Inductive Power Grid Cascading Failure Analysis with GRU-Gated Graph Attention
For power grid operators, this provides a transferable method to identify vulnerable lines in unseen grids without retraining, addressing a key limitation of existing approaches.
The paper tackles the problem of identifying vulnerable transmission lines in power grids before cascading failures occur, with a focus on zero-shot transfer to unseen grids. The proposed GRU-gated Graph Attention Network, trained on combined data from limited grids, successfully transfers to new grids and identifies more vulnerable lines than baselines.
Identifying vulnerable transmission lines in power grids before a cascading failure occurs is challenging: existing methods can learn inter-line failure correlations from cascade data, but they are trained and evaluated on a single grid, and transferring the learned knowledge to an unseen grid remains an open problem. We address this by training a single Gated Recurrent Unit (GRU)-gated Graph Attention Network on combined cascading failure data from limited training grids and applying it directly to any unseen grid without retraining. A GRU gate controls what information each node retains or discards at each cascade iteration. Empirical evaluation shows that the model transfers zero-shot to multiple new grids spanning inter-time and inter-domain settings. Using information extracted from the trained model, we consistently identify more vulnerable lines than established structural and electrical baselines.