An automatic fault diagnosis method for aerospace rolling bearings based on ensemble empirical mode decomposition

• Published in: 2017 8th International Conference on Mechanical and Aerospace Engineering (ICMAE) pp. 502 - 506
• Main Authors: Hong Wang, Hongxing Liu, Tao Qing, Wenyang Liu, Tian He
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.07.2017
• Subjects: Couplings; Empirical mode decomposition; ensemble empirical mode decomposition; envelope demodulation; Fault diagnosis; kurtosis; Resonant frequency; rolling bearing; Rolling bearings; Vibrations; White noise
• ISBN: 1538633051; 9781538633052
• DOI: 10.1109/ICMAE.2017.8038697

This paper presents a dimensionless characteristic indicator for automatic bearing fault diagnosis based on ensemble empirical mode decomposition (EEMD). Firstly, the bearing vibration components called the Intrinsic Mode Functions (IMFs) are obtained by EEMD. Secondly, all IMFs are selected to reconstruct a new signal according to the rule of the kurtosis greater than 3. Then the new signal is processed by the Hilbert envelope demodulation and Fourier transformation to extract the fault characteristic frequencies. A dimensionless characteristic indicator is established to determine faults based on fault characteristic frequencies, and the threshold is given by experiments. Finally, different kinds of faults can be identified by the use of the proposed method. The results show that the proposed method can identify the faults of aerospace rolling bearing automatically and effectively.