A Primer on Vibrational Ball Bearing Feature Generation for Prognostics and Diagnostics Algorithms

Tom, Kwok

doi:10.21236/ada614145

Cited by 10 publications

(7 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Time domain analysis allows detecting defects and assessing their related magnitude by means of several statistical indicators, such as the peaks, peak-to-peak, energy content (Root Mean Square—RMS), the crest factor (CF), kurtosis (KU), skewness or the energy index (EI). In addition, combined time–frequency domain analysis allows capturing the progressive changes in spectrum components of the vibration signals in both frequency and time [ 40 , 41 , 42 , 43 ]. More recently, these classical methods have been replaced or reinforced by AI or ML techniques, used to identify defects and cope with noise or reduced measurement system bandwidth.…”

Section: Vibration-based Rolling Bearings Fault Diagnosis: a Brief Su...mentioning

confidence: 99%

See 1 more Smart Citation

Highly Reliable Multicomponent MEMS Sensor for Predictive Maintenance Management of Rolling Bearings

et al. 2023

View full text Add to dashboard Cite

In the field of vibration monitoring and control, the use of low-cost multicomponent MEMS-based accelerometer sensors is nowadays increasingly widespread. Such sensors allow implementing lightweight monitoring systems with low management costs, low power consumption and a small size. However, for the monitoring systems to provide trustworthy and meaningful data, the high accuracy and reliability of sensors are essential requirements. Consequently, a metrological approach to the calibration of multi-component accelerometer sensors, including appropriate uncertainty evaluations, are necessary to guarantee traceability and reliability in the frequency domain of data provided, which nowadays is not fully available. In addition, recently developed metrological characterizations at the microscale level allow to provide detailed and accurate quantification of the enhanced technical performance and the responsiveness of these sensors. In this paper, a dynamic calibration procedure is applied to provide the sensitivity parameters of a low-cost, multicomponent MEMS sensor accelerometer prototype (MDUT), designed, developed and realized at the University of Siena, conceived for rolling bearings vibration monitoring in a broad frequency domain (from 10 Hz up to 25 kHz). The calibration and the metrological characterization of the MDUT are carried out by comparison to a reference standard transducer, at the Primary Vibration Laboratory of the National Institute of Metrological Research (INRiM).

show abstract

Section: Vibration-based Rolling Bearings Fault Diagnosis: a Brief Su...mentioning

confidence: 99%

“… Example of characteristic rolling bearings frequency, and a typical time–frequency analysis. Left figure from [ 40 ]. …”

Section: Figurementioning

confidence: 99%

Highly Reliable Multicomponent MEMS Sensor for Predictive Maintenance Management of Rolling Bearings

et al. 2023

View full text Add to dashboard Cite

show abstract

“…A wide variety of feature-engineering methods have already been described in the literature [13,24]. The focus of the research conducted here is on the comparison of different feature-engineering methods that consider the temporal past in the context of feature generation.…”

Section: Feature Engineeringmentioning

confidence: 99%

On the Importance of Temporal Information for Remaining Useful Life Prediction of Rolling Bearings Using a Random Forest Regressor

et al. 2022

View full text Add to dashboard Cite

Rolling bearings are frequently subjected to high stresses within modern machines. To prevent bearing failures, the topics of condition monitoring and predictive maintenance have become increasingly relevant. In order to efficiently and reliably maintain rolling bearings in a predictive manner, an estimate of the remaining useful life (RUL) is of great interest. The RUL prediction quality achieved when using machine learning depends not only on the selection of the sensor data used for condition monitoring, but also on its preprocessing. In particular, the execution of so-called feature engineering has a major impact on prediction quality. Therefore, in this paper, various methods of feature engineering are presented based on rolling–bearing endurance tests and recorded structure-borne sound signals. The performance of these methods is evaluated in the context of a regression-based RUL model. Furthermore, the way in which the quality of RUL prediction can be significantly improved is demonstrated, by adding further processed, time-considering features.

show abstract

“…Thus, after Empirical mode decomposition, the vibration signal x(n) can be represented as a sum of "m" IMFs (IMF i , where i=1 to m) and residue (r m ) Histogram: A Histogram or a Discrete Probability Density Function provides a visualization profile for vibration data and can be used as a tool to characterize it. Vibration data from healthy bearing has Normal Gaussian distribution whereas there is proportional increase in the number of high levels of acceleration in the vibration data collected from a damaged or faulty bearing, This results in non-Gaussian PDF [18].…”

Section: A Overview Of Empirical Mode Decompositionmentioning

confidence: 99%

“…(2) Histogram lower bound (HL) is defined as 3where Moments: Four central statistical moments (i.e. moments calculated about the mean) have been used in the feature vector [18].…”

Section: Histogram Upper Bound (Hl) Is Defined Asmentioning

confidence: 99%

Rolling Element Bearing Fault Detection using Statistical Features and Ensemble Classifiers

Grover¹,

Turk²

2020

IJEAT

View full text Add to dashboard Cite

Rolling element bearing health condition is monitored by analysing its vibration signature. Raw vibration signal picked up through suitably placed accelerometers is difficult to analyse hence many signal processing techniques have been proposed and developed by researchers to process the data for suitably extracting an effective signal feature set. Various machine learning techniques have been used for interpretation and accurate fault diagnosis using this extracted feature set. In this study “Empirical mode decomposition” is used for pre-processing the raw vibration data. Six “Statistical features” are extracted from the best Intrinsic mode function obtained through EMD and “Ensemble machine learning classifiers” are used for bearing fault diagnosis. A stacked ensemble of five classifiers is proposed for accurate fault diagnosis and results are compared with conventional ensemble classifiers to prove its effectiveness

show abstract

A Primer on Vibrational Ball Bearing Feature Generation for Prognostics and Diagnostics Algorithms

Cited by 10 publications

References 9 publications

Highly Reliable Multicomponent MEMS Sensor for Predictive Maintenance Management of Rolling Bearings

Highly Reliable Multicomponent MEMS Sensor for Predictive Maintenance Management of Rolling Bearings

On the Importance of Temporal Information for Remaining Useful Life Prediction of Rolling Bearings Using a Random Forest Regressor

Rolling Element Bearing Fault Detection using Statistical Features and Ensemble Classifiers

Contact Info

Product

Resources

About