Abstract
Under complicated conditions, the extraction of a multi-fault in gearboxes is difficult to achieve. Due to improper selection of methods, leakage diagnosis or misdiagnosis will usually occur. Ensemble Empirical Mode Decomposition (EEMD) often causes energy leakage due to improper selection of white noise during signal decomposition. Considering that only a single fault cycle can be extracted when MOMED (Multipoint Optimal Minimum Entropy Deconvolution) is used, it is necessary to perform the sub-band processing of the compound fault signal. This paper presents an adaptive gearbox multi-fault-feature extraction method based on Improved MOMED (IMOMED). Firstly, EEMD decomposes the signal adaptively and selects the intrinsic mode functions with strong correlation with the original signal to perform FFT (Fast Fourier transform); considering the mode-mixing phenomenon of EEMD, reconstruct the intrinsic mode functions with the same timescale, and obtain several intrinsic mode functions of the same scale to improve the entropy of fault features. There is a lot of white noise in the original signal, and EEMD can improve the signal-to-noise ratio of the original signal. Finally, through the setting of different noise-reduction intervals to extract fault features through MOMED. The proposed method is compared with EEMD and VMD (Variational Mode Decomposition) to verify its feasibility.
Highlights
The gearbox in mechanical equipment is one of the most important power transmission components; its health directly affects whether the mechanical equipment can work normally
Entropy 2018, 20, 611 asking the mean of intrinsic mode function (IMF), the literature [7,8,9] confirmed Ensemble Empirical Mode Decomposition (EEMD) is more effective and accurate to the fault feature extraction of rotating machinery; it can self-adaptively resolve a complex signal into several IMFs
This paper explores a new method of fault feature extraction based on Improved Multipoint Optimal Minimum Entropy Deconvolution (MOMED) (IMOMED), which can accurately identify the fault characteristics of the gearbox, and provides a new idea for fault feature extraction of rotating machinery
Summary
The gearbox in mechanical equipment is one of the most important power transmission components; its health directly affects whether the mechanical equipment can work normally. Entropy 2018, 20, 611 asking the mean of IMFs, the literature [7,8,9] confirmed EEMD is more effective and accurate to the fault feature extraction of rotating machinery; it can self-adaptively resolve a complex signal into several IMFs. If the signal frequency band is too wide or the signal–noise ratio (SNR) too low, it will affect the decomposition efficiency of the EEMD [10]. To sum up the above analysis, EEMD has been successfully applied in fault diagnosis, but due to the inappropriate selection for white-noise levels, there is still modal aliasing, which leads to entropy losses; we put forward combined. When multiple faults coexist, coupled with white-noise pollution, the frequency band of complex vibration signals is relatively wide, so it is difficult for traditional FFT to identify each fault feature, and it is likely to cause incorrect diagnosis. This paper explores a new method of fault feature extraction based on IMOMED, which can accurately identify the fault characteristics of the gearbox, and provides a new idea for fault feature extraction of rotating machinery
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
