A substructure approach for fatigue assessment on wind turbine support structures using output-only measurements

Abstract Fatigue constitutes a major and highly-uncertain safety-related factor for wind turbines. In order to ensure a reliable fatigue assessment of such structures, it is essential that stress predictions be based on the actual structural behaviour. The response identification of operational wind turbines in a global framework constitutes a challenging problem due to the uncertainties associated with the variability of the wind loading and the dynamics of the rotor. In reducing these uncertainties, this study proposes a substructuring approach, which abolishes the need for modelling the intricate and time-varying dynamics of the rotor. Instead, response prediction is performed on a substructure model of the tower and the effect of wind loads and servo dynamics is accounted for via the estimated interface forces at the top of the support structure. The application is based on synthetic vibration data generated via the FAST software and an output-only Bayesian filter employing the structural model of the support structure. The effectiveness of the proposed framework is presented in terms of fatigue damage estimates at different locations on the tower.