Aeroengine Rotor Rub-impact Fault Diagnosis Based On Fourier Decomposition Method

Rotor rub-impact faults are frequent failure types of the aeroengine mechanical equipment.In the pursuit of high power,high bypass ratio,and ultra-high speed mechanical performance,the space of between Rotor blades and engine seal case,sealed interval etc other parts are often streamlined to a minimum,which intensifies the occurrence of rotor rubbing faults.In the early stage of rubbing,the vibration of the engine system is intense,and the secondary vibration of the rotor system occurs with the rubbing fault,which has extremely adverse effects on the stable operation of the aeroengine.How to reasonably extract and identify fault feature information from complex rotor rubbing fault vibration signals has been a difficult problem in the field of rotor fault diagnosis.Thus,this paper presents a research on fault diagnosis of aeroengine rotor based on a novel adaptive time-frequency analysis which is called Fourier Decomposition Method.FDM is based on the Fourier transform,developing on the basis of EMD theory,complete,orthogonal and adaptive time-frequency analysis methods,who locality.The mechanical equipment nonlinear non-stationary vibration signal can be achieved the effective decomposition.Using this method to decompose the vibration signal of single-point on casing and whole-cycle on rotor test rubbing positions,then the decomposed signal components analysed by Hilbert envelope,successfully extracted the characteristic frequency of rotor rubbing faults.By comparing and analyzing the rotor rub-impact fault diagnosis results of different adaptive time-frequency analysis methods,the effectiveness of the rotor rub-impact fault diagnosis method based on FDM algorithm is verified.In order to solve the problem that the time-frequency distribution range of the fault feature information cannot be adaptively located due to the fault diagnosis method of the rotor rub-impact method based on the FDM algorithm,the rotor rub-impact fault diagnosis method based on FDM and Fast Kurtogram algorithm been presented by this paper.By using the Fast Kurtogram algorithm to accurately locate the spectral region of the maximum spectral kurtosis component of the signal,the precise positioning of the frequency spectrum region of the information signal component of the rotor rubbing fault feature is achieved.The results show that thismethod can efficiently extract the characteristic frequency of rotor rubbing faults.On the basis of perfecting rotor fault diagnosis method based on FDM algorithm,the reliability of the diagnosis method is verified.The fault diagnosis method based on signal correlation or impact strength criteria can not effectively circumvent the strong noise signal components contained in the fault vibration signal.In order to realize the reasonable diagnosis of the aero engine rotor rub-impact fault and achieve high-efficiency noise reduction purposes,a rubbing fault diagnosis method based on FDM and Singular Value Difference Spectrum is proposed.First the fault vibration signal complete decomposed by the FDM algorithm,and using Singular Value Differential Spectrum Analysis to analyze the advantages of zero phase offset signal decomposition.Efficiently filter the noise signals contained in the fault diagnosis result while properly extracting the feature information of rotor rubbing faults.The effectiveness of rotor rub-impact fault diagnosis based on FDM algorithm has been verified.