Machine Learning-Based Fault Detection And Health Monitoring Of Brushless DC Motors For Electric Propulsion In Civil Aviation Using Experimental Data

Read full text Summarise with VIRAT
Authors: Aria Nazarparvar
Cited by: 0
References: 5

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The increasing adoption of electric propulsion in civil aviation introduces new challenges for the reliable assessment of propulsion system health. Data-driven approaches provide effective tools for identifying deviations in electrical and mechanical behaviour prior to functional failure. This study presents a custom-developed experimental test stand representing a scaled light electric propulsion system, instrumented to measure brushless DC (BLDC) motor parameters including voltage, current, rotational speed, thrust, temperature, and acoustic noise under controlled operating conditions. Motor performance was validated under healthy no-load and propeller-loaded configurations using manufacturer-recommended components. Predefined non-destructive fault scenarios, including battery degradation and propeller damage, were then introduced to generate representative faulty operating states. A supervised Support Vector Machine (SVM) classifier was employed to distinguish between healthy and faulty conditions using multi-sensor experimental data. The proposed approach achieved an overall classification accuracy of approximately 97%, with a fault detection recall of about 98%, demonstrating strong discriminative capability using physically interpretable features. Scenario-based validation under previously unseen operating conditions further indicates the model’s ability to generalize beyond training cases. The developed framework provides a lightweight and experimentally grounded solution for condition monitoring of electric propulsion systems and establishes a foundation for future extensions toward unsupervised learning and predictive maintenance.

Fields of Study & Subjects

Machine Fault Diagnosis Techniques Fault Detection and Control Systems Electrical Fault Detection and Protection

Publication details

Journal / source
International Journal of Advances in Signal and Image Sciences
Volume
12
Issue
3s
Pages
1108–1122
Publication date
9 March 2026
Language
EN