A fault detection method for Auxiliary Power Unit based on monitoring parameters selection

Abstract

Auxiliary Power Unit (APU) is an indispensable component utilized in modern aircraft, which provides electrical and pneumatic power to the aircraft independently. What’s more, APU can help the main engines restart in case of main engine failure during flight. Thus there exists the need of APU monitoring. However, APU has not received sufficient attention in maintenance due to its relatively low cost and safety requirements compared to main engines. Line maintenance shows that APU is likely to fail in the start stage. Based on the Quick Access Recorder data supported by the airline, a fault detection method for APU is proposed in this paper. In order to improve the accuracy of fault detection, feature selection is carried out firstly to determine the optimal subset of monitoring parameters by Recursive Feature Elimination. Then fault detection is carried out by Support Vector Machine. Results show that Exhaust Gas Temperature is the most effective index among all monitoring parameters, and feature selection has a significant improvement on the accuracy of fault detection, which means higher accuracy with less monitoring cost. The method proposed in this paper also disposes that some monitoring parameters should be paid more attention in terms of the condition monitoring of APU.

Air leaks fault detection in maintenance using machine learning

“…Literature is full of studies focused on utilizing ML to enhance maintenance task efficiency and effectiveness for different types of equipment such as aerospace (Adhikari et al, 2018;Deng, 2020), industrial systems (Askari et al, 2023;Vita et al, 2020;Martin-del-Campo and Sandin, 2017), air conditioning (Chen et al, 2022), healthcare (Shamayleh et al, 2020) and energy (Yan et al, 2017;M arquez et al, 2019a). These studies focused on either fault detection (Abela et al, 2022;Chen et al, 2022;Awad et al, 2017;Anis, 2018;Glowacz et al, 2017;Cerrada et al, 2022;Pichler et al, 2016;Jiang et al, 2023;Askari et al, 2023;Abid et al, 2022), or failure classifying (Lei et al, 2008;Poto cnik and Govekar, 2017;Toma et al, 2020;Schneider et al, 2017), or PdM to predict remaining useful life (RUL) (Deng, 2020;Hu et al, 2012;Deutsch and He, 2018). Regardless of the use case, ML models require single or several data streams generated from real-time or offline monitoring of the system.…”

“…Other researchers focused on other pneumatic components such as Jiang et al (2023) who developed a fault detection method for the auxiliary power unit which provide electrical and pneumatic power in modern aircraft using support vector machine (SVM) technique. The same authors used recursive feature elimination for feature selection.…”