Abstracts Submitted: 510
Number of Users: 710
View Abstracts Submitted
Back to home Page
Rolling element bearing is one of the most widely used industrial machinery elements. Their condition is very significant for proper functioning of machines. Over the past few decades, extensive research work has been carried out to investigate the vibration response of rolling element bearings having localized or distributed defects. The localized defect can have a single spall like fault on any of the element of the bearing; extended type fault on the raceways of inner or outer ring; or multiple faults of local in nature with their presence on either a single raceway or simultaneously on both the raceways. The nature of signal generation for single local defect and extended defect has been researched upon elaborately and the techniques to monitor the same are well established. The dearth however still exists in the modelling the signal generation by multiple faults and the techniques or the methodology to identify them have not been well summarized in literature. This paper addresses this gap by presenting a review and critically discussing the current progress of condition monitoring techniques for bearings with single local defect, extended defect and multiple defects. Vibrations generated by a single local fault is known to generate periodic impact force primarily as a function of location of defect, defect size, load on the bearing and shaft speed. The vibration models for a local fault are several and the domain fall into one of the following categories: impulse train models, additional deflection models, stochastic excitation models, multi-event excitation models, finite element models. The techniques to monitor the presence of a local fault have been broadly in time domain, frequency domain or time-frequency domain. The size of defect has been recently researched upon with assistance from time domain signal and the same cannot be predicted in frequency domain using Fast Fourier Transform (FFT). The extended fault has been identified by determining the angular spacing between the low frequency entry event and high frequency exit event. Vibrations generated in deep groove ball bearings due to the presence of multiple defects on their races have been studied in time and frequency domains. The frequency spectra do not provide any information about number of defects; however this information is found in time domain analysis. In order to detect multiple defects on one component of the bearing, a method based on the high frequency resonance technique (HFRT) is introduced. The time constant in the envelope detector is used to find the pattern of the amplitude of defect frequency harmonics (ADFH) in the frequency domain. The experimental results confirm the ability of the proposed method to diagnose multiple defects. An IMF-based adaptive envelope order analysis (IMF-AEOA) is proposed for bearing fault detection under harsh conditions. This approach is established through combining the ensemble empirical mode decomposition (EEMD), envelope order tracking and fault sensitive analysis. In this scheme, EEMD provides an effective way to adaptively decompose the raw vibration signal into IMFs with different frequency bands. The envelope order tracking is further employed to transform the envelope of each IMF to angular domain to eliminate the spectral smearing induced by speed variation, which makes the bearing characteristic frequencies more clear and discernible in the envelope order spectrum. Finally, a fault sensitive matrix is established to select the optimal IMF containing the richest diagnostic information for final decision making. The effectiveness of IMF-AEOA is validated by simulated signal and experimental data from locomotive bearings. The fault diagnosis method of rolling element bearing compound faults based on Sparse No-Negative Matrix Factorization (SNMF)-Support Vector Data Description (SVDD), Support Vector Machines (SVM) and neurofuzzy Min-Max classifiers is summarized. The pre-processing procedure and the classification stage, especially in the case of Min-Max fuzzy networks, do not require demanding computational hardware resources, and as a result, a simple and effective diagnostic system can be designed by feeding the synthesized Min-Max classifiers with the spectral features computed from vibration sensor outputs. The paper discusses the key challenges of previous works and finally a summary of the literature is presented followed by recommendations for future research.
© Copyright 2017 All Rights Reserved