A rolling bearing fault signal denoising algorithm that combines a new adaptive information entropy with a new wavelet threshold function DOI
Min Li, Xuemei Li, Bin Liu

et al.

Engineering Research Express, Journal Year: 2024, Volume and Issue: 6(4), P. 045536 - 045536

Published: Oct. 25, 2024

Abstract Mechanical fault diagnosis is of great significance to industrial automation, and extracting vibration signals one the important tasks in mechanical health monitoring diagnosis. However, due complex working environment rolling bearings, a large amount noise makes it difficult extract signals. Denoising signal bearings can remove interference noise, simplify early identification features, thus improve diagnostic accuracy maintenance efficiency. This paper proposes bearing denoising algorithm, which constructs new feature extraction function. method first decomposes noisy into Intrinsic Mode Functions (IMFs) by Computing Expressive Empirical Decomposition with Adaptive Noise (ICEEMDAN). Secondly, adaptive information entropy threshold function constructed IMFs from it. Then, IMF denoised wavelet Finally, noise-free are reconstructed reconstruct signal. To verify actual performance comparative experiments were conducted on self-collected dataset public dataset, results show that this improves continuity reconstruction various types more effectively accurately, thereby improving detection 2%–9%.

Language: Английский

A robust optimization strategy for brushless DC motor repetitive controller based on H-infinity DOI Creative Commons

Tianli Wang,

Tianqing Yuan,

Jing Bai

et al.

AIP Advances, Journal Year: 2024, Volume and Issue: 14(9)

Published: Sept. 1, 2024

The control of Brushless DC (BLDC) motor has many challenges. A large number harmonics are generated during operation, while it is susceptible to external disturbances such as noise. torque ripple also a constraint the large-scale popularity BLDC motors in some high-end industries. Taking system research object, this paper proposes method H-infinity repetitive control. First, state space expression derived from mathematical model motor. Then, parameters low-pass filter internal mode link obtained for design In addition, constructed and converted into standard form system. Finally, bringing generalized controlled object MATLAB derive compliant compensator. simulation results show that proposed can not only suppress but improve robustness

Language: Английский

Citations

1
A rolling bearing fault signal denoising algorithm that combines a new adaptive information entropy with a new wavelet threshold function DOI
Min Li, Xuemei Li, Bin Liu

et al.

Engineering Research Express, Journal Year: 2024, Volume and Issue: 6(4), P. 045536 - 045536

Published: Oct. 25, 2024

Abstract Mechanical fault diagnosis is of great significance to industrial automation, and extracting vibration signals one the important tasks in mechanical health monitoring diagnosis. However, due complex working environment rolling bearings, a large amount noise makes it difficult extract signals. Denoising signal bearings can remove interference noise, simplify early identification features, thus improve diagnostic accuracy maintenance efficiency. This paper proposes bearing denoising algorithm, which constructs new feature extraction function. method first decomposes noisy into Intrinsic Mode Functions (IMFs) by Computing Expressive Empirical Decomposition with Adaptive Noise (ICEEMDAN). Secondly, adaptive information entropy threshold function constructed IMFs from it. Then, IMF denoised wavelet Finally, noise-free are reconstructed reconstruct signal. To verify actual performance comparative experiments were conducted on self-collected dataset public dataset, results show that this improves continuity reconstruction various types more effectively accurately, thereby improving detection 2%–9%.

Language: Английский

Citations

0