Induction Motor Bearing Fault Diagnosis Using Cascaded EMD and DWT techniques

摘要

The aim of this paper is to develop a method based on a combination of empirical mode decomposition (EMD) and discrete wavelet transform (DWT) for assessment of damage of the induction motor bearing. A machine in standard condition has certain vibration signatures. These signatures are modulated by a number of high frequency harmonic components resulting from structural response to individual impacts. Fault development changes that signature in such a way that can be related to the faults, and EMD is used to separate these intrinsic modes known as intrinsic mode functions (IMFs).Hilbert Transform (HT) is applied to first fourlMFs to get instantaneous amplitude and then applied power spectral density (PSD), to identify the related defect frequencies. Later, DWT has been applied to the IMF, which has higher amplitude and again fault frequencies are obtained from HT and PSD. The work evaluates the detection ability of theappliedmethods. The obtained results show that the proposed method is superior tothe traditional envelope spectrum methodof extracting the incipient faults of roller bearings.