A self-sensing framework for weak fault detection of planetary gearbox

Area of Science:

• Mechanical Engineering
• Condition Monitoring
• Signal Processing

Background:

• Planetary gearboxes are crucial in modern electro-mechanic systems.
• Traditional fault detection methods require costly additional sensors.
• High sensor costs limit the widespread application of fault detection technologies.

Purpose of the Study:

• To develop a self-sensing fault detection framework for planetary gearboxes.
• To detect weak fault impulses without relying on additional sensors.
• To reduce the cost and complexity of planetary gearbox condition monitoring.

Main Methods:

• Capturing preliminary signals from servo control systems.
• Constructing a motor-driven planetary gearbox control model.
• Investigating the feasibility of fault detection using motor servo control signals.
• Proposing a multi-signal assisting adaptive time synchronous averaging method.
• Utilizing periodic enhanced Gini for adaptive parameter selection.

Main Results:

• Successfully demonstrated the feasibility of self-sensing fault detection.
• Identified weak fault impulses in planetary gearboxes using servo control signals.
• Validated the framework through experiments on three distinct component faults.
• Achieved adaptive parameter optimization for enhanced fault detection.

Conclusions:

• The proposed self-sensing framework offers a novel and cost-effective solution for planetary gearbox fault detection.
• This approach enables the detection of weak fault impulses, improving diagnostic capabilities.
• The study provides a viable alternative to traditional sensor-dependent methods, paving the way for wider adoption.