可解释智能电网攻击检测研究：时序泄漏控制下多测试台评估
Interpretable Cyber-Physical Attack Detection in Smart Grids: A Leakage-Controlled Multi-Testbed Evaluation of Fusion, Robustness, and Observability Limits
电网攻击检测大多只在打乱数据上刷分，这篇直接测出时序泄漏让 macro-F1 从 0.96 掉到 0.88，还用 IEEE 14-Bus 和 118-Bus 给出哪些攻击物理上真能做。做工业安全或能源系统检测的工程师可以直接拿开源代码复现基准。
一项多测试台评估研究围绕可解释的时窗融合 Random Forest 展开，在评估协议不一致时采用悲观结果，避免干净、随机打乱数据集带来的乐观偏差。在公开的配对网络-物理数据集上，滚动时序评估使 macro-F1 从 0.96 降至 0.88 +/- 0.06；拓扑无关的 GCN 对照模型在这些划分上表现相当。
研究进一步量化了攻击与观测的边界：adaptive 与 adaptive-joint 特征空间攻击将 macro-F1 分别压至 0.67 和 0.34，对抗训练与随机平滑可部分恢复性能；IEEE 14-Bus 测试台上的物理约束演化策略逃逸呈现相同脆弱模式。真实 IEC-104 流量中，网络侧视图对结构性攻击的召回为 0.87-0.98，但对语义值操纵仅 0.11-0.35；IEEE 14-Bus 可观测性分类显示隐蔽虚假数据注入能同时绕过全部三个观测视图。
在完整 AC 加权最小二乘状态估计测试台上，5% 误报率下预测辅助时序检测会被慢速斜坡逃逸，而速率不变的物理流形检测仍能维持检出：IEEE 118-Bus 端到端评估中，流形测试在 0.05 度/步时可检出全部试验，0.02 度/步时为 0.48，时序测试仅为 0.14。全部代码与基准已开源。
Interpretable Cyber-Physical Attack Detection in Smart Grids: A Leakage-Controlled Multi-Testbed Evaluation of Fusion, Robustness, and Observability Limits
Cyber-physical smart-grid detectors are typically treated as black boxes and validated exclusively on clean, temporally shuffled datasets, which risk overly optimistic performance claims. This work presents a multi-testbed evaluation study built around an interpretable, time-windowed fusion Random Forest, adopting the pessimistic result whenever evaluation protocols disagree. (i) On a public paired cyber-physical dataset, shuffled cross-validation is optimistic: macro-F1 falls from 0.96 to 0.88 +/- 0.06 under rolling-origin temporal evaluation. A topology-free graph convolutional network (GCN) serving as a control confirms that this lightweight model is competitive on these splits. (ii) The adaptive and adaptive-joint feature-space attacks, which are infeasible as realistic grid manipulations, degrade macro-F1 to 0.67 and 0.34, respectively. Adversarial training and randomized smoothing partially restore performance, and a physics-constrained evolution-strategy evasion on the IEEE 14-Bus testbed shows the same vulnerability pattern. (iii) On real IEC-104 traffic, the cyber view reliably catches structural attacks (recall of 0.87 to 0.98) but largely fails on semantic value manipulation (recall of 0.11 to 0.35). An observability taxonomy on the IEEE 14-Bus system reveals that stealthy false data injection defeats all three observation views. (iv) On a full AC weighted least-squares state-estimation testbed, a forecasting-aided temporal test is evaded by slow ramps precisely where a rate-invariant physics-manifold test sustains detection at a common 5% false-alarm target. A full end-to-end evaluation on IEEE 118-Bus confirms the ordering but bounds its strength: the manifold test detects every trial down to 0.05 degree/step and 0.48 at 0.02 degree/step, against 0.14 for the temporal test. All code and benchmarks are released.
This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include 智能电网, Random Forest, GCN, IEEE 14-Bus.
Source attribution: curated from arXiv cs.LG (@Muhammad Ridwanul Hoquea, Sarah Sharif, Reza Saeed Kandezy, Yaser Banad).
