Frequency based Wind Turbine Gearbox Fault Detection applied to a 750 kW Wind Turbine

Abstract

Reliability and availability of modern wind turbines are of increasing importance, for two reasons. The first is due to the fact that power grids around in the world depends at a higher and higher degree on wind energy, and the second is the importance of lowering Cost of Energy of the wind turbines. One of the critical components in modern wind turbines is the gearbox. Failures in the gearbox are costly both due to the cost of the gearbox itself, but also due to lost power generation during repair of it. Wind turbine gearboxes are consequently monitored by condition monitoring systems operating in parallel with the control system, and also uses additional sensors measuring different accelerations and noises, etc. In this paper gearbox data from high fidelity gearbox model of a 750 kW wind turbine gearbox, simulated with and without faults are used to shown the potential of frequency based detection schemes applied on measurements normally available in a wind controller system. This paper shows that two given faults in the gearbox can be detected using a frequency based detection approach applied to sensor signals normally available in the wind turbine control system. This means that gearbox condition monitoring/ fault detection could be included in the standard control system, which potentially can remove the cost of the additional condition monitoring system and the additional sensors used in it.