Comparative Study Between EEMD-MED and VMD-MED Techniques Dedicated to Bearing Fault Detection in Induction Motors

To improve defect detection in rotating electrical machines (REMs), many techniques are used in this field. Bearing defects can be damaged the REMs. This paper compares Ensemble Empirical Mode Decomposition combined with Minimum Entropy Deconvolution (EEMD-MED) to variational mode decomposition with MED (VMD-MED) to detect bearing defects. First, VMD divides the signal into IMFs. In addition, the components with significant kurtosis values were selected for reconstruction. To enhance fault detection, the MED was performed on the reconstructed signal. Meanwhile, EEMD decomposes the bearing signal into IMF components. IMFs of higher kurtosis values were selected for reconstruction, and MED was applied. A final step is to compare the analysis of the results obtained to determine the best technique for detecting defects in bearings. This work is based on experimental results, which allow us to compare the two methods.

• bearing fault detection
• induction motor
• ensemble empirical mode decomposition
• variational mode decomposition
• minimum entropy deconvolution
• rotating electrical machines
• vibration signal analysis

• Machine Fault Diagnosis Techniques
• Electrical Fault Detection and Protection
• Electric Motor Design and Analysis
• bearing fault detection
• induction motor
• ensemble empirical mode decomposition
• variational mode decomposition
• minimum entropy deconvolution
• rotating electrical machines
• vibration signal analysis
• Control and Systems Engineering
• Electrical and Electronic Engineering
• Engineering