物理信息但非物理一致：神经交流潮流中的误差几何与子空间投影
Physics-Informed but Not Physics-Consistent: Error Geometry and Subspace Projection for Neural AC Power Flow
- Friedrich-Alexander-Universität Erlangen-Nürnberg（埃尔朗根-纽伦堡大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
研究神经潮流求解器的预测误差几何，通过奇异值分解和校准子空间投影抑制子空间外分量，显著降低功率平衡残差并提升电压精度，实现物理一致性。
AI中文摘要：
最近的神经潮流求解器，包括新兴的基础模型，实现了准确的电压预测，然而这种准确性并不必然意味着物理上一致的解。即使很小的复数电压误差也可能导致较大的交流功率平衡残差。我们在现实的2224节点英国电网（GBnetwork）场景中，跨PIGNN-GC、GridSFM、gridfm-graphkit和LUMINA研究了这一准确性-一致性差距，并在31个系统上对GridSFM进行了跨网格评估。使用拟合到训练交流潮流解的奇异值分解（SVD）基，我们发现神经预测误差包含大量位于主导解子空间之外的分量。为解决这一不匹配，校准解子空间投影（CSP）在仅训练偏置校准后抑制了子空间外的预测分量，相对于校准预测，PIGNN-GC、GridSFM、gridfm-graphkit和LUMINA的平均PB分别降低了67.0%、37.8%、40.5%和68.9%，同时提高了所有四个模型的电压幅值精度。这些结果将输出误差几何确定为物理一致神经交流潮流中的一个重要因素。代码：此https URL
英文摘要：
Recent neural power-flow solvers, including emerging foundation models, achieve accurate voltage predictions, yet such accuracy does not necessarily imply physically consistent solutions. Even small complex voltage errors can yield large AC power-balance residuals. We study this accuracy-consistency gap across PIGNN-GC, GridSFM, gridfm-graphkit, and LUMINA on realistic 2224-bus Great Britain network (GBnetwork) scenarios, with cross-grid evaluation of GridSFM over 31 systems. Using a singular value decomposition (SVD) basis fitted to training AC power-flow solutions, we find that neural prediction errors contain substantial components outside the dominant solution subspace. To address this mismatch, calibrated solution-subspace projection (CSP) suppresses off-subspace prediction components after train-only bias calibration, reducing Mean PB by 67.0%, 37.8%, 40.5%, and 68.9% for PIGNN-GC, GridSFM, gridfm-graphkit, and LUMINA, respectively, relative to calibrated predictions, while improving voltage-magnitude accuracy in all four models. These results identify output-error geometry as an important factor in physics-consistent neural AC power flow. Code: https://github.com/Kimchangheon/neural-acpf-error-geometry
