A Damage Tolerance and Fatigue Reliability Assessment Method for Flexible Long Wind Turbine Blades

ABSTRACT Wind turbine blades inevitably develop manufacturing‐induced defects during production and operation, while the inherent heterogeneity of composite materials and manufacturing tolerances introduce multisource uncertainty in fatigue failure, directly impacting long‐term service reliability. This paper presents a comprehensive framework for evaluating damage tolerance and fatigue reliability of defect‐containing blades under fluid–structure interaction (FSI) effects, accounting for service conditions, initial imperfections, and parameter uncertainties. Stress responses under various operating conditions are first obtained through FSI simulations, with initial defects positioned at locations of maximum principal stress. Fatigue life is subsequently estimated using S – N curves and Miner's cumulative damage rule. To characterize uncertainties arising from multiple sources, an active learning Kriging (ALK) surrogate model is established for fatigue life prediction, while a probabilistic reliability framework incorporating parameter variability is formulated. Sobol sensitivity analysis is utilized to quantify stochastic effects on damage tolerance, revealing that variations in material properties and defect geometry constitute the primary determinants of blade reliability. The method quantifies the dominant effects of wind speed and defect size on fatigue life (with sensitivity indices of 0.5365 and 0.3480, respectively), and uncovers the amplification mechanism of defect‐stress coupling on damage tolerance under high wind speed conditions. This approach provides a theoretical basis for advancing the understanding of the intrinsic relationships between damage tolerance, fatigue reliability, and uncertainty quantification.