Enhanced Phase Demodulation for Vibration-Based Fault Detection of Wind Turbine Drivetrain

Vibration-based fault detection plays a vital role in monitoring rotary machinery systems such as aircraft Propulsion systems and wind turbines. Detecting their failures early is crucial for ensuring their safety and reliability. One of the main challenges is identifying system faults using non-stationary vibration measurements caused by variable-speed operation. In many cases operating speed has high fluctuations (>±10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup>) that limit the fault detection performance. In this paper, we introduce an improved method for extracting speed profile from a vibration signal utilizing a geometric feature within a rotating machinery system such as gearboxes, actuators and ball-screws. The performance of the enhanced method has been successfully validated using international benchmarking datasets for wind turbine gearboxes.