用物理定律约束图注意力网络，提升供水系统异常检测精度与鲁棒性。
Graph Attention Networks with Physical Constraints for Anomaly Detection
- 基于质量与能量守恒残差构建图注意力特征，融合时空建模。
- 在BATADAL数据集上达F1=0.979，比基准提升3.3个百分点。
- 适合需高可解释性与抗参数误差的工业级水网监控场景。
供水系统面临日益增长的网络物理风险，可靠异常检测至关重要。现有数据驱动模型常忽略网络拓扑结构且难以解释，而模型驱动方法对参数精度依赖过高。本文提出一种考虑水力特性的图注意力网络，以归一化的守恒定律违例作为输入特征。该模型结合质量与能量平衡残差，利用图注意力机制与双向LSTM学习时空模式，并通过多尺度模块聚合节点至全网级别的检测评分。在BATADAL数据集上，模型取得F1=0.979，相比基线提升3.3个百分点，在参数噪声高达15%时仍保持良好鲁棒性。
原文摘要 · Abstract (English)
Water distribution systems (WDSs) face increasing cyber-physical risks, which make reliable anomaly detection essential. Many data-driven models ignore network topology and are hard to interpret, while model-based ones depend strongly on parameter accuracy. This work proposes a hydraulic-aware graph attention network using normalized conservation law violations as features. It combines mass and energy balance residuals with graph attention and bidirectional LSTM to learn spatio-temporal patterns. A multi-scale module aggregates detection scores from node to network level. On the BATADAL dataset, it reaches $F1=0.979$, showing $3.3$pp gain and high robustness under $15\%$ parameter noise.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。