Fault Diagnosis for Rolling Bearing of Combine Harvester Based on Composite-Scale-Variable Dispersion Entropy and...

Area of Science:

• Mechanical Engineering
• Signal Processing
• Machine Learning

Background:

• Rolling bearings in combine harvesters face harsh environments, leading to non-stationary and nonlinear vibration signals.
• Accurate fault diagnosis of these bearings is challenging due to signal complexity.

Purpose of the Study:

• To propose a novel fault diagnosis method for combine harvester rolling bearings.
• To enhance the accuracy and robustness of fault identification in complex vibration signals.

Main Methods:

• Developed a self-optimization variational mode decomposition (SoVMD) for adaptive parameter optimization and multiscale frequency component extraction.
• Established a composite-scale-variable dispersion entropy (CSvDE) based feature learning model to create a multiscale fault feature space (MsFFS).
• Utilized a Softmax classifier for fault category identification using the generated MsFFS.

Main Results:

• The proposed SoVMD method effectively extracts multiscale frequency components from non-stationary signals.
• The CSvDE-based feature learning significantly improves fault feature representation.
• Experimental results show superior and robust fault diagnosis performance compared to existing methods.

Conclusions:

• The combined SoVMD and CSvDE approach offers a powerful tool for rolling bearing fault diagnosis in demanding industrial applications.
• This method effectively addresses the challenges posed by non-stationary and nonlinear vibration signals.
• The study demonstrates significant improvements in diagnostic accuracy and reliability for combine harvester components.