An Automatic Bearing Fault Diagnosis Method Based on Characteristics Frequency Ratio

Area of Science:

• Spacecraft engineering
• Mechanical systems reliability
• Vibration analysis

Background:

• Bearings are critical for satellite actuators like moment wheel assemblies (MWAs) and control moment gyros (CMGs).
• Identifying bearing faults is challenging due to complex vibration signatures and low operational loads in satellites.
• Accurate fault diagnosis is essential for satellite performance and longevity.

Purpose of the Study:

• To develop an automated method for diagnosing bearing faults in satellite components.
• To introduce and validate a novel indicator, the characteristic frequency ratio, for fault identification.
• To improve the reliability and service life of satellite systems through effective bearing monitoring.

Main Methods:

• Acquired vibration signals from MWAs using bearing vibration tests.
• Employed improved ensemble empirical mode decomposition (EEMEMD) for signal envelope demodulation.
• Defined a characteristic frequency ratio and established a fault judgment threshold for automated diagnosis.

Main Results:

• Successfully identified fault characteristic frequencies from vibration signals.
• Developed and validated an automatic diagnosis process using diverse bearing fault data.
• Demonstrated the feasibility and effectiveness of the characteristic frequency ratio method.

Conclusions:

• The proposed characteristic frequency ratio method provides a feasible and effective approach for automatic bearing fault diagnosis.
• This technique enhances the capability for real-time monitoring and maintenance of satellite bearing health.
• Improved bearing fault diagnosis contributes to increased satellite operational performance and extended service life.