Fault Diagnosis of Ball Bearing with Advance Signal Processing Techniques

Abstract

vii newlineABSTRACT newlineRolling element bearings are extensively used in most of the rotating machines to newlinesupport static/dynamic loads. These bearings can take up both radial and axial newlineloads for most of the applications. They have a great influence on the dynamic newlinebehaviour of the rotating machines and act as a source of vibration and noise in newlinethese systems. There is a critical need to increase reliability and fault diagnosis of newlinerolling element bearings to prevent catastrophic failure of the machinery. newlineIn this research, the advance signal processing techniques are used for processing newlineof rolling element bearing fault signals. It s significant advantages over other newlinefrequency domain machine vibration techniques. The effectiveness of the newlineenvelope analysis technique is combined with the flexibility of the wavelet packet newlinetransform, helping in the minimization of interventions by the end user. A timefrequency newlinedecomposition of a vibration signal is provided and the components newlinecarrying the important diagnostic information are selected for further processing. newlineExperimental validation with actual vibration signals measured from bearings newlinewith seeded defects on bearing elements. Experimental set up has been developed newlineto create a database of dynamic responses of rolling bearing at Green Ksv skill newlinedevelopment centre, LDRP-ITR campus, Gandhinagar. The rotation speed has newlinebeen up to 10000 rpm in case of balanced rotor bearing conditions. Four bearing newlineconditions have been analysed, such as: healthy bearing, bearing with outer race newlinedefect, inner race defect and combined defect on outer race, inner race and rolling newlineelement. newlineThe model-based fault diagnosis method has attempted to diagnose incipient fault newlinedetection and classification of bearing with data driven approach. Feature newlineextraction technique has been developed with hybrid signal processing technique newlinebased on the band pass filtering nature of Empirical mode decomposition (EMD), newlinethe resonant frequency bands have allocated in specific mono component signals newlinecalled Intrinsic Mode Functions (IMFs)