Remaining Useful Life Prediction for Proton Exchange Membrane Fuel Cells Including Reversible and Irreversible Losses

Abstract

Today the world is full of time-dependent phenomena in all fields: physics, chemistry, mechanics and many others. Time acts on the performance of any system whatever its nature is. Moreover, proton exchange membrane fuel cells are promising alternatives to conventional power sources due to their high energy density and zero gas emission. However, this technology is still not sufficiently mature to reach large-scale deployment due to its limited lifespan. To extend the lifespan, the “Prognosis and Health Management” discipline has been developed, which is considered to be efficient in improving the reliability, durability and maintainability of fuel cell systems. However, it involves a deep understanding of the reversible and irreversible degradation phenomena and their impacts on fuel cell performance. Based on this, this paper deals with analyses of reversible and irreversible degradation. The criticalities of these losses and their impacts on the fuel cell lifetime are underlined with a useful lifetime estimation based on an autoregressive moving average model. Indeed, to do so, three scenarios are studied. First, the remaining useful life is predicted by taking into account only reversible degradation, and this gives the minimum lifetime. Second, the real remaining useful life is estimated by taking into account both reversible and irreversible degradation. Finally, the maximum lifetime that can be reached is estimated by taking into account only irreversible degradation.