Robust incipient fault identification of aircraft engine rotor based on wavelet and fraction

Signal de-noising and diagnosis of the weak signature are crucial to aircraft engine prognostics in which case features are often very weak and masked by noise. Robust methods are needed to provide more evident information for aircraft engine incipient fault diagnosis and prognostics. This paper develops enhanced and robust prognostic methods for aircraft engine including wavelet based method for weak signature enhanced for adaptive de-noising and correlation dimension based for incipient fault diagnosis. Firstly, the adaptive wavelet de-noising method is used to reduce noise of the vibration signal. Then, correlation dimension of the vibration signal after de-noising is computed, and the correlation dimension is used as the character parameter for identifying the fault deterioration grade. Experiment of the aircraft engine rotor is carried out. The experimental results demonstrate that: (1) the different rotor faults show different kinematics mechanisms; (2) the singular signal of incipient fault on aircraft engine rotor can be effectively extracted by adaptive de-noising; (3) the incipient fault of aircraft engine rotor can be fast distinguished by the correlation dimension; (4) it provides a effective way for robust incipient fault identification of aircraft engine rotor to combine wavelet with fraction.