Vibration-based fault diagnosis in rolling element bearings: ranking of various time, frequency and time-frequency domain data-based damage identi cation parameters

TLDR: Shakya et al. as discussed by the authors performed a comparative study of various vibration signal-based damage identification parameters for rolling element bearings and concluded that the results suggest that the ranking is quite consistent, even with a different bearing type and damage characteristic.

Abstract: Piyush Shakya, Ashish K Darpe and Makarand S Kulkarni are with the Vibration Research Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Delhi, New Delhi – 110016, India. A comparative study of various vibration signal-based damage identification parameters for rolling element bearings is undertaken. Defects of varying severity are seeded on the outer and inner races of a double-row angular contact bearing. The influence of a defect and its severity on the observed identification parameters is investigated using vibration data acquired from the bearing housing. A comparison among the various time domain, frequency domain and time-frequency domain parameters is made based on their robustness, sensitivity to damage change and early detectivity of the bearing faults. An overall ranking of the parameters is attempted with the objective of ascertaining effective damage identification parameters from among those available for diagnosis of the rolling element bearings. Validation of the ranking is carried out with the data obtained on a different test-rig for early detectivity of damage. The results suggest that the ranking is quite consistent, even with a different bearing type and damage characteristic.