Online condition monitoring of floating wind turbines drivetrain by means of digital twin

This paper presents a digital twin (DT) condition monitoring approach for drivetrains on floating offshore wind turbines. Digital twin in this context consists of torsional dynamic model, online measurements and fatigue damage estimation which is used for remaining useful life (RUL) estimation. At first, methods for system parameter estimation are presented. The digital twin model provides sufficient inputs for the load observers designed in specific points of the drivetrain to estimate the online load and subsequently stress in the different components. The estimated real-time stress values feed the degradation model of the components. The stochastic degradation model proposed for estimation of real-time fatigue damage in the components is based on a proven model-based approach which is tested under different drivetrain operations, namely normal, faulty and overload conditions. The uncertainties in model, measurements and material properties are addressed, and confidence interval for the estimations is provided by a detailed analysis on the signal behavior and using Monte Carlo simulations. A test case, using 10 MW drivetrain, has been demonstrated.

Automated summary

This paper presents a digital twin condition monitoring approach for drivetrains on floating offshore wind turbines, which includes torsional dynamic model, online measurements, and fatigue damage estimation for remaining useful life estimation. The model provides inputs for load observers to estimate online load and stress in components, and feeds into a degradation model. Uncertainties in model, measurements, and material properties are addressed, and confidence intervals for estimations are provided through detailed analysis and Monte Carlo simulations.