# Robustness Analysis of AI Models in Critical Energy Systems

arXiv: https://arxiv.org/abs/2406.14361 · HTML: https://mlsanity.com/paper/2406.14361/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 24 references that are in the map.
- First posted on arXiv 20 Jun 2024.

## Builds on (most cited first, 7 of 7)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Physics-Guided Deep Neural Networks for Power Flow Analysis](https://mlsanity.com/paper/2002.00097/) (2020)
- [Physics-Informed Graphical Neural Network for Parameter & State Estimations in Power Systems](https://mlsanity.com/paper/2102.06349/) (2021)
- [Introducing machine learning for power system operation support](https://mlsanity.com/paper/1709.09527/) (2017)
- [Fast Power system security analysis with Guided Dropout](https://mlsanity.com/paper/1801.09870/) (2018)
- [Towards an AI assistant for power grid operators](https://mlsanity.com/paper/2012.02026/) (2020)
- [PowerFlowMultiNet: Multigraph Neural Networks for Unbalanced Three-Phase Distribution Systems](https://mlsanity.com/paper/2403.00892/) (2024)

## Built on by (most cited first, 1 of 1)

- [KCLNet: Physics-Informed Power Flow Prediction via Constraints Projections](https://mlsanity.com/paper/2506.12902/) (2025)

## Abstract (arXiv)

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.%under this security criterion. 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.