT
In order to address the challenge of predicting the Remaining Useful Life (RUL) of Proton Exchange Membrane Fuel Cell (PEMFC), this paper proposes an approach utilizing a Convolutional Neural Network and Bidirectional Long Short-Term Memory (CNN-BiLSTM). The CNN would be employed to capture the effective characteristics of fuel cells, while the LSTM neural network would be utilized to forecast the trend of low-frequency components, thereby achieving accurate predictions of PEMFC performance degradation. Model prediction accuracy evaluation and the formula for remaining useful life prediction error (FE) were chosen as the evaluation metrics for remaining life prediction. The proposed CNN-BiLSTM algorithm and the LSTM algorithm were comprehensively analyzed both theoretically and experimentally. Results indicated that the CNN-BiLSTM method exhibits advantages such as low computational complexity, high precision, and short computational time, validating the feasibility and effectiveness of the CNN-BiLSTM method for online prediction of PEMFC life.
- Lan Yu, Wu Zhankuan, Jiang Qi, Research on proton exchange membrane fuel cell life prediction based on machine learning [J]. Modern Machinery, 2022(05): 1-5.Google Scholar
- Liu DC, Lin R, Feng B, Investigation of the effect of cathode stoichiometry of proton exchange membrane fuel cell using localized electrochemical impedance spectroscopy based on print circuit board[J]. International Journal of Hydrogen Energy, 2019, 44: 7564 -7573.Google ScholarCross Ref
- Yin Liangzhen, Li Qi, Hong Zhihu, PEMFC power generation system FFRLS online identification and real-time optimal temperature generalized predictive control method [J]. Chinese Journal of Electrical Engineering, 2017, 37(11):3223-3235.Google Scholar
- Deng Huiwen, Li Qi, Chen Weirong. Research on HOSM observer suitable for peroxide estimation in PEMFC systems [J]. Chinese Journal of Electrical Engineering, 2017, 37(17):5058-5068.Google Scholar
- Silva RE, Gouriveau R, Jemei S, Proton exchange membrane fuel cell degradation prediction based on adaptive neuro-fuzzy inference systems[J]. International Journal of Hydrogen Energy, 2014, 39(21): 11128-11144.Google ScholarCross Ref
- Yuan H, Dai H, Wei X, Model-based observers for internal states estimation and control of proton exchange membrane fuel cell system: a review[J]. Journal of Power Sources, 2020, 468: No. 228376.Google ScholarCross Ref
- Chen HC, Xu SC, Pei PC, Mechanism analysis of starvation in PEMFC based on external characteristics[J]. International Journal of Hydrogen Energy, 2019, 44(11): 5437-5446.Google ScholarCross Ref
- Hu Zunyan, Xu Liangfei, Li Jianqiu, A reconstructed fuel cell life-prediction model for a fuel cell hybrid city bus[J]. Energy Conversion and Management, 2018, 156: 723-732.Google ScholarCross Ref
- Lu L G, Ouyang M G, Huang H Y, A semiempirical voltage degradation model for a low-pressure proton exchange membrane fuel cell stack under bus city driving cycles[J]. Journal of Power Sources, 2007, 164(1): 306-314.Google ScholarCross Ref
- Jouin M, Gouriveau R, Hissel D, Degradations analysis and aging modeling for health assessment and prognostics of PEMFC[J]. Reliability Engineering & System Safety, 2016, 148: 78-95.Google ScholarCross Ref
- Singh Y, Khorasany R M, Kim W, Ex Situ Characterization and Modelling of Fatigue Crack Propagation in Catalyst Coated Membrane Composites for Fuel Cell Applications[J]. International Journal of Hydrogen Energy, 2019, 44(23): 12057-12072.Google ScholarCross Ref
- Chang Y F, Zhao J, Shahgaldi S, Modelling of Mechanical Microstructure Changes in the Catalyst Layer of a Polymer Electrolyte Membrane Fuel Cell[J]. International Journal of Hydrogen Energy, 2020, 45(54): 29904-29916.Google ScholarCross Ref
- Javed K, Gouriveau R, Zerhouni N, Prognostics of proton exchange membrane fuel cells stack using an ensemble of constraints-based connectionist networks[J]. Journal Power Sources, 2016, 324:745-757.Google ScholarCross Ref
- Silva R E, Gouriveau R, Jemei S, Proton exchange membrane fuel cell degradation prediction based on adaptive Neuro-Fuzzy inference systems[J]. International Journal of Hydrogen Energy, 2014, 39(21): 11128-11144.Google ScholarCross Ref
- Morando S, Jemei S, Hissel D, Proton exchange membrane fuel cell ageing forecasting algorithm based on echo state network[J]. International Journal of Hydrogen Energy, 2017, 42(2): 1472-1480.Google ScholarCross Ref
- Wu YM, Breaz E, Gao F, Nonlinear performance degradation prediction of proton exchange membrane fuel cells using relevance vector machine [J]. IEEE Trans Energy Convers, 2016, 31(4): 1570-1582.Google ScholarCross Ref
- Ma R, Yang T, Breaz E, Data-driven proton exchange membrane fuel cell degradation prediction through deep learning method[J]. Applied Energy, 2018, 231: 102-115.Google ScholarCross Ref
- Gao Jinwu, Jia Zhihuan, Wang Xiangyang, Degradation trend prediction of proton exchange membrane fuel cells based on PSO-LSTM [J]. Journal of Jilin University (Engineering Edition), 2022, 52(09): 2192-2202.Google Scholar
- Zhu Zhenyu, Gao Dexin. Lithium-ion battery health status detection method based on CNN-BiLSTM network [J]. Electronic Measurement Technology, 2023, 46(03): 128-133.Google Scholar
- Mao, Lei; Jackson, Lisa (2016). IEEE 2014 Data Challenge Data. Loughborough University. Dataset. https://doi.org/10.17028/rd.lboro.3518141.v1Google ScholarCross Ref
- Zuo B, Cheng J S, Zhang Z H. Degradation prediction model for proton exchange membrane fuel cells based on long short-term memory neural network and Savitzky-Golay filter[J]. International Journal of Hydrogen Energy, 2021, 46(29): 15928-15937.Google ScholarCross Ref
- Hua Z , Zheng Z , Pahon E , A review on lifetime prediction of proton exchange membrane fuel cells system[J]. Journal of Power Sources, 2022(May 1): 529.Google Scholar
- Hua Z, Zheng Z, Gao F, Remaining useful life prediction of PEMFC systems based on the multi-input echo state network[J]. Applied Energy, 2020, 265.Google Scholar
Recommendations
A Remaining Useful Life Prediction Method for Lithium-ion Battery Based on Temporal Transformer Network
AbstractThe remaining useful life prediction is significant for Lithium-ion batteries to ensure safety and reliability. Due to the advantages of handling time sequence data, recurrent neural network based methods have achieved impressive performance on ...
Remaining Useful Life Prediction for Aero-Engine based on LSTM-HMM1
EITCE '21: Proceedings of the 2021 5th International Conference on Electronic Information Technology and Computer EngineeringThe remaining useful life (RUL) prediction for aero-engine is very complex. There are complex nonlinear characteristics among state variables. Moreover, it is also affected by external factors such as operating environment, which makes it difficult for ...
A two-stage integrated method for early prediction of remaining useful life of lithium-ion batteries
AbstractThis article puts forward a two-stage integrated method to predict the remaining useful life (RUL) of lithium-ion batteries (LIBs). At the first stage, a convolutional neural network (CNN) is employed to preliminarily estimate the ...
Comments