Temperature-dependent performance prediction for cerium oxynitride solid-state symmetric supercapacitor using machine learning
Résumé
The global demand for electricity and dwindling fossil fuel reserves necessitates advanced energy storage solutions like next-generation supercapacitors. This study uses machine learning (ML) to predict the temperature-dependent performance of cerium oxynitride-based solid-state symmetric supercapacitors. Supercapacitors excel in power density and rapid charge-discharge but face challenges under varying temperatures affecting ionic mobility and material stability.
Cerium oxynitride, with unique electronic and thermal properties, is analyzed using Density Functional Theory (DFT) to study temperature-induced changes in the Density of States. ML models trained on experimental and DFT data predict charge-discharge behaviour under thermal and current variations. The supercapacitor exhibits good charge storage (~4.6 mVcm⁻²) and low ohmic resistance (~1.6 Ω) at room temperature. The study also models exponential and linear relationships between time and potential window, highlighting the potential of integrating DFT, experimental insights, and ML to tackle temperature variability in supercapacitors.
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |
| Licence |
|---|