Research on Critical Node Identification in Cyber-Physical Power Systems Based on GCN-IDDA*

View at publisher Summarise with VIRAT
Authors: Yulei Huang, Yan Li, Tianqi Xu, Zhaolei He, Mengmeng Zhu
Cited by: 0
References: 18

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

In recent years, Cyber-Physical Power Systems (CPPS) have increasingly encountered the risk of cascading failures stemming from critical node disruptions, a challenge exacerbated by their intricate topological structures and dynamic interactions.This study proposes the Graph Convolutional Network with Integrated Deformable Depthwise Attention (GCN-IDDA) model to accurately identify critical nodes in CPPS, enhancing system reliability and resilience. The GCN-IDDA model employs a sophisticated integration of multi-scale feature extraction, deformable convolution, depthwise separable convolution, and an attention mechanism to effectively capture both topological and physical attributes of the network, with its performance validated on Barabási-Albert (BA) model-generated networks as well as real-world CPPS networks. Experimental results reveal that GCN-IDDA substantially surpasses traditional methods and baseline graph neural networks in Kendall’s tau coefficient assessments, demonstrating exceptional robustness and generalization capabilities. This research delivers an efficient framework for critical node identification in CPPS, paving the way for future enhancements in model optimization and broader applications to large-scale and dynamic network analyses, thereby advancing the development of resilient smart grids.

Fields of Study & Subjects

Complex Network Analysis Techniques Power System Optimization and Stability Smart Grid Security and Resilience

Publication details

Journal / source
IEEE PES International Meeting (PES IM)
Pages
1–5
Publication date
18 January 2026
Language
EN