Research on Transient Stability Assessment Method for Power Systems Considering the Introduction of Topology Knowledge

Under the goals of “carbon peak and carbon neutrality,” the use of a high proportion of renewable energy has become a factor that must be considered in the transient stability analysis of power systems. The uncertainty of renewable energy makes the operational state of power systems more complex and variable, rendering traditional analysis methods inadequate for accurately assessing the transient stability of power systems. To address this, this paper adopts a graph convolutional neural network (GCN) model enhanced with a Squeeze-and-Excitation (SE) attention mechanism (SE-GCN). This model explicitly considers the influence of network topology by embedding the grid's topological structure, enabling deep feature extraction and learning on the graph, thereby improving assessment accuracy. The proposed method is validated using the IEEE-68 bus test system. The results demonstrate that the SE-GCN-based transient stability assessment method can accurately capture spatial characteristics within the power system, enabling precise evaluation of transient stability.