This paper reports our initial development of a simulation-informed machine learning algorithm for failure diagnostics in solid oxide fuel cell (SOFC) systems. We used physics-based models to simulate electrochemical impedance spectroscopy (EIS) response of a short SOFC stack under normal conditions and under three different failure modes: fuel maldistribution, delamination, oxidant gas crossover to the anode channel. These data were used to train a support vector machine (SVM) model, which is able to detect and differentiate these failures in simulated data under various conditions. The SVM model can also distinguish these failures from simulated uniform degradation that often occurs with long-term operation. These encouraging results are guiding our ongoing efforts to apply EIS as a failure diagnostic for real SOFC cells and short stacks.

No abstract

Surface reactions of ethylene and toluene on a Pt-modified, samarium-doped ceria (Pt/CSO) anode were examined as a function of hydrocarbon to oxygen ratio and cycling of oxy- gen ion current through the electrolyte. Reaction conditions were temperatures of 540 °C (C2H4) and 730 °C (C7H8), hydro- carbon partial pressures of 0.001-0.04 Torr, and oxygen partial pressures of 0.1 Torr (anode) and 5 Torr (cathode). A gen- eral result is that reactions of both ethylene and toluene are strongly sensitive to surface carbon. A step change in oxygen ion current results in electrochemical enhancements of up to 80 and catalytic enhancements as high as 3. Cycling of oxygen ion currents leads to enhanced catalytic activity that remained for 1 hour or more after cycling was terminated. Enhance- ment was sensitive to the ratio of hydrocarbon to oxygen and attributed to a combination of carbon removal and redox ki- netics in the ceria anode.