🤖 AI Summary
Estimating the distribution of relaxation times (DRT) from electrochemical impedance spectroscopy (EIS) data constitutes an ill-posed inverse problem highly sensitive to regularization. This work proposes a physics-informed convolutional autoencoder that directly embeds the discretized EIS–DRT physical relationship into the training process, enabling consistent DRT reconstruction across datasets without per-spectrum hyperparameter tuning. The model demonstrates excellent performance on both synthetic double-ZARC spectra and three independent experimental datasets from solid oxide cells, achieving normalized reconstruction errors below 1.1%. Furthermore, the latent space is found to be naturally organized by relaxation time scales, facilitating interpretable operational monitoring and effectively capturing dynamic changes, hydrogen starvation events, and long-term degradation.
📝 Abstract
Estimating the distribution of relaxation times (DRT) fromelectrochemical impedance spectroscopy (EIS) is an ill-posed inverse problem that is highly sensitive to regularisation choices. We propose a physics-informed convolutional autoencoder that estimates DRT directly from EIS data without spectrum-specific tuning. A discretised relation between impedance and the DRT is embedded in the training process, constraining the network to produce impedance-consistent distributions. The model resolves overlapping relaxation processes in synthetic two-ZARC spectra and accurately reconstructs measurements from three independent solid oxide fuel and electrolysis cell datasets, with range-normalised errors below 1.1%. Decoder-probe analysis shows that the learned latent representation is organised according to relaxation timescale. Distances in this latent space capture operating changes, hydrogen-shortage events, and long-term degradation. The same lightweight architecture is applied across all datasets without modification, providing consistent DRT estimation and an interpretable basis for condition monitoring.