Rotor-Stator Rub-Impact Fault and Position Identification of Aero-Engine Based on VMD-MF-Cepstrum-KNN

Abstract

In order to precisely estimate rub-impact fault between rotor and stator in aero-engine and identify a rub-impact location, the method combining variation modal decomposition (VMD), margin factor (MF) and cepstrum has been proposed (VMD-MF-Cepstrum). Firstly, to effectively separate rub-impact fault information, VMD algorithm has been recruited to decompose original vibration acceleration signals due to its excellent time-frequency locality and the result provides intrinsic mode function (IMF) from different frequency bands. Furthermore, in view of the sensitivity of margin factor to the wear extent of equipment, it has been used to select the sensitive IMF which can best embody the characteristic information of rub-impact faults and is labelled as optimal IMF. Moreover, considering about the consistency and difference of transfer path of rub-impact fault characteristic information shown by the signals obtained by the sensors which are installed in the different or same position when rubbing position is the same (or different), cepstrum has been taken advantage to depict the feature of transfer path and the transfer path from optimal IMF is extracted. Finally, according to the characteristics which correspond to the transfer-path and the k-nearest neighbor (KNN) algorithm, the fault and locations of rub-impact were judged and identified. Check analysis from 2 independent experiments conducted in 2 days and comparison with classical cepstrum method both indicate that the proposed method will be more ideal for unknown samples. Compared with classical cepstrum method, the correctly identified percentage which corresponds to 2 independent experiments has been raised 52.00% and 43.25%.