Automatic Data Curation and Analysis Pipeline for Electrochemical Impedance Spectroscopy Measurements Conducted on Solid Oxide Cell Stacks

In this work, we apply data from Electrochemical Impedance Spectroscopy (EIS) measurements, conducted on a Solid Oxide Cell (SOC) stack, to an automatic data curation and evaluation pipeline. Latter is developed to enable the use of historic data from EIS measurements conducted on SOC stack experiments for Machine Learning (ML) models. We show that the proposed procedure can curate parasitic, inductive impedances, obtained as a common effect of measurements on stack level. In addition, drifts induced by temperature and by steam supply gradients during the EIS measurement can be compensated. The results are experimentally validated on a two-layer SOC stack. For extraction of feature values for subsequent ML models distribution of relaxation times (DRT) deconvolution and equivalent circuit modeling (ECM) are used. To determine a suitable regularization parameter for DRT deconvolution, a variance test is implemented.