Fault Diagnosis of Rolling Bearings Using Data Mining Techniques and Boosting

Abstract

Rolling bearings are key components in most mechanical facilities; hence, the diagnosis of their faults is very important in predictive maintenance. Up to date, vibration analysis has been widely used for fault diagnosis in practice. However, acoustic analysis is still a novel approach. In this study, acoustic analysis with classification is used for fault diagnosis of rolling bearings. First, Hilbert transform (HT) and power spectral density (PSD) are used to extract features from the original sound signal. Then, decision tree algorithm C5.0, support vector machines (SVMs) and the ensemble method boosting are used to build models to classify the instances for three different classification tasks. Performances of the classifiers are compared w.r.t. accuracy and receiver operating characteristic (ROC) curves. Although C5.0 and SVM show comparable performances, C5.0 with boosting classifier indicates the highest performance and perfectly discriminates normal instances from the faulty ones in each task. The defect sizes to create faults used in this study are notably small compared to previous studies. Moreover, fault diagnosis is done for rolling bearings operating at different loading conditions and speeds. Furthermore, one of the classification tasks incorporates diagnosis of five states including four different faults. Thus, these models, due to their high performance in classifying multiple defect scenarios having different loading conditions and speeds, can be readily implemented and applied to real-life situations to detect and classify even incipient faults of rolling bearings of any rotating machinery.

Data mining in predictive maintenance systems: A taxonomy and systematic review

“…In this way, the aim is to simulate behaviors that could occur in a real system to which there is not access or which has not yet failures that can be used to train the models. In this line, there are many works that collect vibration signals from different experimental rotors (Cakir et al, 2021; Li et al, 2019; Liang et al, 2020; Liao et al, 2016; Pang et al, 2020; Qian et al, 2019; Satishkumar & Sugumaran, 2017; Song et al, 2021; Una et al, 2017; Wang et al, 2018; Wang, Zhang, et al, 2020; Yang, Lei, et al, 2019; Zhang, Li, Wang, et al, 2019), fans (Sampaio et al, 2019; Xu et al, 2021; Zenisek, Holzinger, & Affenzeller, 2019), centrifugal pumps (Hu et al, 2020), air compressors (Cupek et al, 2018) or industrial robots (Panicucci et al, 2020). Some works (Venkataswamy et al, 2020; Zenisek, Holzinger, & Affenzeller, 2019; Zenisek, Kronberger, et al, 2019) generate synthetic data from a mathematical approach that models the behavior of their problem, using for that specialized software in simulation.…”

“…Another proposal (Xu et al, 2021) that works with vibration data compares the performance of the FFT, Higher Order Statistics and Structural Co‐occurrence Matrix. In Una et al (2017), a processing closely related to vibration signals, but applied to acoustic signals, is carried out. Thus, feature extraction from the original sound signal is done by enveloping analysis with Hilbert transform and frequency components extraction with FFT and Power Spectral Density.…”

“…Thus, linear SVM has been used in binary classification (Cakir et al, 2021; Gohel et al, 2020; Khodabakhsh et al, 2018; Lasisi & Attoh‐Okine, 2018; Pałasz & Przysowa, 2019; Proto et al, 2020; Shamayleh et al, 2020; Susto et al, 2015) to classify the events according to their proximity to the failure threshold. In multiclass scenarios, linear SVM has been studied together with other methods in both prediction of multiple degradation states (Chen, Hsu, et al, 2021; Cheng et al, 2020; Quatrini et al, 2020; Shafi et al, 2018; Venkataswamy et al, 2020), and several failure modes detection (Una et al, 2017; Xu et al, 2021). Specifically, in Cheng et al (2020) and Shafi et al (2018), SVM achieves the best performance.…”

Degradation Index Extraction and Degradation Trend Prediction for Rolling Bearing

“…Rolling bearings have a widespread use in most industrial applications of rotating machinery such as motors, electrical generators, pumps, and cutting machines [1]. The failure of rolling bearing can break off the production in rotating machinery, causing unforeseen downtime and economic losses.…”

Evaluation for Bearing Wear States Based on Online Oil Multi-Parameters Monitoring

“…Vibration signals processing and lubrication oil analysis are the most often adopted methods to fulfill the task [ 2 , 3 , 4 , 5 ]. Acoustic analysis with classification can also be used for fault diagnosis of rolling bearings as presented in [ 6 ]. However, there is some lack of knowledge about the fault diagnosis method based on vibration or acoustic signals analysis for bearings in terms of high frequency of compound faults [ 7 ] and mechanical systems vibrating intensively with large noise.…”