A Frequency-Weighted Energy Operator and complementary ensemble empirical mode decomposition for bearing fault detection. (1st January 2017)

• Record Type: Journal Article
• Authors: Imaouchen, Yacine; Kedadouche, Mourad; Alkama, Rezak; Thomas, Marc
• Is Part Of: Mechanical systems and signal processing. Volume 82(2017)
• Page Start: 103
• Page End: 116
• Publication Date: 2017-01-01
• Subjects: Complementary Ensemble EMD -- Bearing fault -- IMF selection -- Energy operator -- Demodulation -- Vibration signals -- Frequency-Weighted Energy Operator; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2016.05.009
• Languages: English
• ISSNs: 0888-3270

Abstract

Abstract: Signal processing techniques for non-stationary and noisy signals have recently attracted considerable attentions. Among them, the empirical mode decomposition (EMD) which is an adaptive and efficient method for decomposing signals from high to low frequencies into intrinsic mode functions (IMFs). Ensemble EMD (EEMD) is proposed to overcome the mode mixing problem of the EMD. In the present paper, the Complementary EEMD (CEEMD) is used for bearing fault detection. As a noise-improved method, the CEEMD not only overcomes the mode mixing, but also eliminates the residual of added white noise persisting into the IMFs and enhance the calculation efficiency of the EEMD method. Afterward, a selection method is developed to choose relevant IMFs containing information about defects. Subsequently, a signal is reconstructed from the sum of relevant IMFs and a Frequency-Weighted Energy Operator is tailored to extract both the amplitude and frequency modulations from the selected IMFs. This operator outperforms the conventional energy operator and the enveloping methods, especially in the presence of strong noise and multiple vibration interferences. Furthermore, simulation and experimental results showed that the proposed method improves performances for detecting the bearing faults. The method has also high computational efficiency and is able to detect the fault at an early stage of degradation. Highlights: The Complementary EEMD is used to decompose the signal on a multiple components. The CEEMD not only overcomes the mode mixing, but also eliminates the residual of added white noise and enhance the calculation efficiency of the EEMD method. A new method for choosing automatically the relevant IMFs is introduced. A new A new method called FWEO is presented to demodulate the reconstructed signal. Moreover its simplicity and computational efficiency, the FWEO method is able to detect signal impulsiveness without any pre-filtering process.