Comparison of Selected Point Estimators of Vibration Signals for Purposes of Fault Detection in Rolling Bearings

Most of machine breakdowns relate to bearing failures thus it is very important to diagnose bearing conditions. The main purpose of the study was to classify the condition of bearings and identify defective ones based on visual inspection and the values of classic parameters of the acceleration signal vibration, such as Peak, RMS, Kurtosis. The results were compared to parameters provided by the SPM method. All vibration parameters were estimated for high pass filtered signals where filters had following cut-off frequencies 0.5; 1, 2, 5 and 8 kHz. Bearings were tested on the laboratory test bench being built at the Silesian University of Technology. Based on the signal analysis and visual inspections it can be stated that there is agreement in the assessment of the conditions between the parameters of the SPM method and the Peak and RMS parameters. It was observed that the sensitivity to the existence of low-intensity defects increases when the vibration parameters are determined for signals in the band above 2kHz.