Wind turbine prognostics

Due to the scale of some mechanical systems and the remoteness of some sites, wind turbine repairs can be prohibitively expensive and difficult to co-ordinate resulting in long periods of downtime and lost production.

In response, there has been a growing trend of retro-fitting similar systems on existing wind turbines in order to manage aging assets effectively.

The primary cause of roller bearing failure in turbines is the high contact stress involved, manifesting as abrasive wear, micro pitting, scuffing, and macropitting issues.

[11] Wind turbine bearings also frequently exhibit white etching cracks, a kind of localized damage to the ferrite microstructure of steel.

Traditional content management systems (CMS) typically rely on piezoelectric vibration sensors for gearbox monitoring tasks.

While these sensors are capable of capturing adequate data, they fall short when it comes to detecting low-frequency phenomena, such as rotor imbalance.

This capability enables these advanced sensors to identify critical frequencies related to the input shaft ('1P') and blade passing ('3P'), both of which often fall below 1 Hz and remain undetected by traditional CMS technologies.

Leveraging cutting-edge technology, this advanced monitoring solution excels in tracking vibration trends, including those caused by ice formation on turbine blades.

Additionally, aerodynamic imbalance may occur due to inaccuracies within individual blade profiles, physical damage, or errors in pitch calibration.

LDE is computed based on measurements from condition monitoring system (CMS) and SCADA, and is used to locate and diagnose incipient failure at component level.

Machine learning is also used by collecting and analyzing massive amounts of data such as vibration, temperature, power and others from thousands of wind turbines several times per second to predict and prevent failures.