Gearbox Fault Diagnosis Based on Hierarchical Instantaneous Energy Density Dispersion Entropy and Dynamic Time...

Area of Science:

• Mechanical Engineering
• Signal Processing
• Condition Monitoring

Background:

• Accurate fault diagnosis is crucial for safe and efficient operation of rotating machinery.
• Existing methods may lack the sensitivity or accuracy needed for complex gearbox fault detection.

Purpose of the Study:

• To develop a novel and accurate fault diagnosis method for gearboxes.
• To enhance fault feature extraction and improve automatic fault type recognition.

Main Methods:

• Instantaneous energy density (IED) analysis using singular spectrum decomposition (SSD) and Hilbert transform (HT) to enhance fault features.
• Hierarchical dispersion entropy (HDE) algorithm to quantify the complexity of the IED signal, creating HIEDDE as fault features.
• Dynamic time warping (DTW) algorithm for automatic fault type recognition.

Main Results:

• The proposed HIEDDE method accurately recognizes fault patterns in experimental gearbox data.
• IED analysis effectively reinforces fault features in vibration signals.
• HDE algorithm successfully quantifies signal complexity for feature extraction.

Conclusions:

• The novel HIEDDE method offers accurate gearbox fault diagnosis.
• The approach demonstrates advantages over existing methods like MDE and RCMDE.
• This technique enhances the safety and efficiency of rotating machinery operation.