Periodicity-Impulsiveness Spectrum Based on Singular Value Negentropy and Its Application for Identification of Optimal Frequency Band

Bearing faults are the main contributors to the failure of motors. Periodic harmonic components from the motor rotating and random impulses caused by the electromagnetic interference heavily trouble vibration-based resonance demodulation techniques. This paper presents a method that accurately identifies the optimal frequency band even with complicated interferences from the motor and industrial field. Singular value negentropy (SVN) is originally applied to measure the periodicity of signal without prior knowledge. Based on the SVN, periodicity-impulsiveness spectrum (PIS) that simultaneously takes periodicity and impulsiveness of fault impulses into consideration is constructed. Guided by the period-oriented kurtosis selection criterion, the resonance frequency band excited by the motor bearing fault is located. The proposed method was validated by the simulated motor bearing signal and the real datasets. Compared with the most popular resonance demodulation methods, kurtogram and protrugram, the proposed method is undoubtedly an alternative method for the identification of optimal resonance band.