Life Prediction Based on D-S ELM for PEMFC
Abstract
:1. Introduction
2. Experimental Data
2.1. Raw Experimental Data
- (1)
- The size of monitoring data is large. For better analysis, data preprocessing is applied.
- (2)
- The data set only includes the operation parameters (e.g., inlet hydrogen pressure) and the output parameters (e.g., voltage), but not the internal aging parameters (e.g., catalyst active surface area) of PEMFC. Considering the difficulty of establishing an accurate physical model, the data-driven method is a promising solution.
- (3)
- The data set has the characteristic of non-linear and non-Gauss disturbance, further data tracking in time series contributes to ensuring the accuracy of prediction.
2.2. Data Preprocessing
3. D-S ELM Methodology
3.1. Discrete Wavelet Transform
3.2. Self-Adaptive Differential Evolutionary Algorithm
3.3. D-S ELM
4. Life Prediction Based on D-S ELM
4.1. Selection of Input and Output Layer Data
4.2. Selection of the Number of Hidden Layer Nodes
4.3. Selection of Activation Function
4.4. Simulation Results Based on Two Algorithms
4.4.1. Research on Robustness of D-S ELM
4.4.2. Comparison of Predicted Results
4.4.3. Influence of Training Data Size
5. Conclusions
- (1)
- The proposed D-S ELM algorithm has strong robustness. In order to study the uncertainties of data-driven algorithm, 100 simulations provide the RUL probability distribution of different algorithms. D-S ELM has best stability and smallest RUL deviation.
- (2)
- The proposed algorithm has little requirement for the size of training samples which is suitable for small sample prediction.
- (3)
- D-S ELM overcomes the shortcomings of a traditional neural network, such as a large amount of calculation and slow convergence speed.
- (4)
- D-S ELM improves the prediction accuracy of ELM, with the predicted voltage achieving an RMSE of 0.0036 and a ME of 0.0020.
Author Contributions
Funding
Acknowledgments
Conflicts of Interest
References
- Sharaf, O.Z.; Orhan, M.F. An overview of fuel cell technology: Fundamentals and applications. Renew. Sustain. Energy Rev. 2014, 32, 810–853. [Google Scholar] [CrossRef]
- Jahnke, T.; Futter, G.; Latz, A.; Malkow, T. Performance and degradation of proton exchange membrane fuel cells: State of the art in modeling from atomistic to system scale. J. Power Sources 2016, 304, 207–233. [Google Scholar] [CrossRef]
- Breaz, E.; Gao, F.; Miraoui, A.; Tirnovan, R. A short review of aging mechanism modeling of proton exchange membrane fuel cell in transportation applications. In Proceedings of the Conference of the IEEE Industrial Electronics Society, Dallas, TX, USA, 29 October–1 November 2014; pp. 3941–3947. [Google Scholar]
- Jouin, M.; Gouriveau, R.; Hissel, D.; P´era, M.C.; Zerhouni, N. Prognostics and health management of PEMFC–state of the art and remaining challenges. Int. J. Hydrogen Energy 2013, 38, 15307–15317. [Google Scholar] [CrossRef]
- Jouin, M.; Gouriveau, R.; Hissel, D.; P´era, M.C.; Zerhouni, N. Prognostics of PEM Fuel Cell in a Particle Filtering Framework. Int. J. Hydrogen Energy 2014, 39, 481–494. [Google Scholar] [CrossRef]
- Kim, T.; Kim, H.; Ha, J.; Kim, K.; Youn, J.; Jung, J.; Youn, B.D. A degenerated equivalent circuit model and hybrid prediction for state-of-health of PEM fuel cell. In Proceedings of the International Conference on Prognostics and Health Management, Cheney, WA, USA, 22–25 June 2014. [Google Scholar]
- Lechartier, E.; Laffly, E.; Péra, M.C.; Gouriveau, R. Proton exchange membrane fuel cell behavioral model suitable for prognostics. Int. J. Hydrogen Energy 2015, 40, 8384–8397. [Google Scholar] [CrossRef]
- Zhang, X.F.; Yang, D.J.; Luo, M.H.; Dong, Z.M. Load profile based empirical model for the lifetime prediction of an automotive PEM fuel cell. Int. J. Hydrogen Energy 2017, 42, 11868–11878. [Google Scholar] [CrossRef]
- Ibrahim, M.; Steiner, N.Y.; Jemei, S.; Hissel, D. Wavelet-based approach for online fuel cell remaining useful lifetime prediction. IEEE Trans. Ind. Electron. 2016, 63, 5057–5068. [Google Scholar] [CrossRef]
- Silva, R.E.; Gouriveau, R.; Jemeï, S.; Hissel, D.; Boulon, L.; Agbossou, K.; Steiner, N.Y. Proton exchange membrane fuel cell degradation prediction based on adaptive neuro-fuzzy inference systems. Int. J. Hydrogen Energy 2014, 39, 11128–11144. [Google Scholar] [CrossRef]
- Liu, H.; Chen, J.; Hou, M.; Shao, Z.G.; Ye, S.H. Data-based short-term prognostics for proton exchange membrane fuel cells. Int. J. Hydrogen Energy 2017, 42, 20791–20808. [Google Scholar] [CrossRef]
- Morando, S.; Jemei, S.; Gouriveau, R.; Zerhouni, N.; Hissel, D. Fuel cells remaining useful lifetime forecasting using echo state network. In Proceedings of the 39th Annu. Conference of the IEEE Industrial Electronics Society, Coimbra, Portugal, 27–30 October 2014; pp. 1632–1637. [Google Scholar]
- Morando, S.; Jemei, S.; Hissel, D.; Gouriveau, R.; Zerhouni, N. Proton exchange membrane fuel cell ageing forecasting algorithm based on echo state network. Int. J. Hydrogen Energy 2017, 42, 1472–1480. [Google Scholar] [CrossRef]
- Ma, R.; Yang, T.; Breaz, E.; Li, Z.L.; Briois, P.; Gao, F. Data-driven proton exchange membrane fuel cell degradation predication through deep learning method. Appl. Energy 2018, 231, 102–115. [Google Scholar] [CrossRef]
- Javed, K.; Gouriveau, R.; Zerhouni, N.; Hissel, D. Improving accuracy of long-term prognostics of PEMFC stack to estimate remaining useful life. In Proceedings of the IEEE International Conference on Industrial Technology, Seville, Spain, 17–19 March 2015. [Google Scholar]
- Javed, K.; Gouriveau, R.; Zerhouni, N. SW-ELM: A summation wavelet extreme learning machine algorithm with a priori parameter initialization. Neurocomputing 2014, 123, 299–307. [Google Scholar] [CrossRef] [Green Version]
- Zhou, D.M.; Gao, F.; Breaz, E.; Ravey, A.; Miraoui, A. Degradation prediction of PEM fuel cell using a moving window based hybrid prognostic Approach. Energy 2017, 138, 1175–1186. [Google Scholar] [CrossRef]
- Cheng, Y.J.; Zerhouni, N.; Lu, C. A hybrid remaining useful life prognostic method for proton exchange membrane fuel cell. Int. J. Hydrogen Energy 2018, 43, 12314–12327. [Google Scholar] [CrossRef]
- Cheng, Y.J.; Zerhouni, N.; Lu, C. A prognostic framework for PEMFC based on least squares support vector regression particle filter. In Proceedings of the IEEE Vehicle Power and Propulsion Conference, Belfort, France, 11–14 December 2017. [Google Scholar]
- FCLAB Research. IEEE PHM 2014 Data Challenge; FC Lab: Belfort, France, 2014. [Google Scholar]
- Vetterli, M.; Herley, C. Wavelets and filter banks—theory and design. IEEE Trans. Signal Process. 1992, 40, 2207–2232. [Google Scholar] [CrossRef]
- Huang, G.B.; Zhu, Q.Y.; Siew, C.K. Extreme learning machine: A new learning machine of feedforward neural networks. Neural Netw. Conf. Proc. 2004, 2, 985–990. [Google Scholar]
- Cao, J.W.; Lin, Z.P.; Huang, G.B. Self-adaptive evolutionary extreme learning machine. Neural Process. Lett. 2012, 36, 285–305. [Google Scholar] [CrossRef]
- Qin, A.K.; Huang, V.L.; Suganthan, P.N. Differential evolution algorithm with strategy adaptation for global numerical optimization. IEEE Trans. Evol. Comput. 2009, 13, 398–417. [Google Scholar] [CrossRef]
- Zhao, W.T.; Li, P.; Liu, Q.; Liu, D.; Liu, X.W. Selecting the optimal hidden layer of extreme learning machine using multiple kernel learning. Ksii Trans. Internet Inf. Syst. 2018, 12, 5765–5781. [Google Scholar]
Physical Parameters | Control Range |
---|---|
Cooling water temperature | 20 °C–80 °C |
Cooling water flowrate | 0–10 L/min |
Gas temperature | 20 °C–80 °C |
Gas humidity | 0%–100% |
Air flowrate | 0–100 L/min |
Hydrogen flowrate | 0–100 L/min |
Gas pressure | 0–2 bars |
Fuel cell current | 0–300 A |
Prediction Performance Indicators | ELM | D-S ELM |
---|---|---|
End-of-life Point (h) | 756 | 807.16 |
Deviation (h) | 53 | 1.84 |
ME (V) | 0.0044 | 0.0020 |
RMSE (V) | 0.0061 | 0.0036 |
Training Time (s) | ≤ 0.0156 | 0.0781 |
Testing Time (s) | ≤ 0.0156 | 0.0313 |
© 2019 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 (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zhang, X.; Yu, Z.; Chen, W. Life Prediction Based on D-S ELM for PEMFC. Energies 2019, 12, 3752. https://doi.org/10.3390/en12193752
Zhang X, Yu Z, Chen W. Life Prediction Based on D-S ELM for PEMFC. Energies. 2019; 12(19):3752. https://doi.org/10.3390/en12193752
Chicago/Turabian StyleZhang, Xuexia, Zixuan Yu, and Weirong Chen. 2019. "Life Prediction Based on D-S ELM for PEMFC" Energies 12, no. 19: 3752. https://doi.org/10.3390/en12193752