ANOVA Method Applied to PEMFC Ageing Forecasting Using an Echo State Network

Simon Morando (1) , Samir Jemei (1) , Rafael Gouriveau (1) , Noureddine Zerhouni (1) , Daniel Hissel (1)

1 FEMTO-ST - Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174)

According to the International Energy Agency, an increase of the requests of energy of 40% could arise in the next decades, mainly due to the emergence of developing countries. The problem with the nowaday energy system is the use of fossil energy, which is limited and attempt to disappear in the near future. Thus an energy transition has to begin in order to replace the fossil fuels and anticipate their disappearance. Consequently, in recent years, the promotion and development of renewable energy have been realized. One of this renewable energy, the energy vector hydrogen, appears to be a promising solution, mainly due to interesting performance of Fuel Cells (FC) systems and hydrogen abundance on Earth (it is still important to underline that the hydrogen does not exist in natural form). However, this research area is still subject to scientific and technological bottlenecks. One of these major bottlenecks preventing the industrialization of FC systems is it limited useful lifetime. It is therefore important to develop reliable tools for the diagnosis and prognosis of FC system in order to optimize its efficiency. The aim of this article is to present the results of a sensibility analysis applied to a prognosis tools called Echo State Network.

Domaines: Automatique / Robotique

Simon Morando, Samir Jemei, Rafael Gouriveau, Noureddine Zerhouni, Daniel Hissel. ANOVA Method Applied to PEMFC Ageing Forecasting Using an Echo State Network. 11th International Conference on Modeling and Simulation of Electric Machines, Converters and Systems (ElectrIMACS 2014), Jan 2014, European Union. pp.652 - 657. ⟨hal-01045160⟩