Entropy-Aided Meshing-Order Modulation Analysis for Wind Turbine Planetary Gear Weak Fault Detection under Variable...

Area of Science:

• Mechanical Engineering
• Condition Monitoring
• Signal Processing

Background:

• Wind turbines are crucial energy systems requiring robust monitoring.
• Variable speeds and weak faults challenge traditional fault detection methods.
• Early detection of gear faults is essential for safe wind turbine operation.

Purpose of the Study:

• To propose a novel method for detecting weak gear faults in wind turbines.
• To identify an optimal frequency band containing fault-related information.
• To enhance the reliability of wind turbine condition monitoring strategies.

Main Methods:

• Utilizing a scaling-basis local reassigning chirplet transform (SLRCT) for variable rotational frequency extraction.
• Developing an entropy-aided meshing-order modulation (EMOM) indicator for sensitive frequency area localization.
• Applying order tracking and bandpass filtering based on the EMOM indicator for fault component highlighting.

Main Results:

• The proposed EMOM analysis effectively highlights weak fault components in vibration signals.
• The method successfully detected various gear fault types and levels in experimental data.
• Accurate identification of optimal frequency bands for weak fault detection was achieved.

Conclusions:

• The entropy-aided meshing-order modulation method offers a promising approach for wind turbine gear fault diagnosis.
• This technique improves the sensitivity and accuracy of detecting weak faults under variable operating conditions.
• The method demonstrates significant potential for enhancing the safety and reliability of wind turbine operations.