Answers are grounded in this paper and cite their sources.
Reversible solid oxide cells (rSOC) are promising energy conversion devices due to their bidirectional operation capability, while fault diagnosis under dual-mode operation remains challenging. This study develops a data-driven fault diagnosis framework for rSOC systems by integrating convolutional neural network (CNN), radial basis function support vector machine (RBF-SVM), and Bayesian optimization (BO). Dynamic fault characteristics of fuel-side leakage, air-side leakage, and electrode interface delamination are investigated under both solid oxide fuel cell and solid oxide electrolysis cell modes using simulated operating data. The proposed Bayesian-optimized CNN-SVM model achieves a diagnostic accuracy of 97.55%, outperforming conventional SVM, CNN, and CNN-SVM models. The results demonstrate that the proposed method effectively extracts transient fault features and provides accurate and reliable multi-fault diagnosis for reversible solid oxide cell systems.
No takes yet. Share an insight, caveat, or question.
Ma et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: