Operando Diagnostics of Solid Oxide Fuel Cell Stack Via Electrochemical Impedance Spectroscopy Simulation-Informed Machine Learning

Abstract

In this work, we apply a machine learning approach to solid oxide fuel cell (SOFC) system diagnostics. Instead of fitting electrochemical impedance spectroscopy (EIS) into a physics based model or equivalent circuit, we train machine learning models to recognize failures from a database of simulated EIS. We use a coarse-grained physics-based model to simulate stack EIS under three different failure modes: fuel maldistribution, delamination, and cathode gas crossover to anode channel. Synthesized machine learning classification models successfully recognize these different degradation mechanisms in simulated data across different operating conditions. We are also able to differentiate these failures from the uniform degradation that tends to occur with SOFC over time. These encouraging results prompt our current effort to implement machine learning diagnostics methods on experimental EIS collected on SOFC short stack.

Impedance Analysis of Electrochemical Systems

“…Pal et al 365 machine learning approaches to relate charge-transfer resistances from Nyquist plots to changes in antigen concentration. Le et al 366 and Shao et al 367 applied a machine learning approach to solid oxide fuel cell (SOFC) system diagnostics. They trained machine learning models to recognize failures from a database of simulated EIS.…”

Perovskites Doped with Small Amounts of Noble Metals for IT-SOFCs

“…Fig 6c ) shows the Bode plots of LSFRu/LSGM/LSFPt cell in pure and diluted hydrogen. Fuel dilution affects mainly the low frequency contribution, which can be related to the fuel adsorption and diffusion, according to literature (15), while the high-frequency contribution can be associated to the charge-transfer at the electrode-electrolyte interface, and it is almost not affected by the gas dilution.…”

Predicting Voltammetry Using Physics-Informed Neural Networks

“…Physics-informed neural networks (PINNs), introduced by Raissi in 2018 to solve the Burger, Schrodinger, and Alan–Cahn equations among others, form a class of neural networks that can integrate data and abstract mathematical operators, along with the laws of nature, to provide physically consistent solutions. PINNs have been applied widely in modeling, analysis, and parameter estimation to lithium-ion batteries − and fuel cells. − A PINN can also be informed by chemical kinetics , and thermodynamics to solve the partial differential equations (PDEs) that describe diverse physical chemical models. With the growing application of PINN in scientific and engineering contexts, we note that there are no reports applying PINNs to solve electrochemical problems with coupled diffusional mass transport both in general and in particular in the context of voltammetry, the most generally applied electrochemical methodology. , Simulation of cyclic voltammetry conventionally employs finite difference, finite element, , or random walk algorithms, all needing discretization of simulation spaces leading to the difficulty that the requirements of discretization/simulation grow exponentially as simulations expand to higher dimensions .…”