# Avionic Main Fuel Pump Simulation and Fault-Diagnosis Benchmark

arXiv: https://arxiv.org/abs/2604.22869 · HTML: https://mlsanity.com/paper/2604.22869/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 3 of its 23 references that are in the map.
- First posted on arXiv 23 Apr 2026.

## Builds on (most cited first, 3 of 3)

- [An Evaluation of Anomaly Detection and Diagnosis in Multivariate Time Series](https://mlsanity.com/paper/2109.11428/) (2021)
- [A Research Agenda for AI Planning in the Field of Flexible Production Systems](https://mlsanity.com/paper/2112.15484/) (2021)
- [Artificial Intelligence in Industry 4.0: A Review of Integration Challenges for Industrial Systems](https://mlsanity.com/paper/2405.18580/) (2024)

## Abstract (arXiv)

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 textscMATLAB/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.