Order Tracking: Angular Domain Features Extraction Method for Condition Monitoring of Variable Speed

Abstract

Most of the rotating machines operate at variable speed/cyclo-stationary due to fluctuation of power, load etc. It is difficult to monitor the condition of rotating machines in conventional time/frequency analysis when the fluctuation of speed of rotor is more than 0.5 Hz. In such circumstances, angle/time cyclo-stationary analysis of a machine signal is a fundamental procedure for condition monitoring of rotating machine operating at variable speed. Rolling element bearing is widely used in rotating machines starting from bicycle to more sophisticated machines like cryogenic engines. In the present paper, a hybrid diagnosis scheme based on computed order tracking and envelope analysis is used as diagnosis scheme for condition monitoring of rolling element bearing operating at a fluctuating speed of approximately 5–15 Hz. Experiments were conducted using Spectraquest machine faults simulator. Rolling element bearing with fault in the outer race is taken to show the effectiveness of the diagnosis scheme.