PROGNOSTICS OF HIGH SPEED ROLLING ELEMENT BEARINGS

Abstract

Bearings are the machine elements widely used in rotating machine parts and subjected to dynamic conditions, so are more prone to failure. Their close monitoring is necessary to reduce the down time and catastrophic failure of machine. 1-lence, with increasing demand, the maintenance strategies are nowadays based on the current health state so that unnecessary machine downtime and manufacturing cost are reduced. Diagnostics and Prognostics are used for monitoring the condition of the bearing. While a lot of advancement has been done in field of Diagnostics, the Prognostics field is relatively new. Infact diagnostics is part of prognostics. In prognosis we estimate the remaining useful life based on current health condition. In the present study run time to failure test has been conducted for different SKF 6205 bearings under varying shaft speeds and loading condition using a test rig as bearing prognostics simulator. The data is captured online and saved for further analysis using DAQ system & Vibra Quest Software. Various features are extracted from these signals. The extracted features are then used for fault diagnostics and prognostics. The bearings after complete failures have been disassembled completely to look for the faults and confirm with the results using time domain and frequency domain techniques.