arXiv:2604.22869cs.LGcs.AI2026-04

构建飞机主油泵高保真仿真系统,提供带故障标签的数据集。

Avionic Main Fuel Pump Simulation and Fault-Diagnosis Benchmark

论文配图:Avionic Main Fuel Pump Simulation and Fault-Diagnosis Benchmark
图 1 · 摘自论文原文
  • 基于物理模型在Simulink中构建油泵系统仿真,支持故障注入。
  • 生成带健康状态与故障模式标注的时间序列数据,共包含12种故障类型。
  • 可用于异常检测与运行模式识别,适合航空系统可靠性研究者使用。

在许多网络物理系统中,特别是在飞机等关键应用中,由于数据保护问题和可观测性不足,用于训练异常检测与诊断算法的数据极为稀缺。为应对这一数据缺失问题,我们引入了一个高保真、基于物理的协同仿真系统,针对常见飞机主燃油泵进行建模,采用 extsc{MATLAB/Simulink Simscape Fluids} 构建。该系统生成了带有健康状态和故障模式标注的时间序列数据。为验证基准的有效性,我们应用无监督的循环变分自编码器(RNN-VAE)进行异常检测,并使用SOM-VAE实现运行模式离散化,训练模型以区分正常与故障状态。结果表明该仿真系统可有效支持故障诊断算法的开发与评估。

原文摘要 · Abstract (English)

In many cyber-physical systems, especially in critical applications such as aeroplanes, data to train anomaly detection and diagnosis algorithms is lacking due to data protection issues and partial observability. To combat this inherent lack of data, we introduce a high-fidelity, physics-informed co-simulation of a common aircraft main-fuel-pump system modelled in \textsc{MATLAB/Simulink Simscape Fluids}. We also describe its generated time-series data with health and fault mode annotations. To show feasibility of our benchmark, we apply an unsupervised Recurrent Variational Autoencoder (RNN-VAE) for anomaly detection and a SOM-VAE for operating mode discretization, trained to separate healthy and faulty conditions.

故障诊断仿真系统航空工程时间序列

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。