Adaptive Feature Extraction Using Sparrow Search Algorithm-Variational Mode Decomposition for Low-Speed Bearing Fault...

Area of Science:

• Mechanical Engineering
• Signal Processing
• Condition Monitoring

Background:

• Extracting fault features at low speeds is challenging due to weak signals and environmental noise.
• Traditional variational mode decomposition (VMD) requires manual parameter setting, limiting its effectiveness.

Purpose of the Study:

• To develop a parameter-adaptive VMD method for enhanced fault feature extraction in low-speed rotating machinery.
• To overcome the limitations of manual parameter selection in traditional VMD.

Main Methods:

• Proposed a parameter-adaptive VMD method using the sparrow search algorithm (SSA).
• SSA optimizes VMD parameters (number of modes, penalty factor) based on mean envelope entropy.
• Intrinsic mode functions (IMFs) are decomposed, selected using the kurtosis criterion, and reconstructed.
• Envelope analysis is applied to the reconstructed signal for fault diagnosis.

Main Results:

• The proposed method effectively reduces noise interference in low-speed bearing fault signals.
• Accurate extraction of fault characteristic frequencies and their harmonics was achieved.
• Demonstrated superior performance compared to other advanced signal processing methods.

Conclusions:

• The parameter-adaptive VMD method significantly improves the diagnosis of low-speed bearing faults.
• This approach offers a robust solution for identifying subtle fault signatures in noisy environments.