Multi-Fault Diagnosis of Gearbox Based on Improved Multipoint Optimal Minimum Entropy Deconvolution

Area of Science:

• Mechanical Engineering
• Signal Processing
• Condition Monitoring

Background:

• Gearbox multi-fault diagnosis is challenging due to complex signal characteristics and limitations of current methods.
• Ensemble Empirical Mode Decomposition (EEMD) can suffer from energy leakage, while Multipoint Optimal Minimum Entropy Deconvolution (MOMED) has limitations in extracting compound fault features.
• Existing techniques often lead to misdiagnosis or incomplete fault identification in complex gearbox systems.

Purpose of the Study:

• To develop an adaptive method for gearbox multi-fault feature extraction.
• To enhance the accuracy and reliability of fault diagnosis in complex gearbox conditions.
• To overcome the limitations of EEMD and MOMED in multi-fault scenarios.

Main Methods:

• Adaptive decomposition of gearbox signals using Ensemble Empirical Mode Decomposition (EEMD).
• Reconstruction of intrinsic mode functions (IMFs) to mitigate mode-mixing and improve fault feature entropy.
• Application of Improved Multipoint Optimal Minimum Entropy Deconvolution (IMOMED) with adaptive noise reduction for feature extraction.

Main Results:

• The proposed IMOMED method effectively extracts multi-fault features from gearbox signals.
• Adaptive signal decomposition and reconstruction improved the signal-to-noise ratio and fault feature representation.
• Comparative analysis demonstrated the feasibility and superiority of the proposed method over traditional EEMD and Variational Mode Decomposition (VMD).

Conclusions:

• The developed adaptive gearbox multi-fault-feature extraction method offers a robust solution for complex diagnostic challenges.
• IMOMED provides enhanced accuracy in identifying multiple faults compared to existing EEMD and VMD techniques.
• This approach significantly improves the reliability of gearbox condition monitoring and fault diagnosis.