A Bearing Faults Diagnosis Enhancement Using EMD and MEDA

The growing importance of induction motor is attributed to the rapid progress in modern manufacturing. Consequently, a faults diagnosis of rotating machinery using vibration signals is extremely important for the reliability of mechanical systems. In this regard, we applied Empirical Mode Decomposition technique (EMD) combined with Minimum Entropy Deconvolution Adjusted (MEDA). Firstly, we analyze the bearing faults signals to Intrinsic Mode Functions (IMF's) using an EMD technique. To enhance faults diagnosis, we applied the Minimum Entropy Deconvolution Adjusted (MEDA) technique to the reconstructed signals. Subsequently, the results were comprehensively scrutinized to illustrate the capabilities of this approach in diagnosing bearing faults. In this study, we used the CWRU dataset and highlights the effectiveness and appropriateness of the EMD-MEDA method, evidently in diagnosing bearing faults and extracting fault features from the vibration signals.