Page 125 - IJOCTA-15-4

Rolling bearing fault diagnosis method based on GJO–VMD, multiscale fuzzy entropy, and GSABO–BP...
Table 6. Diagnostic accuracy of different signal decomposition methods under varying conditions

Signal decomposition Diagnostic accuracy under varying testing sample sizes (%)
method 50 (10) 40 (20) 30 (30) 20 (40) 10 (50)
EMD 95.50 96.88 96.67 98.75 100.00
EEMD 88.00 86.67 85.56 86.67 93.33
GJO–VMD 99.00 100.00 100.00 100.00 100.00
Note: ( ) represents the number of training samples.
Abbreviations: EEMD: Ensemble empirical mode decomposition; EMD: Empirical mode decomposition;
GJO–VMD: Golden jackal optimization–variational mode decomposition.

ranging from 40 to 10. Compared with other AI tools statement
commonly used classifiers, such as KNN, SVM,
All authors confirm that no AI tools were used in
random forest, decision tree, and standard BP
the preparation of this manuscript.
neural networks, the proposed method exhibits
superior diagnostic accuracy and robustness.
In this study, the rolling bearing vibration sig- References
nals were investigated. It is anticipated that the
proposed method will also perform well when ap- 1. Song X, Wang H, Liu Y, Wang Z, Cui Y. A fault
plied to other signals. In future work, this method diagnosis method of rolling element bearing based
should be extended from rolling bearings to other on improved PSO and BP neural network. J In-
tell Fuzzy Syst. 2022;43(5):5965-5971
types of rotating machinery, such as gears and ro-
https://doi.org/10.3233/JIFS-213485
tors, to enhance the universality of diagnosis. At
2. Hu R, Zhang M, Xiang Z, Mo J. Guided deep sub-
the same time, by integrating the method with
domain adaptation network for fault diagnosis of
real-time monitoring systems, it will be possible
different types of rolling bearings. J Intell Manuf.
to develop fault diagnosis technologies suitable for
2023;34(5):2225-2240
online monitoring, thereby addressing the grow- https://doi.org/10.1007/s10845-022-01910-7
ing demand for real-time monitoring and rapid 3. Lei Y, Li N, Guo L, Li N, Yan T, Lin J. Machinery
diagnosis in industrial applications. health prognostics: a systematic review from data
acquisition to RUL prediction. Mech Syst Signal
Acknowledgments Process. 2018;104:799-834
https://doi.org/10.1016/j.ymssp.2017.11.016
None. 4. Liu R, Yang B, Zio E, Chen X. Artificial intelli-
gence for fault diagnosis of rotating machinery: A
Funding review. Mech Syst Signal Process. 2018;108:33-47
https://doi.org/10.1016/j.ymssp.2018.02.016
None.
5. Zhou Y, Jin Z, Zhang Z, Geng Z, Zhou L. Ad-
versarial subdomain adaptation method based on
Conflict of interest multi-scale features for bearing fault diagnosis.
Math Found Comput. 2024;7(4):485-511
The authors declare they have no competing in-
https://doi.org/10.3934/mfc.2023024
terests.
6. Kanneg D, Wang W. A wavelet spectrum tech-
nique for machinery fault diagnosis. J Signal Inf
Author contributions Process. 2011;2(4):322-329
https://doi.org/10.4236/jsip.2011.24046
Conceptualization: Zhang Jing Song, Zhou Xiao
7. Feng H, Chen R, Wang Y. Feature extraction for
Long
fault diagnosis based on wavelet packet decompo-
Investigation: Zhang Jing Song, Wang Yan Zhen
sition: An application on linear rolling guide. Adv
Methodology: Zhou Xiao Long, Zheng Bin
Mech Eng. 2018;10(8):1687814018796367
Formal analysis: Jing Zhe, Li Hao Yu
https://doi.org/10.1177/1687814018796367
Writing – original draft: Zhang Jing Song 8. Zhou X, Wang X, Wang H, et al. Method for de-
Writing – review & editing: Zhang Jing Song, noising the vibration signal of rotating machinery
Zhou Xiao Long, Yang Soo Sing, Tan Min Keng through VMD and MODWPT. Sensors (Basel).
2023;23(15):6904
Availability of data https://doi.org/10.3390/s23156904
9. Huang NE, Shen Z, Long SR, et al. The empir-
The raw data supporting the findings of this study ical mode decomposition and the Hilbert spec-
are available at https://github.com/yyxyz/Case trum for nonlinear and non-stationary time series
WesternReserveUniversityData. analysis. Proc R Soc Lond A Math Phys Eng Sci.
667