学术论文：学习型 AC 最优潮流代理模型的蜕变测试揭示其表征依赖性
Metamorphic testing of learned AC optimal power flow surrogates reveals representation dependence
AI 导读
一项 arXiv 论文用 13 条蜕变关系审计学习型 AC-OPF 代理模型的编码表征依赖性，覆盖 17 个模型与流水线配置，每个配置至少违反 6 条关系。对 GridSFM-Open v1.1 的源码检查发现一个索引 bug 和三处非规范或不稳定的预处理选择；case500 数据增强实验中 4 个模型有 3 个降低变换视图误差，但所有预测仍 AC 不可行。作者建议将表征一致性纳入常规评测报告。
正文 · 原文
Abstract:Evaluating a learned AC optimal power flow (AC-OPF) surrogate on one grid serialization cannot show whether its accuracy depends on that encoding. We test this dependence with thirteen metamorphic relations, each specifying when a transformation is valid and how to compare its outputs. Our audit covers 17 model and pipeline configurations: two released trained checkpoints and 15 random-weight architecture or mechanism controls. Every configuration violates at least six relations, even at reporting tolerances ten times tighter or looser. For GridSFM-Open v1.1, source inspection and intermediate-tensor interventions point to an indexing bug and three noncanonical or numerically unstable preprocessing choices; a repaired checkpoint remains untested. In a separate case500 training experiment, augmentation lowers transformed-view error in three of four models and median discrepancy and constraint-violation P95 in all four. These benefits depend on the training protocol: some relations worsen, and every evaluated prediction remains AC infeasible. A GPU/CPU eigensolver mismatch accounts for most of the large raw-Laplacian-baseline gain. We recommend reporting representation consistency alongside accuracy, feasibility, and cost. These results establish neither deployment readiness nor universal invariance.
| Comments: | Accepted to the NeurIPS 2026 Workshops on (i) Trustworthy AI Evaluation, (ii) AI Foundations for Power Grids, (iii) Tackling Climate Change with Machine Learning, (iv) Sim2Science, and (v) Representations for the Physical Sciences |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Systems and Control (eess.SY); Optimization and Control (math.OC) |
| Cite as: | arXiv:2611.13566 [cs.LG] |
| (or arXiv:2611.13566v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2611.13566 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Vineet Jagadeesan Nair [view email]
[v1]
Tue, 10 Nov 2026 04:33:01 UTC (655 KB)
来源：arXiv cs.AI · arxiv.org