Blind Separation of Vibration Sources using Deep Learning and Deconvolution

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Makienko, Igor, Grebshtein, Michael, Gildish, Eli
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917672388657152
author Makienko, Igor
Grebshtein, Michael
Gildish, Eli
author_facet Makienko, Igor
Grebshtein, Michael
Gildish, Eli
contents Vibrations of rotating machinery primarily originate from two sources, both of which are distorted by the machine's transfer function on their way to the sensor: the dominant gear-related vibrations and a low-energy signal linked to bearing faults. The proposed method facilitates the blind separation of vibration sources, eliminating the need for any information about the monitored equipment or external measurements. This method estimates both sources in two stages: initially, the gear signal is isolated using a dilated CNN, followed by the estimation of the bearing fault signal using the squared log envelope of the residual. The effect of the transfer function is removed from both sources using a novel whitening-based deconvolution method (WBD). Both simulation and experimental results demonstrate the method's ability to detect bearing failures early when no additional information is available. This study considers both local and distributed bearing faults, assuming that the vibrations are recorded under stable operating conditions.
format Preprint
id arxiv_https___arxiv_org_abs_2405_12774
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Blind Separation of Vibration Sources using Deep Learning and Deconvolution
Makienko, Igor
Grebshtein, Michael
Gildish, Eli
Machine Learning
Artificial Intelligence
Audio and Speech Processing
Signal Processing
Vibrations of rotating machinery primarily originate from two sources, both of which are distorted by the machine's transfer function on their way to the sensor: the dominant gear-related vibrations and a low-energy signal linked to bearing faults. The proposed method facilitates the blind separation of vibration sources, eliminating the need for any information about the monitored equipment or external measurements. This method estimates both sources in two stages: initially, the gear signal is isolated using a dilated CNN, followed by the estimation of the bearing fault signal using the squared log envelope of the residual. The effect of the transfer function is removed from both sources using a novel whitening-based deconvolution method (WBD). Both simulation and experimental results demonstrate the method's ability to detect bearing failures early when no additional information is available. This study considers both local and distributed bearing faults, assuming that the vibrations are recorded under stable operating conditions.
title Blind Separation of Vibration Sources using Deep Learning and Deconvolution
topic Machine Learning
Artificial Intelligence
Audio and Speech Processing
Signal Processing
url https://arxiv.org/abs/2405.12774