Reservoir Computing optimisation for PEM fuel cell fault diagnostic

Fuel cell (FC) is considered as one of the most interesting solutions to overcome future energy crisis announced by the International Energy Agency. However, various bottlenecks, whether technological or societal, slow the industrial interest for this technology and therefore the mass production of fuel cells. One of these bottlenecks is related to the limited lifetime of the FC system. To counter this bottleneck, the implementation of fault diagnostics and identification methods is relevant. This paper presents an original and experimentally compatible diagnostics approach, named Reservoir Computing. This paradigm, coming from the Artificial Intelligence domain, is an evolution of traditional Artificial Neural Networks, with a reservoir of neurons instead of the succession of different neuronal layers. Targeted fault types on the fuel cell are stoichiometry values faults, pressure loss, temperature drop and problem on the cooling circuit. Experimental results show the simplicity and effectiveness of RC method to detect these faults under a dynamic load profile.