Abstract
Prognostic techniques play a critical role in predicting upcoming faults and failures in machinery or a system by monitoring any deviation in the operation. This paper presents a novel method to analyze multidimensional sensory data and use its characteristics in bearing health prognostics. Firstly, detrended fluctuation analysis (DFA) is exploited to evaluate the long-range correlations in ball bearing vibration data. The results reveal the existence of the crossover phenomenon in vibration data with two scaling exponents at the short-range and long-range scales. Among several data sets, applying the DFA method to vibration signals shows a consistent increase in the short-range scaling exponent toward bearing failure. Finally, Kendall’s tau is used as a ranking coefficient to quantify the trend in the scaling exponent. It was found that the Kendall’s tau coefficient of the vibration scaling exponent could provide an early warning signal (EWS) for bearing failure.
Highlights
Rolling element bearings (REBs) are one of the essential elements in rotary machinery, such as pumps, compressors, gearboxes, and electric motors
The results show a persistent increase in the average scaling exponents by moving moving from from the the first first time time series series
Using detrended fluctuation analysis (DFA), it was shown that long-range correlations exist in bearing vibration signals before slowing down in the vibration signals toward their failure
Summary
Rolling element bearings (REBs) are one of the essential elements in rotary machinery, such as pumps, compressors, gearboxes, and electric motors. REBs are susceptible to catastrophic failures when they are operating under abnormal conditions, including heavier loading, corrosion, overheating, etc. As a result of frequent wear and friction, they will experience a gradual degradation and functionality loss during their operation. REBs go through different stages (mild, moderate, and serious) of functionality loss before their failure. Prognostics and diagnostics have been active research areas in recent years [1,2]. Diagnostic techniques evaluate the fault severity during operation to help in planning immediate reactive maintenance and avoiding any future failure. Once a bearing fault has been detected, the time remaining before bearing failure is relatively very short. Prognostic techniques provide an early warning for bearing breakdown to plan the needed maintenance ahead of time. That permits a continuous and reliable operation of the mechanical system while enabling proactive maintenance
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
