In the gear is fault diagnosis

the key is how to extract the fault feature from random signals by proper methods. In fact

the signal we get will be polluted by noise. Some methods are sensitive to the noise and difficult to distinguish the faults

such as the power-spectrum. Spectral entropy based on the statistics indicates the orderlessness and is robust to noise. In this paper

spectral entropy is introduced into the gear transmission system

the fault features of the gear with crack and abrasion are extracted

and compared to that of normal gear. The results of analysis and experiment show that the precisions of identification are higher than 90%. It is shown that the spectral entropy is surely an effective method for the fault diagnosis of gear transmission system.