Sparsity-enhanced signal decomposition via generalized minimax-concave penalty for gearbox fault diagnosis

Gaigai Cai, Ivan Selesnick, Shibin Wang, Weiwei Dai, Zhongkui Zhu

Research output: Contribution to journalArticle

Abstract

Vibration signals arising from faulty gearboxes are often a mixture of the meshing component and the periodic transient component, and simultaneously contaminated by noise. Sparsity-assisted signal decomposition is an effective technique to decompose a signal into morphologically distinct components based on sparse representation and optimization. In this paper, we propose a sparsity-enhanced signal decomposition method which uses the generalized minimax-concave (GMC) penalty as a nonconvex regularizer to enhance sparsity in the sparse approximation compared to classical sparsity-assisted signal decomposition methods, and thus to improve the decomposition accuracy for gearbox fault diagnosis. Even though the GMC penalty itself is nonconvex, it maintains the convexity of the GMC regularized cost function to be minimized. Hence, similar to the classical L1-norm regularization methods, the global optimal solution can be guaranteed via convex optimization. Moreover, we present and validate a straight-forward way to choose transforms and set parameters for the proposed method. Through simulation studies, it is demonstrated that the proposed sparsity-enhanced signal decomposition method can effectively decompose the simulated faulty gearbox signal into the meshing component and the periodic transient component. Comparisons with the classical L1-norm regularized signal decomposition method and spectral kurtosis show that the proposed method can accurately preserve the amplitude of the periodic transient component and provide a more accurate estimation result. Experiment and engineering case studies further verify that the proposed method can accurately estimate the periodic transient component from vibration signals, which demonstrate that the proposed method is a promising tool for gearbox fault diagnosis.

Original languageEnglish (US)
Pages (from-to)213-234
Number of pages22
JournalJournal of Sound and Vibration
Volume432
DOIs
StatePublished - Oct 13 2018

Fingerprint

transmissions (machine elements)
penalties
Failure analysis
Decomposition
decomposition
norms
Convex optimization
Cost functions
convexity
vibration
kurtosis
optimization
engineering
costs

Keywords

  • Convex optimization
  • Gearbox fault diagnosis
  • Generalized minimax-concave penalty
  • Signal decomposition
  • Sparse representation

ASJC Scopus subject areas

  • Condensed Matter Physics
  • Acoustics and Ultrasonics
  • Mechanics of Materials
  • Mechanical Engineering

Cite this

Sparsity-enhanced signal decomposition via generalized minimax-concave penalty for gearbox fault diagnosis. / Cai, Gaigai; Selesnick, Ivan; Wang, Shibin; Dai, Weiwei; Zhu, Zhongkui.

In: Journal of Sound and Vibration, Vol. 432, 13.10.2018, p. 213-234.

Research output: Contribution to journalArticle

@article{8b530ddee53047af828fd8856ce6e6eb,
title = "Sparsity-enhanced signal decomposition via generalized minimax-concave penalty for gearbox fault diagnosis",
abstract = "Vibration signals arising from faulty gearboxes are often a mixture of the meshing component and the periodic transient component, and simultaneously contaminated by noise. Sparsity-assisted signal decomposition is an effective technique to decompose a signal into morphologically distinct components based on sparse representation and optimization. In this paper, we propose a sparsity-enhanced signal decomposition method which uses the generalized minimax-concave (GMC) penalty as a nonconvex regularizer to enhance sparsity in the sparse approximation compared to classical sparsity-assisted signal decomposition methods, and thus to improve the decomposition accuracy for gearbox fault diagnosis. Even though the GMC penalty itself is nonconvex, it maintains the convexity of the GMC regularized cost function to be minimized. Hence, similar to the classical L1-norm regularization methods, the global optimal solution can be guaranteed via convex optimization. Moreover, we present and validate a straight-forward way to choose transforms and set parameters for the proposed method. Through simulation studies, it is demonstrated that the proposed sparsity-enhanced signal decomposition method can effectively decompose the simulated faulty gearbox signal into the meshing component and the periodic transient component. Comparisons with the classical L1-norm regularized signal decomposition method and spectral kurtosis show that the proposed method can accurately preserve the amplitude of the periodic transient component and provide a more accurate estimation result. Experiment and engineering case studies further verify that the proposed method can accurately estimate the periodic transient component from vibration signals, which demonstrate that the proposed method is a promising tool for gearbox fault diagnosis.",
keywords = "Convex optimization, Gearbox fault diagnosis, Generalized minimax-concave penalty, Signal decomposition, Sparse representation",
author = "Gaigai Cai and Ivan Selesnick and Shibin Wang and Weiwei Dai and Zhongkui Zhu",
year = "2018",
month = "10",
day = "13",
doi = "10.1016/j.jsv.2018.06.037",
language = "English (US)",
volume = "432",
pages = "213--234",
journal = "Journal of Sound and Vibration",
issn = "0022-460X",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - Sparsity-enhanced signal decomposition via generalized minimax-concave penalty for gearbox fault diagnosis

AU - Cai, Gaigai

AU - Selesnick, Ivan

AU - Wang, Shibin

AU - Dai, Weiwei

AU - Zhu, Zhongkui

PY - 2018/10/13

Y1 - 2018/10/13

N2 - Vibration signals arising from faulty gearboxes are often a mixture of the meshing component and the periodic transient component, and simultaneously contaminated by noise. Sparsity-assisted signal decomposition is an effective technique to decompose a signal into morphologically distinct components based on sparse representation and optimization. In this paper, we propose a sparsity-enhanced signal decomposition method which uses the generalized minimax-concave (GMC) penalty as a nonconvex regularizer to enhance sparsity in the sparse approximation compared to classical sparsity-assisted signal decomposition methods, and thus to improve the decomposition accuracy for gearbox fault diagnosis. Even though the GMC penalty itself is nonconvex, it maintains the convexity of the GMC regularized cost function to be minimized. Hence, similar to the classical L1-norm regularization methods, the global optimal solution can be guaranteed via convex optimization. Moreover, we present and validate a straight-forward way to choose transforms and set parameters for the proposed method. Through simulation studies, it is demonstrated that the proposed sparsity-enhanced signal decomposition method can effectively decompose the simulated faulty gearbox signal into the meshing component and the periodic transient component. Comparisons with the classical L1-norm regularized signal decomposition method and spectral kurtosis show that the proposed method can accurately preserve the amplitude of the periodic transient component and provide a more accurate estimation result. Experiment and engineering case studies further verify that the proposed method can accurately estimate the periodic transient component from vibration signals, which demonstrate that the proposed method is a promising tool for gearbox fault diagnosis.

AB - Vibration signals arising from faulty gearboxes are often a mixture of the meshing component and the periodic transient component, and simultaneously contaminated by noise. Sparsity-assisted signal decomposition is an effective technique to decompose a signal into morphologically distinct components based on sparse representation and optimization. In this paper, we propose a sparsity-enhanced signal decomposition method which uses the generalized minimax-concave (GMC) penalty as a nonconvex regularizer to enhance sparsity in the sparse approximation compared to classical sparsity-assisted signal decomposition methods, and thus to improve the decomposition accuracy for gearbox fault diagnosis. Even though the GMC penalty itself is nonconvex, it maintains the convexity of the GMC regularized cost function to be minimized. Hence, similar to the classical L1-norm regularization methods, the global optimal solution can be guaranteed via convex optimization. Moreover, we present and validate a straight-forward way to choose transforms and set parameters for the proposed method. Through simulation studies, it is demonstrated that the proposed sparsity-enhanced signal decomposition method can effectively decompose the simulated faulty gearbox signal into the meshing component and the periodic transient component. Comparisons with the classical L1-norm regularized signal decomposition method and spectral kurtosis show that the proposed method can accurately preserve the amplitude of the periodic transient component and provide a more accurate estimation result. Experiment and engineering case studies further verify that the proposed method can accurately estimate the periodic transient component from vibration signals, which demonstrate that the proposed method is a promising tool for gearbox fault diagnosis.

KW - Convex optimization

KW - Gearbox fault diagnosis

KW - Generalized minimax-concave penalty

KW - Signal decomposition

KW - Sparse representation

UR - http://www.scopus.com/inward/record.url?scp=85048946960&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048946960&partnerID=8YFLogxK

U2 - 10.1016/j.jsv.2018.06.037

DO - 10.1016/j.jsv.2018.06.037

M3 - Article

VL - 432

SP - 213

EP - 234

JO - Journal of Sound and Vibration

JF - Journal of Sound and Vibration

SN - 0022-460X

ER -