Skip to main navigation Skip to search Skip to main content

Within and among Wavelet-subband Sparse Decomposition for Bearing Fault Diagnosis

  • Xi'an Jiaotong University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Extracting fault components from structural noise is one of the key challenges in bearing fault diagnosis. In this paper, we research the sparsity of bearing fault signals in the wavelet domain and propose Within and Among Wavelet-subband Sparse Decomposition (WAWSD) for extracting the fault signal. Based on the sparse distribution of fault signals in the wavelet domain, two constraints based on the sparsity within and among wavelet-subband prior are proposed respectively. Then, an optimization algorithm based on the thought of Iterative Atomic Decomposition Thresholding (IADT) is proposed to solve the proposed highly non-convex model. The performance of the WAWSD is verified by the simulation study and the run-to-failure experiment.

Original languageEnglish
Title of host publicationIEEE International Instrumentation and Measurement Technology Conference, I2MTC 2025 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798331505004
DOIs
StatePublished - 2025
Event2025 IEEE International Instrumentation and Measurement Technology Conference, I2MTC 2025 - Chemnitz, Germany
Duration: 19 May 202522 May 2025

Publication series

NameConference Record - IEEE Instrumentation and Measurement Technology Conference
ISSN (Print)1091-5281

Conference

Conference2025 IEEE International Instrumentation and Measurement Technology Conference, I2MTC 2025
Country/TerritoryGermany
CityChemnitz
Period19/05/2522/05/25

Keywords

  • Sparse prior model
  • fault diagnosis
  • non-convex

Fingerprint

Dive into the research topics of 'Within and among Wavelet-subband Sparse Decomposition for Bearing Fault Diagnosis'. Together they form a unique fingerprint.

Cite this