Damage detection in gear systems using empirical mode decomposition

A method for monitoring the evolution of gear faults based on the newly developed empirical mode decomposition scheme is presented. A theoretical model for a gear pair with a tooth root crack was developed. Experimental vibration signals from a test rig were decomposed into oscillatory functions called intrinsic mode functions. An empirical law, which relates the energy content of the intrinsic modes to the crack magnitude, was established. The modal energy is thus associated with the deterioration in gear condition and can be utilized for system failure prediction. In addition, it is shown that the instantaneous frequency of the vibration signal is a sensitive indicator of the existence of damage in the gear pair. (C) 2004 Elsevier Ltd. All rights reserved.