A Rolling Bearing Fault Feature Extraction Algorithm Based on IPOA-VMD and MOMEDA

Area of Science:

• Mechanical Engineering
• Signal Processing
• Machine Condition Monitoring

Background:

• Vibration signals from rolling bearings are often corrupted by significant background noise.
• Accurate extraction of fault features is crucial for effective condition monitoring and predictive maintenance.
• Existing methods struggle to isolate fault signatures amidst high noise levels.

Purpose of the Study:

• To develop a robust algorithm for rolling bearing fault feature extraction in noisy environments.
• To improve the accuracy and reliability of fault diagnosis in rotating machinery.
• To enhance the detection of transient shock components indicative of bearing failure.

Main Methods:

• An improved pelican optimization algorithm (IPOA) was developed using reverse learning strategies.
• Variable modal decomposition (VMD) was applied to decompose the noisy signal.
• Multipoint optimal minimum entropy deconvolution adjustment (MOMEDA) was used for optimal deconvolution.
• The kurtosis-square envelope Gini coefficient criterion selected optimal modal components.
• Teager energy operator (TEO) was employed for signal demodulation and analysis.

Main Results:

• The optimization performance of IPOA was validated.
• The proposed method successfully extracted fault features from simulated and actual bearing signals.
• Effective enhancement of transient shock components was achieved.
• Accurate fault characteristic extraction was demonstrated even with strong background noise interference.

Conclusions:

• The developed IPOA-VMD-MOMEDA algorithm offers a superior solution for rolling bearing fault diagnosis.
• The method effectively mitigates the impact of background noise on fault feature extraction.
• This approach significantly improves the reliability of condition monitoring for rolling bearings.