Industry-Education Integration Case for an Enhanced Time-Frequency Analysis Method: Bearing Fault Diagnosis

Read full text Summarise with VIRAT
Authors: Weidong Li, Renhao Xu, Binbin Qiu
Cited by: 0
References: 6

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

Bearing vibration signals are fundamental to the condition monitoring and fault diagnosis of rotating machinery, serving as key indicators of mechanical health. However, due to their strong nonlinearity and non-stationary dynamic behavior, conventional signal processing and feature extraction techniques often struggle to capture meaningful fault-related information with sufficient clarity and robustness. To overcome these limitations, this paper proposes an enhanced time-frequency analysis framework, rigorously validated using experimental datasets acquired from a machinery fault simulation test rig. The proposed method incorporates multiple data augmentation techniques—specifically, noise injection, gain adjustment, and time reversal—to enrich training diversity and improve feature representation. These augmentations effectively enhance time-frequency resolution, suppress background noise and cross-term interference, and sharpen the definition of transient fault signatures. Consequently, the approach yields significantly more discriminative and robust fault indicators compared to baseline methods. Comprehensive comparative experiments further demonstrate the superiority of the proposed framework and provide valuable practical insights for the deployment of deep learning models in real-world intelligent fault diagnosis systems.

Fields of Study & Subjects

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

Publication details

Journal / source
Automation and Machine Learning
Volume
7
Issue
1
Publication date
1 January 2026
Language
EN