Detection of particle contaminants in rolling element bearings with unsupervised acoustic emission feature learning

Martin-del-Campo, Sergio; Schnabel, Stephan; Sandin, Fredrik; Marklund, Pär

doi:10.1016/j.triboint.2018.12.007

Cited by 17 publications

(5 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…77,[83][84][85][86] Also, the number of particles seems to correlate to the hit-rate of the contamination initiated signals. 77,83 Besides these conventional characterisation approaches, machine learning algorithms -such as sparse dictionary learning 87 and a convolutional neural network 88 -have also been applied in early-stage studies to differentiate between contaminated and uncontaminated lubrication.…”

Section: Introductionmentioning

confidence: 99%

Acoustic emission monitoring of naturally developed damage in large-scale low-speed roller bearings

Scheeren

Kaminski

Pahlavan

2023

Structural Health Monitoring

View full text Add to dashboard Cite

This article presents an approach to identify naturally developed damage in low-speed bearings using waveform-similarity-based clustering of acoustic emissions (AEs) under fatigue loading. The approach is motivated by the notation that each recorded AE signal from a particular damage is defined by the convolution of the source signal, transfer function of the propagation path and transfer function of the utilised sensor, and may thusly be used to identify consistent AE sources, for example due to crack growth. A sequential clustering procedure is proposed, that is based on waveform cross-correlation. The supporting theoretical background of waveform similarity, rooted in an analytical formulation of waveform propagation and transmission in complex structures, is discussed. The presented methodology is evaluated through application to AE data obtained in a low-speed run-to-failure experiment utilising a densely instrumented purpose-built linear bearing segment. The implemented sensor system comprises arrays of three types of AE transducers, that is relatively low - (40–100 kHz), mid - (95–180 kHz) and high-frequency (180–580 kHz), that are situated on both the raceways and supporting substructures of either side of the bearing. Over the course of 225,000 cycles of extension and retraction, wear has been developed. A total of about ∼2,300,000 AE signals have been recorded. Analysis of the recorded data suggests the rate of degradation increases from around 70,000 cycles onwards. Highly consistent structures of clusters indicative of a localised defect in the raceway have been identified from around 170,000 cycles onwards. These clusters are characterised by hit-rates in the range of 1–2 hits per cycle and an average similarity of 93%, they comprise about half the AE activity for the periods they have been identified for. These results highlight that the proposed cross-correlation-based clustering of AE waveforms and identification of multi-channel formations in said clusters compose a suitable methodology for assessment of damage in low-speed roller bearings.

show abstract

Section: Introductionmentioning

confidence: 99%

Acoustic emission monitoring of naturally developed damage in large-scale low-speed roller bearings

Scheeren

Kaminski

Pahlavan

2023

Structural Health Monitoring

View full text Add to dashboard Cite

show abstract

“…A few instances of its app in fault diagnosis can be found in the literature. [4][5][6][7][8][9][10][11][12] Samanta and Al-Balushi 4 described the procedure for the app of artificial neural networks in bearing fault detection. The method comprises five node inputsroot mean square (RMS), variance, kurtosis, skewness and normalized sixth central moment derived from time signal.…”

Section: Introductionmentioning

confidence: 99%

“…The technique employed wavelet analysis and artificial neural network. Martin-del-Campo et al 12 used AE and unsupervised learning technique to study the particle contamination in a ball bearing.…”

Section: Introductionmentioning

confidence: 99%

Classification and detection of cavitation, particle contamination and oil starvation in journal bearing through machine learning approach using acoustic emission signals

Poddar

Tandon

2021

Proceedings of the Institution of Mechanical Engineers, Part J:

View full text Add to dashboard Cite

The ability to classify condition-monitoring data and make a decision can be imparted to a computer through the machine learning approach. In this article, the acoustic emission signals emerging from journal bearings under normal operating conditions and faulty states, namely cavitation, particle contamination and oil starvation, have been classified to develop fault-prediction model using the machine learning approach. Furthermore, an application has been developed that takes acoustic emission data as input and diagnoses the category of faults besides triggering an alarm under faulty states.

show abstract

“…The AE technique is therefore suitable for predicting the fatigue of bearings and for evaluating the mode of damage. Several studies have been conducted on the diagnosis of abnormalities in rolling bearings and in bearing steels by using the AE technique [3][4][5][6][7][8][9]. However, most of these were based on a comparison between the operating time of the rolling bearing and changes in AE signals.…”

Section: Introductionmentioning

confidence: 99%

Early Detection and Identification of Fatigue Damage in Thrust Ball Bearings by an Acoustic Emission Technique

Hase

2020

Lubricants

View full text Add to dashboard Cite

As rolling bearings are widely used in various machines, there is a strong need to detect any problems as early as possible. Although vibration analysis is commonly used in the diagnosis of rolling bearings, it is possible that the failure of such bearings might be detected earlier by an acoustic emission (AE) technique. Methods for detecting potential fatigue damage in a thrust ball bearing by AE signal analysis and by vibration analysis were compared. For the AE signal analysis, the maximum amplitude and the frequency spectrum were used to detect and identify fatigue damage in the bearing. Features of AE signals detected when a defect was artificially formed on the raceway surface of a bearing by using a Vickers hardness tester were also examined. The AE technique detected initial cracks due to fatigue damage earlier than the vibration technique. Additionally, AE signals were always detected during bearing fatigue tests, but the AE signals detected during the running-in process, crack initiation, crack propagation, and flaking all contained different frequency components. Furthermore, the correlation map between the frequency spectra of AE signals and deformation and fracture phenomena (friction and wear modes) was updated by adding the new findings of this study.

show abstract

Detection of particle contaminants in rolling element bearings with unsupervised acoustic emission feature learning

Cited by 17 publications

References 24 publications

Acoustic emission monitoring of naturally developed damage in large-scale low-speed roller bearings

Acoustic emission monitoring of naturally developed damage in large-scale low-speed roller bearings

Classification and detection of cavitation, particle contamination and oil starvation in journal bearing through machine learning approach using acoustic emission signals

Early Detection and Identification of Fatigue Damage in Thrust Ball Bearings by an Acoustic Emission Technique

Contact Info

Product

Resources

About