Research on Grid Topology Leakage Threat and Defense Under Open Data Correlation

The digital transformation of power systems has accelerated the open sharing of data; however, it has also introduced significant security risks that require systematic assessment. Existing studies have demonstrated that attackers can exploit open data—combined with a priori knowledge and physical constraints—to infer sensitive information, such as power grid topology. This form of topology inference, driven by data correlation, bypasses traditional physical isolation mechanisms and can trigger a cascade of cybersecurity threats. In this paper, we focus on the threat of power grid topology leakage, which results from the analysis of open power system data. We propose a three-stage progressive research framework comprising threat modeling, experimental verification, and defense assessment. First, we analyze publicly disclosed data—such as electricity prices and line blocking information—and integrate this with limited node location data to construct a correlation path between public data and grid topology. Second, based on this correlation path, we develop a grid topology inference model driven by public data correlations and validate its inference capabilities through simulation experiments. Finally, drawing on the empirical findings of the inference model, we systematically examine the technical characteristics of existing grid data protection mechanisms and evaluate the effectiveness of different defense strategies against the proposed topology leakage threat.