Estimation of Remaining Useful Life of a Fatigue Damaged Wind Turbine Blade with Particle Filters

Abstract

Structural maintenance operations in wind energy sector are steering towards condition based maintenance (CBM) which requires prognostic estimates of existing condition of the wind turbine (WT) structural systems that is damage propagation and remaining useful life (RUL). WT blades are highly vulnerable structural components that are subjected to continuous cyclic loads of wind and self weight variation. A method for estimation of RUL of wind turbine blades considering the fatigue mode of failure is proposed in this paper. Stochastic life expectancy methods that use Bayesian updating with measurements of evolving damage for damage propagation estimation have proven to be reliable in RUL estimation. In this study probability density functions for the RUL of WT blades are estimated for diffident initial crack sizes and particle filtering method is used for forecasting the evolution of fatigue damage addressing the non-linearity and uncertainty in crack propagation. The stresses on a numerically modeled life size onshore WT blade subjected to turbulence are used in computing the crack propagation observation data for particle filters.