An analytical method for the Elastic Supporter Dynamic Stress Signals applied to Aero-engine fault diagnosis

Abstract

A new method based on analyzing Elastic Supporter dynamic stress signals used to diagnose the rotor system faults in small and medium-sized aero-engine is proposed. Firstly, singular value decomposition (SVD) was used to de-noise the elastic supporter dynamic stress signals, and the theory of the difference energy entropy of singular values was achieved. This method was used for determining the proper number of useful singular value and compared with the singular value sequence and the difference spectrum of singular value. Secondly, the de-noised dynamic stress signals were decomposed into a finite number of IMFs by EMD, and in order to remove the fictitious IMFs, the spectral ratio method was utilized to judge the fictitious IMFs. Finally, the fault frequency of the rotor system can be identified accurately by its frequency spectrum. Practical application shows that this method is efficient to recognize the rotor system faults of the Aero-engine.