A Fault Diagnosis Method for Aircraft Engine Bearings Based on Variational Mode Decomposition and Fused Features

View at publisher Summarise with VIRAT
Authors: Shuoyan Zhang
Cited by: 0
References: 12

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

As the core rotating component of an aircraft engine, bearings play a crucial role in ensuring operational reliability. Bearing failures can easily lead to cascading malfunctions, posing a significant threat to flight safety. Therefore, achieving accurate fault diagnosis of bearings is of great engineering importance. To overcome the limitations of traditional diagnostic methods-such as vulnerability to noise under complex operating conditions, insufficient feature extraction, and low classification accuracy in imbalanced datasets-this study proposes a novel approach for aircraft engine bearing fault diagnosis. First, variational mode decomposition (VMD) is applied to adaptively decompose the raw vibration signals and suppress noise, effectively mitigating the issue of background interference that obscures fault-related information. Second, multi-domain feature extraction is performed on the reconstructed signals to comprehensively capture dynamic characteristics across multiple scales and fault types. Subsequently, t-distributed stochastic neighbor embedding (t-SNE) is used to fuse the high-dimensional feature vectors, removing redundant information while preserving key discriminative features to enhance inter-class separability. Finally, an equilibrium optimization-based support vector machine (EO-SVM) is developed as the fault classification model. The equilibrium optimization strategy improves SVM parameter tuning, thereby enhancing overall fault recognition performance. Experimental validation using real aircraft engine bearing fault datasets demonstrates that the proposed method effectively addresses challenges in noise suppression, feature representation, and classification optimization, providing reliable technical support for the health management of aviation equipment.

Fields of Study & Subjects

Machine Fault Diagnosis Techniques Fault Detection and Control Systems Gear and Bearing Dynamics Analysis

Publication details

Journal / source
IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
Pages
917–921
Publication date
28 January 2026
Language
EN