Bearing Faults Classification Using THH and Neural Network
The induction machine has many advantages: its specific power, its strength, relatively low manufacturing cost and minimal maintenance. But despite all these qualities, a number of faults can affect the life of the machine showing premature failures. The purpose of preventive maintenance in real time, we introduce a new signal processing technique based on Hilbert-Huang Transform (HHT) and marginal spectrum. Firstly, the current signals are decomposed into several intrinsic mode function (IMFs) using the empirical mode decomposition (EMD). The Hilbert Huang spectrum for each IMF is an energy representation in the time-frequency domain using the instantaneous frequency. The marginal spectrum of each IMF can then be obtained. The next step is the classification of faults detected by the application of neural network on IMFs. Tests on real signals show that the marginal spectrum of the second IMFs can be used for the detection and classification of bearing faults. The proposed approach provides a viable signal processing tool for an online machine health status monitoring.
Keywords: Signal processing, bearing faults, Hilbert-Huang transform, empirical mode decomposition, neural networks.
Download Full-Text
ABOUT THE AUTHORS
Aida Kabla
Department of Electrical Engineering, University of Bejaia,06000, Bejaia Algeria
Karim Mokrani
Department of Electrical Engineering, University of Bejaia,06000, Bejaia Algeria
Aida Kabla
Department of Electrical Engineering, University of Bejaia,06000, Bejaia Algeria
Karim Mokrani
Department of Electrical Engineering, University of Bejaia,06000, Bejaia Algeria
