Empirical Mode Decomposition Based Method for Fault Diagnosis in a Helical Gearbox

Vibration measurement and analysis are widely used for machine fault diagnosis because it carries the rich information
of machine conditions. However, in practice, vibration signals are often interfered by noise or influenced by
transmission paths and operating conditions, such as loading. These factors make it difficult to distinguish and identify
the characteristic frequencies result from the faults.
In this paper, the empirical mode decomposition (EMD) based methods was proposed to extract feature based on the
intrinsic mode functions corresponding to the faults so as to eliminate the interferences’ influences. This method was
applied to gearbox fault diagnosis and the experimental results show that this method can extract the time-frequency
features effectively for different severity gear faults independent of the operating conditions.