MVMD-MOMEDA-TEO Model and Its Application in Feature Extraction for Rolling Bearings

Area of Science:

• Mechanical Engineering
• Signal Processing
• Condition Monitoring

Background:

• Rolling bearings are critical components in machinery, and their operational state directly impacts equipment reliability.
• Effective fault feature extraction is essential for timely condition monitoring and preventing catastrophic failures.
• Existing methods often struggle with noise interference, hindering accurate fault diagnosis.

Purpose of the Study:

• To propose an improved method for extracting fault features of rolling bearings.
• To enhance the characterization of rolling bearing operational states.
• To provide a novel solution for condition monitoring and fault diagnosis.

Main Methods:

• Modified Variational Mode Decomposition (MVMD) to decompose vibration signals into intrinsic mode functions (IMFs).
• Selection of effective IMF components based on energy ratio (≥90%) for signal reconstruction.
• Multipoint Optimal Minimum Entropy Deconvolution Adjusted (MOMEDA) to extract submerged fault impulse components.
• Teager Energy Operator (TEO) demodulation to calculate the Teager energy spectrum.
• Matching dominant frequencies with fault characteristic frequencies for fault feature extraction.

Main Results:

• The proposed MVMD-MOMEDA-TEO method demonstrated superior performance compared to MVMD-TEO and MOMEDA-TEO.
• The method effectively extracts fault features even when submerged in noise.
• Experimental validation on two datasets confirmed the method's effectiveness.

Conclusions:

• The MVMD-MOMEDA-TEO method offers a robust and superior solution for rolling bearing fault diagnosis.
• This approach significantly improves the accuracy and reliability of condition monitoring systems.
• It presents a valuable new tool for industrial fault diagnosis applications.