Hierarchical GNNs for power flow: letting physics shape the hierarchy
Hierarchical latent communication improves the generalization of a power-flow model, shared across three grids, to new operating scenarios, with one set of learned parameters shared across grids.