With more dynamic condition information, the built-in encoder signal is more superior and convenient than the traditional vibration signal in the mechanical system fault diagnosis. However, the early incipient fault features are apt to be submerged in the raw encoder signal due to the predominant feature of increasing. In this paper, the raw encoder signal is derived firstly to obtain the instantaneous angular speed. Then, IAS as the input series of the algorithm of maximum second-order cyclostationarity blind deconvolution (CYCBD), is used to enhance the fault features and identify the condition of the planetary gearbox. Through the simulation and experiment case, the feasibility of CYCBD in IAS signal can be verified.

Encoder signal, Planetary gearbox, Feature enhancement, Maximum second-order cyclostationarity blind deconvolution (CYCBD)