Physics-informed GNN for medium-high voltage AC power flow with edge-aware attention and line search correction operator
This work addresses the need for fast and accurate power flow solvers in electrical grid operations, offering an incremental improvement over existing PIGNN methods.
The paper tackles the accuracy limitations of physics-informed graph neural networks (PIGNNs) for AC power flow by introducing PIGNN-Attn-LS, which combines edge-aware attention and a line-search correction operator, achieving a test RMSE of 0.00033 p.u. in voltage and 0.08° in angle, with 2–5× faster inference than Newton–Raphson solvers.
Physics-informed graph neural networks (PIGNNs) have emerged as fast AC power-flow solvers that can replace classic Newton--Raphson (NR) solvers, especially when thousands of scenarios must be evaluated. However, current PIGNNs still need accuracy improvements at parity speed; in particular, the physics loss is inoperative at inference, which can deter operational adoption. We address this with PIGNN-Attn-LS, combining an edge-aware attention mechanism that explicitly encodes line physics via per-edge biases, capturing the grid's anisotropy, with a backtracking line-search-based globalized correction operator that restores an operative decrease criterion at inference. Training and testing use a realistic High-/Medium-Voltage scenario generator, with NR used only to construct reference states. On held-out HV cases consisting of 4--32-bus grids, PIGNN-Attn-LS achieves a test RMSE of 0.00033 p.u. in voltage and 0.08$^\circ$ in angle, outperforming the PIGNN-MLP baseline by 99.5\% and 87.1\%, respectively. With streaming micro-batches, it delivers 2--5$\times$ faster batched inference than NR on 4--1024-bus grids.