For effective identification of rub-impact faults between rotor and stator of equipment, the paper has contributed the method integrating dual complexity parameters and variational mode decomposition (VMD). Firstly, to effectively separate the fault characteristics involved in signals, VMD algorithm was applied to decompose vibration signals and the component signals were obtained; secondly, taking account of the large difference of fault feature information involved in different component signals and in order to make options of sensitive fault component signals and reduce the loss of fault features, multi-scale Lempel–Ziv complexity and complexity parameter in Hjorth parameters were brought. Starting from 2 different perspectives of complexity evaluation, choose from the sensitive component signals containing more fault characteristics with these 2 complexity parameters; thirdly, signals were reconstructed based on selected sensitive component signals, and meanwhile, singular value difference spectrum algorithm was used to denoise reconstructed signals to further lessen the influence of noises; finally, the rub-impact fault between rotor and stator was identified by square demodulation spectrum (SDS) of denoised signal. The effectiveness of the proposed method has been proved by comparative analysis with other approaches as well as validation analysis of rub-impact fault signals in multiple situations.

As it is difficult to extract the combined faults from rolling bearings of aero-engine in strong noise, a fault diagnosis method based on wavelet transform (WT), principal component analysis (PCA) and self-correlation noise reduction is proposed to solve this problem. The proposed method is then compared with the target matrix composed of maximum component of kurtosis, the largest and the second largest kurtosis value. The result of comparative analysis reveals that the 2D target matrix proposed in this paper works better to extract the characteristic frequency of combined faults of rolling bearings in terms of accuracy and effectiveness.

No abstract

Inter-shaft bearing is an important component of dual-rotor aeroengine and it is difficult to extract fault characteristics and identify a fault type. Therefore, the paper has proposed a method (CCF–Complexity–VMD–SVD) combining information fusion, Complexity parameter of Hjorth parameters, variational mode decomposition (VMD), and singular value decomposition (SVD). Firstly, to fully embody fault characteristic information, the method has established the cross-correlation function (CCF) of homologous acceleration information from the casing of one of the aeroengines (acceleration signals collected from the same section at the same moment in orthogonal directions) to blend homologous information, reinforce fault characteristic components, and lessen the influence of noise. Secondly, considering the great influence of signal complexity on the extraction of fault characteristics and the Complexity parameter of Hjorth parameters can express the complexity of signals better, the Complexity parameter of Hjorth parameters is placed into the fault characteristic extraction of inter-shaft bearing. The Complexity parameter is applied to optimize the number of layers and central frequency of VMD. Thirdly, signals are denoised through SVD to further lessen the influence of noise on characteristic extraction. Finally, frequency spectrum (FS) of signal denoised is used to make characteristic extraction and fault identification of compound faults of inter-shaft bearing. The result of comparative analysis with other methods has further illustrated the proposed method CCF–Complexity–VMD–SVD that can actualize the effective characteristic extraction of compound faults and precise identification in the early stage of fault based on vibration signals from casing of aeroengine. The performance of engineering appliance of the proposed method has been further verified through fault analysis of disassembly for the aeroengine.

No abstract