图注意力网络释放：面向微电网的快速且可解释的脆弱性评估框架
机器学习
2025-06-09 v2 人工智能
摘要
独立微电网通过结合分布式能源和负荷在孤立岛屿和野战等场景中供电至关重要。对微电网抵御故意攻击或自然灾害的脆弱性进行快速准确的评估，对于有效的风险预防和设计优化至关重要。然而，传统的蒙特卡洛模拟（MCS）方法计算成本高且耗时，而现有的基于机器学习的方法往往缺乏准确性和可解释性。为应对这些挑战，本研究提出了一种快速且可解释的脆弱性评估框架，将 MCS 与通过自注意力池化增强的图注意力网络（GAT-S）相结合。MCS 生成训练数据，而 GAT-S 模型学习微电网的结构和电气特征，并进一步智能地评估其脆弱性。GAT-S 通过动态分配关键节点的注意力权重来提升可解释性和计算效率。在各种微电网配置上的全面实验评估表明，所提框架提供了准确的脆弱性评估，均方误差低至 0.001，1 秒内实时响应，并提供可解释的结果。
引用
@article{arxiv.2503.00786,
title = {Graph Attention Networks Unleashed: A Fast and Explainable Vulnerability Assessment Framework for Microgrids},
author = {Wei Liu and Tao Zhang and Chenhui Lin and Kaiwen Li and Rui Wang},
journal= {arXiv preprint arXiv:2503.00786},
year = {2025}
}
备注
Since we have found that there are still several issues in this article. Some statements in the article are not rigorous, and the language and structure of the article still have a lot of room to polish. Moreover, the experiment of the article is not sufficient, and the experimental conclusion is not convincing enough. We sincerely hope to withdraw this article for further revision