An EVO-VMD-CNN-based method for bearing fault detection

View at publisher Summarise with VIRAT
Authors: Xuefeng Deng, Yadong Wang, Lulu Zheng, Yadong Pang, Yifan Shao
Cited by: 0
References: 25

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

Abstract Bearing fault detection plays a vital role in ensuring the reliability and safety of rotating machinery. However, traditional approaches often face difficulties in determining suitable parameters for Variational Mode Decomposition (VMD), such as the number of modes and penalty factor, which can lead to unstable signal decomposition and degraded diagnostic performance. To address these challenges, this paper proposes a hybrid fault detection framework integrating Energy Valley Optimization (EVO), VMD, and Convolutional Neural Networks (CNN), referred to as EVO–VMD–CNN. The EVO algorithm, inspired by physical energy decay mechanisms, adaptively optimizes key VMD parameters to enhance the stability and efficiency of signal decomposition. The optimized intrinsic mode function (IMF) components are then fed into a CNN classifier for automatic feature learning and fault identification. Experimental results on the CWRU bearing dataset demonstrate that the proposed method achieves near-perfect classification accuracy while exhibiting faster convergence and more stable training behavior compared with conventional optimization strategies such as PSO and SOA. Furthermore, noise injection experiments under low signal-to-noise ratio conditions verify the strong robustness of the proposed framework. These results highlight the effectiveness and practical potential of the EVO–VMD–CNN method for intelligent condition monitoring and fault diagnosis of rotating machinery.

Fields of Study & Subjects

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

Publication details

Journal / source
Engineering Research Express
Volume
8
Issue
7
Pages
075226–075226
Publication date
25 March 2026
Language
EN