Wind turbines work in highly variable operational conditions and harsh environments, especially offshore, that can cause expensive operation and maintenance, reaching up to 30% of the overall energy generation cost. Lowering the levelized cost of electricity is the main challenge faced by the wind turbine industry; hence, condition monitoring is a key enabler to avoid shutdowns and reduce operational and maintenance costs, providing a high availability.

The most “challenging” component of a wind turbine is the main bearing, responsible for up to 30% failures over 20-year lifetime. Furthermore, current roller main bearing technology faces challenges in scalability, as wind turbines increases in size. To address these challenges, SGRE continuous to develop and explore new technologies, for which its condition monitoring system likewise needs to evolve, as classical vibration monitoring no longer necessarily can be used.

Our consortium (SGRE DK, SGRE I&T, AAU) will develop a first ever set of algorithms and methods (A&M) for condition monitoring of new disruptive main bearing technology, aimed for the next generation offshore wind turbines. We will use parametric & non-parametric anomaly detection models and recent advances in machine learning techniques. The condition monitoring system to be developed, which will be implemented in a new full-scale wind turbine prototype, is expected to increase availability, and significantly lower the operational and maintenance cost.