Bearing fault feature extraction based on Hankel-AFMD-ILTP
Measurement Science and Technology · 10.04.2026 · 37 (15) · 156104
DOI: 10.1088/1361-6501/ae4cb3 ↗Kurz gefasst
Fachartikel in der Zeitschrift Measurement Science and Technology, online erschienen am 15.04.2026. Verfasst von Ziru Ma und 3 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Gefördert von: Natural Science Foundation of Liaoning Province, Department of Education of Liaoning Province, Aeronautical Science Foundation of China.
Abstract (Originalsprache)
Since the one-dimensional local ternary pattern (1D-LTP) method is sensitive to noise and has difficulty recognizing bearing faults, a method combining phase space reconstruction, adaptive feature model decomposition and improved 1D-LTP (Hankel-AFMD-ILTP) is proposed. First, a Hankel matrix of a signal is constructed to dig out the hidden feature information of low-dimensional space. Second, the covariance matrix of the Hankel matrix is calculated and related reconstructed signals are acquired to reduce the noise interference and strengthen the periodic features of signals. Third, fractional order permutation entropy is introduced to adaptively determine the optimal mode number and filter length of FMD, and AFMD is proposed; furthermore, the reconstructed signals (in place of the original signals) are decomposed by AFMD, and the feature components are obtained. Then, the improved 1D-LTP (ILTP) method is proposed. The absolute value of difference between the average value of whole signal and the local average value within a window is taken as the quantization threshold to solve the problem of inaccurate signal quantization of 1D-LTP. Next, feature components are quantized with the ILTP, and new decimal sequences are obtained. Finally, optimal decimal sequences are chosen according to the square envelope Gini coefficient, the features of the compound faults of rolling bearings and intershaft bearings are extracted, and the faults are determined according to the power spectrum of the optimal feature component. The results indicate that the proposed Hankel-AFMD-ILTP method can successfully extract the features of compound faults of rolling bearings and intershaft bearings of real aeroengines, and accurately determine faults.
Förderung
- Natural Science Foundation of Liaoning Province
- Department of Education of Liaoning Province
- Aeronautical Science Foundation of China
- National Natural Science Foundation of China
- Steady fundamental supporting project phase II for scientific research institute of military industry
Zitieren
APA
Ma, Z., Yu, M., Ge, X., & Yu, L. (2026). Bearing fault feature extraction based on Hankel-AFMD-ILTP. Measurement Science and Technology, 37(15), 156104. https://doi.org/10.1088/1361-6501/ae4cb3
Vancouver
Ma Z, Yu M, Ge X, Yu L. Bearing fault feature extraction based on Hankel-AFMD-ILTP. Measurement Science and Technology. 2026;37(15):156104. doi:10.1088/1361-6501/ae4cb3
BibTeX
@article{Ma2026,
author = {Ma, Ziru and Yu, Mingyue and Ge, Xiangdong and Yu, Lipei},
title = {Bearing fault feature extraction based on Hankel-AFMD-ILTP},
journal = {Measurement Science and Technology},
year = {2026},
volume = {37},
number = {15},
pages = {156104},
doi = {10.1088/1361-6501/ae4cb3}
}
EUVIMED