基于GNN的电力系统复杂网络拓扑识别与脆弱性算法分析  

Analysis of Complex Network Topology Identification and Vulnerability Algorithm in Power System Based on GNN

作　　者：朱天淳 ZHU Tianchun(Jiangsu Power Transmission and Transformation Co.,Ltd.,Jiangsu 210028,China)

机构地区：[1]江苏省送变电有限公司,江苏210028

出　　处：《电子技术(上海)》2024年第12期395-397,共3页Electronic Technology

摘　　要：阐述一种基于图神经网络(GNN)的算法,以识别输变电系统的复杂网络拓扑并分析其脆弱性。通过构建网络模型,利用图卷积网络(GCN)和图注意力网络(GAT)提取特征,实现拓扑识别和脆弱性评估。实验结果表明,该算法在提高系统稳定性和优化运行策略方面效果明显。This paper expounds an algorithm based on Graph Neural Network(GNN)to identify the complex network topology of power transmission and transformation systems and analyze their vulnerability.It achieves topology recognition and vulnerability assessment by constructing a network model and utilizing Graph Convolutional Networks(GCN)and Graph Attention Networks(GAT)to extract features.The experimental results show that the algorithm has significant effects in improving system stability and optimizing operational strategies.

关 键 词：图神经网络 网络拓扑 脆弱性分析 输变电系统 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TM75[自动化与计算机技术—控制科学与工程]