Robustness Analysis of AI Models in Critical Energy Systems
Abstract
This paper analyzes the robustness of state-of-the-art AI-based models for power grid operations under the $N-1$ security criterion. While these models perform well in regular grid settings, our results highlight a significant loss in accuracy following the disconnection of a line. Using graph theory-based analysis, we demonstrate the impact of node connectivity on this loss. Our findings emphasize the need for practical scenario considerations in developing AI methodologies for critical infrastructure.
Cite
Text
Dogoulis et al. "Robustness Analysis of AI Models in Critical Energy Systems." ICML 2024 Workshops: NextGenAISafety, 2024.Markdown
[Dogoulis et al. "Robustness Analysis of AI Models in Critical Energy Systems." ICML 2024 Workshops: NextGenAISafety, 2024.](https://mlanthology.org/icmlw/2024/dogoulis2024icmlw-robustness/)BibTeX
@inproceedings{dogoulis2024icmlw-robustness,
title = {{Robustness Analysis of AI Models in Critical Energy Systems}},
author = {Dogoulis, Pantelis and Jimenez, Matthieu and Cordy, Maxime and Ghamizi, Salah and Le Traon, Yves},
booktitle = {ICML 2024 Workshops: NextGenAISafety},
year = {2024},
url = {https://mlanthology.org/icmlw/2024/dogoulis2024icmlw-robustness/}
}