Deep learning based prognostic framework towards proton exchange membrane fuel cell for automotive application

Abstract Currently, the larger-scaled commercialization of fuel cell technology is considerably impeded by the limited durability of fuel cells. Prognostics and health management (PHM) is one of the most widely researched technologies used to improve the durability of fuel cell devices. More recently, the combination of deep neural network approaches and PHM techniques shows a broad research prospect. Attention mechanisms can enhance their data processing ability, which helps to extract useful features more efficiently. Herein, we propose an attention-based Recurrent neural network (RNN) model to improve the prognostics of PHM, which enables a more accurate prediction of the output voltage degradation of proton exchange membrane fuel cell (PEMFC) based on the original long-term dynamic loading cycle durability test data. In particular, the prediction results with different prediction models, namely, long short-term memory (LSTM), gated recurrent unit (GRU), attention-based LSTM, and attention-based GRU are obtained and compared. For dynamic test data (dataset 1), the root mean square error results for the attention-based LSTM and GRU models are 0.016409 and 0.015518, respectively, whereas for the LSTM and GRU model the corresponding error results are 0.017637 and 0.018206, respectively. The same effects are demonstrated and proved for the pseudo–steady dataset (dataset 2). The attention-based RNN model achieves a high prediction accuracy, proving that it can help improve the prediction accuracy and may further help the implementation of PHM in the fuel cell system.

[1]  Noureddine Zerhouni,et al.  Phm of Proton-exchange Membrane Fuel Cells - a Review , 2013 .

[2]  Noureddine Zerhouni,et al.  Prognostics of Proton Exchange Membrane Fuel Cells stack using an ensemble of constraints based connectionist networks , 2016 .

[3]  Jian Chen,et al.  Remaining useful life estimation for proton exchange membrane fuel cells using a hybrid method , 2019, Applied Energy.

[4]  Sepp Hochreiter,et al.  The Vanishing Gradient Problem During Learning Recurrent Neural Nets and Problem Solutions , 1998, Int. J. Uncertain. Fuzziness Knowl. Based Syst..

[5]  Tong Zhang,et al.  A review of automotive proton exchange membrane fuel cell degradation under start-stop operating condition , 2018 .

[6]  Shixiang Xia,et al.  The application of orthogonal test method in the parameters optimization of PEMFC under steady working condition , 2016 .

[7]  Shaker Haji,et al.  Analytical modeling of PEM fuel cell i–V curve , 2011 .

[8]  Dario R. Dekel.  Review of cell performance in anion exchange membrane fuel cells , 2018 .

[9]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[10]  Guo-Bin Jung,et al.  Study of high voltage applied to the membrane electrode assemblies of proton exchange membrane fuel cells as an accelerated degradation technique , 2011 .

[11]  Jürgen Schmidhuber,et al.  Learning to Forget: Continual Prediction with LSTM , 2000, Neural Computation.

[12]  Anusorn Kongkanand,et al.  The Priority and Challenge of High-Power Performance of Low-Platinum Proton-Exchange Membrane Fuel Cells. , 2016, The journal of physical chemistry letters.

[13]  Israel Cohen,et al.  A deep architecture for audio-visual voice activity detection in the presence of transients , 2018, Signal Process..

[14]  Noureddine Zerhouni,et al.  Degradations analysis and aging modeling for health assessment and prognostics of PEMFC , 2016, Reliab. Eng. Syst. Saf..

[15]  Abdellatif Miraoui,et al.  Degradation prediction of PEM fuel cell using a moving window based hybrid prognostic approach , 2017 .

[16]  Huicui Chen,et al.  Lifetime prediction and the economic lifetime of Proton Exchange Membrane fuel cells , 2015 .

[17]  Weirong Chen,et al.  Remaining useful life prediction of PEMFC based on long short-term memory recurrent neural networks , 2019, International Journal of Hydrogen Energy.

[18]  Huicui Chen,et al.  Main factors affecting the lifetime of Proton Exchange Membrane fuel cells in vehicle applications: A review , 2014 .

[19]  Georgios Tsotridis,et al.  A comparison of Fuel Cell Testing protocols – A case study: Protocols used by the U.S. Department of Energy, European Union, International Electrotechnical Commission/Fuel Cell Testing and Standardization Network, and Fuel Cell Technical Team , 2013 .

[20]  Hui Li,et al.  A review of polymer electrolyte membrane fuel cell durability test protocols , 2011 .

[21]  Noureddine Zerhouni,et al.  Proton exchange membrane fuel cell ageing forecasting algorithm based on Echo State Network , 2017 .

[22]  Hongye Su,et al.  Data-based short-term prognostics for proton exchange membrane fuel cells , 2017 .

[23]  Walter Sextro,et al.  PEM fuel cell prognostics using particle filter with model parameter adaptation , 2014, 2014 International Conference on Prognostics and Health Management.

[24]  Salah Laghrouche,et al.  Degradation model of proton exchange membrane fuel cell based on a novel hybrid method , 2019, Applied Energy.

[25]  Razvan Pascanu,et al.  Understanding the exploding gradient problem , 2012, ArXiv.

[26]  Rachid Outbib,et al.  Adaptive Prognostic of Fuel Cells by Implementing Ensemble Echo State Networks in Time-Varying Model Space , 2020, IEEE Transactions on Industrial Electronics.

[27]  Maher A.R. Sadiq Al-Baghdadi,et al.  Modelling of proton exchange membrane fuel cell performance based on semi-empirical equations , 2005 .

[28]  Nicu Bizon,et al.  Energy efficiency of multiport power converters used in plug-in/V2G fuel cell vehicles , 2012 .

[29]  Jun Shen,et al.  A review of PEM fuel cell durability: Degradation mechanisms and mitigation strategies , 2008 .

[30]  Jian Chen,et al.  Prognostics of Proton Exchange Membrane Fuel Cells Using A Model-based Method , 2017 .

[31]  Fei Gao,et al.  Data-driven proton exchange membrane fuel cell degradation predication through deep learning method , 2018, Applied Energy.

[32]  Suk Joo Bae,et al.  Degradation pattern prediction of a polymer electrolyte membrane fuel cell stack with series reliability structure via durability data of single cells , 2014 .

[33]  Alberto Traverso,et al.  Operating strategies to minimize degradation in fuel cell gas turbine hybrids , 2017 .

[34]  Walter Mérida,et al.  Designed experiments to characterize PEMFC material properties and performance , 2014 .

[35]  Daniel Hissel,et al.  Proton exchange membrane fuel cell degradation prediction based on Adaptive Neuro-Fuzzy Inference Systems . , 2014 .

[36]  R. Mukundan,et al.  Spatially resolved degradation during startup and shutdown in polymer electrolyte membrane fuel cell operation , 2019, Applied Energy.

[37]  Ki-Young Kim,et al.  Optimal operation of a 1-kW PEMFC-based CHP system for residential applications , 2012 .

[38]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[39]  Mathias Schulze,et al.  Evaluation of reversible and irreversible degradation rates of polymer electrolyte membrane fuel cells tested in automotive conditions , 2016 .

[40]  Christophe Varnier,et al.  Decision process to manage useful life of multi-stacks fuel cell systems under service constraint , 2017 .

[41]  Noureddine Zerhouni,et al.  Prognostics of PEM fuel cell in a particle filtering framework , 2014 .

[42]  T. Morawietz,et al.  Local impact of load cycling on degradation in polymer electrolyte fuel cells , 2020, Applied Energy.

[43]  Ke Zhang,et al.  Online remaining useful lifetime prediction of proton exchange membrane fuel cells using a novel robust methodology , 2018, Journal of Power Sources.